4. Collaboration with Calzabigi.

A decisive event for Gluck's future was the arrival in Vienna early in 1761 of the Tuscan poet Ranieri Calzabigi. A relatively minor literary figure with little practical experience as a librettist, Calzabigi had nevertheless, while in Paris during the 1750s, edited (with the author's cooperation) a prestigious complete edition of the works of Metastasio, which he prefaced with a ‘Dissertazione’ mixing high praise with subtle criticisms (his own, and those of French critics). He was probably also the author of a more direct attack on the imperial poet's operatic system, in an anonymously published Lettre sur le méchanisme de l'opéra italien (Paris, 1756; see Heartz, C1995), which proposed a fusion of the best features of French and Italian serious opera. While his employment was in the Netherlands Finance Ministry, Calzabigi quickly came into the orbit of Kaunitz and his protégé Durazzo, through whom he presumably was introduced to Gluck.

The first product of their collaboration was not an opera but a pantomime ballet: Don Juan, ou Le festin de pierre, to choreography by Angiolini, first given on 17 October 1761. Calzabigi was probably considered useful for the project on account of his familiarity with Parisian debates over the nature and purpose of theatrical dance, and with practical innovations in dance in Parisian theatres. Just as significant was his strong classical orientation (evident from the time of his first published writing, an explication of two Etruscan inscriptions), and in fact the ballet was one of the first attempts since ancient times to stage a complete theatrical action using pantomime and dance alone. The programme essay in French for Don Juan was signed by Angiolini, but Calzabigi later claimed authorship, and it is likely that he was primarily responsible for the discussions in it of ancient writings on dance and pantomime, discussions that were continued in two later programmes for ballets by Angiolini (Citera assediata, 1762, and Sémiramis, 1765). In the Don Juan programme Gluck received high praise for his music, and particularly for his masterful handling of the terrifying dénouement. The ballet created a sensation in Vienna, and was quickly and widely imitated. But it should be emphasized that the work was the product of a long-term effort in the Habsburg capital towards the reform of theatrical dance, and of the several years' experience that Gluck already had as a ballet composer. Although innovative in its adoption of a three-act structure, and in being accompanied by a polemical essay, Don Juan was well within the norms of Viennese ballet in terms of its proportions, formal procedures and function in the repertory.

A performance of Angiolini and Gluck's ballet Don Juan, with its fiery finale (Gumpenhuber's chronicle, C1758–63, lists 44 torches as props and 29 furies), was long thought to have caused the destruction by fire of the Kärntnertortheater on 3 November 1761, but as Croll (H1976, pp.12–15) has demonstrated, the ballet had been given in the French (court) theatre, and the Don Juan at the German theatre that night (already concluded when the fire broke out) was actually a spoken comedy by Prehauser. This disaster had direct consequence for Gluck as music director, in that the two theatres' orchestras were temporarily consolidated into one, their repertories were mixed, and additional concerts were scheduled in order to raise funds for the rebuilding of the German theatre. Despite the turmoil, Gluck's next opéra comique, a ‘Turkish’-genre piece called Le cadi dupé, was ready for performance for the emperor's birthday on 8 December 1761. Gluck was involved in the preparation of two other works for the court in the time between Don Juan and Orfeo, though not as a composer of fully original scores. He was responsible for arranging borrowed (and in some cases parodied) numbers – described in the libretto as ‘most recently composed and applauded in Italy’ – and for composing the recitatives for a one-act pasticcio by Migliavacca, Arianna, presented on 27 May 1762 during the court's annual sojourn at Laxenburg palace. The roles of Ariadne and Bacchus were taken by the future protagonists of Orfeo ed Euridice, Marianna Bianchi and Gaetano Guadagni (engaged as court virtuosi since Easter). In mid-September Angiolini presented Citera assediata, a ballet-pantomime version of Gluck's earlier opéra comique after Favart; in a later essay (Lettere … a Monsieur Noverre, 1773) he noted that he had not introduced a single foreign note into Gluck's music, though he had considerably abridged the score. Only on account of the need to prepare decorations for Orfeo (according to Gumpenhuber) were performances of the highly successful ballet curtailed.

