The Origins of Daphnis et Chloe (1912)

The Origins of Daphnis et Chloe (1912)Author(s): Simon MorrisonSource: 19th-Century Music, Vol. 28, No. 1 (Summer 2004), pp. 50-76Published by: University of California PressStable URL: http://www.jstor.org/stable/10.1525/ncm.2004.28.1.50 .

Accessed: 21/11/2013 04:29

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

University of California Press is collaborating with JSTOR to digitize, preserve and extend access to 19th-Century Music.

http://www.jstor.org

This content downloaded from 164.15.128.33 on Thu, 21 Nov 2013 04:29:08 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=ucal

http://www.jstor.org/stable/10.1525/ncm.2004.28.1.50?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


50

19TH

CENTURYMUSIC

19th-Century Music, XXVIII/1, pp. 50–76. ISSN: 0148-2076, electronic ISSN 1533-8606. © 2004 by the Regents of the Univer-sity of California. All rights reserved. Please direct all requests for permission to photocopy or reproduce article content through

the University of California Press’s Rights and Permissions website, at http://www.ucpress.edu/journals/rights.htm.

I am grateful to the staffs of the London Theatre Museum,the Paris Bibliothèque Nationale, the StockholmDansmuseet, and especially the Russian National Libraryand State Theatre Library in St. Petersburg for their in-valuable assistance with my research. I am also grateful tomy anonymous readers as well as Stephanie Jordan, AnnaPakes, and Rebecca Harris-Warrick for their counsel onthe first and second drafts of this article, and to SindhuRevuluri, Britta Gilmore, and Baty Landis for their helpwith the French-language translations. Funding for theproject was provided by the American Council of LearnedSocieties and Princeton University.

creative methods, specifically his approach, inhis first ballet, to writing for dance.

No systematic comparison between the mu-sic and choreography of the original Daphnis etChloé can, however, be made, for that choreog-raphy was neither notated nor filmed—a factthat has enabled musicologists to imagine thatthe music, popular in concert performance,bears little relation to its own stage concep-tion. Commentary on the interaction of thetwo media, including that which follows inthis article, is ineluctably provisional. Mattersare further complicated by the fact that, for the8 June 1912 Paris premiere, the choreographywas rehearsed in haste, leaving some sectionslooking unrefined. Though reviewers of boththe Paris premiere and the 9 June 1914 Londonpremiere mentioned the stylistic and technicalinfelicities, their assessments of the choreogra-phy were positive.1 So, too, was the assessment

The Origins of Daphnis et Chloé (1912)

SIMON MORRISON

Besides Maurice Ravel’s score, the remnants ofthe original production of Daphnis et Chloé—one known stage photograph, an assortment ofstudio photographs, seven known costumes,brief reviews, anecdotal memoirs, and a bundleof pencil and pastel drawings—constitute cho-reographer Mikhail Fokine’s draft and revisedscenarios. There also exist proof pages for ashorter version of the 1910 piano score, musi-cal evidence to suggest that Fokine conceivedthe ballet in 1907 for another composer, andreproductions of Léon Bakst’s stage décor.Though interrelated, these materials are scat-tered across the globe, preserved in librariesand museums in Russia, Sweden, France, En-gland, and the United States. Their contentsdetail the conception and realization of Daphniset Chloé while also offering insight into Ravel’s

For Feets

1Robert Brussel, “Les Théâtres,” Le Figaro, 9 June 1912:“Fokine, the author and director, offers new evidence ofhis extraordinary talent in Daphnis et Chloé. Though hehas perhaps conceived more spectacular tableaux, he hasnever realized any more delicate or expressive. The danceshe designed for Daphnis, Chloe, and the grotesque Darkonare of a prodigious sort: Daphnis is light; Chloe is tenderand affected; Darkon is clumsy. And all of this is expressedin the most diverse, appropriate and sympathetic man-ner”; Gustave Samazeuilh, “Les Ballets Russes,” Le



51

SIMONMORRISONOrigins ofDaphniset Chloé

by the prima ballerina of the Ballets Russes.Tamara Karsavina, who took the part of Chloeat the Théâtre du Châtelet (with VaslavNijinsky as Daphnis), opined that the ballet,“in spite of a certain sketchiness in the lastAct, due to the pressure under which Fokineworked (he only half finished Nijinsky’s andmy final dance on the morning of the perfor-mance), was to my mind his masterpiece.”2

Irrespective of the accolades, Fokine (1880–1942) looked back on Daphnis et Chloé with aheavy heart, recalling that the project sufferedowing to his creative and personal disputes withSergey Diaghilev (1872–1929), the impresarioof the Ballets Russes, the troupe that had per-formed the ballet.3 Ravel, too, was disappointedwith the project, especially its London incarna-tion, which omitted an essential feature of thescore: the offstage, wordless mixed chorus.4

Even before the production, Ravel began to re-cast his score, by far the longest in his œuvre,as a pair of orchestral suites;5 afterwards, Fokinereworked and then abandoned his dances. Bakst(1866–1924), meantime, authorized the sale ofhis sample images for the ballet—those con-sulted by the costume makers and backdroppainters.6

Though longevity is the exception rather thanthe rule in ballet history, Daphnis et Chloé didnot fade away after 1914. There have been sev-eral restagings, all involving Ravel’s score (withor without the chorus) and at least part ofFokine’s scenario. Two of these—the first cre-ated by Frederick Ashton in 1951, the secondby Graeme Murphy in 1982—will be discussedat the conclusion of this article. My main fo-cus, however, will be on the origins of Daphniset Chloé and the differences between Ravel’sand Fokine’s approach to the ballet’s Hellenicsubject matter. Loosely tracing the chronologyof the ballet’s creation, my discussion will be-gin with the first and second versions of thescenario, progress to the first and second ver-sions of the score, and conclude with the dance.

Courrier musical 12:15 (15 June 1912), 364–65: “It may bethat in other circumstances M. Fokine had conceived morefrenetic whirls, and movements with more forceful im-pact. However, if in this case he was seeking a more re-served feel and one better suited to the natures of hissubjects, he managed to tap his powers of precision andapply restraint in creating dances of such varied effect asthose of Chloe, Daphnis, and the cowherd, Darkon”; EmileVuillermoz, “Les Théâtres: La Grande Saison de Paris,” LeMercure musical 8:6 (15 June 1912), 68: “Thanks toDaphnis et Chloé, the Russian season ended in an apo-theosis, and what remained of the adorable vision of thisveritable masterpiece—a vision so quickly evaporated, atoo-fleeting apparition—was a sort of dreamlike transpar-ency that will enchant our memories for a long time tocome.”2Tamara Karsavina, Elsa Brunelleschi, Tony Mayer, andAlexander Bland, “Four Opinions on ‘Daphnis and Chloe’,”Ballet 2:5 (June 1951), 5. The article compiles four assess-ments of Frederick Ashton’s 1951 choreography, whichKarsavina compares favorably to Fokine’s original.3Michel Fokine, Memoirs of a Ballet Master, trans. VitaleFokine (Boston: Little, Brown and Company, 1961), pp.200–04. Fokine argues that Diaghilev, having decided toreplace him as ballet master of his company with hisprotégé Nijinsky, sabotaged the production of Daphnis etChloé by scheduling insufficient rehearsal time, recyclingcostumes from another ballet (Narcisse), and billing it first,rather than second, on the final program of the 1912 sea-son. Diaghilev moved the ballet to second place after Fokinevowed to publicize Diaghilev’s sexual relationship withNijinsky. Fokine’s account of the squabble is corroboratedby Serge Lifar, Serge Diaghilev: His Life, His Work, HisLegend: An Intimate Biography (New York: G. P. Putnam’sSons, 1940), pp. 191–94.4As Arbie Orenstein summarizes, “When it was disclosedthat the London production would take place without thechorus. Ravel was incensed, and protested in an open let-ter to four London newspapers. . . . In his reply, Diaghilev

stated that he had recently produced Daphnis et Chloé atMonte Carlo without chorus and was unaware of any pro-test from the composer. Furthermore, he denied that thealternate orchestral version was intended solely for smalltheaters. When the chorus had been included in Paris, hecontinued, its participation was found to be detrimental.Therefore, he concluded, M. Ravel was asked to write asecond version. In a long statement to the editor ofComoedia, Ravel convincingly exposed Diaghilev’s mis-representations point by point, and observed that hence-forth the impresario would be bound by written agree-ment to include the chorus in all major productions”(Ravel: Man and Musician [New York: Columbia Univer-sity Press, 1975], pp. 69–70). Diaghilev had previously an-gered Ravel by voicing dissatisfaction with the music. SeeJacques Durand, Quelques souvenirs d’un éditeur demusique (Paris: A. Durand et fils, 1924), p. 16.5Orchestral Suite No. 1 (“Nocturne,” “Interlude,” “Danseguerrière”) and Orchestral Suite No. 2 (“Lever du jour,”“Pantomime,” “Danse générale”) date from 1911 and 1913respectively.6These watercolor images, which include a sketch of thedécor of scenes 1 and 3, a detail thereof, a study of analternate version of the décor of scene 2, and costumedesigns for Chloe, the brigands, and the shepherds, areitemized on p. 36 of Maurice Ravel (Paris: BibliothèqueNationale, 1975), the catalogue of an exhibit curated byFrançois Lesure and Jean-Michel Nectoux. The list of buy-ers includes the Bibliothèque Nationale, Musée Nationald’Art Moderne, London Theatre Museum, and private col-lectors. The notebook of Bakst’s early images is housed inthe Dance Collection of the New York Public Library.



52

19TH

CENTURYMUSIC

ILong before Ravel’s and Bakst’s Hellenic vi-

sions, there was Fokine’s. The choreographerconceived Daphnis et Chloé while he was stilla member of the Imperial Ballet in St. Peters-burg, a company he joined in 1898. In 1907 hesubmitted a detailed two-act scenario toVladimir Telyakovsky (1861–1924), the Direc-tor of the Imperial Theatres, in hopes that asuitable composer would be commissioned towrite the music, and that the completed balletwould be included in the Imperial Theatre rep-ertoire. As evidenced by the absence of paper-work in the official archive,7 Telyakovsky nei-ther critiqued nor even considered the scenariofor possible development. Besides the scenario,Fokine forwarded to Telyakovsky a bulletin-point list of general proposals for ballet reform,the precursor to a manifesto he published inthe London Times on 6 July 1914. These pro-posals centered on enhancing the emotionaland expressive content of ballet through theabolition of the number format, suppression ofathletic display from the solo variations, andincreased verisimilitude in the choice of décorand dress.8 Like the scenario, the proposals were

not adopted by Telyakovsky,9 who, despite be-ing innovative in his selection of designers andscenarists and interested in updating familiarballets, was obligated in his programming totake account of conservative tastes, especiallythose of the Imperial family.10

In his memoirs, Fokine declared that boththe scenario and the proposals dated from 1904,three years before they were actually written.The change in dating was first observed by theeminent dance historian Vera Krasovskaya, whopointed out that Fokine’s 1905 ballet Acis etGalatea, a work for graduating student dancersthat received at best neutral reviews,11 bore nosigns of his plans for choreographic reform. Thechange in dating, Krasovskaya reasoned, re-flected Fokine’s desire to conceal the stronginfluence (and strong anxiety of influence) thatthe pioneering American choreographer IsadoraDuncan (1877–1927) had on his creative activi-ties.12

7Rossiyskiy gosudarstvennïy istoricheskiy arkhiv, fond 497,opis’ 5, yed. khr. 1316, “O sluzhbe skripacha K. Kadlets;fond 497, opis’ 13, yed. khr. 1126, “O sluzhbe baletmeysteraMikh. Fokina”; fond 497, opis’ 14, yed. khr. 23, “Dogovors M. M. Fokinïm o postanovke im baletov na stsene Imp.Teatrov.”8A handwritten draft of the list of reforms is preserved inSt. Petersburg at the Russian National Library (Rossiyskayanatsional’naya biblioteka, henceforth RNB), fond 820, no.2, pp. 2–3; a typed and considerably expanded copy is pre-served at the State Theatre Library (Gosudarstvennayateatral’naya biblioteka, henceforth GTB), no. 19996, pp. 1–2. The first document, “Towards a staging of the balletDaphnis and Chloe” (K postanovke baleta “Dafnis iKhloya”) is held in the Mikhail Fokin collection at RNB;the second, “On staging the ballet Daphnis and Chloe” (Opostanovke baleta “Dafnis i Khloya”) is held in the collec-tion of the violinist/conductor Andrey-Karl Kadletz (1859–1928), Fokine’s close colleague at the Imperial School ofBallet. An annotated composite of the original and revisedversions of the list of reforms can be found in M[ikhail]Fokin, Protiv techeniya: vospominaniya baletmeystera.Stat’i, pis’ma, ed. Yu. I. Slonimskiy (Leningrad: Izdatel’stvo“iskusstvo,” 1962), pp. 567–68. The truncated second edi-tion of this book, published in 1981, does not include thelist of reforms, nor does the truncated English-languageedition, Memoirs of a Ballet Master. The Russian-languageeditions accept 1904 as the date of the list, though it wasalmost certainly written later (see n. 12).

