19

Speech

When does speaking become marked in opera? We are not talking about dialogue opera such as opéra comique or German-language Singspiel, in both of which speaking is an accepted part of the genre; we mean those occasions in all-sung opera when a character, usually in crisis, simply talks, or shouts, or half shouts and half sings. This is a phenomenon of the late nineteenth century and beyond, and sometimes the composer seems to encourage it, sometimes not. Carmen’s final word to Don José – ‘Tiens!’ (Take it) – is almost always shouted rather than sung, and probably better that way. Risë Stevens’ recording is a prime instance. The composer wrote notes, but the performers take some dramatic licence and speak instead. A more complex example might be Tosca’s last words in Act 2 of Puccini’s opera, ‘E avanti a lui tremava tutta Roma!’ (And all Rome trembled before him!). This is written to be intoned on low C♯, but can sound far more contemptuous when (as in Maria Callas’s famous recordings) it’s spoken. In this case, though, one could argue that Puccini wrote the line on one note not as an invitation for speech-like declamation, but for a precise reason. Tosca is just about to enact a little religious ritual, placing candles on either side of Scarpia’s corpse – she has just stabbed him to death – and then balancing a crucifix on his chest; in other words, Puccini wrote those low repeated notes because he wants her to sound ecclesiastical at the end, as if delivering a sliver of Gregorian chant. What’s more, if we rewind a bit in the Tosca example, to the moment where Scarpia has just been stabbed, we see in the score that Puccini does indeed specifically ask for speaking – words like ‘Maledetta!’ and ‘Questo è il bacio di Tosca!’ (This is Tosca’s kiss!) are written without note heads, just stems and flags to indicate the rhythms. Singers interpret this notational instruction with a motley assemblage of gasps and rattles.

These are isolated examples from a century and more ago. But soon after the phenomenon would become less of a fluke. Speech in opera, and operatic styles that veer towards speech, marks an important development in the earlier twentieth century. The invasion of speech tells us that, true to the spirit of the age, operatic composers were experimenting with new styles and effects, just as some of them borrowed from new media, like recording, radio and film. But it also suggests that the entire business of opera, of people singing out their feelings with lyrical abandon, was becoming increasingly problematic, and at certain moments seemed hard to sustain. And real speaking in opera gets much more common when we approach 1945, as the borders between speech and singing become blurred in so many twentieth-century operas. Modern iconoclasm welcomed greater intrusions of speech and half-speech into opera as a bulwark against old-fashioned lyricism, despite the fact that the phenomenon had existed, and was extensively theorized, in the nineteenth century.

Ordinary speaking within otherwise sung operas is always special, and sometimes very special. In Schoenberg’s Moses und Aron (written 1930–32; first performed 1954), Moses’ pronouncements are spoken, as if mere singing were for idolaters and the unserious. A deep voice is of course specified. In Berg’s Lulu, when the heroine asks Dr Schoen’s son, Alwa, ‘Isn’t this the sofa where your father bled to death?’ her speaking voice conveys dreadful matter-of-factness. There is also the phenomenon of Sprechstimme, which means literally speech-voice, sometimes also called Sprechgesang, speech-song, an eerie vocal style involving pitched speaking in which the singer hoots and intones rhythmically – with the swoops up and down guided by special notation on the page. In Berg’s Wozzeck, the protagonist’s drowning scene is written entirely in this way, Wozzeck’s rhythmic wailing sounding like eighteenth-century Melodram from the expressionist Inferno. Sprechgesang is an old word. In the nineteenth century it was even used to refer to ‘Wagner’s song-speech’, where lines are fully sung but, in the words of one early critic, ‘the musician subordinates himself entirely to the poet; a free declamatory element prevails’.¹ One of the earliest books written about the Ring says that Sprechgesang is its basic idiom, as opposed to ‘songs and ensemble singing’.² But usually the term refers to a mix of singing and speaking. A nineteenth-century biography of Carl Maria von Weber treats Sprechgesang as synonymous with Melodram, speaking over the orchestra.³ In the Allgemeine musikalische Zeiting of 1877 an article on sacred music notes that ‘Verdi recently used Sprechgesang twice in his Requiem, to excellent effect’, presumably a reference to the soprano’s intonations in the ‘Libera me’ movement.⁴

When we consider half-speech or sung-speech in early twentieth-century opera, we need to remember that, in spoken drama, actors often half sang their lines. They were intoning, using actual pitched notes, getting the throb in the voice that way. We know this not only through early recordings, where one famous example is Sarah Bernhardt,⁵ but also because the style was parodied for its exaggerations or recalled as antique grandeur in early sound films. Some idea of what serious professional actors sounded like in, say, 1900 can be gleaned from the opening scene of the German film comedy Viktor und Viktoria (1933), in which an aspiring tragedian (played by Hermann Thimig) chews up the scenery in an audition. He sounds far more like Caruso than Olivier, because every syllable is delivered to a sung, voiced musical pitch. The phenomenon of Sprechstimme and singing declamation thus existed outside opera long before it became a special, non-singing, operatic effect. One can even make the case that Wagner’s radical changes in melody, the long stretches of freely sung declamation in the Ring, were in part modelled on a sonorous elocution common in German acting of an earlier time, a style put into his ear by singers who passed over into half-speech as a quirk, like Wilhelmine Schroeder-Devrient.⁶