Although in his earlier Italian operas Gluck had made some efforts towards the integration of chorus (L'innocenza giustificata) and the simplification of vocal writing (e.g. in certain numbers in Tetide), his and Calzabigi's Orfeo ed Euridice marked a dramatic break with operatic practice, even taking into account its genre (azione teatrale, traditionally based on a mythological plot and including chorus and spectacle). For one thing, Gluck's complete abandonment of coloratura in this opera constituted a drastic change in his relations with singers. Previously, as with most composers of drammi per musica, his successes had been closely associated with those of the singers; he had often travelled from production to production with them, and had catered to their specific talents, particularly as regarded passage-work. His new manner of composition necessarily brought other aspects into greater prominence, and Durazzo ensured that all collaborators worked towards a single end. The reduced role of vocal athleticism was due in part to the nature of Calzabigi's text, which largely eschewed similes and metaphors (favourite provocations of word-painting) as well as aria structures conducive to da capo returns (and thus to improvised ornamentation).

More than two decades later, in a letter to the Mercure de France (dated 15 July 1784, published August 1784), Calzabigi claimed near-total credit for the innovations in Orfeo, writing that he had read his libretto aloud to Gluck,

showing him … the nuances that I put into my declamation and that I wished him to make use of in his composition: the pauses, the slowing down, the speeding up, the sound of the voice now strong, now weaker and in an aside. At the same time, I begged him to forgo passage-work, cadenzas, ritornellos and all that is gothic, barbaric, and extravagant in our [Italian] music.

He claimed also to have notated for Gluck the proper inflections in signs written between the lines of his text, both for this opera and for the later Alceste, the composer supposedly having only an imperfect knowledge of Italian. No such manuscripts have come to light, and Calzabigi's account is highly suspect in any case, dating as it does from a time when the poet was incensed at what he saw as the misappropriation by Gluck and his protégé Salieri of his Ipermestra text, in the latter's Parisian opera Les Danaïdes. A document more nearly contemporary with Orfeo, and more credible, is Calzabigi's letter of 6 March 1767 to Kaunitz with regard to plans for his Alceste (still many months off), in which he pleads for singers appropriate for the expressive purposes of his text and worthy of ‘the sublime gifts of Sig. Gluck’, and states that ‘Orfeo went well because we came across Guadagni, for whom it seemed to have been made expressly, and it would have succeeded miserably in other hands’ (Helfert, A1938). Indeed, the castrato Guadagni, who had received acting lessons from David Garrick, frequently incurred the wrath of the public on account of his refusal to break theatrical illusion by acknowledging applause (a concern of Durazzo's even in opera buffa): according to Burney, ‘a few notes with frequent pauses, and opportunities of being liberated from the composer and the band, were all he wanted’ (BurneyH, ii, 876). In a sense, then, Gluck's reliance on singers was just as close as before, merely altered in its nature.

Fundamental to the novelty of Orfeo ed Euridice was Calzabigi's thoroughly classical orientation, something evident in the simplicity of the basic plan of the work and of the diction, and in its extensive use of pantomime. That he managed to communicate his vision of the work to his collaborators is clear from the stage directions for the opening ballet by Angiolini, which recreate ancient funerary rites, and (according to the diaries of Count Karl von Zinzendorf) from the spectators' comparisons of the décors to their sources in Virgil. Zinzendorf himself found the music ‘divine, completely pathetic, [and] completely suited to the subject’. Preparations for the opera, which was to be given for the emperor's nameday (the première was on the following day, 5 October 1762), apparently stretched over many months; Gluck was free to devote his full energies to the project, having been freed from his ballet-composing duties in the German theatre at Easter 1761. On 8 July 1762 and again on 6 August 1762, Zinzendorf (c1747–1813) reports encountering Gluck at aristocratic dinner gatherings, performing excerpts from his score, miming the Furies in the choruses. Gumpenhuber lists as many as 12 rehearsals of the groups involved in the performance, in various combinations; in a conversation with Burney a decade later, the composer recounted that he had never ‘suffered them [the performers] to leave any part of their business, till it was well done, and frequently obliged them to repeat some of his manoeuvres twenty or thirty times’ (BurneyGN, i, 344).