9In his monograph on Fokine, written with thechoreographer’s permission and involvement, theballetomane Cyril Beaumont notes that Telyakovskyadopted only one of the reforms and applied it not toballet but opera: “In view of the necessity for preservingthe illusion and theatrical impression, artistes were for-bidden to bow during the performance” (Michel Fokineand His Ballets [1935; rpt. London: Dance Books, 1996],p. 24).10These points come from Roland John Wiley, A Centuryof Russian Ballet: Documents and Eyewitness Accounts,1810–1910 (Oxford: Clarendon Press, 1990), pp. 405–07,420.11The most detailed review of Acis et Galatea comes fromValeryan Svetlov, who provided an assessment of eachperformer’s mastery of classical technique. See“General’naya repetitsiya baletnogo ekzamenatsionnogospektaklya,” Birzhevïye vedomosti, 12 April 1905, p. 5.The music was composed by Fokine’s colleague Kadletz(see n. 8) for another version of the ballet in 1896. At thattime, N. L. O-skiy complained that the music “was ex-tremely long-winded, dry, unbearably dull and sometimeswholly unrelated to the plot. It is truly strange, for ex-ample, to encounter piano music in those passages depict-ing [the ogre] Polypheme’s fury and, likewise, fortissimomusic for [the shepherd] Acis and [sea-nymph] Galatea’sloving embraces, when it should be expressing the lovers’tender feelings” (“Teatral’nïy kur’yer,” Peterburgskiylistok, 23 January 1896, p. 3).12In Krasovskaya’s words: “Fokine backdated his scenarioto 1904 in order to show his independence [from] Duncan.In addition, recalling his first ballet Acis and Galatea,given at the graduation performance on 20 April 1905,Fokine openly admitted: ‘Until then it hadn’t occurred tome that I might have some ability in this area.’ The entireaccount of Acis and Galatea corroborates Fokine’s admit-ted lack of preparation for a ballet master’s work and proves



53


Fokine stubbornly dismissed the assertion,first made by Diaghilev, that Duncan cast anirresistible spell on him when he saw her per-form, the outcome being the fundamental trans-formation in his technique.13 He insisted thathis interest in ancient designs, and the pros-pect of fabricating movements from them,stemmed from independent research. In early1905, according to his memoirs, he approachedthe director of the Imperial Public Library,Vladimir Stasov (1824–1906), for sources onsculptural art. Studies of such sources, he be-lieved, would provide him with creative fodderand enable him to create an edifying ballet on apar with the tragedies of Euripides andSophocles, which were being staged at the timein the Russian capital. Stasov agreed to assistFokine and brought several illustrated volumesto his attention, in particular George Perrotand Charles Chipiez’s Histoire de l’art dansl’Antiquité. This multivolume set was issuedbetween 1882 and 1914. The five volumes aboutGreek art, however, all date from after 1905.Those published earlier concern only Egyptianand Chaldean art and would thus have been oflittle use for Daphnis et Chloé.14

Taking into account the factual detours inhis memoirs, it seems clear that Fokine’s dramaderived as much or more from the fanciful Hel-lenic illusions in Duncan’s dancing than fromhis own imagination. Duncan was, as LynnGarafola puts it, “the shining star of hisyouth.”15 Though too proud to acknowledgeher influence, he heaped praise on her in inter-

views. The earliest of these, “Barefoot Danc-ers” (Bosonogiye tantsovshchitsï, 1909), at oncelauds her “response to the one-dimensionalityof ballet” and iterates that “costume and dancereform” were crucial to the revitalization ofthe art.16 Enraptured by Duncan, Fokineachieved this revitalization in Daphnis et Chloé.

Fokine devised the scenario of the ballet froman 1895 Russian-language translation of a pas-toral romance by Longus, an obscure second- orthird-century writer who is thought to havelived on the Eastern Aegean Island of Lesbos,the idyllic setting of Daphnis et Chloé. Thetranslation was made by the first-generationSymbolist poet Dmitriy Merezhkovsky (1865–1941), who prefaced it with an elaborate ar-ticle, “On the Symbolism of Daphnis andChloe” (O simvolizme “Dafnisa i Khloi”), thepossible influence of which on Fokine has notyet been explored. This article revolves aroundMerezhkovsky’s belief that Longus’s tale pro-vided the basis for centuries of writing on thetheme of rustic love,17 including that of theFrench and Russian Symbolists. Merezhkovskybases this claim not on specific plot details, buton

a nuance, a first suggestion, a first gleam of thatpeculiar morning-twilight duality that is especiallyendemic and precious to us, the contemporaries of thecommon European Symbolist movement, which somecall a decline (Décadence) and others a Renaissance,but in which there is actually both of these things:decay and rebirth, the end of the old and beginning ofthe new, descent and ascent, decline and resurrection,decadence and symbolism. These qualities are found,however, in all comparable artistic movements, as Ihave tried to corroborate by using the example of theearly, pre-Raphaelite, Florentine Rinascimento.18

Merezhkovsky supplements his descriptionwith quotations from Goethe, whose enthusi-asm for the pastoral romance further evincedits semantic and stylistic novelties. The trans-lator also repeatedly underscores that the text

that the ideas expounded in the libretto of Daphnis andChloë (1912) did not precede the production of Acis andGalatea (1905) but came later, not before 1907” (VeraKrasovskaya, Russkiy baletnïy teatr nachala XX veka,1971, quoted in Tim Scholl, From Petipa to Balanchine[London: Routledge, 1994], pp. 59–60).13For the specifics of Diaghilev’s assertion, see ArnoldHaskell, Balletomania: The Story of an Obsession (NewYork: Simon and Schuster, 1934), pp. 126–29.14Fokine, Memoirs of a Ballet Master, pp. 87–88. Imagesfrom L’Histoire de l’art dans l’Antiquité and the sketchesthat Fokine derived from them for his later, “Eastern”ballets are housed in GTB. See E. G. Pogosova,“Dyagilevskiye materialï v arkhive Mikhaila Fokina,”Zapiski Sankt-Peterburgskoy gosudarstvennoy teatral’noybiblioteki 1 (1997), 45.15Lynn Garafola, Diaghilev’s Ballets Russes (New York:Da Capo Press, 1998), p. 40. For additional information onFokine and Duncan, see pp. 39–42.

16Fokin, “Bosonogiye tantsovshchitsï,” in Protiv techeniya:vospominaniya baletmeystera. Stat’i, pis’ma, p. 336.17D[mitriy] S. Merezhkovskiy, Dafnis i Khloya: Povest’Longusa (St. Petersburg: Izdaniye M. V. Pirozhkova, 1904),pp. 16–17.18Ibid., p. 18.



54

19TH

CENTURYMUSIC

is not “Hellenic” but “Hellenistic,” a copy of apastoral romance rather than the thing itself.19

Recognizing that the ideals and mores of thepre-Christian era had perished, its author(whether Longus or a predecessor) conceived ameta-narrative, an Arcadian fable aboutArcadian fables, one that eulogized a deceasedculture while also heralding a new one. Sup-planted by the austere morality of the earlyBiblical epoch, “The pagan spirit no longer ex-isted; it was a ghost, unreal, and more or lesspart of the far distant past.”20

To ascertain what Fokine may have absorbedfrom these remarks, we can turn to his draftscenario for Daphnis et Chloé, copies of whichare preserved in two St. Petersburg archives21

together with his list of ballet reforms. Thestructure of this text highlights Longus’s duali-ties of “decay and rebirth, the end of the oldand beginning of the new.” Fokine cast thedraft scenario in two acts of equal length (act Iis partitioned into six scenes; act II is not parti-tioned), with certain events in the first halfreprised in the second. The draft scenario isalmost twice the length of the final scenario,which consists of just one act of three scenes,and which was published by Durand just prior

to the ballet’s Parisian premiere.22 Fokine likelyshortened the text after he had left St. Peters-burg for Paris in the late spring or early sum-mer of 1909, the period when he met Ravel. Tothe best of my knowledge, no substantive de-scriptions of his and the composer’s discus-sions exist, but a letter from Ravel to one of hisbenefactors, Madame René de Saint-Marceaux,suggests that they were something less thancordial: “I have to tell you that the last weekhas been insane: preparing a ballet libretto forthe next Russian season. [I’ve been] working upto 3 a.m. almost every night. To confuse mat-ters, Fokine does not know a word of French,and I can only curse in Russian. Irrespective ofthe translators, you can imagine the timbre ofthese conversations.”23

Act I of the draft scenario furnishes all of theplot events of the final product, albeit with agreatly expanded cast and some lines of spokendialogue. (Fokine may have planned to includethese in the printed program and have themmimed onstage, though the effect would havebeen somewhat old-fashioned, since the use ofspoken dialogue had dwindled in the late nine-teenth century.) Excluding the corps de ballet,the draft scenario identifies sixteen characters:Lamon, an old goatherd; Mirtala, his wife;Daphnis, their adopted son; Drias, a sage andshepherd; Nape, his wife; Chloe, their adopteddaughter; Dionisofan, a magnanimous lord;Klearista, his wife; Gnafon, a buffoon; Filetas,an old herdsman; Darkon, a young herdsman;First Nymph; Second Nymph; Third Nymph;the God Pan; and Blioksis, leader of the brig-

19Ibid., p. 7 (emphasis in the original).20Ibid.21RNB fond 820, no. 2, pp. 4–13 (Mikhail Fokine collec-tion); GTB P 10/30, unpaginated (Valeryan Svetlov collec-tion) and GTB No. 19996, unpaginated (Andrey Kadletzcollection). RNB fond 820, no. 2, “‘Dafnis i Khloya.’ MuzïkaM. Ravelya. Libretto [1912],” merges three separate draftsof the scenario. Page 4 contains a list of potential castmembers; pp. 5–9 contain a typescript of act I, scs. 1–6,with numerous handwritten corrections; pp. 10–11 con-tain an abbreviated typescript of the same act. Pages 12–13, finally, include a handwritten draft of act II. GTB P 10/30, “Dafnis i Khloya, balet v 2-x d.,” includes a typed andbound copy of RNB fond 820, no. 2, pp. 4–9 (cast membersand act I). Act II is absent, though the frontispiece to thefile states that the ballet is in two acts. GTB No. 19996,titled “‘Dafnis i Khloya.’ Muzïka Kadlitsa,” includes atyped, bound copy of RNB fond 820, no. 2, pp. 4–9. Act I,scs. 1–6 are preserved intact, save for the deletion of thelast two sentences: “Daphnis stands amongst his herd,stroking a billy-goat with one hand and a nanny-goat withanother, and so pledges his love to Chloe. Everyone, see-ing the lovers’ bliss, dances happily. Darkon, tormentedwith jealousy, leaves alone.” The file also contains a typedand reordered version of act II. The final version of thescenario, the one used by Ravel, compresses the six scenesof act I into three and altogether dispenses with act II. Theassociation of Kadletz, the composer of Acis et Galatea,with Daphnis et Chloé is explained below.

22Daphnis et Chloé: Ballet de Michel Fokine (Paris: Durandet Cie, 1912). The same text was published in 1910 in thepiano version of Ravel’s score, and in 1913 in the orches-tral version. The 1989 reprint of the orchestral version byDover Publications includes an English-language transla-tion of the text.23Maurice Ravel: Lettres, écrits, entretiens, ed. Orenstein(Paris: Flammarion, 1989), p. 105. The letter is dated 27June 1909. Diaghilev was somewhat more positive aboutthe meetings. In a 25 June 1909 letter to the artist AlexandreBenois, he wrote: “Through the shared efforts of Bakst,Fokine and Ravel we have worked out a detailed program—so things are well under way” (Sergey Dyaghilev i russkoyeiskusstvo: Stat’i, otkrïtiye pis’ma, interv’yu, perepiska.Sovremenniki o Dyaghileve, ed. I. S. Zil’bershteyn andV. A. Samkov, 2 vols. [Moscow: “Izobrazitel’noyeiskusstvo,” 1982], II, 108).



55


ands.24 The final, published scenario of the bal-let omits Mirtala, Drias, Nape, Dionisofan,Klearista, Gnafon, and Filetas. One secondarycharacter, a fieldworker who flirts with Daphnisin a brief opening episode, is unnamed in thedraft scenario but identified as Lycenion in therevision. Pan maintains his presence in the re-vision, but does not physically appear onstage.Instead, his awesome, oversized profile is pro-jected against the rocks and cliffs in the décorof the middle scene.