Perhaps, then, this use of speech is just another example of the way in which opera is so highly porous. There is always a temptation to account for its evolution, in the twentieth or any other century, as a self-contained system, glaring at its own past and grappling with problems of its own aesthetics. But to do this is to falsify one of its hallmarks. Sounds and tricks from other genres were always re-forming opera, and continued to do so in the modern era. We do of course occasionally pay tribute to spoken theatre and its relationship to modern opera, usually by citing libretti that were inspired by spoken actors like Gertrud Eysoldt (the heroine in the first German performances of Oscar Wilde’s Salomé) who caught Strauss’s attention on behalf of a dramatic subject. Puccini is another good instance. He got the first creative ideas for several of his operas by seeing versions of their stories in the spoken theatre: in the case of David Belasco’s play Madame Butterfly, he enthused about a London performance even though with hardly any knowledge of the language in which the actors were speaking.

But not long into the twentieth century there appeared another significant influence, as cinema began to talk to opera. Film historians have of course dealt extensively with the influence of opera on early film. Operatic music was used in live cinema accompaniments, and there were opera-centred extravaganzas like Phantom of the Opera (1925). There were biopics of Wagner and Verdi, and silent-film operas such as those in which Geraldine Ferrar starred, or the 1925 Der Rosenkavalier. Cecil B. DeMille saw performances of Saint-Saëns’ Samson et Dalila in his impressionable youth, and plumbed his memories of French Orientalist operas for many a Babylonian processional. There is, though, greater resistance to seeing influence move in the opposite direction: to the idea that film – low and technological and brand-new as it was – could materially change something as hallowed and long-standing as opera. Yet film belongs to the cultural history of modern opera, as much as does expressionist theatre, or recording technology or the rise of the stage director.

Let us pose a non-canonic question: did opera composers go to the movies? Maurice Ravel was certainly an enthusiastic filmgoer, and his experience in the cinema may well have shaped his scenario for the ballet Ma Mère l’oye (Mother Goose, 1912), in which the protagonist, Sleeping Beauty, is shown apparitions – various fairy-tale stories – while in a drowsy state. Each story is introduced by an inter-title, two Moors bringing out a scroll. In 1912, where would the composer have experienced visual phantasmagoria, been in a dark place conducive to sleep, attended a flickering fantasy world where effects could seem magical, and where the story was explained by inter-titles? In the cinema, of course. Cinematic influence on modern music in general was widespread: the famous silent-film sequence in Erik Satie’s ballet Parade (1918) is a single instance that can stand for scores of others. Two characters in the ballet suddenly start enacting an American silent-movie parody (cowboy, chase, gunshot, train rescue) and Satie adds sound effects to the music, shots and whistles accompanying the action, just as they were often added when silent films were screened. Ravel’s operas of course drew from more than just a single well of influence. His one-act opera L’Enfant et les sortilèges (The Child and the Spells, 1925) began as a concept for a ballet. There are significant stretches with no singing but much dancing: teacups dance the foxtrot with teapots; wallpaper shepherds and shepherdesses tread a measure; there is a long waltz for winged insects and other fluttering creatures. The opera is arranged as a series of vignettes, in Part One involving objects and books the Child (mezzo-soprano) has destroyed in a fit of temper. They come alive and reproach him. In Part Two, after a scene change from the Child’s room to the garden, it’s the turn of the animals he has teased and caged. Ravel’s brother, at least, seems to have noted the affinity with film, claiming that ideally the way to present the opera would be along the lines of Disney’s Snow White, which the composer saw in 1937.⁷ Dances gave Ravel a sonic background for many of the scenes, the characters singing along with and in between their music. The most heartrending of these is the lament of the Squirrel (mezzo-soprano) in Part Two, sung against a carefree waltz danced by the insects, and tallying up what he lost once caged: ‘Le ciel libre, le vent libre, mes libres frères’ (The free sky, the free air, my free companions). What is extraordinary is that as the Squirrel’s words become more and more impassioned, his musical arc carries the waltz further and further afield. At the end he sings, ‘Regarde donc ce qu’ils reflétaient, mes beaux yeux, tout mirotants de larmes’ (See then what they reflect, these pretty eyes of mine, all gleaming with tears). The cadence and crescendo on these lines, which shifts the waltz into the minor mode to mark the anguish, proves what was once said of Ravel by a sage philosopher: you don’t need funeral marches when you can compose devastating minuets.⁸

This technique – composing lyrical vocal lines against the backdrop of dance numbers – alternates in the opera with freer outbursts of barely sung lines. The Child’s opening tantrum is an example and, as so often during this period, the reversion to naturalistic declamation is earmarked for rage or loss of control. ‘Plus des leçons!’ (No more lessons!), shrieks the Child, and although he sings to fixed pitches, they are written in such a way as to be anti-singing, anti-melody. But the most spectacular and enigmatic use of speech is reserved for the end of the opera, when the Child faints after having been menaced by the animals. Once his consciousness is removed, the animals can suddenly no longer sing, and the orchestra itself can barely play, uttering just a few strange tones amid long silences. The Child’s imagination, it would seem, has alone given forth all the music. Then the animals start to speak, haltingly, and in a slow, brilliantly choreographed musical crescendo they teach themselves how to shout out a coherent word, ‘Ma … man!’ (Mum … my!). As they do so, they simultaneously learn how to add pitch and rhythm to their sounds, to get back to singing. Their reward is to reach the grandest of musical heights, since at the end they are able to sing a fugue, a gentle contrapuntal chorus to the words ‘Il est bon, l’Enfant, il est sage’ (He is good, the Child, he is wise).

