¹ I wish to thank Buford Norman, David Schulenberg and Cynthia Verba for helpful comments on the oral version of this essay.Google Scholar

² For instance, the introductory monologue in ternary form, or the binary-form ‘maxim air’ sung by a confidant, ending with an aphoristic couplet. Though somewhat dated, Masson's, Paul-MarieL'Opéra de Rameau (Paris, 1930; rpt. New York, 1972) remains a useful introduction to the style.Google Scholar

³ Throughout this study, scene numbering and punctuation follow the edition of the libretto in Recueil général des opera, XV (Paris, 1739), 318–80; facs. edn (Geneva, 1971), 471–87.Google Scholar Regarding versions and editions of the score and libretto, see the introduction and critical notes by Malherbe, Charles to the modern edition, ed. d'Indy, Vincent, in Rameau, Jean-Philippe, Œuvres complètes, VI (Paris, 1900; rpt. New York, 1968);Google ScholarMassip, Catherine, Rameau et l'édition de ses œuvres: bref aperçu historique et méthodologiqué, in Jean-Philippe Rameau: Colloque international organisé par la Société Rameau. Dijon, 21–24 septembre 1983, ed. Gorce, Jérôme de La (Paris and Geneva, 1987), 145–57;Google ScholarSadler, Graham, ‘Rameau, Pellegrin, and the Opéra: the revisions of “Hippolyte et Aricie” during its first season’, The Musical Times, 124 (1983), 533–7;CrossRefGoogle Scholar and Sadler, Graham, ‘Vincent d'Indy and the Rameau œuvres complètes: a case of forgery?’ Early Music 21 (1993), 415–21.CrossRefGoogle Scholar

⁴ The first pair of lines is stated once; the latter pair is stated twice, the first time with an inconclusive ending (in this case a half-cadence) and the second time with full closure.Google Scholar

⁵ Literally, ‘air in tempo’. The term, used in the eighteenth century to refer to operatic airs, had been introduced in the seventeenth century to distinguish a metrically regular song from the metrically shifting type of air de cour that was popular at the time (the latter being an important precursor of récitatif).Google Scholar

⁶ In any event, Buford Norman mentions (in an unpublished remark) that variants in punctuation from source to source abound in livrets of the period.Google Scholar

⁷ See Schérer, Jacques, La Dramaturgie classique en France (Paris, n.d.), 214–84.Google Scholar

⁸ For instance (in a prescription for improving the tragédie en musique, ‘[…] preserve the Scènes […], while cutting all that is not part of the action of the Poem; add in a consistent style the [instrumental?] airs, the songs, and the. fêtes […]’ Blainville, Charles-Henri de, L'esprit de l'art musical(Geneva, 1754), 111–12. While the term ‘divertissement (or ‘fête’ ), in the sense of the obligatory episode for collective characters within each act of a French Baroque opera, is defined in only general terms by eighteenth-century writers, the distinction I mean to draw here is structural: the divertissement is the scene, or portion of a scene, in which the librettist provides for dancing (in Rameau's day, generally with the explicit instruction ‘on danse’ ); it might contain poetry that differs in structure from that of the rest of the drama (e.g. Lully's strophic dance songs). The resumption of dialogue by principal characters marks the end of the divertissement. In Rameau's operas there is a sharp distinction between continuous tonal motion in the scènes and tonal stasis in the divertissements. By this definition, spectacular events such as tempests and sommeils, though associated with divertissements, are not restricted to them.Google Scholar

⁹ See Rosow, Lois, ‘The articulation of Lully's dramatic dialogue’, Lully Studies (Cambridge University Press, in press);Google Scholar and Legrand, Raphaëlle, ‘Ricercar sopra fa la mi: préludes de basse et articulation des scènes dans l'opera français de Lully à Rameau’, D'un Opéra l'Autre: Hommage à Jean Mongrédien, ed. Gribenski, Jean, Mussat, Marie-Claire and Schneider, Herbert (Paris, 1996), 209–15. Legrand points out (p. 214) that of all the composers of French opera from Lully to Rameau, only Marc-Antoine Charpentier — in Médée, his one work performed at the Paris Opéra — combined the function of the modulatory gesture and the entrance prelude by writing preludes that modulate.Google Scholar

¹⁰ This feature will be exemplified by Act II scene 2, to be discussed below.Google Scholar

¹¹ For an unannounced entrance in an opera by Lully, also following a monologue but one without postlude, see Act II scene 2 of Amadis; Lully, J.-B., Œuvres complètes, Les Opéra, ed. Prunières, Henry, III (Paris, 1939; rpt. New York, 1966), 80.Google Scholar

¹² In an unpublished paper, ‘What recitatives owe to the airs: a look at the dialogue scene, Act I, scene 2, of Rameau's Hippolyte et Aricie — version with airs’, Cynthia Verba offers an interesting discussion of the complementary harmonic behaviour of recitative and airs in this scene. She correctly points out that in a scene that begins with modulatory recitative, as does this one, it might well be the first air that clearly establishes a key. On the other hand, in many scenes Rameau does provide his character with a conventional entrance prelude; in those cases it is the prelude that establishes the initial key of the scene. I wish to thank Dr. Verba for sharing her paper with me.Google Scholar