Act I, scenes 1 to 6 of the draft scenariofurnish a well-proportioned tale of amorous rus-tics in an Edenic grove. Nymphs lay gifts at thealtar of Pan, the ruler of their enclosed world;the goatherd Daphnis and shepherdess Chloeappear, devoted to one another but soon temptedby others. A fieldworker plays a guessing gamewith Daphnis, inciting Chloe’s jealousy. Mean-time, the oafish herdsman Darkon, hoping tomake Chloe his wife, challenges Daphnis forher hand. There ensues a dance contest be-tween the two suitors, which Daphnis wins,his reward being a kiss on the cheek. Cloudsappear, lightning flashes, and the valley is over-run by brigands who abduct the shepherdessdespite tearful protests from the gatherednymphs and pledges of divine retribution. Hav-ing lost his beloved, Daphnis collapses in de-spair and falls into a fitful sleep. The brigandsload their ship with booty and prepare to setsail, but the storm intensifies, preventing theirdeparture. Bound, Chloe prays for her rescue.Her pleas, echoed by the nymphs, compel Pan,the rustic god of fertility, to appear; he signalshis intent to protect her by placing a wreath onher head. The brigands are commanded to freeher, and they in turn plead for her mercy.

Here something unexpected occurs: Daphniswakes in the woods and realizes that the deusex machina was a dream. The preceding stageevents are reinterpreted, or re-envisioned, asfantasy.25 The dream, however, proves to be a

prophetic vision of Pan’s actual imminent res-cue of Chloe. She reappears, and Daphnis joy-fully embraces her. Filetas, a grizzled herds-man who does not figure in the final, publishedversion of the scenario, brings the act to a closeby pantomiming the tale of Pan, who had onceloved a maiden who found him repugnant andrejected him. Unable to accept the rejection, hekept up his pursuit, causing the maiden to fleeinto a reed-bed that Pan cut down and fash-ioned into a flute. “The maiden was transformedinto Syrinx,” the draft scenario concludes,“Pan’s sweet-sounding pipe.”26

These are the essential plot details of theballet as it is known today. The final, pub-lished version of the scenario ends here. Thedifferences between the draft version and itsrevision are relatively modest. Pan, as men-tioned, appears only in ominous silhouette andthus does not offer Chloe a wreath. The tale ofPan is pantomimed not by Filetas, but by thetitle characters.

Act II of the draft scenario provides the back-ground detail that act I lacks, but owing to itsrepetitiveness, it had little likely chance of be-ing staged. The curtain rises to reveal a gardenbrimming with a superabundance of exoticfruits and flowers, reverently tended byDaphnis. His adopted father, the goatherdLamon, resolves to petition his master,Dionisofan, to bless Daphnis and Chloe’s mar-riage. Receipt of his approval, however, is con-tingent on his being satisfied by the state ofaffairs in his land. Daphnis’s rival Darkon reap-pears and, seeking to prevent the marriage, van-dalizes Daphnis’s garden, breaking stems andtrampling on petals. Seeing the carnage, Daphniswails “Woe is me! What will the master think,seeing such neglect? He will hang my poor oldman from a pine tree, and will also hang me.”27

There occurs a second deus ex machina. TheFirst Nymph arrives and, with a wave of herhand, restores the garden to life. Dionisofanarrives to bestow his blessing on the nuptialcouple, but is interrupted by the sage Drias,who reveals that Chloe (his adopted daughter)is actually of noble birth and thus cannot marry

24RNB fond 820, no. 2, p. 4.25Upon reading the final, published version of the sce-nario, which retains the waking episode, Jean Cocteauasked: “Do the brigands truly exist, or are they the mani-festations of a nightmare? And the metal-clad goddesses?And the great god Pan? Might not the whole thing in factbe a figment of Daphnis’s fantasy?” (Beaumont Scrapbooks,London Theatre Museum, vol. 4, pt. 1, p. 12).

26RNB fond 820, no. 2, pp. 10–11.27RNB fond 820, no. 2, p. 12.



56

19TH

CENTURYMUSIC

a common gardener. Lamon at this point inter-cedes with the disclosure that Daphnis is like-wise of noble birth, and none other thanDionisofan’s lost son. Dionsofan, humbled bythe news, blesses the wedding and orders acelebration. His minions dance.

Act II provides the biographies of the heroand heroine, describing how one pagan deity,Mother Nature, saw them through childhoodwhen their parents abandoned them in thewoods, and how another, the God of Love, es-corted them to the altar. If one were to look forSymbolist traits in the scenario, traits ofMerezhkovsky’s poetics, one might focus onthe collusion of dream and reality in act I andthe stress on scent, touch, color, and sound inact II. One would also, however, have to lookat the cluster of references in the scenario toother Anacreontic and mythological ballets,which serve, in typical Symbolist fashion, toenhance the impression of timelessness. Theworks in question involve woodland and grottoscenes, benevolent deities, malevolent forces,nymphs, hunters, peasants, and shepherds.28 Toenrich the scenario further, Fokine likewisereferred to the “ballet blanc” episodes in suchmystical, nocturnal ballets as Filippo Taglioni’sLa Sylphide (1832) and Jean Coralli and JulesPerrot’s Giselle (1841). Daphnis and Chloe’smagical helpers also call to mind the “LilacFairy” of Marius Petipa’s Sleeping Beauty (1889)and the “White Lady” in his Raymonda (1898);these figures set everything to rights in theirrespective plots by dispatching villains and

resuming interrupted wedding celebrations.29

From this litany, it becomes clear that Fokinedrafted a two-act scenario that is indeed less“Hellenic” than “Hellenistic,” a multivalenttext, in short, that documented the use and re-use of transcendental characters throughoutballet history. Though his choreographic ap-proach need not have matched his literary ap-proach, it seems that he conceived his dances astributes to ever-lasting ideals of human beauty,natural wonder, and the sublime. The “sculp-tural groups and processions”30 of the openingand closing scenes of Daphnis et Chloé wereretrospective only insofar as the performersdonned white tunics and sandals and recalled,in their held positions, Greek-style bas-reliefs.

Bakst, who created the décor for severalFokine ballets, brought a similar timelessnessto Daphnis et Chloé. One critic noted that thebackdrop for the outer scenes evoked “a hypo-thetical Greece, but not a false one—that Greecewhich was created in the imaginations of art-ists and writers of the last century, but whichwe love all the same, like the fruits of our ownspiritual creativity.”31 The title characters

28Sylvia, ou La Nymphe de Diane (1876), a ballet choreo-graphed by Louis Mérante to music by Léo Delibes, wasperhaps the strongest influence on Fokine. Its scenarioconcerns the abduction of a nymph (Sylvia) by a sinisterhuntsman (Orion), her rescue by a goddess (Diana), andher return to her beloved (Amnytas), a young shepherdwhose happiness is guaranteed by a god (Eros). When herewrote the scenario for Daphnis et Chloé, Fokine removedperiod details, redundant situations, and incidental char-acters; the result was a story line closely following that ofSylvia, ou La Nymphe de Diane. To give one example: inbook 1 of Longus’s text, Daphnis is taken prisoner by“Tyrian pyrats, in a Carian vessel”; in book 2, Chloe iscaptured (and her lover beaten) by the “Methymnaeans,” agroup of merchants exacting revenge for the loss of theirprovisions. (Longus, Daphnis & Chloe, trans. GeorgeThornley [New York: Rarity Press, 1931], pp. 53 and 80.)Fokine conflated the two abductions, making Chloe, likeSylvia, the sole victim.

29For the scenarios of The Sleeping Beauty, Raymonda,and a further influence on Daphnis et Chloé—Le Pavillond’Armide—see Wiley, A Century of Russian Ballet: Docu-ments and Eyewitness Accounts, 1810–1910, pp. 360–72,392–401, and 422–28. Fokine choreographed Le Pavillond’Armide in 1907, the same year he began work on Daphniset Chloé. The visuals include a lavish garden, a tapestrythat comes to life, and various mythological figures.30Elizabeth Souritz, “Isadora Duncan and Prewar RussianDancemakers,” in The Ballets Russes and Its World, ed.Lynn Garafola and Nancy Van Norman Baer (New Haven:Yale University Press, 1999), p. 114.31Svetlov, “Russkiy sezon v Parizhe: ‘Dafnis i Khloya’[Korrespondentsiya iz Parizha],” Peterburgskaya gazeta, 3/16 [June 1912], p. 11. Pierre Lalo was less generous in hisassessment, quipping that “the manner in which M. Bakstviews and represents Greece is decidedly unintelligible:for him, it is a dismal, cold, and sunless place” (“LaMusique,” Le Temps, 11 June 1912, p. 3). The backdrop ofscenes 1 and 3 shows a cypress grove, with two shroudedfigures tucked into the foreground. In the distance a min-iature Parthenon is visible and an outsized effigy of threenymphs marks the entrance to a grotto. Behind (or atop)the Parthenon, Bakst includes a lake with a glittering mir-rored surface. Throughout the image, perspective is dis-torted: the deciduous trees in the foreground and back-ground are too tall and too narrow, their trunks elongatedto twice their height, their lower branches stripped away.Depth in the image, moreover, is made to look like height:the lake in the distance less frames than crowns theParthenon, which adopts the likeness of a porcelain orna-ment, akin to one of the trinkets that filled Ravel’s hill-



57


moved in plain white tunics, a direct expres-sion of their sexual innocence; the corps deballet, in contrast, wore childlike prints. Thesurviving cloaks and smocks, acquired in 1969by the Stockholm Dansmuseet, feature largecircles, small ovals, checker- and zigzag-pat-terns in fanciful colors: chocolate brown, steelgrey, periwinkle blue, saffron yellow, mustardyellow, crimson, and scarlet.32 Though Bakstand Fokine subscribed to conceptions of origi-nal Hellenic purity, these garments did not beara retrograde look.33 Their neoclassicism wasabstracted and estranged, a reimagining of thecreative mindset, rather than the creativemethod, of the ancients.

IIRavel’s conception of Daphnis et Chloé, how-

ever, clashed with that of his collaborators.The result was a ballet of contrastingneoclassicisms, its visual layer composed offlowing lines and bright colors, its aural layerof floating motives and blended timbres. Thecomposer’s imagination was fired neither bythe sensuous dancing of Duncan, whom he ac-companied at a salon recital,34 nor by the pre-

cious objects unearthed at Crete, Mycenae,Tiryns, and Troy by Sir Arthur Evans, andFederico Halbherr, the archaeologists singledout by Bakst in his well-known essay on neo-classicism.35 Ravel’s meditations on the Janus-faced nature of art—the reverberation of thecreative past in the present—occupied differentterrain. In his oft-quoted autobiographicalsketch, he remarked that, upon receivingDiaghilev’s ballet commission, he resolved tocompose “a vast musical fresco, less concernedwith archaism than with faithfulness to theGreece of my dreams, which is similar to thatimagined and painted by French artists at theend of the eighteenth century.”36 Ravel wasprobably referring to the depiction of Daphnisand Chloe by François Boucher (1703–70), to-gether with his Loves of the Gods tapestries,and perhaps also to the rustic scenes of AntoineWatteau (1684–1721), though the latter con-cerned Roman rather than Greek deities. SergeLifar, a principal with the Ballets Russes, feltthat these artists were neither more nor lessauthentic than Fokine and Bakst in their depic-tions of Lesbos. “It is possible,” he surmised,“that Ravel and the French painters of the eigh-teenth century, were nearer to Longus, and thedecadence of his epoch.”37

Ravel claimed that he received the Daphniset Chloé commission from Diaghilev in 1907,two years before the first season of the BalletsRusses in Paris. The accuracy of this date hasbeen much debated in the secondary sources,with most historians concluding that the com-poser was mistaken, since he also claimed—inhis aforementioned letter to Madame René deSaint-Marceaux—that he only started seriouswork on the score in June 1909. Arbie Orensteinrefers to the controversy surrounding the dateof the commission, citing primary sources for1907 and 1909.38 Orenstein refers to Lifar, who

side house in Montfort l’Amaury. The backdrop for scene2, in contrast, shows a barren, rocky canyon, the domainnot of the nymphs, but of the brigands. The browns, greens,and yellows that govern the backdrop for scenes 1 and 3give way to orange and red.32The costumes are shown in Överdådets konst: Kostymerfrån Diaghilews Ryska Baletten i Paris (Stockholm:Dansmuseet, 1996), pp. 64–69, the catalogue of an exhibitcurated by Erik Näslund. Svetlov (“Russkiy sezon v Parizhe:‘Dafnis i Khloya’ [Korrespondentsiya iz Parizha],” p. 11)considered the costumes inappropriate, but blamedDiaghilev rather than Bakst for the problem: “It is ratherstrange that all these antique shepherds were dressed inhighly varied costumes with bright coloring that at timesactually affected the design of the dance. Modest costumesof simple colors and patterns were required. It is verydisconcerting that within this purely artistic enterprise—and what has always been Diaghilev’s enterprise—quali-ties of carelessness, of haphazardness, have become mani-fest this season. I regard this as a very serious conditionand would sincerely like to draw the attention of thetroupe’s administration to these defects.”33Bakst documented his impressions of Greek art, bothancient and contemporary, following his visit to Greece in1907. See L[éon] Bakst, Serov i ya v Gretsii: Dorozhniyezapisi (Berlin: Knigoizdatel’stvo “slovo,” 1923).34Richard Langham Smith, “Ravel’s Operatic Spectacles:L’Heure and L’Enfant,” in The Cambridge Companion toRavel, ed. Deborah Mawer (Cambridge: Cambridge Uni-versity Press, 2000), p. 202.