ZEITOPER, OPERA UP TO DATE

At one point Ravel entertained the idea of introducing film projection into L’Enfant et les sortilèges, and the device was not uncommon in opera in the 1920s and 1930s. Berg, a film-crazy composer, used it in Lulu, which includes a silent-film interlude, to be shown between Act 2 scenes 1 and 2, and depicting Lulu’s arrest, imprisonment and release. Theodor Adorno and Hanns Eisler’s 1947 book Composing for the Films argues that Berg’s operas used film-music tricks, mentioning that the twelve-note chord which accompanies Lulu’s death ‘produces an effect very much like that of a modern motion picture’.⁹ There were other modernist composers who incorporated film episodes, such Paul Hindemith (1895–1963), whose Hin und zurück (There and Back, 1927) is like a film run forwards and then in reverse, since it first tells its story from beginning to end, and then tells it backwards. Hin und zurück is a so-called Zeitoper or ‘Timely Opera’, a sub-genre that arose in Germany during the Weimar Republic (1919–33) and attracted self-consciously iconoclastic young composers. Its attractions were fashionable technology (including sound recording and film), plots revolving around modern communication and the frantic pace of life, and arms flung open to popular music.

Ernst Krenek (1900–1991) puts the matter in a nutshell in his Jonny spielt auf (Johnny Strikes Up, 1927). The main characters are a serious composer, an opera singer, an African-American jazz musician and a violin virtuoso; as one review from 1929 tallied it, ‘Everything that is typical in contemporary life finds a place in [Krenek’s] opera … movies, radio, loud speakers, foxtrotting, exoticism, revues, grand hotels.’¹⁰ The plot takes place in Krenek’s present, and is farcical and complicated, with scenes in a railway station, a hotel corridor, on a glacier and in urban streets. Typical productions featured flashing neon signs and film projection. Max the composer (tenor) is a sighing bore (perhaps a caricature of the arch-serialist and Schoenberg pupil Anton von Webern) who is transformed when he meets the opera singer Anita (soprano) on a glacier. He vows to lighten up and descend to the modern world, a sphere personified by Jonny, the jazz musician (baritone). As Krenek said around that time, making the message of the opera clear:

No one can get around the fact that the existence or nonexistence of symphonies is of absolutely no consequence to the members of today’s bourgeoisie. On the other hand, if the output of dance music were to cease for some reason, they would demand, through their newspapers or in some other way, the immediate resumption of its production.¹¹

Now very much a curiosity, Jonny spielt auf was the most-often performed opera in Germany of the Weimar period – and over forty productions were mounted worldwide in its first season, including one at the Metropolitan Opera in New York, with mostly German singers. The New York Sun wrote: ‘Krenek has set most of it to music of a modernist German opera type. This is the inevitable result of his design. His aim was to travesty the kind of recitative so often heard in very grand opera, which disguises commonplace and even silly remarks in lofty sounding phrases.’¹² But for ‘recitative’ one should read ‘conventional operatic vocal writing’ in general, since what Krenek does is compose unstructured, freely sung dialogue to a background that often includes hints of ragtime, dance-hall music and jazz. His own, much later memories underline his operatic hinterland:

[Jonny] was labelled a ‘jazz opera’, which I felt to be a misnomer, for whatever jazz here occurs is brought in to characterize the professional sphere of the protagonist, Jonny, leader of an American combo. The music attached to the other characters, which to me were at least as important, is conceived in that early romantic idiom I had chosen as my model, occasionally touched up with dissonant spices and Italianizing Pucciniesque vocal exuberance.¹³

In one scene, though – the one in which Jonny, fleeing for America, drops his train ticket on the street – the vocal writing edges close to Sprechstimme, the half-spoken sounds perhaps expressing Jonny’s agitation or perhaps making a sharp contrast with what comes next. Feeling lonely for home, Jonny breaks into ‘Swanee River’, a Stephen Foster minstrel song. But since ‘Swanee River’ was written in 1851, it represents the historical past, the ancient yesteryear of Jonny’s own world.

WHEN EVERYDAY LIFE IS SUNG

Musicalized speech appears in Zeitoper but hardly sets it apart from other operas written at the same time; what’s more, the device says little either about Zeitoper’s irreverence or its love affair with the up to date and the prosaic. Such speech was in fact Zeitoper’s most serious debt to the past and high culture, since the idiom’s ancestry includes richly pedigreed elements – something from Wagner, something from the realist composers, Puccini’s ‘vocal exuberance’ not excluded, and something from Literaturoper – in which setting a spoken drama more or less intact dictates certain freedoms in the vocal writing and its accompaniment. And we should add to these the connection between the half-sung quality of musicalized speech and classical spoken theatre’s sonorous elocution in performance.