Despite some complaints (e.g. about the simplicity of Orpheus's air ‘Che farò senza Euridice’, and about certain moral ambiguities in the text), Orfeo was a tremendous success in Vienna, being given 19 times in 1762 alone; it was revived with equal success during Carnival 1763, with further performances during February and 11 more between July and September. Ever anxious to publicize his and the Viennese theatres' productions, Durazzo commissioned his agent Favart to arrange for publication in Paris of an engraved full score (in a manner common for French operas but rare for Italian works), which appeared in 1764. Although few copies were sold, the circulation of the manuscript of Gluck's opera aroused considerable interest among musicians and composers in the French capital. (The music so impressed itself upon Philidor, whom Favart had asked to proofread the score, that, probably unconsciously, he plagiarized certain passages in Le sorcier and Ernelinde.) During the summer of 1763 Durazzo had planned to send Gluck himself (then in Bologna) to Paris to check on the progress of the edition, but (according to Favart's friend Dancourt; see Favart, C1808) the count recalled him to Vienna upon hearing that the Paris Opéra had burnt down. Durazzo's action points towards an ulterior motive for the trip: an attempt to gain a foothold for Gluck in Paris (presumably with a performance of a French version of Orfeo), already a decade before Iphigénie en Aulide.

On 21 January 1763, with the success of Orfeo still echoing, Count Durazzo petitioned the empress for an annual pension of 600 gulden for Gluck, and also for Angiolini and the dancer Louise Joffroy-Bodin, principally as a way of retaining their services in the face of offers from other courts or theatres. (All three pensions were granted by imperial decree on 26 April 1763.) The danger of losing these artists was real, for already during the previous autumn Gluck had been in negotiation with representatives of the new Teatro Comunale in Bologna for a dramma per musica – his first foreign commission in seven years. The surviving correspondence concerning the opera between Count Bevilacqua (intendant of the Comunale) and his Viennese agent Lodovico Preti shows Gluck to have been a hard bargainer. Gluck had apparently proposed Metastasio's L'olimpiade, one of the poet's most famous and pathetic texts, but Bevilacqua instead suggested a setting of his more recent Il trionfo di Clelia, which had greater scenic possibilities. That connoisseurs would inevitably compare Gluck's version with the original setting by Hasse, written for Vienna only a year before, was probably another advantage of the latter piece in Bevilacqua's eyes. The preparations and performances (beginning on 14 May) of Il trionfo di Clelia, as well as such distractions as visits to the retired singer Farinelli and to Padre Martini, are vividly described in the memoirs of Gluck's protégé Ditters (Dittersdorf, C1801), who accompanied him on the trip. They departed after the conclusion of the Lenten concerts (for whose direction Gluck was still responsible), travelling via Venice in order to accommodate a young singer, Chiara Marini, who was returning there with her mother. (During Gluck's absence his place as musical director in the Burgtheater was taken by F.L. Gassmann; from Easter 1763 Gassmann also shared with Gluck the job of composing ballets for the French theatre.) Among the cast in Bologna were the castrato Giuseppe Manzuoli, for whom Gluck had already written in Tetide, and the tenor Giuseppe Tibaldi, who four years later created the role of Admetus in Gluck's Alceste. The singer cast as Cloelia, Antonia Girelli Aguilar, pressed the composer to write an aria with obbligato oboe for her to perform with her husband, but Gluck declined to do so. (Dittersdorf notes that Gluck composed exclusively during the morning and evening, the afternoon being reserved for visits and the café.) Gluck's score for Il trionfo di Clelia, which apart from the overture contained only a single borrowed number, did not fully live up to Bevilacqua's expectations; Gluck himself (according to Dittersdorf) was dissatisfied with the orchestra's performance, which, despite ‘seventeen large-scale rehearsals’, ‘lacked … the ensemble and precision that we had long been accustomed to hearing from the Viennese orchestra’.