35Robert Johnson, “Bakst on Classicism: ‘The Paths of Clas-sicism in Art’,” Dance Chronicle 13 (1990), 180.36“Une esquisse autobiographique de Ravel” (1928) La Re-vue musicale 19 (1938), 17–23; quoted in Mawer, “Balletand the Apotheosis of the Dance,” in Cambridge Com-panion to Ravel, p. 143. The author quotes Orenstein’strans. (A Ravel Reader [New York: Columbia UniversityPress, 1990], pp. 29–37).37Lifar, Serge Diaghilev, p. 192.38Maurice Ravel: Lettres, écrits, entretiens, p. 490, n. 22.



58

19TH

CENTURYMUSIC

mentioned the encounter in his 1938 essay“Maurice Ravel et le ballet” and in his 1940Diaghilev biography. In the second source, Lifaradded that plans for the ballet evolved between1907, the year Fokine drafted the scenario, and1909. Diaghilev, hoping at first to stage Daphniset Chloé in 1910, “ventured to include a clausein his contract with Karsavina, stipulating herappearance as Chloë in alternation withPavlova.”39

The interregnum between 1907 and 1912 fu-eled the nascent disputes between the artistsinvolved in the work. Fokine’s “literal archa-ism,” which aspired, in Lifar’s view, to “‘recap-ture, and dynamically express, the form andimage of the ancient dancing depicted in redand black on Attic vases’,”40 struck a looseparallel with Bakst’s primordial stylizations,which aligned the past with the unbridled sen-suality, “nervous dynamic,”41 and malaise ofthe fin de siècle. Their methods, however, de-parted from Ravel’s: the fabrication of an amor-phous score from a shape-shifting assemblageof tonal, modal, and whole-tone syntax. WithRavel, representation yields to an elucidationof the creative fantasy behind the representa-tion.

Doubtless the principal expression of thedream state, with its indeterminate and unin-telligible character, is the vocal music of thescore. It emanates from an invisible (offstage)mixed chorus, whose buzzing and hummingsounds are sometimes articulated fortissimowith mouths open, sometimes pianissimo withmouths closed, thus negotiating the boundarybetween the audible and inaudible, the innerself and outer world. The chorus is heard dur-ing the ballet’s opening ritual scene, in whichadolescents present gifts to the nymphs; thelove scene between the title characters; thedawn episode; and the concluding Bacchanale.Jean-David Jumeau-Lafond defines the some-times distant, sometimes intimate vocal ulula-tions in these passages as “an oneiric halo that

envelops the work and confers upon it a pan-theistic dimension.” It embodies, he adds, “acollective force that unites human beings, gods,and the embodiments of nature” and transcendsthe orchestral music in its “aspiration to exalta life-giving flux without limit.”42 Jumeau-Lafond’s provocative reading places the choruswithin the extensive European tradition ofwordless mystic singing. It overlooks, however,the affect of the music: the impression of dream-like innocence that Ravel paradoxically createsby complex means.

The chorus’s entrance in scene 1 is precededby the instrumental presentation of a ladderlikearrangement of perfect fifths, A, E, B, F�, C �,G�—six of the seven pitches of the A-majorscale. These pitches are distributed in the open-ing measures over several instrumental regis-ters; their synchronic overlay supplants theirdiachronic unfolding. Through an exquisitemanipulation of orchestral color, Ravel pre-serves each pitch at the very moment of itsdissolution, the very moment it turns into re-membered, as opposed to real, sound. The lis-tener is drawn into the ambiguous interspacedifferent experiential domains. The substanceof the pianissimo tremolo A, its material incar-nation as struck and bowed pitch in the tim-pani and string basses, gives way to its embel-lishment in the altos, second violins, and harps.Repeated and overlaid, the array of perfect fifthsassumes the attributes of a harmonic series.43

In m. 8, the “muted” and “almost impercep-tible” first violins introduce real harmonics:the interval of a perfect fourth, played high onthe bridge. The upper pitch of this interval, D,rounding out the A-major scale, would appearto confirm A major as the governing tonality.

39Lifar, Serge Diaghilev, p. 192.40Mawer, “Ballet and the Apotheosis of the Dance,” pp.143, 44.41S. V. Golïnets, Lev Bakst (Moscow: “Izobrazitel’noyeiskusstvo,” 1992), p. 38. Quoted from Lifar, Serge Diaghilev,p. 265.

42Jean-David Jumeau-Lafond, “Le choeur sans paroles oules voix du sublime,” Revue de musicologie 83 (1997),275.43In her discussion of this passage, to which mine is in-debted, Danielle Cohen-Levinas remarks that “the shim-mering sonority of the timpani’s surface, placed in a stateof vibration, has the function of ‘rubbing’ with the stringsas if, literally, the bow had this surface for support.” “Theaesthetic” of the passage, she adds, is that of “spectralmusic” (“Daphnis et Chloé ou la danse du simulacre,”Musical: Revue du Théâtre musical de Paris-Châtelet 4[1987], 90). On the points in this and the next paragraph,see also Mawer, “Ballet and the Apotheosis of the Dance,”pp. 144–45.



59


This pitch, however, is preceded by a D� in theflute and harp, a modal (Lydian) inflection.

One measure after rehearsal number 1, thechorus enters with doubled perfect fourths: B–F�, C �–F�. The vocal texture derives from thedoubled perfect fourths in the French hornsone measure before rehearsal number 1. Thissyntactical relationship, which involves themetamorphosis of an instrumental sound (as-sociated with the nymphs) into a vocal sound(associated with the spirit world), is subse-quently paralleled by another. Right at rehearsalnumber 1, a solo flute introduces the tunealigned with Syrinx, the nymph that the covet-ous, amorous Pan changed into a reed pipe. Inthe ensuing measures, the flute and the chorusmerge. Jointly, they “sing out the rapture ofspirit and senses.”44

The tonal-modal hybrid is thereafter dis-solved, the tonic pedal being replaced at re-hearsal number 2 with G and then, four mea-sures later, with F. At rehearsal number 3, thepoint at which the corps de ballet begins toassemble onstage, the chorus and orchestra in-troduce the five pitches of the chromatic col-lection—F, G, C, E�, B �—missing from the pas-sage preceding rehearsal number 1. Had theopening ladder of perfect fifths not ended atG�—had it extended, in other words, beyondthe boundaries of A major—these five pitcheswould have been intoned by m. 12.

The effect of the entire passage is to presentthe orchestral music, marked by changes intimbre and syntax, as the domain of shadowstates, and the vocal music as the essentialspiritual realm behind these states. Similar re-lationships are evident in the ballet’s scenario,which changes one mode of being into another.At the start of scene 3, for example, Chloe’srescue by Pan is presented first as reality, thenagain as dream, then once more as prophecy.At the conclusion of scene 3, Daphnis and Chloeexpress thanks to Pan for his kindness by act-ing out the events of his love affair with Syrinx.Their pantomime, however, becomes a grandpas de deux; it metamorphoses into a celebra-

tion of their own love affair. Both on an auraland a visual level, the ballet swaps metaphorsfor the metaphorical, objects for the objective.The score in particular embraces this technique,approximating, through the intermingling ofsyntaxes, what the French Symbolist poetCharles Baudelaire called the “universal corre-spondence.”45

The conclusion of scene 1, where the chorusmarks a shift in the drama between whatMerezhkovsky called “morning” and “twilight”states, clarifies these points. The singers hereperform for thirty-two measures a cappella,translating the preceding orchestral passagesinto desemanticized vocal exhalations. Thewordless (but not vowel-less) utterances con-jure a noncognitive state: the prelinguistic,presubjective envelope of utterances that psy-choanalytic theorist Julia Kristeva terms “thechora.” Kristeva adopts the term from Plato,who used it to identify the “eternal, unchang-ing ‘Forms’” of the cosmos “and their unstable‘reflections’ in the physical, perceptible worldof ‘becoming’.”46 In the Timaeus, a dialogueabout the advent of the world, Plato describesthe chora as a “receptacle” whose spatial ele-ment affords “a location for all things that cometo be.”47 Though invisible to our eyes and inau-dible to our ears, the chora makes itself knownin dream states, the nocturnal vistas whereshapes and sounds arise before us as thoughfrom a dark void, an alien ocean.

Kristeva redefines the chora by using it as ametaphor for the “receptacle” of the primalforces (or drive energies) that precede the for-mation of subjectivity. She describes it as “anessentially mobile and extremely provisionalarticulation constituted by movements andtheir ephemeral stases.”48 The chora, Kristevaexplains, antedates cognitive perceptions ofspace and time. It “can be designated and regu-lated,” but “it can never be definitely posited.”

44Charles Baudelaire, “Correspondences,” 1857, quoted inJulia Kristeva, Tales of Love, trans. Leon S. Roudiez (NewYork: Columbia University Press, 1987), p. 334.

45Baudelaire, “L’Art romantique,” 1868, quoted in Tales ofLove, p. 330.46Plato, “Timaeus,” in Complete Works, ed. John M. Coo-per (Indianapolis: Hackett Publishing, 1997), p. 1224. Thequotation comes from Cooper’s introduction to this text.47Ibid., pp. 1254 (51a) and 1255 (52b).48Kristeva, Revolution in Poetic Language, trans. MargaretWaller (New York: Columbia University Press, 1984),p. 25.



60

19TH

CENTURYMUSIC

For this reason, “one can situate the chora and,if necessary, lend it a topology, but one cannever give it axiomatic form.”49 The SATB vo-cal writing in Daphnis et Chloé offers an ap-proximation of the breaks, surges, and pausesof this space.50 The singing undulates by tonesand semitones, the inner parts misaligned withthe upper and lower parts. Pedal tones anchorthe music less than they enhance its disso-nance; major and minor triads, enlarged bythirds above and below, break down intononharmonic tones. The music does, however,bear structural logic, which divorces it fromthe disintegrative, wayward character of thechora. Disorder’s mirage replaces disorder it-self. The strong vibrato makes the vocal linessound microtonal, but they are in factsemitonal, doubled at the third, and destinedfor consonant resolution. The C–E–F� (G�) andF�–E–B�–D sonorities that brace the a cappellapassages at rehearsal numbers 83 + 4 and 88 ofthe 1913 orchestral score make up “whole-tone”metamorphoses of the B–D–F–G “dominant-seventh” chord in the exact middle of the epi-sode, at rehearsal number 85. This chord findsbelated partial resolution in the C pedal intro-duced by the cellos as the singing concludes(ex. 1).

Beyond the chora, Ravel’s offstage chorusbrings to mind the Hellenic dithyramb, thetype of vocal ode performed in honor ofDionysus at pre-Christian-era religious festi-vals. The dithyramb was first identified by theseventh-century BC bard Archilochus of Paros,who described it as a prominent element ofAthenian ceremonies. Other sources reveal thatit was originally performed in unison by smallgroups of adolescent males and later by mixedchoirs engaged in antiphonal exchanges between

leader and ensemble. The term “dithyramb”refers etymologically to the “double birth” ofDionysus in his mother Semele’s womb and,after Semele’s death, his father Zeus’s side;eventually it came to mean a fervid hymn ofpraise. From a choreographic perspective, onemight add that the middle syllable “amb” im-plies a “step” or “movement.” (“Amb” comesfrom the words iambos and thriambos, whichidentify one- and three-part foot sequences re-spectively.) By the first century CE, the word“dithyramb” had also come to designate a text-setting practice. The examples by Bacchylides,Philoxenus, and Timotheus exhibit frequentshifts in meter and stress. Anecdotal evidencesuggests that the melodies and instrumentalaccompaniments to which their verses wererecited modulated for expressive effect. Accord-ing to the rhetorician Dionysius of Halicar-nassus, the dithyrambic poets used “Dorian,Phrygian, and Lydian [modes] in the same poem,and they varied the melodies, making themsometimes enharmonic, sometimes chromatic,and sometimes diatonic, and they fearlesslygave themselves complete rhythmic indepen-dence.”51

Daphnis et Chloé exhibits a similar clusterof effects. The choral lines transform the “some-times enharmonic, sometimes chromatic, andsometimes diatonic” syntax allotted to thedithyramb into a hybrid of whole-tone, chro-matic, and tonal syntax amid much metric li-cense. Between rehearsal numbers 83 and 84,the bass voices intone the pitches of a whole-tone collection (G, A, B, C�, D�, E�), while be-tween rehearsal numbers 87 and 88, they in-tone a G-major collection. The upper (soprano)line and inner (contralto and tenor) lines,marked by half- and whole-step motion, movein and out of phase with each other. The poly-phonic texture is saturated with appoggiaturas,suspensions, and upper and lower neighbornotes. In a wonderful discussion of the ballet,Deborah Mawer describes the larger intervallicleaps in the texture—the perfect and dimin-

51Dionysius of Halicarnassus, Laws, quoted in ArthurPickard-Cambridge, Dithyramb, Tragedy and Comedy, rev.T. B. L. Webster (Oxford: Clarendon Press, 1962), p. 43.The preceding information in this paragraph comes frompp. 1–9.