But the strangest of the half-spoken operas before the Second World War were not operas written for live performance. They were the early sound films that both imitated opera – particularly Zeitoper and operetta – and in turn provided opera with inspiration for some of its special iconoclastic moves. Consider, for instance, René Clair’s 1931 film comedy Le Million. Clair knew the classical-music world from the inside and had collaborated with Satie in 1924 on the short film Entr’Acte. In Le Million, ordinary people – charladies, policemen, thieves and poor artists – frequently take it into their heads to deliver their lines by singing them freely. Sometimes they do it just by rhythmicizing their speech, sometimes by erupting into quasi-melody, sometimes by joining in an impromptu song-like ensemble or by arguing in tango rhythm. As soon as they do anything remotely musical they are magically accompanied by the Invisible Orchestra from Nowhere. Considered as opera, this surreal state where everyday life is sung, is familiar and unsurprising. In the context of early sound cinema – usually a more rigidly realistic affair – the film constituted an elaborate, avant-garde piece of absurdity. In Germany after 1933, much of this lively cinematic experimentation was lost. The Nazi cultural crackdown effectively suppressed both the German varieties of such absurdist films, where modern-day contemporary life turned into an opera, and the operatic world they reflected and refracted. Krenek, for instance, would see Jonny spielt auf condemned as ‘degenerate music’ (entartete Musik) and banned from performance; he fled to America in 1938.

In France the crosswinds blew a little longer. During the war years, Francis Poulenc (1899–1963), a protégé of Satie, wrote an absurdist comic opera based on a play by Guillaume Apollinaire, Les Mamelles de Tirésias (Tiresias’ Breasts, finished 1944, premiered 1947). The heroine Thérèse (soprano) is tired of being a housewife and turns herself into a man by releasing her breasts – two helium balloons – into the air. Renamed Tirésias, she heads off for adventure, leaving her husband (the baritone) behind. A series of improbable incidents, unrelated to one another, completes the first act. The opening of Act 2 is a musical high point: after some couples dance a sarabande, accompanied by a kitschy dance-hall orchestra, the husband arrives (with no fewer than 40,049 offspring somewhere in tow). The orchestral musicians in the pit – burly men among them, no doubt – are then required to impersonate this infant juggernaut by bawling out ‘La, Lala, Lala, La-Lalala-La!’ to a kind of circus-slip-on-the-banana-peel number complete with penny-whistle slide, while the husband – in desperation – urges ‘Silence! Silence! SilenceSilenceSilenceSilence!’ In general, Poulenc ballasts the score with vocal noise as unrelated to operatic singing as he can devise: cackling, talking, sneezing and whistling. Reverting to the obvious, he reserves one great operatic gesture – with a touch of old-fashioned French Orientalism – for Tirésias’ Act 2 entrance in disguise as the Cartomancer, complete with turban, robe and high soaring voice.

By placing Les Mamelles de Tirésias and Le Million side by side, we get a sense of the affinities between comic and sardonic opera and the 1930s films that made real life into opera – so-called ‘operetta’ films but for the most part having nothing in common with traditional, theatre-based operetta. In one famous case, though, an opera was remade immediately as a film: Die Dreigroschenoper (The Threepenny Opera), by the playwright Bertolt Brecht (1898–1956) and composer Kurt Weill (1900–1950), was premiered in 1928 as a wildly successful theatre piece and was then filmed in 1930–31 as a multilingual production, with both a German and a French cast directed by G. W. Pabst. Adapted from an eighteenth-century English ballad opera by John Gay, Die Dreigroschenoper is a work with a purpose; artists with strong principles put it together. In line with Brecht’s thinking, artifice – people breaking into song, for example – is all important as a reminder to the audience that they are watching something invented, and must not get sentimentally involved. The work thus returns to the conservative tradition of dialogue opera, with spoken scenes interrupted by numbers that are often given parodic titles: ‘Das Lied von der Unzulänglichkeit menschlichen Strebens’ (The Song About the Insufficiency of Human Endeavour) or ‘Eifersuchtsduett’ (Jealousy Duet). Weill’s musical style draws from German cabaret but with strange distancing effects, often brought about by complicated harmonic tricks. He also freely parodies earlier operatic styles, and employs contradictory effects: a sarcastic, brutal text will be set to pleasantly attractive music. The actor-singers were directed to perform their numbers without real involvement, by assuming attitudes. Rather than smoothing over the abyss between speaking and singing, Brecht and Weill revel in its estranging quality.

