4. Last works (1831–5).

With the commission in 1830 of two operas that were to receive their premières at La Scala during the next two years (La sonnambula and Norma), Bellini reached his goal of financial pre-eminence, earning an unprecedented L12,000 for Norma alone. The December 1831 première of Norma in Milan was, however, no more than a moderate success (although far from ‘fiasco, fiasco, fiasco’ reported in one of the letters falsified by Florimo); several of the singers were below par, and it was not until the following summer in Bergamo, slightly-revised and with a superior cast, that the opera achieved the success it still enjoys today. The plot is drawn from a play by Alexandre Soumet, which had its première in Paris only the year before; Romani and Bellini tempered Soumet's boulevard-theatre excesses by omitting the sensational ending, in which Norma goes mad and throws herself off a precipice. Although the Classical restraint with which Bellini and Romani approached their melodramatic source might suggest a retreat from the Romantic experiments in both plot and music of Il pirata and La straniera, it is probably more accurate to see Norma, like Sonnambula, as a step towards a different brand of Romanticism, its passions all the more forceful for being collapsed into an understated dramatic design. Indeed, the pair's very next collaboration, Beatrice di Tenda, written in 1833 for La Fenice, embraces precisely those violent and tragic aspects of Romantic opera they had played down in adapting Norma and in substituting La sonnambula for Ernani. The plot of Beatrice is Romantic, but in the most conventional way: set in the generic Renaissance castle that provides the background for so many operas of the period, it concerns a tyrant who wrongly acccuses his wife of adultery and has her supposed lover tortured, then put to death. Even Romani himself called the subject ‘horrible’: he had already begun work on a Cristina di Svezia based on Dumas, but Giuditta Pasta, Bellini's first Amina and Norma, convinced Bellini to set Beatrice after she and Bellini together saw a ballet on the subject. This was probably far from an unwelcome or isolated instance of intervention from Pasta: throughout his career Bellini was unusually open to, even dependent on, advice from friends, and Pasta was foremost among both friends and artistic advisors. During the early 1830s Bellini paid extended visits to the Pastas' house at Lake Como and even at times entertained hopes of marrying their daughter, Clelia.

In this case, however, listening to Pasta turned out to be a mistake: not only was Beatrice received with utter coolness in Venice, it became the occasion for the disagreement that ended Bellini's relationship with Romani. As usual, Romani was late in delivering the poetry, so much so that La Fenice's impresario, Alessandro Lanari, called out the police to pressure him. The première was delayed by a month and the Venetian public became impatient, groundlessly suspecting that Bellini was devoting his energies to composing a new opera for a foreign theatre while neglecting their commission. Relations between the collaborators had deteriorated into hostility by the March 1833 première, but the irrevocable rift came just after this, when Romani and an anonymous defender of Bellini's exchanged vitriolic letters in Venice and Milan newspapers, blaming each other for the delays and for the opera's failure. Romani's complaints focussed on the selection of the subject, accusing Bellini of holding things up by dithering too long over the choice and of being too swayed by his ‘Minerva’ – Pasta. Bellini's champion, of course, tossed the blame back to Romani, and countered charges that Beatrice was merely a cynical retread of Norma (a claim that seems incomprehensible when the two scores are placed side by side). By 1834 Bellini and Romani had renewed their correspondence and were discussing future projects; the relationship might have been repaired had Bellini's sudden death not intervened.

After the Beatrice imbroglio, Bellini embarked on a period of rest and travel. In spring 1833 he paid an extended visit to Naples and Sicily, returning home for the first time in six years. The following spring he spent four months in London in connection with performances at the Italian opera there; extended periods of both 1833 and 1834 were spent in Paris. All this travel may partly have been motivated by a break with Giuditta Turina late in 1833: her marriage had reached a point of crisis and she wanted to live openly with Bellini. He refused, perhaps no longer as much in love as he had been five years earlier, but also – always – wedded to his career. As he confided to Florimo, ‘such a relationship would be fatal to me, because it would take away my time and my peace as well’. In London and Paris Bellini was much in demand in society circles, despite speaking no English, and French only (as Heine put it) ‘world-destroyingly, … [breaking] French words on the wheel like an executioner’. Although he clearly glittered in salon society, he found the carnival atmosphere of Paris tiresome at times, and passed long periods of retreat at the house of his friend S. Levy at Puteaux just outside Paris.