Possibly on account of Gluck's absence from Vienna (until 6 June), the commission for an opera for the emperor's nameday went to Traetta; the choice of the piece – an Ifigenia in Tauride to a text by Coltellini – presumably came from Durazzo. In composing the work Traetta clearly benefited from the singers' and dancers' recent experience of performing in such an integrated spectacle as Orfeo, which in fact was revived again in late July. Around this time Durazzo asked Gluck to compose what later turned to be his final opéra comique (apart from revisions of earlier works for Paris), a setting of Les pèlerins de la Mecque. Based on an old classic of the vaudeville repertory, the work had been adapted as a modern comédie mêlée d'ariettes and purged of its original doubles entendres by L.H. Dancourt, an actor friend of Favart's whom Durazzo had brought to Vienna mainly in order to shorten French plays for performance at Laxenburg. The new version of Les pèlerins de la Mecque (whose ‘Turkish’ plot has an ancestry in common with that of Die Entführung aus dem Serail) was conceived in part as a showpiece for a recently acquired haute-contre by the name of Godard (playing Ali), who had sung premiers rôles at the Paris Opéra and who far outshined the singer-actors available to Gluck in the French troupe up until then. The opera was nearly ready for its première when the illness and subsequent death (on 27 November 1763) of Archduke Joseph's wife, Isabella of Parma, closed the Viennese theatres. The opera was finally given on 7 January 1764, with the new title La rencontre imprévue and with textual revisions that were meant to mask the similarities between Princess Rezia's feigned death in the original libretto and the murky circumstances of Isabella's demise. Despite its extremely episodic plot, the opera quickly became popular through much of Europe, both in French and in translation.

Even before the première of La rencontre imprévue, another opera by Gluck reached the stage of the Burgtheater, on 26 December (the beginning of Carnival): an almost completely new setting of Metastasio's Ezio (some 14 years after his first version, for Prague). Guadagni sang the title role; the prima donna (Fulvia) was Rosa Tartaglini-Tibaldi. Besides the desire to tailor the music to the singers at hand, the advisability of modernizing the forms of the set pieces worked to discourage Gluck from making large-scale borrowings from the earlier setting – though he did parody five arias from his more recent Il trionfo di Clelia. A largely favourable review of Ezio in the Wienerisches Diarium (7 January 1764; perhaps written at the instigation of Durazzo) sought to link the opera to reforms of the sort seen recently in Orfeo, but the work still clearly adheres to the norms of dramma per musica.

Gluck's compositions over the next two years were almost all written at the behest of the imperial court. (Few Italian theatres were as yet prepared to essay his and Calzabigi's brand of Greek-inspired musical tragedy, as the poet himself later lamented on several occasions.) In spring 1764 Gluck and a retinue of musicians including Guadagni and Ditters accompanied Durazzo to Frankfurt for the coronation of Archduke Joseph as King of the Romans (a preliminary step towards election as Holy Roman Emperor). A celebratory cantata Enea e Ascanio, to a text by Coltellini, was performed for the occasion; although the music does not survive, and the libretto names no composer, circumstances suggest that it was Gluck who composed the music. Before the festivities Gluck and Durazzo stopped in Paris to consult with Favart. Another noteworthy event during the trip was the dismissal of Count Durazzo, on instructions from the empress. His demise was apparently due to intrigues by Favart and Dancourt, and to his relationship to the dancer Joffroy-Bodin (who also was let go). Durazzo's departure (for Venice, where he became imperial ambassador) did not immediately jeopardize Gluck's position, but it did mean the end of a period of firm support and guidance from the court theatre director. Durazzo's replacement, Count Wenzel Sporck, did not commission any further opéras comiques from Gluck, relying on Parisian imports instead. With regard to more ambitious projects, Gluck and Calzabigi largely bypassed Sporck, dealing directly with Kaunitz.

For what was to be the last nameday of Emperor Francis Stephen (4 October 1764), Gluck and Angiolini produced an impressive ballet entitled Les amours d'Alexandre et de Roxane. It has long been assumed that an explanatory programme, now lost, accompanied the work at its première, but this is unlikely; a passing knowledge of Plutarch's life of Alexander was probably sufficient to acquaint most spectators with the plot, and for a subsequent performance of the ballet in Moscow, under Angiolini's direction, no programme was issued. For the next major court occasion, the remarriage of Archduke Joseph to Maria Joseph of Bavaria in January 1765, Gluck was called upon to produce three works: an azione teatrale by Metastasio, Il Parnaso confuso, to be performed for the couple as a surprise by four of Joseph's sisters, with Archduke Leopold directing from the harpsichord (24 January); a concluding ballet by Hilverding and Gassmann, Le triomphe de l'amour, was performed by the younger imperial children); a full-scale dramma per musica, Telemaco, ossia L'isola di Circe, on a text by Coltellini (30 January); and a new pantomime ballet with choreography by Angiolini, Sémiramis (31 January). If in the first piece (the performance of which at Schönbrunn is depicted in two large paintings by J.F. Greipel; fig.5) Gluck was constrained by the thoroughly encomiastic text and the necessity of writing for amateur (though skilful) singers, for the other two works he had fully professional performers at his disposal. Nevertheless, the production of Telemaco suffered from the haste with which the closely spaced festivities were arranged, in that (judging from a remark by the diarist Count J.J. Khevenhüller, later Prince Khevenhüller-Metsch; see R. Khevenhüller-Metsch and H. Schlitter, C1907–25) its concluding ballet (copiously described in the libretto) was either omitted or replaced by a ballet from the repertory. Two years later ( see Helfert, A1938) Calzabigi wrote privately to Kaunitz that

