¹ This article was presented at ‘Dentro l'opera: Livelli di lingua e di stile nel melodramma’, a conference held in September 1989 at the Fondazione Giorgio Cini in Venice. I am grateful to Gianfranco Folena, Giovanni Morelli and Maria Teresa Muraro for permission to publish an expanded version in English.

² Petrobelli, Pierluigi, ‘Music in the Theatre (à propos of Aida, Act III)’, in Redmond, James, ed., Themes in Drama 3: Drama, Dance and Music (Cambridge, 1980), 129–42Google Scholar, here 129. Petrobelli refers to Noske, Fritz, The Signifier and the Signified: Studies in the Operas of Mozart and Verdi (The Hague, 1977CrossRefGoogle Scholar; rpt. Oxford, 1990), especially to Appendix 1, ‘Semiotic Devices in Musical Drama’.

³ For a somewhat more rigorous attempt to formalise these ideas, see my ‘Affetto e azione: Sulla teoria del melodramma italiano dell'Ottocento’, Atti del XIV congresso della Società internazionale di musicologia (Bologna, 27 agosto-1 settembre 1987). Trasmissione e recezione delle forme di cultura musicale, III: Free Papers (Turin, 1990), 395–400.Google Scholar

⁴ Mila, Massimo, L'arte di Verdi (Turin, 1980), 17.Google Scholar

⁵ It is true, as Mila asserts, that ‘the basic contrast of the action…is not so much that of individual passions as of peoples and religions. Two peoples are in combat, the oppressor and the defeated, the Assyrians and the Hebrews, and through the choral masses they speak a language full of dignity’ (p. 16). But the examples he cites pertain to the Hebrews, and a little further on he observes that ‘in general the choral writing for the Assyrian masses is less controlled and appropriate than that for the Hebrew chorus’ (p. 17).

⁶ See esp. Chusid, Martin, ‘Rigoletto and Monterone: A Study in Musical Dramaturgy’, in Verdi: Bollettino dell'Istituto di studi verdiani, III. 9 (Parma, 1982), 1544–58.Google Scholar On the variety of stylistic levels in Verdi's operas, and in Rigoletto in particular, see Weiss, Piero, ‘Verdi and the Fusion of Genres’, Journal of the American Musicological Society, 35 (1982), 138–56.CrossRefGoogle Scholar

⁷ The most representative example of this tendency is the discussion of the opera (reissued in 1958) in Mila, (see n. 4), 187–200Google Scholar, where – it seems to me – many still useful observations on the function of exoticism contained in the original 1933 book (see 53f. and 77f.) tend to be weakened. For a similar position, to which I shall return, see Casini, Claudio, Verdi (Milan, 1981), 301–7.Google Scholar

⁸ A convincing attempt in this sense, though limited to outlining the problem, is in Pinagli's, PalmiroRomanticismo di Verdi (Florence, 1967), 145–52Google Scholar, which proceeds from a gentle criticism of Mila: ‘before ignoring those episodes as purely decorative and extraneous to the central inspiration of Aida, one should perhaps seek out what new essentials the composer might have pursued, or – more simply – what function these episodes might have assumed in the structural framework of the drama’ (p. 146).

⁹ Given the wide-spread familiarity with the opera, it seems unnecessary to specify each reference to the score. I follow the newly revised orchestral score, Ricordi P. R. 153 (1980). The libretto is cited from Baldacci, Luigi, Tutti i libretti di Verdi (Milan, 1975), 449–71.Google Scholar

¹⁰ The characters of Ramfis and Amonasro are also symmetrical by virtue of Ramfis's vicarious paternal function towards Amneris (and also Radamès). Compare the variant, subsequently discarded, contained in an ‘outline’ sent by Verdi to Ghislanzoni, in which Amneris directly asks the king her father to pardon Radamès. See Carteggi verdiani, ed. Luzio, Alessandro, IV (Rome, 1947), 19.Google Scholar

¹¹ In the Act II Finale Radamès seems for a moment to enter into conflict with Ramfis; on this occasion he does not act as an autonomous subject, but rather as a means through which Aida – and thus indirectly Amonasro – operates.