Die Dreigroschenoper would seem not to fit the mould of this chapter in one sense. There are musical numbers, and there is talking to link them together. The work largely avoids modern opera’s slide towards speech, towards the meandering, continuous music and singing, speech-turned-into-music, blurred boundaries between vocalizations that sing and those that speak. But Brecht and Weill’s opera is not entirely deaf to the phenomenon. For instance, in Act 1 there is a ‘Liebeslied’ (Love Song) between ingénue Polly Peachum (soprano) and her murderous, unfaithful, violent husband Macheath (aka Mackie Messer or Mack the Knife; tenor or baritone). On the page, the libretto looks like a conventional opera duet, starting with dialogue and ending with simultaneously sung verse. But the dialogue is not in fact sung: it is spoken over the music, and with stiff rhythms and swooping intonation: Do you see the moon over Soho, asks Macheath. I see it, beloved, answers Polly. And this mixture – which would be called Sprechstimme were a Berg or a Schoenberg to have written it – was also characteristic of German cabaret in the Weimar era, where a famous singer like Gussy Holl would be called a ‘diseuse’ as opposed to a ‘chanteuse’ – a ‘sayer’ rather than ‘singer’ – because she half spoke her songs.

OLD STRAUSS

It might seem that twentieth-century opera’s many ways of sliding towards speech mostly go with two prior conditions. One is a libretto based in comedy, satire or farce. The other is a librettist and composer unfazed by or even enamoured of movies, popular music, flappers, things generally that come from America and other elements never to be found, say, in French grand opéra or in Wagner. But that conclusion ignores the fact that the phenomenon also has distinguished European roots in Literaturoper, where a composer sets a play without trying to pass it through the intermediate stage of becoming a libretto.

As we saw in Chapter 17, it was through operas such as Debussy’s Pelléas and Strauss’s Salome that some early twentieth-century composers managed to renew an operatic tradition then thought by many to be mired in Wagnerism. They did so by writing operas whose text emerged more or less unchanged from a pre-existing spoken drama. In both cases, and with others that followed, a prose play proved liberating. It encouraged jettisoning the formality and regular musical periods that seemed dictated by conventional libretto poetry, and led to experiments with more intense styles of declamation – not to mention descents into the labyrinth of the soul that were pursued with an enthusiasm unknown even to Wagner. Prolonged explorations of this labyrinth also encouraged composers to use their previous expertise as masters of instrumental music, allowing the orchestra to take much of the burden of expression, even making it the prime mover. Both Pelléas and Salome might seem more like continuations of Wagner than rebellions; but they were so primarily in their most trivial aspects, in that both use leitmotifs in ways related to his practice. Both were also tributes in sound to Wagner’s orchestral innovations, and Strauss sailed close to the Wagnerian wind in his harmonic language. But their subject matter – their composers’ submission to verbal imaginations very distant from Wagner’s – marked the important breakthrough.

In both cases, the experiment was unsustainable. The peculiarity and intensity of the situations Debussy and Strauss explored in these two operas were impossible to emulate, let alone surpass. After attempting just such a bigger-and-scarier sequel in Elektra (1909), Strauss – as we have seen – then took a very different operatic route in his eighteenth-century comic extravaganza Der Rosenkavalier. Debussy’s case was, predictably, more tortured and hesitant. He was certainly fastidious enough to abhor the idea of repeating himself, and in 1903 wrote to fellow composer André Messager:

Those who are kind enough to hope that I will never be able to forsake Pelléas are carefully averting their eyes. They simply don’t understand that if that were to happen, I would immediately turn to growing pineapples in my bedroom, believing as I do that repeating oneself is the most tiresome thing.¹⁴

But – one of the eternal questions for twentieth-century opera composers – where was he to go next? His later life, with its series of extraordinarily inventive orchestral and solo piano works, was haunted by abandoned dramatic projects. The most extensive was a version of Edgar Allan Poe’s story ‘The Fall of the House of Usher’, a subject which had some Pelléas-like features (dream-like oddity in particular), but which never got past the sketch stage. Debussy blamed his lack of success on failed musical innovation, saying that ‘everything strikes me as boring and empty. For a single bar that is alive, there are twenty stifled by the weight of what is known as tradition, whose hypocritical and shameful influence I nonetheless recognize there, despite my efforts.’¹⁵ The early years of the twentieth century, with tradition ever more heavy and the need to avoid it ever more exigent, were – as we have seen again and again – hard times in which to write an opera with ease.

Even though Debussy faltered the vogue continued, and it became a critical turn. For so long, melodious singing – the arc of the voice, the live amalgam of the beautiful line in the singer’s performance – had been almost a definition of opera. The burden of expressiveness lay in that amalgam. Yet now operas in which the orchestra took over the emotional expression, and in which characters became increasingly prone to natural speech rather than lyrical songs, were becoming commonplace. Small surprise, then, that an association between speech-like opera and displeasure surfaces in the 1920s:

Why, it might be asked, should it be considered more true to life to declaim and utter sounds which more often than not give no pleasure, than singing a beautiful melody? To have cut-and-dried airs all through an opera of today would be quite as tiresome as to have none at all; yet to exclude them from a score which it is reasonable to suppose is written with a view to pleasing the majority of the public is surely just as annoying. Nothing can make any theatrical performance absolutely real, and yet truth can be conveyed through beautiful melody (or airs, or tunes if you will) just as it can by beautiful language … the public wants what gives it pleasure. Within the last thirty years or so it would be really very interesting to know how many dozens – nay hundreds – of operas have been produced in Italy, Germany, and France, not to mention other countries, and how few have survived, and the underlying reason. This would probably be found to be lack of melody, though by this I would not be misunderstood as implying that it is owing to the absence of the old modelled type of vocal air, which would be out of place in opera of today.¹⁶