Telemaco, with the best of poetry and singularly divine music, went most badly because [Travaglini] Ti[b]aldi was no actress, Guadagni was a scoundrel and the famous [Elisabeth] Teyber was unsuited for the role of Circe and had insufficient voice for a sorceress and for doing justice to music worthy of an enchantress and an enchantment.

Significantly, in the same letter Calzabigi lists those Viennese operas created since his arrival – Orfeo, Ifigenia in Tauride (by Traetta), Telemaco and Alceste – that in contrast to Metastasian works required ‘actresses who sing what the composer has written, without inserting a trunkful of notes by repeating 30 or 40 times either a “Parto” or an “Addio” decked out in musical hieroglyphics’.

The protagonist of Telemaco was probably meant to be seen at least on some levels as a representation of Archduke Joseph; the same hero had been the subject of an elaborate, specifically allegorical decoration for a Burgtheater concert given for the archduke's birthday in 1758. The court may have helped determine the subject also of Angiolini and Gluck's pantomime ballet, for a setting by Andrea Bernasconi of Metastasio's Semiramide riconosciuta – a text closely associated with the empress – had been given in Munich during the earlier part of the wedding festivities, beginning on 7 January. But Angiolini's ballet was based on a very different telling of the story, the tragedy by Voltaire of 1748, in which the motive forces were regicide, matricide and incest, and which, like Don Juan, featured a ghost returning to exact revenge. (In deciding to stage a tragic ballet Angiolini put himself into direct competition with his rival Jean-Georges Noverre in Stuttgart.) Khevenhüller reported that Sémiramis ‘found no approval at all, and indeed was also far too pathetic and sad for a wedding feast’.

More even than in the case of Don Juan, Calzabigi was involved with the conception and explication of Sémiramis. The essay or programme accompanying it was entitled Dissertation sur les ballets pantomimes des anciens, on the model of the Dissertation sur la tragédie ancienne et moderne with which Voltaire had prefaced his play. Much of the pamphlet's theoretical orientation, as well as various Parisian topical references, betrays the hand of Calzabigi, who in his Lettera al signor conte Vittorio Alfieri of 1784 specifically claimed to have written it (see Bellina, A1994). His authorship is suggested also by his proud assertion, contradicted by all other witnesses, that the ballet ‘succeeded sublimely’, more so even than Viennese performances of Voltaire's original tragedy. But the powerful effect of the work (whether appropriate for the occasion or not) was ultimately due to the ability of Angiolini and Gluck to translate Calzabigi's rigorous and concise plan into effective stage spectacle and music. Much of Gluck's score is starkly functional, with musical gestures intimately tied to the pantomime; in several numbers Gluck also strives for a specifically antique tone. Though conceived with this particular plot in mind, more than half of the music in Sémiramis found its way into Gluck's later Iphigénie en Tauride.

The next composition of Gluck's of which we have notice is his music to a ballet by Angiolini, given at Laxenburg on 19 May 1765, ‘taken from the tragedy of Iphigenia and … better than that of Semiramis’, according to Khevenhüller. The ‘tragedy’ in question was no doubt Racine's Iphigénie, familiar to Burgtheater spectators from many performances over the years; in his Lettere … a Monsieur Noverre (Milan, 1773) Angiolini called his ballet Ifigenia in Aulide, and admitted that the outcry against his Sémiramis ballet had forced a ‘prudent silence’ upon him and caused him to compose this one ‘with a lieto fine, in order to show the public my respect’. The music of the Iphigenia ballet does not survive, but we know that it pleased Archduke Leopold (who also named its composer explicitly). When Khevenhüller heard it again at Innsbruck an 18 August he found it ‘as long as [it is] sad’. That Innsbruck performance was part of the wedding festivities for Archduke Leopold and Maria Ludovica of Spain; while leaving the theatre after the conclusion of the ballet, Emperor Francis Stephen suffered a fatal stroke, among the results of which were the closing of all theatres in the Habsburg realms and the disbanding of the French troupe with which Gluck had worked for over a decade.