49Ibid., p. 26.50Lawrence Kramer offers the following insight: “The cho-rus reduces the human voice from a vehicle of agency andsignification to a pure and mobile materiality. Voice inthis form is meant to give the impression of ‘speaking’ thesemiotic from outside the symbolic. It mirrors a kind ofsubjectivity, sometimes called ‘ideodynamism,’ that pre-occupied the nouvelle psychologie of late-nineteenth-cen-tury France: an unconsciously determined flight of ideas(nervous excitations) associated variously with hypnotism,suggestion, and dreams” (Classical Music and PostmodernKnowledge [Berkeley and Los Angeles: University of Cali-fornia Press, 1995], p. 208).



61

SIMONMORRISONOrigins ofDaphniset Chloé� 34 � � � � ��

� 34 � � � � � �� 34 �� 34 ��

Sop.

Contr.

Ten.

Bass

��

��

�

� � � ��

Sop.

Contr.

Ten.

Bass

�

� ��

� ��

Sop.

Contr.

Ten.

Bass

�

� � ��

�

�

��

��

�

�

83 N.B.Derrière la Scène, on entend des voix, très lointaines d’abord.Même mouv.t

N.B.

84

85

�

Example 1: The a cappella chorus.

52Mawer, “Ballet and the Apotheosis of the Dance,” pp.146 and 148.

production: the invisible singing recoded themusic assigned to the visible dancing.

There is, however, a third, much less am-biguous source for Ravel’s offstage chorus: thewordless singing in those French and Russianfin-de-siècle compositions with fantastic andspiritual subject matter, an extensive repertoirethat Jumeau-Lafond has plumbed to greatdepths. Ravel’s chorus has an attitudinal kin-ship to Hector Berlioz’s Le Ballet des ombres:

ished fourths and fifths—as “a poignantextemporisation” of the “legato waltz-themeassociated with Chloe” four measures after re-hearsal number 29.52 This point underscoresone likely reason why Ravel berated Diaghilevfor eliminating the chorus from the London



62

19TH

CENTURYMUSIC � � ��

� � � ��

Sop.

Contr.

Ten.

Bass

��

��

��

� � � � � � � � � � ��

� � � � � � ��

� � ��

�

�

1. Hn.

1. Trpt.

Sop.

Contr.

Ten.

Bass

Str.

� ��

� ��

� ��

�

� ��

�

�

�

�

86 � 87

88loin derrière la scène

Des appels de trompes au loin.

N.B.

à l’orchestreUnis(1)

Les voix se rapprochent

� � �loin derrière la scène

3434

Example 1 (continued)

53Jumeau-Lafond, “Le choeur sans paroles ou les voix dusublime,” pp. 264–65.

54Marcel Marnat, Maurice Ravel (Paris: Fayard, 1986), p.338, n. 44.

Formez vos rangs (1829), a “nocturnal round”for chorus and piano that was included in thedramatic symphony Romeo et Juliet (1839), andthe same composer’s La Mort d’Ophélie (1842),a ballade arranged in 1848 for female voices.53

Other wordless choruses heard—and in someinstances written about—by Ravel come fromVincent D’Indy (Fervaal, 1895), Paul Dukas(Ariane et Barbe-bleue, 1907), Florent Schmitt(La Tragédie de Salomé, 1907), and Claude

Debussy, whose work list would have beenintimately familiar to Ravel. (Ravel would alsohave known that in 1895 Debussy consideredsetting Pierre Louÿs’s Daphnis et Chloé li-bretto.54) In “Sirènes,” the third of Debussy’sNocturnes (1899), soprano and contralto voicesinterweave, furnishing, through an eerie ma-nipulation of “ahs” and “ohs,” distorted trans-lations of orchestral sounds. The voices repre-sent the nocturnal incantations of Neptune’s



63


daughters, the orchestra the agitated ocean. Inact I, scene 3 of Pelléas et Mélisande (1902),Debussy likewise uses wordless male singingto suggest the advance of a mist-shrouded ship.55

On the Russian side, Ravel likely also foundinspiration in Nikolai Rimsky-Korsakov’sopera-ballet Mlada (1892), a score that Jumeau-Lafond does not mention, but that includesboth a wordless “ghost” chorus and a hummed“infernal” chorus. This work, like Rimsky-Korsakov’s Capriccio espagnol (1887) andShéhérazade (1888), served as a creative modelfor Ravel, notably in the realm of orchestra-tion.56 Finally, Alexander Scriabin’s tone poemPrometheus (1911), composed at approximatelythe same time as Daphnis et Chloé, meritscitation. The thick score reaches a sensual ze-nith with a mixed chorus intoning “ah,” “e,”and “oh,” a pattern that increases in intensityand that symbolizes the rapturous utterancesof a collectivized reborn humanity. The pianosymbolizes Prometheus—a shining star in thefirmament—and the orchestra the four elementsof matter.

To listen to Ravel’s offstage chorus is thusto be escorted through a gallery of musical arti-facts. To revert back to Merezhkovsky’s termi-nology, the music is “Decadent” as opposed to“Symbolist”; it provides decorative acousticbackground for the ballet, an artifice of thenature worship and religious iconicity endemicto Symbolist art. The high- and low-pitch pho-nemes are devoid of extramusical meaning.Ravel achieves a nihilistic inversion with hischorus, inscribing the presumed inability ofthe audience to interpret the sound into thesound itself. The extramusical program is theabsence of an extramusical program. By forgingthe illusion of intertextual complexity, the off-stage singing lures us into thinking that it har-bors hidden codes, secret conceits, but theseprove to be little more than chimeras. Each

“oh” and “ah” is a concatenation of metaphor,a mélange of allusions to various compositionalgenres—French and Russian orientalism, thedithyramb, the Platonic “chora”—that leads ourimagination back to an implied but not im-plicit primal source, as though the singing borethe trace of psychomythical events. Followingthe trail of musical crumbs, we reach the van-ishing point of hermeneutic understanding.

Besides the chorus, Ravel dwelled on theballet’s heaven-storming finale, composing ittwice, first (primarily) in 34 time and then (pri-marily) in 54 time, the latter a meter that Fokinestruggled with during the rehearsals. There sur-vives intact an engraved piano score of the 3

4version of the finale—which Ravel recalled forrevision from his publisher in October 1910,though not before it had been reproduced andseveral copies distributed to music dealers—and sufficient anecdotal evidence to suggest amultilayered creative process. The French mu-sicologist Jacques Chailley, the first to com-ment on the 3

4 version, recalls finding it byaccident at a used bookshop in 1964. Puzzledby his acquisition, he approached Ravel’s pub-lisher, Durand, about its provenance. Chailleyrecollects that the firm’s director, RenéDommange, “desired to look into the matterhimself and, in a letter of November 24, 1964,confirmed to me that, as I had assumed, thisscore had been ‘issued from the printed proofsin advance of Maurice Ravel’s corrections’.”Chailley had stumbled across a rare draft, onethat predated the composer’s “submission ofthe final orchestral manuscript to the editor,”and his “inclusion, at the proof stage, of thoseimportant alterations that were included in theorchestral manuscript.” “Normally, all trace ofthis first version would have disappeared,”Chailley adds; “chance alone preserved the evi-dence.”57

Beyond identifying alterations in barring,rhythm, and phrase length between the first

55Jumeau-Lafond, “Le choeur sans paroles ou les voix dusublime,” p. 267.56Orenstein, introduction to Maurice Ravel: Lettres, écrits,entretiens, p. 38. While familiar with Rimsky-Korsakov’sMlada, Fokine would have been better acquainted withMarius Petipas and Ludwig Minkus’s ballet of the samename, which premiered in 1878 and, after 1896, retained afoothold in the Russian Imperial Ballet repertoire.

57Jacques Chailley, “Une première version inconnue deDaphnis et Chloé de Maurice Ravel,” in Mélangesd’histoire littéraire (XVIe-XVIIe siècle): offerts à RaymondLebègue (Paris: A. G. Nizet, 1969), pp. 371–72. This docu-ment is preserved at the Département de la musique,Bibliothèque Nationale, as LoVm 6 17: 1910 D[urand] &F[illes] 7748. Ravel’s revision is preserved as Rés. Vma.222: 1910 D. & F. 7748.



64

19TH

CENTURYMUSIC

and final versions of the “Danse générale,”Chailley also notes changes in the proper namesof the characters, changes strongly suggestingthat Fokine revised his original scenario onlyafter he met Ravel in 1909. The buffoon“Darkon” was first named “Darion,” theingénue “Lycenion” “Lyceia,” and the villain“Bryaxis” the “Leader of the Pirates.” Some—but pace another French musicologist,58 notall—of the original spellings came from thechoreographer’s Russian-language draft. Oth-ers accord with the French-language transla-tion of Longus’s pastoral romance by JacquesAmyot (Les amours pastorales de Daphnis etChloé, 1559). This was the version of the ballet’ssource text consulted by Ravel.

Beyond the name changes, Chailley also iden-tifies a change in plot detail. The stage direc-tion for the apparition of Pan at the end ofscene 2 first reads “Fearsome, Pan appears in achariot drawn by savage beasts,” but was sub-sequently recast as “Fearsome, Pan’s shadow isprofiled on the hills in the background, gestur-ing menacingly.”59 As mentioned earlier, thisalteration led to the removal of Pan from thecast list. It also marked a significant departurefrom Longus’s pastoral romance, one that at-tests to a dispute between Fokine and Ravelabout the ballet’s visceral content. In the draft,Pan lives and breathes onstage. The deity “sud-denly appears” amid the brigands and mimes anone-too subtle warning to them: “I’ll kill youall, if you don’t release Chloé and her flock.”60

In the final scenario, Pan becomes an abstrac-tion, a bas-relief on stone.

Though Chailley does not mention it, thereoccurred another significant change to the miseen scène. The direction for the flirtatious ex-change between Daphnis and Lycenion in scene1 first reads: “But [Daphnis] recognizes Lyceiaand playfully turns away”; it was rewritten as:“But [Daphnis] recognizes Lycenion and tries

to pull away.”61 The alteration reduces the emo-tional content of the drama. The goatherd nolonger expresses any feelings for the coquettishfieldworker, but simply removes himself fromher physical orbit. In this subtle detail, oneglimpses how, as the ballet came into being,Fokine’s literal approach to the subject matterclashed with Ravel’s stylized approach.

It is not certain why Ravel decided to re-write the finale, causing Fokine headaches withhis dancers in the process, but competition fromIgor Stravinsky, whose The Firebird andPetrushka were choreographed by Fokine andstaged by the Ballets Russes in 1910 and 1911,respectively, is the most likely cause. ArnoldBennett, an acquaintance of Ravel, noted in hisdiary that the composer, upon completingDaphnis et Chloé, started to worry that it wouldsound old-fashioned to Parisian connoisseursof chic and so decided to recalibrate the finale.Louis Aubert, a Ravel disciple, adds that thecomposer was tinkering with the finale as lateas April 1911.62 The example of Stravinsky,specifically his extensive reliance, in his balletscores, on machinelike ostinati and asymmetri-cal accent and pulse patterns, probably inspiredRavel to expand and extend his own finale. Themusical language of the 3

4 version, composedbefore Ravel had heard The Firebird andPetrushka, includes advanced (chromatic andwhole-tone) syntactical relationships but notadvanced temporal ones. The exotic braid ofeffects recalls Alexander Borodin’s PolovtsianDances and Alexander Glazunov’s Bacchanale.Fokine choreographed these ballets in 1909 and1910, shortly before the Stravinsky ballets.