¹² This hierarchy is slightly different from that outlined by Petrobelli, Pierluigi, who placed Aida at the lowest level, ‘Un conflitto tra individuo e potere’, in Aida di Giuseppe Verdi, Programme Guide, Teatro Comunale di Bologna, 1981–1982 season, 9–15.Google Scholar The difference probably derives from the fact that Petrobelli gives prominence to Aida's social position, whereas I emphasise her function in the dramatic process, at which level she undeniably tends to dominate Radamès. Whatever the case, Petrobelli's essay is important in underscoring the ‘political’ content of the opera. The dramatic significance of the thematic opposition in the prelude, in which Aida's love theme does not represent one individual so much as the value of the individual as opposed to that of power, remains incontestable. See the suggestion by Hussey, Dyneley, Verdi, rev. ed. (London, 1963), 192.Google Scholar

¹³ This parallelism is also symptomatic of the two rivals' community in defeat, of which Amneris is faintly conscious in an aside commenting on the distress of Aida, whom she is none the less preparing to crush: ‘Ah, quel pallore…quel turbamento / Svelan l'arcan – febbre d'amor…/ D'interrogarla – quasi ho sgomento…/ Divido l'ansie – del suo terror’ (Ah, that pallor, that turmoil reveals the mysterious fever of love; I am almost afraid to question her; I share the anxiety of her terror).

¹⁴ The basic stages in the formation of the libretto are Auguste Mariette's scenario in French, published facing the Italian version by Giuseppe and Giuseppina Verdi in Jean Humbert, ‘A propos de l'égyptomanie dans l'oeuvre de Verdi: Attribution à Auguste Mariette d'un scénario anonyme de l'opéra Aida’, Revue de musicologie, 62 (1976), 228–55Google Scholar; and the synopsis of scenes in French prose by Camille Du Locle and Verdi, currently available only in English translation in Busch, Hans, Verdi's Aida. The History of an Opera in Letters and Documents (Minneapolis, 1978), 448–71.Google Scholar Other documents concerning the genesis of the opera are in Cesari, Gaetano and Luzio, Alessandro, eds., I copialettere di Giuseppe Verdi (Milan, 1913), 635–82Google Scholar; Abbiati, Franco, Giuseppe Verdi, III (Milan, 1959), 336ff.Google Scholar; Abdoun, Saleh, ed., ‘Genesi dell' Aida, con documentazione inedita’, Quaderni dell' Istituto di studi verdiani, 4 (Parma, 1971)Google Scholar; Günther, Ursula, ‘Zur Entstehung von Verdis Aida’, Studi musicali, 2 (1973), 15–71.Google Scholar All the documents, with the addition of many unpublished ones in English only, are collected in the volume by Busch. The most accessible scholarly synthesis is Budden, Julian, The Operas of Verdi. III: From ‘Don Carlos’ to ‘Falstaff’ (London, 1981), 161–259Google Scholar, which also provides excerpts from the Disposizione scenica per l'opera ‘Aida’ (Milan, 1872).Google Scholar

¹⁵ For a perceptive comparison of the two versions, see Ferrero, Mercedes Viale, ‘Scene e costumi di Aida al Cairo (1871) e a Milano (1872)’, in ‘Aida’ al Cairo, Banca nazionale del lavoro (Milan, 1982), 139–44.Google Scholar According to the author, ‘the modifications added to the scenario of Aida in Milan were dictated by theatrical and psychological concerns, aimed at increasing the expressive force and the dramatic coherence of the scenes, so that at Milan the historical-archaeological apparatus became a means of communication and not – as in Cairo – its goal’ (p. 140).