The emperor's sudden death left unperformed two already completed works by Gluck. Already before the court's departure for Innsbruck Gluck had begun rehearsing another azione teatrale to a text by Metastasio for the archduchesses, La corona, intended for Francis Stephen's nameday. (The commission is a sign of the favourable reception of Il Parnaso confuso.) Another work, the ballet Achille in Sciro, based on the libretto by Metastasio, was simultaneously being readied for performance at Innsbruck, according to a later account by its choreographer, Angiolini (Lettere … a Monsieur Noverre), and likewise fell victim to the closure of the theatres. (A note of 13 November by Sporck likewise mentions ‘various well-decorated ballets’ for Innsbruck, ‘of which two were not performed at all’.) Gluck's music for Achille survives in a set of parts at the former Schwarzenberg archive at Český Krumlov; the composer salvaged its concluding Passacaille for use in Paride ed Elena, Iphigénie en Aulide and the revised version of Cythère assiégée.

The grief of Empress Maria Theresa for her consort was extreme, since (unusually among sovereigns) her marriage had been a love-match. The protracted period of mourning led to the dispersal of many musicans, actors and dancers from the court theatres and enforced inactivity on the part of those who had stayed, including Gluck. During this period one opportunity for Gluck both to compose and to conduct presented itself in Florence, in Habsburg-controlled Tuscany, where (after a decent interval) Traetta's Ifigenia in Tauride was being given during Carnival 1767. Gluck was asked to set the purely celebratory prologue by Lorenzo Ottavio del Rosso that opened the performance, which he also led, in the absence of Traetta (who was busy producing a new opera in Munich). The text was so slight (consisting of only a few choruses and one aria) that Gluck was able to set it in a mere two weeks (not even having received it until 6 February, nearly a week after his arrival), and the first performance followed on 22 February, a day after the single rehearsal.

In Vienna, there was considerable turmoil after 1765 with regard to the managerial arrangements and repertory of the theatre. The court relinquished some – though by no means all – control to a series of impresarios (including Hilverding), and the increased necessity of pleasing a paying public led to the recruitment of an opera buffa troupe, which reopened the Burgtheater in October 1766 with Il viaggiatore ridicolo by Gassmann. By then Calzabigi and Gluck were already at work on a more dignified piece – the opera Alceste – with which to console the grieving empress and tempt her back into the court theatre, a main point of contact with her people. (Metastasio had made a similar attempt at consolation the previous year with his poem ‘I voti pubblici’.) The story of Alcestis, the subject of a famous tragedy by Euripides, was practically synonymous with conjugal devotion. The resonances of the protagonist's noble sacrifice for the sake of her husband – who is fated to die unless someone can be found to take his place – with Maria Theresa's own situation were obvious, and were specifically emphasized in numerous verses of the libretto and in Calzabigi's dedication of it to her.