Behind the original 34 version exists an addi-tional work that has not been mentioned in thesecondary literature and that, owing to its evi-dent incompleteness, cannot be performed. Thiswork, dating from 1907 or 1908, is the originalDaphnis et Chloé. It was written by AndreyKadletz, Fokine’s former colleague at the Impe-

58Roland Manuel. See Chailley, “Une prèmiere versioninconnue de Daphnis et Chloé de Maurice Ravel,” p. 375,n. 4.59Ibid., p. 374. A letter of 3 May 1910 to the music criticMichel Calvocoressi finds Ravel seeking the correct spell-ings of Lycenion and Lammon along with the name ofPan’s flute, which, he quipped, “neurasthenia” caused himto forget (Maurice Ravel: Lettres, écrits, entretiens, p. 112).60RNB fond 820, no. 2, p. 7.

61LoVm 6 17: 1910 D. & F. 7748, p. 32 (“Mais il reconnaitLyceia et se détourne capricieusement”); Rés. Vma. 222:1910 D. & F. 7748, p. 32 (“Mais il reconnaît Lycenion etveut s’éloigner”).62Arnold Bennett, The Journal of Arnold Bennett, 1932,and Louis Aubert, Souvenirs confiés à Jean Guitard, LaTable Ronde, quoted in Marnat, Maurice Ravel, p. 333.



65


� � � 38 �� 38 ��

� � � � � � � � � � � � ��

� celli�

alti

rial Ballet and the composer with whom hecollaborated on his first ballet, Acis et Galatea.To date, all that I have uncovered of the musicis a fragment of the ending in triple meter (38),which Kadletz copied out as a gift to two mid-ranking members of the Russian nobility on 9April 1909.63 The rest of the manuscript maybe lost or, as Kadletz’s granddaughter Muzasuggested to me,64 may have been left unfin-ished, since Fokine pulled out of the project inmedias res once he received word that the Im-perial Theatres had rejected his draft scenarioand would not be staging the finished ballet.This was approximately the same time thatDiaghilev, according to Ravel, became inter-ested in the scenario.

There are no correlations between Ravel’sscore and Kadletz’s fragment; rather, there areexplicit departures. Fokine proudly recalled that“it was unnecessary for me to ask Ravel torefrain from the traditional forms of the oldballet music, as I had to do in the case of myfirst music collaborator, A[ndrey] Kadletz [. . . .]Having seen a series of my productions, [Ravel]was already well aware that polkas, pizzicatos,waltzes, and gallops—so indispensable in theold ballet—were completely out of place in thenew.”65 Fokine glosses over the facts some-what: his “new” ballets did include set pieces,

albeit only where they suited the drama.66 Thechoreographer does, however, identify a knotof tension between the two Daphnis et Chloés.Kadletz’s finale is nothing if not a waltz, whileRavel’s finale—at least in its revised version—has no precedent in “the old ballet” (ex. 2).

The opening of the 34 version, bearing the

subheadings “A group of young males rusheson stage” and “Joyous tumult” is brief.67 Thepassage is a harmonic and melodic microcosmof the more extensive “Danse générale” thatfollows. Example 3a highlights the structuralpitches of the extremely angular bass line: thetonic pitch A, C�, G, and D�, the last interpret-able as the dominant (E) displaced by semitone.Three sonorities, each articulated in the middlerange of the texture, comprise whole-tone clus-ters: C�–E�–G–B (m. 2), B�–D–F� (m. 6), and B–D�–F–G (m. 11). The conflict between B and A,the pitches stressed in the upper and lowerregisters, is audible throughout the finale(ex. 3).

The middle section, featuring solo variationsby Daphnis, Chloe, and Darkon, inaugurates asteady chromatic descent in the bass from F� toE�, each pitch functioning as the root of a ninthchord with a missing interval: F�–C–E–G, F–A–E�–G, E–B�–D–F, and E�–B�–D�–F. The ending ofDarkon’s solo inaugurates a further extensionof the chromatic descent, first to D and then toC�, at which point the corps de ballet combineswith the soloists for the apotheosis. Eventu-

Example 2: Fragment from the finale of Andrey Kadletz’s Daphnis et Chloé,Russian National Library, St. Petersburg.

63Aleksandr Gavrilovich Maksimov and Mariya Konstan-tinovna Maksimova, Sobraniye avtografov raznïkh lits1900–1917 gg., Al’bom 6 [4 apr. 1909 g.—29 yanv. 1911 g.],p. 9. The volume is preserved in RNB fond 459.64Interview, 4 March 2002, St. Petersburg. Ms. Kadlets is amusical commentator for Radio Russia (Radio Rossii).65Fokine, Memoirs of a Ballet Master, p. 196. Both this andthe following quotation from the English-language editionof the memoirs are identical in content to the equivalentpassages in the two Russian-language editions.

66For an overview of Fokine’s choreographic advances, seeGarafola, Diaghilev’s Ballets Russes, pp. 3–49.67LoVm 6 17: 1910 D. & F. 7748, p. 92 (“Un groupe dejeunes hommes envahit la scène. Joyeux tumulte”). Forthe following technical points, I am indebted to Christo-pher Matthay.



66

19TH

CENTURYMUSIC

� � � � 4 � � � �� 4 �� 4 ��

� �

��

� � � � � � ��

��

��

� � � � � � � ��

��

� � � � � � � ��

��

� � � � � � � � ��

��

�� ff

�

� � � � � ��

��

�128128128

333( )( )( )

��

��

N.B.Animé

Un groupe de jeunes hommes envahit la scène. Joyeux tumulte.

�

subito m.g.

N.B.

m.g. � subito

N.B.

&&

&&

4

8

12

16

20

a. The first two pages of the original (34) finale, Bibliothèque Nationale, Paris.

Example 3



67

SIMONMORRISONOrigins ofDaphniset Chloé� � � � � ��

� � � � ��

� � � � ��

��

� � � � � � � � � � � �

� �Danse générale

23

27

��

a. (continued)


ally, Ravel establishes E, the dominant, in thebass, but cadential closure is postponed whenan augmented-sixth chord on B� fails to resolveto A major. This chord subsequently recurs(enharmonically respelled), but resolves onemeasure afterward with the fortissimo final ap-pearance of the chorus. In the second to lastmeasure, the contraltos push B, the pitch em-phasized in the upper register throughout thefinale, to C�, the third degree of the tonic chord.

In terms of texture, the principal differencebetween the first and second versions of thefinale concerns the chorus, whose role is ex-panded from a mere six measures in the formerto over eleven pages in the latter. The 34 finaleoccupies eleven pages of the piano score, the 54finale twenty-two pages. The combined maleand female voices most often enter on the sec-ond beat of the measure, thus blurring themeter, and rise and fall in contrary motionwith the instrumental lines. Only at the veryend, in the brief duple meter coda, do the voicesenter on the first beat. In terms of syntax, thereare no manifest differences between the firstand second versions, though the expansion ofthe score enhances the psychological effect ofthe chromatic runs. Certain chords, amongthem the unresolved augmented sixth, becomemore prominent, but the changes are rhythmicrather than syntactic.

In the second version, the Stravinskian 54 epi-sodes—which, in Karsavina’s recollection, Ravel

told the dancers to count “1 2 3–1 2 3 4 5–12”68—are interrupted by duple and (less fre-quently) triple meter passages, most notably atthe outset prior to the “Danse générale” (ex.3b). The accordion-like augmentation and dimi-nution of meter, together with the displace-ment of tonal regions, produce the impressionof narcosis and delirium. In Daphnis et Chloé,“Bacchante” existence differs from “Shepherd”and “Brigand” existence; it inhabits an alteredstate in which “time images,” segmented, sus-pended, and interlocked durations identifiedwith altered consciousness, supplant “move-ment images,” cognitive and empirical mea-sures of time.69 Interpreted as a choreographicprescription, the frequent changes of meter in-carnate the template for multilayered ensemblework, with different groups of dancers conform-

68Tamara Karsavina, Theatre Street (1948; rpt. London: Co-lumbus Books, 1988), pp. 238–39. Strangely, Lifar, whodanced in the Daphnis et Chloé premiere, claimed thatthe meter was danced in a 2 + 3 pattern: “Rehearsals wereconducted with difficulty, and in an atmosphere chargedwith ill-feeling, the finale, in particular, giving endlesstrouble to the corps de ballet, written, as it was, in five-four time. However, by omitting actual counting, and sub-stituting the syllables ‘Ser-ge[i]-Dia-ghi-lev’ in its place,the corps de ballet, after humming it over an infinity oftimes, finally succeeded in getting it right” (Lifar, SergeDiaghilev, p. 193).69The terms “time image” and “movement image” comefrom the philosopher Gilles Deleuze. For detailed defini-tions, see D. N. Rodowick, Gilles Deleuze’s Time Ma-chine (Durham: Duke University Press, 1997), pp. 8–14.



68

19TH

CENTURYMUSIC

� � � � 44 54�� 44 54� (( � � � � � 44 54� ��

� �

��

�

��

��

� �

�

��

� � � � 24�� 24 � � ��

��

� �

� � � � 24 54� �� )� �� 24 54� ��

��

� � � � 24� �� 24 ��

��

� � � � 54� ��

� � � � 54 ��

��

��

� � � � 54 � �� 54 ��

� � �� ff

�

Lent = 50*

��

��

��

expressif

Animé

Daphnis et Chloé s’enlacent tendrement. Un groupe de jeunes hommes envalit la scène. Joyeux tumulte.

� � � � � � � � � � � � � � � � ��

� � � � � � � � � � � � � �

�subito

�

subito

�

&

&

b. The first two pages of the revised (54) finale.




69

SIMONMORRISONOrigins ofDaphniset Chloé� � � � � ��

� � � � � ��

� � � � �� + � + + � +

��

�Danse générale

b. (continued)


ing to different pulse patterns. Moving in andout of phase with each other and with the ac-companiment, the dancers would express col-lective abandon, suffusing the finale with adreamlike or oneiric ambiance.

In reality, Fokine devised a “unique meth-od”70 for staging the finale, one that, like Ravel’smusic, had a Stravinskian quality. In his re-splendent, kaleidoscopic choreography for theostinato-driven crowd sequences in the firstand fourth scenes of Petrushka, Fokine scriptedthe steps in such a manner as to convey theillusion of disorder.71 In Daphnis et Chloé, heoffered his seasoned dancers interpretive license,thus creating the impression that the choreo-graphic “text” had somehow escaped his con-trol. The result was a simulacrum of Dionysianecstasy, a hallucinatory flux. In the final dressrehearsal, which found Fokine still construct-ing the finale, he

sent one bacchante across the stage, then another,then two at a time, then three together, then anentire group with interwoven arms reminiscent ofGreek bas-reliefs. They rushed across the stage again,singly and in groups. I gave each a short but differentcombination of steps. Each dancer was required tolearn only her own brief passage.

Having thus led everyone upstage from one wingto the other, I then had the remaining mass emergefrom the downstage wing. The entire ensemblelurched together in a whirlpool of a general dance,and—the biggest part of the most difficult finale wasready! It only remained to stage a small passage forDaphnis and Chloë, a solo for Darkon, and the gen-eral end.72

Fokine’s ensemble technique integrated hisown kinesthetic thought with that of his expe-rienced dancers while also allowing for the in-clusion of chance elements, those unplannedfeatures of physical expression that derive fromthe circumstances of actual performance. Inhis 1914 manifesto, Fokine described his en-semble technique as a commingling of “theexpressiveness of the individual body [with]the expressiveness of a group of bodies and theexpressiveness of the combined dancing of acrowd.”73 This fusion of affects found its equiva-lent not only in the accelerating orchestral mu-sic of the finale, but also in the wordless cho-rus, the primordial wellspring of the score.Ravel’s decision to recompose the finale and

72Fokine, Memoirs of a Ballet Master, p. 210. On p. 213,Fokine adds that for the scene 2 episode in which “the godPan, to save Chloë from the pirates, puts terror into them”he also “permitted a somewhat collective creation.”73Fokine, “Letter to ‘The Times,’ July 6th, 1914,” in WhatIs Dance? Readings in Theory and Criticism, ed. RogerCopeland and Marshall Cohen (Oxford: Oxford UniversityPress, 1983), p. 260.

70Fokine, Memoirs of a Ballet Master, p. 210.71See the discussion of Fokine’s ensemble technique inGarafola, Diaghilev’s Ballets Russes, pp. 21–25.



70

19TH

CENTURYMUSIC

his insistence on tinkering with it after itscompletion may have frustrated Fokine,74 butthe finale became the one section of the balletwhere the difference between his and thecomposer’s conception of antiquity was erased.