In Paris much energy was directed toward securing a contract at the Opéra, negotiations that were complicated by his insistence on a fee equal to Rossini's. Things were easier at the Théâtre Italien, where Bellini signed a contract early in 1834 for the opera that would become I puritani. The première in January 1835 featured the group of principal singers that would become famous as the Puritani quartet: Giulia Grisi, Rubini, Tamburini, and Luigi Lablache. The libretto was by exiled republican poet Carlo Pepoli, who made up for his utter lack of experience with infinite good humour and flexibility in the face of Bellini's incessant reprimands and requests for revision. Thanks to Pepoli's novice status, Bellini was obliged to spell out what he wanted more explicitly than ever before, and their correspondence is the most complete statement we have of Bellini's views on opera. Particularly interesting are Bellini's remarks concerning the tone he sought for I puritani. He hoped to capture something of the sentimental pastoral tone of Paisiello's Nina, another opera with a mad heroine and a happy ending; I puritani's unusual fusion of a misty historical atmosphere vaguely reminiscent of Walter Scott and a gentler style seems typical of the pastel-tinged, fluidly structured plots Bellini preferred in these last years, and might also be seen as a nostalgic return to the Neapolitan roots of his earliest style, now infused with a Romantic air.

At the same time as Bellini fought to match Rossini's earning power in Paris, he also sought both the older composer's friendship and his official stamp of approval as director of the Théâtre Italien. The relationship between the two had been fraught with tension, at least on Bellini's side, since an 1829 meeting in Milan when Rossini had complimented Il pirata, but remarked that Bellini's style was so ‘philosophical’ as to lack orchestral brilliance. Orchestration continued to be a main area of artistic disagreement, and Rossini's comments on the score of I puritani were orientated towards broadening the timbral palette and generally increasing the role of the orchestra to cater to the Parisian enthusiasm for complex instrumental passages. Rossini was encouraging, but Bellini wanted more: he wrote to Florimo that he longed for the affection ‘of a father for his son, of a brother for his brother’, a phrase that – even if one ignores its Oedipal resonances – speaks of the enduring and complex connection Bellini felt with the older composer, perhaps stretching back to the half-heeded prohibitions of his Naples teachers against Rossinian ornament.

Given the concentration with which Bellini absorbed French operatic style during these years, it seems surprising that I puritani marks no more of a break with Bellini's previous style than does, for example, La sonnambula or Norma. In fact, among the last operas it is Beatrice di Tenda that stands out, set apart by the florid style assigned to the heroine and by its predictable approach to aria and ensemble forms. One reason for this may be that Bellini was already experimenting long before the move to Paris, absorbing cosmopolitan trends and setting his sights on international success. Becoming more ‘French’ primarily meant formal experimentation, exploring alternatives to the lyric prototype and blurring or expanding the still-dominant double aria structures, but it also entailed scenic innovation: giving the chorus a larger role and using stage space more imaginatively. I puritani, with its rich exploitation of offstage effects, does this with virtuosity, but the tendency can already be felt in the final scene of La sonnambula, where the sight of Amina sleepwalking along a narrow beam above the mill-wheel is not only a striking stage picture, but a crucial plot ingredient, serving as wordless proof of her innocence.

Like La sonnambula, the last operas move freely between syllabic and florid style. Norma's dominant idiom is mostly syllabic ‘long melody’, but in moments of fury or religious transport she sings floridly (‘Oh non tremare, o perfido’ in the Act 1 finale; ‘Casta diva’); similarly, Elvira's cavatina, ‘Son vergin vezzosa’, presents her as a giddy, virtuoso bride-to-be, but in her Act 2 mad scene a more declamatory style predominates. Both Norma and I puritani offer numerous examples of ‘long’ melodies in the manner of ‘Ah! non credea mirarti’, built of speech-like fragments strung together by harmonic or orchestral threads of continuity with few intermediate cadences or melodic repetition. In this last period the lyric prototype finally becomes the default mode for both slow movements and cabalettas, but at the same time French-accented alternatives become more common, especially ternary and strophic designs. Pollione's dream narration (‘Meco all'altar di Venere’) from Act 1 of Norma gestures toward both lyric prototype and ternary form, returning to the opening pair of phrases after an extended departure; however, in an effect that anticipates Verdi's treatment of such climactic moments in dream and vision scenes, as Pollione quotes Norma's curse the movement opens out into a loosely-organized passage of monotone declamation and chromatic harmony, never returning to its opening material.