The poet's extraordinarily candid and (with respect to Metastasio) acerbic letter of 6 March 1767 to Kaunitz on the subject of Alceste (see Helfert, A1938), quoted in part above, attacked the old manner of composing and performing drammi per musica mainly in order to emphasize the quite different requirements of those performing in works on the ‘new plan’. It has usually been assumed that Calzabigi's request for Kaunitz to guarantee a proper cast for Alceste indicated that Gluck's music for the opera was already essentially complete by this time. But it is equally possible that his letter – sent during the Lenten break between theatrical seasons, when most decisions on personnel were made – was an attempt to procure singers for whom Gluck could write just the sort of music he desired. The eventual choices of protagonists were controversial – Leopold Mozart scoffed that the opera was being performed by ‘mere opera buffa singers’ (letter of 30 January – 3 February 1768) – but were vindicated in the lengthy, enthusiastic critique of the opera by Joseph von Sonnenfels, in his (initially) anonymously published Briefe über die Wienerische Schaubühne (Sonnenfels, C1768). The singer cast as Alcestis, Antonia Bernasconi (the German-born stepdaughter of the Munich composer Andrea Bernasconi), had a small voice, and indeed was a specialist more in comic than in tragic roles, but she ‘acted … with a truth, feeling, and participation [in the role] that [were] marvelled at’. Giuseppe Tibaldi, who sang Admetus, Sonnenfels remarked, gestured more expressively and consistently, now that the upper part of his range was fading, and Gluck aided the cast generally in that ‘he put fewer [musical] difficulties in the way of the talent of his actors than any other [composer]’. The expressive pantomime ballets by the star choreographer Noverre, newly engaged for the Viennese theatres, were much appreciated by spectators as well. His recent arrival (during autumn of that year), too, speaks against Gluck's score having been completed by March; in fact, the music reflects Noverre's separation of sections of expressive pantomime (called balli pantomimi in the libretto) and movements of pure dance (balli ballati), which is much stricter than in Angiolini's works.

The staging of Alceste was undoubtedly delayed by the deaths first of Joseph's second wife Maria Josepha of Bavaria on 28 May and then of the Archduchess Maria Josepha (newly betrothed to the King of Naples) on 15 October, both events entailing the closure of the theatres. Under the circumstances, it is hardly surprising that Khevenhüller found the opera Alceste, once it was finally performed, ‘pathetic and lugubrious beyond all measure’; only the concluding ballet by Noverre in the grotesque vein found general approbation, he reported. The opera reached a respectable number of performances, but a year later, when its score was being prepared for publication (this time in Vienna), Calzabigi insisted (as he explained to Antonio Greppi on 12 December 1768) that it be accompanied by a preface explaining ‘the enormous change we have made in dramatic compositions such as this one’, acknowledging in effect the difficulties that such works posed for audiences (see Donà, A1974). Although the preface (dedicated to Archduke Leopold, since 1765 Grand Duke Pietro Leopoldo of Tuscany; see §11 below) is signed by Gluck, the librettist told others that he himself was the author; the ruse allowed him more easily to include a considerable amount of praise for his own efforts as librettist.

The extent of Gluck's output during 1768 in unclear. No score survives to show exactly what revisions of L'innocenza giustificata (1755) he undertook for its restaging during the summer as La vestale, with Bernasconi in the leading role. (In the libretto (A-Wn), Gluck is for the first time called a member of the Arcadian Academy, under the name ‘Armonide Terpsicoreo’.) Calzabigi's claim (in the above-cited letter to Greppi) that Gluck had already set his Paride ed Elena must be viewed with some scepticism, as that opera was not staged until November 1770. Presumably Gluck was occupied by early 1769 with Le feste d'Apollo, a piece of three loosely related acts with a prologue on the model of the French opéra-ballet, for the wedding of Archduchess Maria Amalia with the Duke Ferdinand of Parma – Gluck's last opera commissioned for a Habsburg festivity (if one discounts the revised L'arbre enchanté). The celebrations in Vienna were on a small scale, with most of the theatrical productions, including Gluck's, taking place in Parma. The wedding was postponed by several months on account of delays over a papal dispensation (since the parties were related) and then the death of the pope; as a result of the confusion, Gluck made the trip to Parma twice. Only about half of Gluck's music for the opera was new; the final ‘Atto d'Orfeo’, included at the request of the archduchess, was a compressed version of Orfeo ed Euridice, and much of the rest was borrowed from earlier works, especially Telemaco. The performances of Le feste d'Apollo, particularly of the ‘Atto d'Orfeo’, were well received; indeed, no court in Italy could have been more receptive of a Gluckian work of this sort – or capable of performing it – than that at Parma, where choral and balletic forces had been employed in Italian opera for more than a decade. At Parma Gluck met and became friendly with the soprano castrato Giuseppe Millico (whose vocal range required adjustments in the music originally written for the alto Guadagni). This singer was to create the role of the male protagonist in Gluck's next opera, Paride ed Elena, and he became the singing teacher of Gluck's young niece and ward, Marianne (Nanette).