The only other hint of parallel thinking con-cerns the brigands. The eclectic poet JeanCocteau observed that at the start of scene 2Parisian audiences were offered “a fearsomeglimpse (between two mossy glades straightout of the eclogues) of the rugged creek wherethe pirates chain up poor Chloe while theyindulge in their rough horse-play.” “Nothing,”Cocteau added, “could provide a more delight-ful contrast to [the pirates’] gruesome gambolsthan Daphnis’s dance, tinkling and twinklinglike dew, or the ever-breaking garland of themost Latin of farandoles.”75 This remark is cryp-tic, to say the least, but it offers some insightinto the relationship between décor, dance, andmusic in the ballet. Against the rock-strewnbackdrop of scene 2, the brigands lumbered toand fro, while against the verdant backdrop ofscene 3 the positive characters performed afarandole, a chain dance dating from the four-teenth century. Assuming that one of the Lon-don critics was correct when he remarked thatFokine’s corps de ballet tended to move in syncwith Ravel’s score,76 one could further deducethat the brigands struggled to follow therhythms of the music, while the positive char-acters—tracing the curved, arched figures typi-cal of the farandole—adhered to the melodicphrasing. The basic difference between the“Danse légère et gracieuse de Daphnis” (sc. 1,rehearsal numbers 43 to 51) and the brigands’“Danse guerrière” (sc. 2, rehearsal numbers 92to 104) is, in fact, one of melody and its ab-sence.

IIIThe preceding quotation from Fokine’s mem-

oirs is his only substantive description of thechoreography of the original Daphnis et Chloé.This choreography is, by all standards of mea-surement, lost. It was recorded in neither writ-ten nor visual form and cannot be significantlyreconstructed. Like Daphnis after Chloe’s cap-ture, the traces of the choreography are entitiesin repose.

Of the preserved visuals, the palm-sized im-ages by the artist Valentine Hugo (neé Gross,1887–1968) of Nijinsky, Karsavina, AdolmBohm (who danced the part of Darkon), andLudmila Schollar (who danced the part ofLycenion) provide the most information aboutthe Paris premiere. Sketched with Diaghilev’sknowledge and agreement in the darkenedThéâtre du Châtelet, the images offer both dra-matic detail and an imprint of striking posesand gestures. Richard Buckle, who analyzedtwenty of the images, hypothesized that “asthe artist’s eyes follow[ed] the figure of Nijin-sky on stage, she [did] not know whether herhand [was] drawing head, neck, arm or cos-tume.” Usually it etched “none of these, butinvent[ed]—with the speed born of necessity—a symbol for movement.”77 Each pencil andpastel flourish, in effect, preserved a fleetingphysical reaction, the physical production ofemotional or psychological effects. As in allballets, the choreography was “allographic,”scripted and shaped by Fokine for his dancers,but also “autographic,”78 bearing the dancers’own creative signatures. Hugo, accordingly,sought to represent both stage events and theparticularities of the individual performers’styles.

Buckle identifies grand jetés and épaule-ments in some of Hugo’s pastels; in others,those not published in his monograph on the

74On this point, see Clive Barnes, “Daphnis and Chloë,”Dance and Dancers 9:7 (July 1958), 16–17.75Beaumont Scrapbooks, vol. 4, pt. 1, p. 12.76“The one drawback in the design of many of the con-certed dances seems to be a tendency to reproduce theexact pattern of the melodic phrases in the steps, much asM. Jacques Dalcroze’s students do, and that is apt to pro-duce an impression of artificiality as compared with reallyrhythmic dancing” (Unsigned, “Ballet of Butterflies: ‘Pap-illons’ at Drury Lane,” Times, 12 June 1914, p. 12). Muchof this review concerns, as its title implies, Papillons; thequoted passage, however, concerns Daphnis et Chloé.

77Richard Buckle, introduction to Nijinsky on Stage: Ac-tion Drawings by Valentine Hugo of Nijinsky and theDiaghilev Ballet Made in Paris between 1909 and 1913(London: Studio Vista, 1971), p. 13.78Joseph Margolis, “The Autographic Nature of the Dance,”Journal of Aesthetics and Art Criticism 39:4 (1981), 419–27. The terms “allographic” and “autographic” were coinedby Nelson Goodman (Languages of Art: An Approach to aTheory of Symbols [Indianapolis: Bobbs-Merrill, 1968]),whom Margolis critiques.



71


79Valentine Hugo Archive, London Theatre Museum. Thefirst and second images, tagged 159 and 160 respectivelyin the archive, are reproduced in Nijinsky on Stage, pp.109–10. Three of the four other images are tagged 6, whilethe fourth is tagged 175 and reproduced in Nijinsky onStage, p. 129.

artist, the curves, lines, and squiggles aremore ambiguous, like Rorschach tests forTerpischore. Ballet audiences of the early twen-tieth century were accustomed to reading sto-ries into physical gestures, however, and wouldhave been able to extract plot details fromHugo’s images. Seen in this light, her work isvaluable not only as a technical glossary, aguide to the formal elements of Fokine’s dances,but also as a storybook. In one image (see plate1a), a male figure leans on a missing staff; in asecond, another male is seen movingungracefully, right arm stretched skyward, righthand fingers splayed, left arm and hand dan-gling, knees bent, and torso twisted, breakingthe natural line of the spine. In four other im-ages, the profile of a female figure graduallycomes into view, first in a fetal position, thenin the general shape of a boulder (see plate1b).79 The first image can be interpreted as

Daphnis tending his flock in isolation in scene1 or 3, the second as Darkon performing hisgrotesque duple meter variation in the scene 1dance contest, and the four others as Chloecrumpled in abject despair in scene 2 followingher molestation by the brigands.

The studio photographs of the ballet aresometimes more, sometimes less provocativethan Hugo’s stage drawings. Many of these werepublished in the French and British illustratedpress—Comoedia illustré, Le Théâtre, Illus-trated London News, the Sketch, and theTatler—between 1912 and 1914; others arehoused in the Mikhail Fokine Archive at theState Theatre Library in St. Petersburg. Withone exception, these images complete the cho-reographic record of the original Daphnis etChloé. The exception is a stage photograph,dating from 1914, that shows the corps de bal-let reaching skyward in praise of Pan in scene1. It is “a rare example in this period of aphotograph taken from the stage.”80

80Souritz, “Isadora Duncan and Prewar RussianDancemakers,” p. 115. The photograph is reproduced onthis page.

Plates 1a and 1b: Two of Valentine Hugo’s drawings, as reproduced in Nijinsky on Stage.


http://www.jstor.org/action/showImage?doi=10.1525/ncm.2004.28.1.50&iName=master.img-000.jpg&w=471&h=273



72

19TH

CENTURYMUSIC

the chorus-less London version, and inscribed“M[ikhail] M[ikhailovich] in the role of Daphnisin 1914.” File 13/8, finally, contains two pho-tographs of Mikhail and Vera in “antique balletcostumes.”

The images in files 13/7 and 13/8 resemble,but do not duplicate, those published in the 15June 1912 issue of Comoedia illustré, whichfeatured a photo essay on the 1912 season ofthe Ballet Russes. The images in 13/6 are, tothe best of my knowledge, unknown. Three ofthem were made in a professional London stu-dio by Saul Bransburg, who took many photo-graphs of Diaghilev’s troupe in 1914, several ofwhich were published in The Sketch; the re-mainder are amateur stills made in the court-yard of a Paris apartment building. In one of thelatter, Fokine is shown balancing a crossbow ata forty-five degree angle over his shoulders. Hestands on his right leg, with his right foot ondemi-pointe; his left leg is raised and bent atthe knee, with the left foot pointed straightdown. The choreographer’s costume consistsof sandals, white tunic, headband, and sash.The décor includes French patio doors, a Rus-sian floor carpet, and an Indian silk sari—truecosmopolitanism. Two other images depictFokine, flat-footed, holding a staff. In the firstof these, he holds the staff with his outstretchedright hand; in the second, he holds the staffwith both hands across the back of his neck.The prop takes on two distinct functions, onenarrative (it informs us that Daphnis is agoatherd) and the other practical (it provides ahorizontal axis for his sculptural poses).

In a fourth image, Vera lies atop the carpet,torso raised and turned, gazing serenely sky-ward. She has become two-dimensional, muchlike the coterie of nymphs in Nijinsky’s L’Après-midi d’un faune, the ballet that stole Daphniset Chloé’s Anacreontic and mythological thun-der during the 1912 season of the Ballets Russes.Fokine complained in his memoirs thatNijinsky borrowed from his technique withoutacknowledgment, but critics were divided as tothe significance of the overlap, and many be-lieved that Nijinsky was a much more radicalinnovator than Fokine. In L’Après-midi d’unfaune, as in Daphnis et Chloé, the corps deballet took on the guise of bas-reliefs, raisedfigures in white against dark backdrops. The

Made in different locations and perhaps atdifferent times, the sepia photographs in theFokine archive are the least familiar objects inthe Daphnis et Chloé curio cabinet. Unlike thelone stage photograph, these stills offer an ex-ternal rather than an internal perspective onthe ballet. File 13/6 of the archive containsthirteen photographs (with one duplicate) ofthe choreographer and his wife, the ballerinaVera Fokina, in Greek-style garb. The plain-ness of their outfits forms a radical contrast tothe colorful, childlike prints fashioned by Bakstfor the corps de ballet. File 13/7 contains elevenadditional photographs of Mikhail and Vera inthe roles (for the London staging) of Daphnisand Chloe. Of these, three are gathered in aletter-sized envelope dated 1914, the year of

Plate 2: Photograph of Mikhail and VeraFokine, State Theatre Library, St. Petersburg.

Reproduced by permission.




73


dance critic for the Pall Mall Gazette enthusedthat “M. Fokine’s choreography strikes a newnote. While rejecting M. Nijinsky’s revolution-ary aesthetic he has not disdained to profit byhis plastique.”81 The dance critic for the DailyMail appended that the chief pleasure ofDaphnis et Chloé was “the shepherds’ and shep-herdesses’ final dance of jubilation when thelovers are reunited, thanks to the interventionof Pan—a dance that seemed a miraculous vivi-fication of the garlanded figures of some Gre-cian urn.”82 The flattened perspectives in theballet furnished a view of reality—our hum-drum, quotidian reality—from the richer, mul-tidimensional realm of the Greek and Romandeities, of sprites and fairies, and (according tothe idealist philosophers) of music. If we couldlook at our world from the vantage point ofthis other realm, the place that the French Sym-bolists called the au delà and the Russian Sym-bolists realiora, it would appear colorless,depthless, and uniform.83

A fifth, much more beguiling image fromthe St. Petersburg archive likewise offers a flat-tened perspective (see plate 2). It shows Vera ondemi-pointe with her torso arched precariouslyback, arms stretching outward and upward, andhands cupped. Fokine grasps her forehead andchin, gently pressing them down even as shetries to reach up to break the frame of theimage. Her spouse, jutting his right leg out-ward and back and rotating his hips, seems tostand two-dimensionally. The image beguiles

insofar as it appears to express the individualwill of the dancer struggling against physicalconstraints, an idealized Romantic perceptionof movement that posits dance as wayward,semiotic force and choreography as the sym-bolic script that binds it. Though antitheticalto neoclassicism, the pose would have beensuitable for the episodes in the ballet whereChloe submits to the gathered brigands or ani-mates the accents of the reed pipe Syrinx.

Though these photographs can be related tothe scenario, they cannot be related to the mu-sic with much precision. The difficulty is notjust one of insufficient information. Except fortheir heated 1909 discussions about the sce-nario, Fokine and Ravel seldom consulted dur-ing the creative process. Diaghilev’s mean-spir-ited time pressures and Ravel’s last-minute de-cision to recompose the finale of the balletmeant that Fokine did not in fact hear theorchestral score until the night of the premiere.He developed the choreography using the pianoscore alone. For this reason, Daphnis’s andChloe’s solo variations, Darkon’s and Lycen-ion’s respectively comical and seductive char-acter dances, and the nymph and brigand en-sembles all evolved with reference to the meterand rhythm of the music but not to its texture.The omission is unfortunate, since Ravel paidspecial attention to texture in all of his musicfor dance.