¹⁶ Ferrero, Viale, 143.Google Scholar

¹⁷ Humbert, (see n. 14), 252.Google Scholar

¹⁸ See Busch, (n. 14), 463Google Scholar, 467 and, here, 468, the latter available in the original in the Carteggi verdiani (see n. 10), IV, 13.Google Scholar

¹⁹ ‘The allusion to the enslavement of the Ethiopian Aida generally passes unnoticed’ – Pirrotta, Nino, ‘Semiramis e Amneris, un anagramma o quasi’, in Il melodramma italiano dell'Ottocento. Studi e ricerche per Massimo Mila (Turin, 1977), 5–12, here 7.Google Scholar

²⁰ The contrast between ‘two distinct exotic ambiences’, the ‘pompous ambience of the Egyptian court, with its intrigues and its martial and religious ceremonies’, which ‘is presented in the form of brilliant and resounding kitsch, in contrast with the nostalgic evocation of a distant and mysterious region, the Ethiopia of “foreste imbalsamate” and “azzurri cieli”, which is the real exotic, imaginary and invented locus of the opera’, has been noticed by Casini (see n. 7), who also underscores the connection between this place and ‘another place, no less imaginary – the realm beyond the grave towards which Radamès and Aida move together’. This insight, however, is vitiated by Casini's reductive reading, according to which the exoticism of Aida, whose spectacular and intimate moments alternate through the opera, results in something completely separate from what he designates as the ‘dramaturgy’ (which turns out to be nothing other than the psychological characterisation of the dramatis personae). From this preconception there follow judgments on the ‘elementary dramatic conception’, the ‘irremediable break between the dramaturgical undertaking, of a linearity that recalls eighteenth-century opera seria and its rigid conflicts, and a musical language that expresses itself with the virtuosity, and also the morbidity of decadence’ – whence the paradox of a ‘separation in which the author is neither dramatist nor – in his usual dimension – musician’ (p. 306), as if for an operatic composer the two were not identical.

²¹ For a historical survey of the problem, see Maehder, Jürgen, ‘Die musikalische Realisierung altägyptischen Lokalcolorits in Verdis Aida’, Programmheft der Bayerischen Staatsoper (Munich, 1979), 54–66Google Scholar; and, more generally, Wolf, Hellmuth Christian, ‘Der Orient in der französischen Oper des 19. Jahrhunderts’, in Die ‘couleur locale’ in der Oper des 19. Jahrhunderts, ed. Becker, Heinz, Studien zur Musikgeschichte des 19. Jahrhunderts, 42 (Regensburg, 1976), 371–85.Google Scholar

²² The use Verdi makes of these characteristics has been discussed on several occasions by Wolfgang Osthoff; see ‘Musikalische Züge der Gilda in Verdis Rigoletto’, Verdi: Bollettino dell'Istituto di studi verdiani, III. 8 (Parma, 1973), 950–79, esp. 956–72Google Scholar; and Il Sonetto nel Falstaff di Verdi’, in Il melodramma italiano dell'Ottocento (see n. 19), 157–83Google Scholar, esp. 179. It is a pleasure to recall a statement of Stravinsky, who, asked in 1967 to compose a short piece to be used in association with an eye painted by Picasso as a logo for a new colour channel on English television, said, ‘an eye means transparency, and consequently the sound should be produced by very high instruments, possibly flutes, compared with which oboes are fat and clarinets oily’. The anecdote, taken from an article by Craft, Robert (1969)Google Scholar, is cited by Vlad, Roman, Stravinsky, 2nd ed. (Turin, 1973), 349.Google Scholar

²³ Bellaigue, Camille noted this early on, in Verdi (Paris, 1912), 71Google Scholar: ‘The flutes in particular give certain scenes of Aida, that on the banks of the Nile, those of the temple, an oriental and sacred colour’.

²⁴ See the correspondence between Verdi, and Ricordi, from 10 1871 in Busch (n. 14), 238–46Google Scholar, where it seems that what Verdi wanted from such an instrument was not loudness but fullness of sound, to obtain which it would have been necessary to increase the number of flutes to ten or twelve.