It was during the latter part of 1769 that Gluck became involved in the financial and managerial sides of the Viennese theatrical enterprise, at that time leased to the Venetian adventurer Giuseppe d'Afflisio (or d'Affligio). This episode (on which see Grossegger, C1995) nearly cost Gluck his personal fortune, and does not reflect well on his character. On 11 October 1769 Gluck and Franz Lopresti (son of another former entrepreneur of the Viennese stages) signed a contract with d'Afflisio whereby, investing 30,000 gulden apiece, they became ‘economic directors’, each with a 25% share of the profits (minus d'Afflisio's considerable debts). According to Grossegger, Gluck was motivated to participate not only by the hope of financial gain, but also in order to assure a venue for his opera Paride ed Elena; that d'Afflisio was a close friend of Calzabigi certainly encouraged him as well. Kaunitz required that the directors agree to six conditions, including notably the re-engagement of a French troupe and exclusivity for Noverre as choreographer. French theatre and ballet were by far the costliest of the theatrical spectacles, and when not long afterwards d'Afflisio left on a recruiting trip to Italy, leaving Gluck to administer the theatres, the partnership was threatened with ruin. Aiming solely for higher box-office receipts, Gluck vigorously fought efforts towards the reform of the German repertory, instead promoting the often indecent improvised fare; in this he was opposed by Gottlieb Stephanie the younger (later the librettist of Die Entführung aus dem Serail). When the financial situation of the theatres worsened, Gluck's petitions for relief to Kaunitz earned him the enmity of his and Calzabigi's former supporter. On 30 March 1770 Gluck withdrew entirely from the enterprise, without economic loss, as d'Afflisio ceded his lease to Count Johann Nepomuk Koháry. Gluck's personal credit with the court and the aristocracy was much damaged, however, and another result of the débâcle was the departure of Antonia Bernasconi, who was likewise a friend of Calzabigi and d'Afflisio.

In 1770 two of Gluck's operas were revived in the Burgtheater: Orfeo ed Euridice, in May, with Millico in the role of Orpheus winning applause ‘on account of his voice and acting’ (Khevenhüller, 13 May 1770), and Alceste, on 21 October 1770 (later also in the Kärntnertortheater), Millico taking over the role of Admetus. The latter work was thus no longer, as originally, the unusual phenomenon of ‘a serious opera without castratos’ (Sonnenfels, C1768, p.10). Less than two weeks later, on 3 November, Paride ed Elena received its première, with Millico opposite the Viennese-born Catharina Schindler (later replaced by Clementina Chiavacci). The extensive ballets were again by Noverre. The first performance seems to have been a court occasion only in that Calzabigi had dedicated the libretto to Archduke Leopold, who had been staying in Vienna for several months. Calzabigi's later claim, in his mock-picaresque Risposta … [di] Don Santigliano di Gilblas … (1790; see Bellina, A1994, p.379), that the opera's lack of success was due to the court's having ordered a ‘festive’ work (thus precluding the necessary contrasting passions of fear and compassion) contradicts his statement to Greppi in December 1768 that Paride ed Elena was already complete, only awaiting an opportunity for performance (see Donà, A1974). The truth probably lies somewhere in between; the publication of a full score (by Trattner of Vienna) before the end of 1770 points to the opera having been finished a considerable time before its production.

Although both Count Zinzendorf (C1747–1813, entry for 3 November 1770) and the court-sponsored Wienerisches Diarium (7 November 1770) reported favourably on the première of Paride ed Elena, performances of which continued into 1771, the apologetic tone that Calzabigi and Gluck took in their writings on the opera tends to confirm Khevenhüller's statement that it ‘did not find particular approval, on account of its uneven and somewhat strange taste’ (3 November 1770). The unusually long dedication of the score to Duke Juan Carlos de Braganza (one of the composer's earliest supporters, living in exile in Vienna), signed by Gluck but probably drafted by Calzabigi, again mentions the lack of contrasting passions in the opera and explains its ‘strange taste’ as the result of an attempt to depict ‘the different character of the Phrygian and Spartan nations, contrasting the roughness and savagery of the one with the delicacy and softness of the other’. Gluck was later to use similar musical means in Iphigénie en Tauride to contrast the Greeks and their Scythian captors.

Gluck, Christoph Willibald Ritter von