The construction of the double aria as a whole was similarly diversified. Less willing to suspend conflict simply to make room for lyrical expansion, Bellini occasionally omits slow movements in duets for characters who are at odds dramatically, such as Romeo and Tebaldo in I Capuleti or Agnese and Orombello in Beatrice; in Norma, Adalgisa and Pollione simply do not take the time for a cantabile as they decide to flee together to Rome. The edges of individual sections within a double aria can also be blurred, as in the first duet for Norma and Adalgisa, where dialogue in versi sciolti evoking Norma's memories of past love (‘O rimembranza’) is layered over the slow movement's orchestral introduction; the cantabile proper begins only after ten bars when Adalgisa begins to tell of the birth of her love for Pollione (‘Sola, furtiva, al tempio’, ex.8). This double exposition is, in a sense, an extension of Bellini's lifelong practice of beginning set pieces with versi sciolti set as arioso, but here the blurring of aria and recitative also creates the sense of a musical discourse moving simultaneously on two levels: Norma's past and Adalgisa's present, Norma's fragmented dream of happiness and Adalgisa's melodically present reality.

A more radical challenge to the symmetries of the double aria is posed by Norma's Act 2 finale, which constantly spills over the confines of the standard aria finale model to become a free succession of lyric numbers, as much ensemble as aria. In the autograph score, Bellini labelled this finale only with a tempo marking, but he described it to a friend as composed of ‘a pezzo concertato and a stretta’, both pieces ‘of a completely new type’. In addition to replacing the cantabile-cabaletta sequence with a pair of slow movements (‘Qual cor tradisti’ and ‘Deh! non volerli vittime’), the Norma finale eliminates the plot tension that usually propels a double aria: the central conflict is resolved during the previous duet with Pollione, leaving these last two lyric movements as sites of pure vocal pleasure. This impulse towards loose structure and sonority for its own sake reaches an apex in the third act of I puritani. The act runs without interruption through a romanza for tenor, a duet and a finale. But, in a curious parallel to the idea of a ‘long, long, long melody’ built from separate tiny motifs, the effect of large-scale continuity is tempered by repeated fragmentation and interruption of individual numbers: in the romanza, the offstage soprano and onstage tenor toss fragments of the melody to each other; during the duet the soprano slides in and out of madness, her psychological shifts accompanied by recalled themes; and platoons of soldiers pass by offstage, their martial music breaking into the lyric movements of the lovers' duet. Paradoxically, the I puritani finale introduces so much scenic activity and so many plot reversals that suspense is almost erased: the tight balance between plot events and static set pieces disintegrates, replaced by a continuous dramatic discourse that, rather than becoming more like spoken drama, spills over into sheer song.

This final act of the final opera is an intriguing document of Bellini's late development and perhaps a hint of the qualities that might have emerged had he lived longer. His style would almost inevitably have become even more French, building on the dramatic continuity of moments like this third act of I puritani and on the flexible shaping of individual movements. The real innovation of I puritani is its use of the stage, in the emphasis it throws on spatial effects and especially its ongoing dialogue between onstage and offstage music; this seems a particularly Bellinian conception of the scenic emphasis of French grand opéra and one that might well have been developed further. Finally, the delicate musical tinta of I puritani, when seen in the context of the various attractions and resistances to Romantic plots evident in the negotiations over Zaira, Ernani, Norma and Beatrice, suggests that Bellini was moving toward a new vision of operatic Romanticism, a style less based on the sharply articulated conflicts and shocking stage effects of many of Donizetti's plots (and, indeed, of Victor Hugo's plays), but one that was more fluid, and perhaps as concerned with reviving a vaguely remembered pastoral mode as with engineering sharp breaks or steps into innovation.

Bellini died on 23 September 1835 after several weeks of illness. Much speculation has centred around the fact that he apparently died alone in Levy's house at Puteaux, and that during his illness an acquaintance who visited the house was sent away by servants without being allowed to see him. These events are sad but hardly sinister: Bellini died of an attack of the amoebic dysentery that had plagued him intermittently since 1831. A requiem mass was held at Les Invalides on 2 October, with Rossini, Paer, Carafa, and Cherubini as pall-bearers; Rossini served as executor of the estate. The remains rested in Père Lachaise cemetery in Paris until 1876, when they were transported by train to Catania in a grandiose procession met at every station with patriotic speeches, musical performances, sentimental spectacles and copious tears.

Bellini, Vincenzo