²⁵ Thomas Mann introduces the sombre sound of the flute in Aschenbach's dream in chapter 5 of Death in Venice with just this ritual characteristic, and with orgiastic connotations.

²⁶ Gino Roncaglia observes that in the last scene of the opera ‘Aida's love is no longer that of the first scene, and her first theme no longer needs to exist. It represents earthly love, in which Ethiopia was vanquished; but now Aida is finally united with Radamès in death; Amneris no longer exists; death for love makes a victor of Aida’. ‘Il tema cardine’, in his Galleria verdiana (Milan, 1959), 48.Google Scholar

²⁷ Letter of 5 August 1871, from Verdi to Ghislanzoni, , in I copialettere (see n. 14), 674.Google Scholar

²⁸ References to Aida's father and brothers, i.e., to the more immediate sentimental and patriotic motivation of her nostalgia, which appeared in Verdi's own draft in the letter cited in n. 24 (p. 675), were subsequently deleted. An analogous case of a late addition that changes the meaning of an already planned opera has been examined by Osthoff, Wolfgang, ‘Il Sonetto’ (see n. 22).Google Scholar

²⁹ Budden (see n. 14), for example, sees in this theme an ‘Ethiopian contour’ (236), whose history he traces from ‘Il tuo bel cielo vorrei ridarti’ (203) through ‘I sacri nomi di padre, d'amante’ (209f.). Without precluding the possibility of a cumulative significance, it seems to me that above all in the last instance the melodic revolving within the space of a few notes is intended rather to suggest an unsuccessful attempt to find a means of escape.

³⁰ Doubtless with the significance of ‘open space’; but there is also an analogous pedal that runs through the sacred dance of Act I.

³¹ Regarding this passage Guido Paduano notes that ‘the repetition and functional opposition of pitches can convey other oppositions corresponding to the basic conflicts of the action: the relationship between father and daughter, between an authoritarian subjugation and a repressed passion, between a future occupation and an antiphrastic certainty of defeat’. Noi facemmo ambedue un sogno strano: Il disagio amoroso sulla scena dell'opera europea (Palermo, 1982), 17.Google Scholar On the function of timbre, see my ‘Verdi’ in the Dizionario universale della musica e dei musicisti, ed. Basso, Alberto, Le biografie, VIII (Turin, 1988), 204.Google Scholar

³² See Gerhard, Anselm, ‘L'eroe titubante e il finale aperto. Un dilemma insolubile nel Guillaume Tell di Rossini’, Rivista italiana di musicologia, 19 (1984), 113–30, esp. 113f.Google Scholar

³³ See n. 29. Toye, Francis, Giuseppe Verdi. His Life and Works (1931; rpt. London, 1962), 403Google Scholar, praises ‘the subtle manner in which the low register of the flutes is used to suggest the tropical fragrance of Ethiopia’.

³⁴ Verdi's insistence in the letters to Ghislanzoni, from 8 to 22 10 1870Google Scholar on the antithesis between the positions of Radamès and Aida is especially interesting, for example that between ‘Lasciar la patria, i miei Dei, i luoghi ove nacqui, ove acquistai la gloria’ (To leave my homeland, my Gods, the places where I was born, where I attained glory) and ‘La patria è dove si ama’ (The homeland is where one loves). I copialettere (see n. 14), 653.Google Scholar Shortly after Verdi found that ‘“Il ciel de’ nostri amori / come scordar potrem” è felice assai, assai, assai’ (‘How could we forget the skies of our love’ is very, very, very apt), whereas he was displeased that the librettist ‘non abbia conservato: “L'are de’ nostri Dei”, colla risposta d'Aida: “…nel tempio stesso / Gli stessi numi avrem”’ (did not retain ‘the altars of our gods’ with Aida's response, ‘in the same temple we shall have the same deities’) (p. 654f.). On the correct dating of these letters, see Gossett, Philip, ‘Verdi, Ghislanzoni, and Aida: The Uses of Convention’, Critical Inquiry, 1 (1974), 291–334, esp. 298.CrossRefGoogle Scholar