¹³ The libretto is taken from Recueil général (see n. 3), XV, 343–5. The excerpt from the engraved score in Appendix 1 — Hippolyte et Aricie, tragédie (Paris: l'Hauteur, 1733), 46–52 — comes from the copy in Chicago, Newberry Library, Case VM1500.R17h.Google Scholar A rough guide to interpreting the changing meters in the recitative is C= 3= 2. On this topic in general, see Rosow, Lois, ‘The metrical notation of Lully's recitative’, in Jean-Baptiste Lully: Actes du colloque/Kongressbericht: Saint-Germain-en-Laye-Heidelberg 1987, ed. Schneider, Herbert and Gorce, Jérôme de La (Laaber, 1990), 405–22.Google Scholar

¹⁴ The associations of key area with affect are of course Rameau's, from his well-known analysis of passages from Lully's, Jean-BaptisteArmide, in Observations sur notre instinct pour la musique (Paris, 1754; facs. edn, New York, 1967), 52–8, 78–82.Google Scholar

¹⁵ Pellegrin was fond of this sort of compression. The first four lines of Thésée's ‘Puissant maître des flots’ (Act III scene 7) are an alexandrine, followed by two octosyllables, followed by a hexasyllable; cited in Peyronnet, Pierre and Landy, Rémy, ‘Dramaturgie textuelle et dramaturgie musicale’, in L'Opéra au XVIII^e siècle: Actes du Colloque organisé à Aix-en-Pnvence … 1977 (Aix-en-Provence, 1982), 387.Google Scholar

¹⁶ Marmontel, Jean-François, Poétique française (Paris, 1763), II, 94.Google Scholar

¹⁷ As Appendix 1 shows, only the second is actually labelled ‘air’ in the score engraved under Rameau's direction in the year of the premiére. In ‘Towards an Understanding of the Dialogue Air in the Style of Jean-Philippe Rameau’ (M.A. thesis, Ohio State University, 1994), Margaret Butler argues that Rameau used genre labels as directions to performers: when an air grows relatively imperceptibly out of the preceding recitative, Rameau tends to include a label; when the switch from recitative to air is immediately self-evident from the score, the label might be omitted. Regarding the eighteenth-century aesthetician Pierre Estève's curious identification of ‘Sous les drapeaux de Mars’ (and other such songs) as ‘récitatif mesuré’,Google Scholar see Dill, Charles, ‘Eighteenth-century models of French recitative’, Journal of the Royal Musical Association, 120 (1995), 242–4.CrossRefGoogle ScholarDill's thesis (pp. 248–50) is that in the eighteenth century the critical language for discussing the music of French opera was poetic and dramatic in conception.Google Scholar

¹⁸ In the production score, Paris, Bibliothèque de l'Opéra, Rés. A.128.a, p. 50, the composer wrote, ‘Recomandes au Bassons de nourir les Sons, sans les détacher, et faites chanter de mesure, on me la promi’. As Appendix 1 shows, the engraved score of 1733 contains just one bassoon part, not two as in d'Indy's edition.Google Scholar

¹⁹ Graham Sadler points out (personal communication) that the change from E major to C minor is particularly extreme for Rameau.Google Scholar

²⁰ The word is Charles Dill's; he bases his observations on Rameau's theoretical writings. Dill, Charles, ‘Rameau reading Lully: meaning and system in Rameau's recitative tradition’, this journal, 6 (1994), 1–17.Google Scholar

²¹ Dill, Charles, ‘The reception of Rameau's “Castor et Pollux” in 1737 and 1754’, Ph.D. diss., Princeton University (1989), 164–96; Rosow, ‘The articulation of Lully's dramatic dialogue’ (see n. 9); Verba, ‘What recitatives owe to the airs’ (see n. 12).Google Scholar

²² See Norman's, Buford paper, ‘Remaking a cultural icon: Phédre and the operatic stage’, in this issue.Google Scholar

²³ Stressed by Rameau himself. See Christensen, Thomas, Rameau and Musical Thought in the Enlightenment (Cambridge, 1993), 113–23.Google Scholar

²⁴ Howarth, William D., ‘L'alexandrin classique comme instrument du dialogue théatral’, in Dramaturgies: langages dramatiques (Paris, 1986), 349–53.Google Scholar

²⁵ Michel Bernard, ‘Esquisse d'une théorie de la théatralité d'un texte en vers à partir de l'exemple Racinien’, ibid., 285. For a discussion of different types of French rhyme, see Barzun, Jacques, An Essay on French Verse for Readers of English Poetry (New York, 1991), 25–43.Google Scholar

²⁶ I have transcribed them from Le Mercure de France (September 1742), p. 1073.Google Scholar

²⁷ Rameau, , Œuvres complètes, VI, 57–8. This passage is discussed at length in Buder, ‘Towards an Understanding’ (see n. 17), 44–8.Google Scholar

²⁸ Rameau's assertion that Lully ‘thought on a large-scale’ (i.e., had a large-scale sense of tonal hierarchy) appears in Observations (see n. 14), 78.Google Scholar