This, from a critic in The Musical Times, is not just an old lay, sung by many a bard. Here, ‘melody’ is standing in for a sense that expressive truth in opera is easier to fashion when the human voice is central; and that, in performance, the art or power of a particular singer will go a long way towards moving the listener in the direction of that truth. To compound the problem, speech-like opera was particularly attractive to composers who were otherwise instrumental-music specialists. As we have mentioned before, this led to a situation in which the professional opera composer – the type who spent virtually his entire life in the theatre, as had Verdi and Wagner – became a rarity. And this in turn led to twentieth-century opera becoming primarily a tale of experimental works, solitary attempts by composers whose hearts were given to purely instrumental music, to reassemble opera from the wreckage of late-nineteenth-century excess.

One of the very few who managed a continuous career in operatic composition was Richard Strauss. In a series of works after Der Rosenkavalier and Ariadne, he succeeded largely through an uneasy and partial allegiance with melody. For a time, until the poet’s untimely death in 1929, he did this in collaboration with Hofmannsthal. They veered between grandiose works that seemed to return to Wagnerian seriousness and complexity, such as Die Frau ohne Schatten (The Woman without a Shadow, 1919), which Strauss hoped would be ‘the last Romantic opera’,¹⁷and Die ägyptische Helena (The Egyptian Helen, 1928); and lighter pieces such as Arabella (1933), in which he strove to rekindle the tone of Der Rosenkavalier. In all these operas, and in the five post-Hofmannsthal works that followed, Strauss maintained certain Literaturoper attitudes he had learned as early as Salome, in particular a relaxed attitude to long sections in which musical elaboration was sacrificed to convoluted dialogues in quasi-recitative. Many have enjoyed selected highlights from these works, and some have enjoyed complete performances in the theatre; few, though, have wished any of them wordier.

Strauss’s last opera, premiered when the composer was in his late seventies, went deep into speech, in fact took words and music in opera as its central theme. This was Capriccio (1942), which Strauss called a Konversationsstück für Musik (Conversation Piece for Music), and which he wrote in stubborn antithesis to the destruction of the Second World War raging around him. Set in eighteenth-century France, the basic dramatic conflict is simple: the poet Olivier (baritone) and the composer Flamand (tenor) are invited to the birthday celebration of a beautiful young widow, the Countess Madeleine (soprano); both become rivals for her affection. They and various other characters go over this conundrum in allegorical terms: when words are set to music, which is more powerful? Which is the master? But added to this core are a series of small events and much – very much – discussion. In a synopsis, it is astonishing how often certain phrases recur, ‘they discuss’, ‘they engage in a discussion’, ‘an argument ensues’, ‘the conversation grows heated’, ‘servants enter and comment’, ‘she wonders out loud’. And so the opera, which is in a single act, grew to well over two hours in performance and thus became a monumental speech opera in which conversation is elaborately celebrated. The opening music is a string sextet, played by the first desk principals, and perfectly sets the tone. Its outer sections are imbued with an elaborate nostalgia reminiscent of Der Rosenkavalier, while in the middle comes a sudden injection of heaving theatrical passion. The final scene of Capriccio, in which the Countess meditates on her conundrum in a famous soliloquy, is preceded by ‘Mondescheinmusik’ (Moonlight Music), an orchestral peroration that seems in some ways to return, as does much of the opera, to the energies and certainties of Strauss’s earliest persona. But the Moonlight Music is – and not just in the context – imbued with unbearably poignant nostalgia for a German operatic world much of which was fast disappearing into the rubble of history. Two years later, in 1944 and with catastrophe imminent, the music critic Willi Schuh described a visit to the composer’s study:

During a conversation about our era, he took down a volume of Goethe from the bookcase to read the passage from the last letter to Wilhelm von Humboldt, written five days before Goethe’s death: ‘Confusing conclusions about confusing deeds dominate the world, and I have nothing more pressing to do than if possible to increase that which remains and is left to me and to keep my originality in hand.’¹⁸

But the Capriccio sextet had been given its first performance, in May 1942, at a private gathering in the Vienna house of Baldur von Schirach, former head of the Hitler-Jugend and now Gauleiter of Vienna.¹⁹ A few months later, von Schirach would give a speech in Vienna in which he said that the deportation of the Jewish population from the city would ‘contribute to European culture’.²⁰ With such as this, Strauss had made his rotten compromise.

BROUGHT IN FROM THE COLD

Serious, non-farcical speech-opera could be a black hole for composers in Germany, or indeed in France and Italy. Yet speech-opera thrived in the hands of composers and singers outside that circle. For this reason, the earlier twentieth century marks the moment when opera from outside, written in Hungarian or Russian or Czech or English, joins a repertory High Table once restricted to those west of the German linguistic border with an Eastern hinterland. To some extent, this was because these composers’ nationalist, almost anti-colonialist impulses permitted lingering salutes to folk music and popular idioms, and to nature. This is what one critic, hauling in the Slavs to defeat the moderns, called the ‘path of nature that will drive away the thick atmosphere of the theatre which surrounds the more sophisticated productions of modern European cosmopolites’ – although this comment in fact looks far back in time, to the Czech composer Bedřich Smetana’s Prodaná nevěsta (The Bartered Bride, 1866), for an antidote to Ariadne auf Naxos.²¹