³⁵ The perplexity concerning this cabaletta goes back at least as far as the review of Filippo Filippi in La perseveranza, a large part of which is reprinted in Verdi intimo. Carteggio di Giuseppe Verdi con it conte Opprandino Arrivabene (1861–1886), ed. Alberti, Annibale (Verona, 1931), 138–43Google Scholar, esp. 140: ‘it is a real cabaletta cast in the old mould, constructed above a high, spasmodic motive that Verdi has tried to cover with some harmonic turns, but without succeeding in making it please’. For a partially positive evaluation of this passage, see Mila, (n. 4), 198.Google Scholar

³⁶ Verdi wanted to remove all indication of free will in the betrayal of Radamès so that it would appear less ‘odious’ – thereby increasing the character's dramatic passivity. See the letter to Locle, Du of 15 07 1870, Carteggi verdiani (n. 10), IV, 15.Google Scholar I cannot agree with Budden's statement that the scene confirms ‘Verdi's bias towards a world of masculine values’ and ‘Aida reveals herself for the mere pawn that she is’, while ‘the two men dominate the picture, musically and dramatically’. The Operas of Verdi (see n. 14), 244.Google Scholar

³⁷ Ghislanzoni was anxious to eliminate Radamès' ‘Godiamo un istante di felicità’ (Let us enjoy one moment of happiness), because it might ‘occasion an erotic interpretation, which certainly would not correspond to the intentions of the author’. Letter of 31 October 1870 in Abbiati, (see n. 14), 402.Google Scholar This did not prevent Thomas Mann from perceiving that ‘The condemned one protested, quite properly, against the sacrifice of the precious life; but in his lender, despairing, “No, no, troppo sei bella” was the intoxication of final union with her whom he had thought never to see again. It needed no effort of imagination to enable Hans Castorp to feel with Radamès all this intoxication and gratitude.’ The Magic Mountain (Chapter VII: ‘Fullness of Harmony’), trans. Lowe-Porter, H. T. (1944; rpt. New York, 1953), 645.Google Scholar

³⁸ The second attribute comes from the Disposizione scenica. See Ferrero, Viale (n. 15), 143.Google Scholar

³⁹ Ferrero, Viale, 143.Google Scholar

⁴⁰ Some of the devices employed in this passage, especially the rapid violin tremolo and the flute arpeggios, were reused by Verdi in the ‘Lux aeterna’ of the Requiem.

⁴¹ Mila, (see n. 4), 199Google Scholar, but see also Pinagli, (n. 8), 151f.Google Scholar The most explicit interpretation of the Finale and that of La forza del destino in transcendent terms is by Loschelder, J., Das Todesproblem in Verdis Opernschaffen (Cologne, 1938), 56f.Google Scholar

⁴² According to Du Locle, the scenic division in two levels was expressly asked for by Verdi. See the Carteggi verdiani (n. 10), IV, 5.Google Scholar

⁴³ Petrobelli (see n. 12) and Viale Ferrero (see n. 15) emphasise this reading; compare also Marchesi, Gustavo, Giuseppe Verdi (Turin, 1970), 425.Google Scholar

⁴⁴ Letter of 8 September 1870, I copialettere (n. 14), 644.Google Scholar

⁴⁵ Letter of 22 August 1870, I copialettere, 642.Google Scholar

⁴⁶ Letter to Ghislanzoni, , 7 10 1870, I copialettere, 650.Google Scholar

⁴⁷ This, incidentally, casts doubt on readings of Aida that are too fashionable, those that stress imperialistic expansionism (to which Budden [see n.14], 258, alludes), as well as those in an anticolonial vein (which Ferrero, Viale [see n. 15], 144Google Scholar, cautiously suggests). Casini, (see n. 7), 304Google Scholar, ventures to see in the betrayal of Radamès ‘a just retaliation for Egyptian oppression of the vanquished Ethiopian people’.

⁴⁸ See Pinagli, (n. 8), 145.Google Scholar