Beginning with this point, the musicologistDanielle Cohen-Levinas argues that Ravel sub-limated physical gestures or “pulsations” intothe score of Daphnis et Chloé, thereby trans-forming the music into a simulacrum of a hy-pothetical choreography.84 The famous anec-dote about Ravel deriving musical inspirationfrom Nijinsky’s breathtaking leaps—in whichthe dancer either hovered in the air, as thoughunable to come down, or extended his bodysideways through space85—furnishes the subtext

81Unsigned review dated 10 June 1914, quoted in NestaMacDonald, Diaghilev Observed By Critics in Englandand the United States 1911–1929 (New York: Dance Hori-zons, 1975), p. 113.82R[ichard] C[apell], “‘Daphnis and Chloe’: Return of theRussian Ballet,” Daily Mail, 10 June 1914, p. 5.83Nijinsky, as noted, danced the role of Daphnis in thepremiere production of Daphnis et Chloé. By this time,however, the relationship between Nijinsky and Fokinehad deteriorated to the point that the former evidentlyignored the latter’s choreographic wishes. In an interviewwith a pseudonymous Russian theater critic, Fokine grievedthat, while the billboards showed that he had created theballet’s choreography, “in actuality, Nijinsky allowed him-self to mount a dance that was wholly at odds with theballet. Nijinsky devised some kind of puppet dance, intro-ducing wholly inappropriate stylization and contradictingthe manner of my staging” (Teatral’, “M. M. Fokinobvinyayet Nizhinskogo,” Peterburgskaya gazeta, 25 Au-gust 1913, p. 10).

84Cohen-Levinas, “Daphnis et Chloé ou la danse dusimulacra,” p. 90.85Michel-Dmitri Calvocoressi recalled: “The very first barsof music which Ravel wrote were inspired by the memoryof a wonderful leap sideways which Nijinsky (who was tobe Daphnis) used to perform in a pas seul in Le Pavillond’Armide” (Ravel Remembered, ed. Nichols [London:Faber, 1987], p. 187; quoted in Mawer, “Ballet and theApotheosis of the Dance,” p. 143).



74

19TH

CENTURYMUSIC

for Cohen-Levinas’s assertion that the “Danselégère et gracieuse de Daphnis” (rehearsal num-ber 43) translates physical action into acousticaction:

In a ternary 68 rhythm, Ravel integrates an ascendingpulsation in the pizzicato strings into two harpglissandi of sixteenth-note sextuplets. The rhythmis concentrated in full on the first beat of the mea-sure. Instead of “naturally” resolving on the second,it pauses “in the air,” suspended at a pedal point,leaving to the silence, the mirror of [acoustic] space,the charge to complete the descent of the instru-mental gesture.86

Cohen-Levinas’s claim that a “pedal point”—in fact, a fermata—underscores the rhythmicsuspension reflects a central feature of the score:just as Nijinsky’s leaps appeared to defy grav-ity, landing not on Earth but in another realmof experience, so too does the rhythmic activ-ity in Ravel’s score dash musical logic. Therhythmic “antecedent,” the rising sextupletpattern, finds its “consequent” not in a de-scending rhythmic pattern, but in a hollow out-line in the orchestration.87

The author touches on additional inter-changes between the temporal and spatial di-mensions of the score at rehearsal numbers 44,49, and 50. In these passages, as at number 43,rhythmic gestures are translated into hollowoutlines in the instrumentation. Interpreted asexpressions of dream logic, the outlines be-come the paradoxical markings of a hearablerather than a seeable choreography. Much inthe same way that dreams revive distant memo-ries, these passages revive those aspects ofdance—poetic Hellenic dance—that have beenlost to history. They represent the nocturnal,slumbering Greece described by Ravel in hisautobiographical sketch.

In this reading, the music of Daphnis et Chloébecomes a preparatory study for the process ofchoreographic sublimation that characterizes

Ravel’s La Valse (1920), which Diaghilev de-scribed as “a masterpiece” before noting that“it is not a ballet. . . . It is the portrait of a ballet. . . the painting of a ballet.”88 Cohen-Levinas’selegant and probing reading of Daphnis et Chloéassumes that Ravel privileged music over dance.This assumption is reasonable, but also prob-lematic, for it implies that the composer de-nied physical gestures historical being. By trans-forming bodies into symbols of negation, Ravel’sscore becomes a meditation on the historicalfragility of ballet, specifically the erasurethrough time of its visual facet. In this concep-tion, paradoxically, the first performance ofDaphnis et Chloé was already the performanceof a lost work.

Though choreography may not endure in his-tory in the manner that music, at least notatedmusic, does endure, Fokine stressed the perma-nent rather than the transient attributes of hiscraft. Placing his dancers in sandals or barefooton the stage, he rejected the ethereal Romanticballet tradition in favor of grounded earthiness.He privileged, in short, tactile, materialpresence. Studio and stage photographs of bal-let always exclude the aura or autographicplentitude of performance. The available im-ages of Daphnis et Chloé nonetheless manageto convey something of the substance ofFokine’s choreography because this choreogra-phy was in and of itself derived from images:black and white reproductions of sculptures,engravings, and bas-reliefs. The visual traces ofthe ballet embody, in effect, what the originalchoreography incarnated. In an essay aboutDuncan, André Levinson made the obviouspoint that pictures from a ballet—any ballet—provide little information about the choreogra-phy, since they show gestures in isolation, mostoften from the beginning or end of lengthy se-quences.89 Fokine subverted this principle bycreating a ballet of pictures. His choreographyembraced sculptural principles of balance andharmony. It highlighted the fixed over the fluid,the spatial over the temporal, and relied on

86Cohen-Levinas, “Daphnis et Chloé ou la danse dusimulacra,” pp. 92–93.87Cohen-Levinas’s remarks bring to mind those of JeanMarnold, who argued that the narrative of Daphnis etChloé was “transposed completely into the music”(“Daphnis et Chloé,” 1917, rpt. in a special Ravel editionof La Revue musicale 6:5 [1925], 101).

88Nichols, Ravel Remembered, p. 118; quoted in Mawer,“Ballet and the Apotheosis of the Dance,” p. 151.89André Levinson, “The Art and Meaning of IsadoraDuncan,” 1917, in What is Dance? Readings in Theoryand Criticism, p. 439.



75


natural laws: the curve of the spine, the align-ment of the arms and legs. “Except in the pi-rate and final scenes” of Daphnis et Chloé,another dance critic related, “there is not agreat deal of dancing, elastic poses and pos-tures—happily not of the Après-Midi d’unFaune type—playing a large part in the devel-opment of the action.”90

IVThe problem with Fokine’s (and Bakst’s) ap-

proach to Daphnis et Chloé was, once again, itsdisconnection from Ravel’s approach, and vice-versa. To various degrees, recent re-creationshave addressed the conceptual divide in theballet and in some instances have managed tobridge it. As noted at the outset of this article,each of these re-creations uses Ravel’s music,with or without the wordless chorus; each ofthem likewise abridges Fokine’s scenario andwhat little is known of his choreography.

One such Daphnis et Chloé was performedin 1982 by the Sydney Dance Company witherotic choreography by Graeme Murphy.91 Para-phrases of Fokine’s asymmetrical progressionsabound in the staging, as do paraphrases ofNijinsky’s flattened perspectives, with the ap-parent aim of showing the historical influenceof the former on the latter. Murphy allows themagical figures in Fokine’s scenario to travel inand out of the visual frame of reference byfantastic means. Pan descends and ascends bycloud machine; the leather-clad brigands skate-board from stage right to left as though dashingto catch the subway trains whose clatter can beheard in the distance between scenes; thenymphs travel by roller blades and on pointe.The dancing makes reference to dancingthroughout the ages. Murphy’s lexicon expandsto include elements of ballroom, folk, and jazz.The playful mixture of these genres enhancesthe ballet’s qualities of unreality.92

The production is thus a stylistic and ge-

neric mélange, but in Chloe’s opening varia-tion it does assume an attitudinal relationshipto what we know of the 1912 and 1914 chore-ography. Performed by Victoria Taylor, thevariation forms a rapid-fire series of multifac-eted interactions between music and gesture.The initial envelopées, the interceding hoppingpas de bourrées, and the following steps inattitude are executed at twice the pulse rate ofthe music. Upon moving into flexed-foot pir-ouettes, Taylor slows to the pulse of the or-chestra, then returns to her envelopées, whichbegin to flow in- rather than out-of-sync withthe music, the pace of which now equals hers.In the subsequent chassés, the dance and mu-sic enter into a new dialogue. Taylor moves herlegs in lockstep with the accompaniment, butholds her arms firmly above her head in alongéeduring a series of leaps. These are followed byfouettés, in which the tempo of Taylor’s portde bras differs markedly from that of her steps.The misalignment is somewhat unusual, forthe arm and leg actions in this pattern shouldbe closely related; in other words, the incon-gruence between her upper and lower bodymovements extends beyond their anatomicallynatural incongruence. Taylor’s upper bodyseems to be divorced from her lower body. Herarms and legs move in and out of sync with oneanother just as the dance and music did a fewmeasures earlier. The sight-sound dialogue in-volves asyndeton, digression, and rhetoricalflourish.

Toward the end of the variation, the tempoof the dance escapes that of the music again asTaylor performs a series of abbreviated jetédéveloppé jumps followed by step-over turns inplace. Her arms and legs begin actions of daz-zling height and speed, but these are not ac-complished, leaving the audience with the im-pression of a marionette tangled in strings. Cru-cially, the routine comprises a scripted ratherthan an improvised marking of another dance,a dance implied by the music but devoid ofstatus beyond the music. The stunted prances,limp arms, and flexed feet suggest a ballerinapracticing rather than performing a variation.The half-completed leaps and turns likewisesuggest the presence of another set of gesturesexisting in another time and space. Their do-main is neither the island of Lesbos nor the

90Unsigned, “Beecham Opera: First Performance of the Bal-let,” Standard, 10 June 1914, p. 5.91Maurice Ravel’s Daphnis and Chloe, prod. CharlesThompson, dir. Jolyon Wimhurst, 60 min., Home Vision,1982, videocassette.92For this point, and the technical points in the followingtwo paragraphs, I am very grateful to Emily Abruzzo.



76

19TH

CENTURYMUSIC

locale of the Delphic oracle. Rather, it is theGreece of Ravel’s “dreams.”

Like Murphy’s, Frederick Ashton’s 1951 re-creation of Daphnis et Chloé for the Sadler’sWells Ballet aimed to give a contemporary feelto the title characters and the gods who regu-late their world. His production enhanced thetepid eroticism of Fokine’s scenario through astronger focus on Daphnis’s entanglement withLycenion in the first scene and intense physi-cal contact in the scene 3 pas de deux. JulieKavanagh points out that in the Adagio of thepas de deux, Margot Fonteyn, as Chloe, “slowlybourrée[d]” to her partner “through a frieze ofcorps”; then, in a repeated, ecstatic cascade,she put her arms around his neck and began towhirl, eventually “flying around him horizon-tally.”93 A further strength of Ashton’s produc-tion was Chloe’s animation, in scene 3, of theaccents of the reed pipe Syrinx. Fonteyn per-formed a poignant variation that had less incommon with neoclassicism than with Roman-ticism in its gestures and that translated Ravel’smusic into dance.94 The variation flowed intoand out of static poses, the port de bras dissolv-ing into fluttering gestures approximating theflute line’s fleeting, floating ornamentation.95

Fonteyn’s lyrical movements filled the middleground between the conflicting musical andchoreographic conceptions of Daphnis et Chloé.

In effect, to be fully realized, Ravel andFokine’s ballet had to wait for the dissolutionof the Ballets Russes and for different compa-nies with different performers to emerge.Though the ballet was conceived in 1907 andpremiered in 1912, it has only recentlycome into being.

Abstract.Beyond Maurice Ravel’s 1910 score, the remnants ofthe original production of Daphnis et Chloé—oneknown stage photograph, an assortment of studiophotographs, seven known costumes, brief reviews,anecdotal memoirs, and a bundle of pastel draw-ings— constitute choreographer Michel Fokine’s 1907scenario. These materials are scattered across theglobe, preserved in libraries and museums in Russia,Sweden, France, England, and the United States. Theycompose less a ballet, even the archival detritus of aballet, than a haunting absence. This article as-sembles all of these materials in an assessment ofthe differences between Ravel’s and Fokine’s con-ceptions of Hellenic antiquity. The discussion fo-cuses on the draft and revised versions of the literaryscenario and the draft and revised versions of thefinale of the score, cast in 34 and 54 meter respectively.The ending of the article offers brief remarks on themusic-dance relationship in the original and twosubsequent productions of the ballet.

l

knowing that it’s impossible [to] get it back. It begins withtrying to unwind the hands, [to] unbind yourself, but notwith any hope that it will ever happen. It’s just an auto-matic reaction” (Stephanie Jordan, “Antoinette Sibley andMichael Somes on Daphnis and Chloe,” Ballet Review 27[1999], 23).

93Julie Kavanagh, Secret Muses: The Life of FrederickAshton (London: Faber & Faber, 1996), p. 366. Informationin the preceding sentences comes from pp. 365–67.94Archival footage of her performance is scarce; part of a1958 rehearsal of this episode, however, is included onMargot Fonteyn, a tribute to the dancer produced and broad-cast by the BBC in 1991.95The same trait is evident in Ashton’s choreography ofthe “Danse suppliante,” which Antoinette Sibley describedas “yearning, begging, longing for what you did have and