The starring non-German proponent of ‘speech melody’, one of the oddest and most remarkable, was Janáček. We discussed his Jenůfa in Chapter 17 as an opera that runs parallel to Pelléas and Salome. As mentioned there, although Jenůfa was premiered in Brno in 1904, Janáček remained little more than a local celebrity in his Moravian homeland until the opera’s Prague revival in 1916, when he was sixty-one years old. Even then it took another decade for the work to find a place in the international repertory (mostly via performances in German translation). But Janáček, encouraged by the gradual emergence of Jenůfa and further inspired by an autumnal infatuation for a young married woman called Kamila Stösslová, produced in his last decade five major operas, Výlety páně Broučkovy (The Excursions of Mr Broucek, 1920), Kát’a Kabanová (1921), Příhody lišky Bystroušky (The Cunning Little Vixen, 1924), Věc Makropulos (The Makropulos Affair, 1926) and Z mrtvého domu (From the House of the Dead, premiered posthumously in 1930). These works were thought eccentric and even amateurish by many at the time, yet they gradually assumed repertory status in the second half of the twentieth century, belatedly crowning Janáček as one of the greatest opera composers of the last hundred years.

Janáček’s compositional use of ‘speech melodies’ was already evident in parts of Jenůfa, and faithful transcriptions of the melodic and rhythmic shapes of tiny verbal phrases would become the basic material of his late operatic language. The drama in these late works typically proceeds by means of a constantly changing series of speech-melody events, fashioned into blocks in which these melodic and rhythmic shapes (inspired by a character’s verbal utterance) would not so much be developed as endlessly repeated and varied in ever-changing orchestral textures. In Jenůfa the technique was in its infancy and was interrupted by more conventional passages such as arias and ensembles. In his late works the alternation of blocks becomes almost continuous. The resulting fragmentation was made more complex by a well nigh constant sense of rhythmic unpredictability, caused by the fact that, in the Czech language, stress and length of syllable frequently fail to coincide, thus leading speech melodies to be based on syncopated rhythmic figures. Also characteristic is a leaner orchestral sound: there are fewer string melodies, more brass and percussion, in particular the soaring trumpet and pounding timpani known to many from the opening Fanfare of Janáček’s Sinfonietta (1926).

The vocal lines are similarly sharpened: after Jenůfa, Janáček rarely indulged in prolonged lyricism. It is easy to see how this style confused audiences at the time. For those listening with symphonic ears, waiting for recurring themes and leisurely development, Janáček seemed wilfully to thwart expectations. The physical look of his autograph manuscripts hardly inspired confidence, being full of erasures, changes of mind, inconsistencies of notation and rhythmic obscurities – further testimony to the sheer difficulty of operatic composition in the post-Wagnerian world. But Janáček’s strangeness was testament to the originality of his imagination. To bring into sound his special musical world required him to do battle with nothing less than the limitations of musical notation.

One of the reasons these late operas succeed is that their subject matter, unconventional by any earlier operatic standards, matches so well Janáček’s highly individual musical language. The Cunning Little Vixen is a fantastical tale of a Vixen (soprano) who is captured by a Forester (baritone), escapes after laying waste to his hens, finds true love with a Fox (soprano) but is at the last killed by a Poacher (bass). In the final scene, years after the Vixen’s death, the Forester reminisces, at first sadly but then with gathering fondness, about the countryside and its eternal rhythms. He thinks he sees the Vixen once more, but a Frog (child soprano) reminds him unsentimentally that generations have now come and gone. Although the scene is one of the most expansive in all Janáček, with an almost Straussian vocal climax (soaring horn and all) as the Forester celebrates the countryside in its ‘month of love’, Janáček in fact remains loyal to his technique of small musical building blocks. The relentlessly repeated orchestral ostinatos and variation points are profoundly at one with the idea of nature, endlessly prolific and teeming with life and new energy.

A very different but equally apt marriage of dramatic theme and musical technique occurs in The Makropulos Affair, based on a play by that great Czech original Karel Čapek (1890–1938). Čapek’s play is sui generis, a good illustration of the quirky literary imagination that pioneered an early form of science fiction as well as writing some of the most telling anti-Nazi propaganda of the 1930s. At the centre of the opera is a mysterious diva called Emilia Marty (soprano), who fascinates every man she meets but who treats them all with cold indifference. After many twists in the plot, she reveals herself at the end as Elina Makropulos, the daughter of a sixteenth-century alchemist who gave her a potion that prolonged her life for 300 years. The potion is now at last wearing off. Elina/Emilia, although she succeeds in rediscovering the formula and so could repeat the dose, ultimately decides not to do so. The boredom she feels, constantly seeing those she is attached to age and die, have convinced her that immortality is a curse rather than a blessing. Makropulos is in some ways Janáček’s most complex opera, since its tortuous plot, much of it unravelling incidents that happened generations before, is matched by music that creates formidable difficulties for the performers (both orchestral and vocal). As early as the Overture – which is as formal an opening as any Janáček wrote – the idea of conflict is highlighted, with fragments of lyricism in the strings battling with uncertain, hesitant ostinato figures and a recurring, dissonant offstage trumpet fanfare. Once Makropulos began to be performed extensively in the West, some lamented that ‘folk music’ – a special, operatic gift from the Slavic hinterlands – had mostly disappeared from an opera where ‘the abstract nature of the story is matched by a corresponding disinterested quality in the music’.²² But such austerity is precisely the point, for is Emilia not indifferent, and moreover unanchored in time and place? The rebarbative surface that continues through the piece – as so often with Janáček – is redeemed by a gathering focus on the principal character, and her macabre bond with a time-honoured operatic prop: the elixirs, the potions, the poisons in the cup.

SINGING POINT BLANK

Russian opera in the nineteenth century darted among the competing realisms of Musorgsky and Tchaikovsky. The first wrote his own kind of proto-Literaturoper on to the historical canvas of Boris Godunov; the second created a realist urban opera in Yevgeny Onegin (Eugene Onegin), one in which intense personal emotions are channelled via a background of salon music. Musorgsky knew little of Wagner and, like Tchaikovsky, was much more obviously influenced by Meyerbeerian grand opéra. But soon enough, as elsewhere, Wagnerism became a commanding force, albeit not the only one. The most important figure during the Russian fin-de-siècle and beyond was Nikolay Rimsky-Korsakov (1844–1908), who presided over a period in which opera developed from a sporadic entertainment, under the exclusive control of the Imperial theatres at St Petersburg (the Mariinsky) and Moscow (the Bolshoy), into a lively competitive culture involving several private companies and further venues in provincial cities.

The sheer range of subject matter among Rimsky-Korsakov’s fifteen operas underlines the absence of a stable Russian tradition – let alone a consensus about which among the competing styles of musical drama were most suitable in a Russian context. On the one hand are historical early works such as The Maid of Pskov (various versions between 1873 and 1901), which continue the manner of Boris with realistic word declamation, avoidance of conventional forms and vivid crowd scenes. At the opposite extreme are a series of fairy-tale operas, the most famous being Rimsky-Korsakov’s last opera, The Golden Cockerel (1909), in which intense individual emotions are shunned in an anticipation of the overt anti-Romanticism of Stravinsky and Prokofiev. There are also sui generis works such as Mozart and Salieri (1898), which takes the Mozartian pastiche of Tchaikovsky’s The Queen of Spades and imbues it with something close to neo-classical irony.

Scattered among these stylistic zigzags are a series of epic works that look back to the earlier nineteenth century, Russian and otherwise. At least one of them illustrates the perilous attractions of Wagnerism. The Ring was first performed by a travelling German company in St Petersburg in the late 1880s, and there were Russian-language performances, again at the Mariinsky, after the turn of the century; on both occasions Rimsky-Korsakov attended rehearsals with a score.²³ Although initially he seems to have embraced only Wagner’s orchestral practice, we can see deeper influence in a work already mentioned in Chapter 17, The Legend of the Invisible City of Kitezh (1907). This strange concoction, with a plot far too complicated for more than telegraphic summary (Exotic marauding Tartars – Divine intervention – City made invisible – Tartars foiled), is often called ‘the Russian Parsifal’ and in places sounds like the missing link between Wagner and the early Stravinsky of Firebird. But the Parsifal-like descending-fourth bell sounds of the final scene (so unlike the noisy, dissonant, Boris-like bells of so many nationalist Russian operas), and the sprinkling of leitmotifs and occasional advanced harmonies, make odd bedfellows with the modal choruses of much of the rest. Rimsky-Korsakov was uncertain how to progress: Wagner, who might have seemed a saviour, was – as so often – fatally easy to follow but impossible to emulate.

Wagner anxiety was, though, hardly a major factor in the larger picture of Rimsky-Korsakov’s eclectic operatic manners. In later life he strove above all to evade what he had earlier done so much to consolidate: the idea (unfortunately still with us, sometimes in blithely unreconstructed form) that Russian opera is profoundly different from the central European variety, and that its difference springs from some ineffable strain of national character. Near the end of his life, Rimsky was unequivocal about this:

In my opinion, a distinctly ‘Russian music’ does not exist. Both harmony and counterpoint are pan-European. Russian songs introduce into counterpoint a few new technical devices, but to form a new, unique kind of music: this they cannot do. And even the number of these devices is probably limited. Russian traits – and national traits in general – are acquired not by writing according to specific rules, but rather by removing from the common language of music those devices that are inappropriate to the Russian spirit … To create a characteristically Russian style I avoid some devices, for a Spanish style others, and for a German style others again.²⁴

Rimsky’s deflation of Russianness in music only goes so far: it still upholds the idea of a ‘Russian spirit’, even if that very concept was largely responsible for the operatic ghettoization he seems to criticize. But his prosaic dismantling of an old cliché is still worth bearing in mind, as is his casual equivalence of a ‘Russian’ style with – of all things – a ‘German’ one, which is best seen as an engagingly mischievous tilt at the increasingly common identification of German music as the ‘universal’ musical language.

Rimsky’s last years coincided with Russia’s so-called ‘Silver Age’, a period in which cultural influences from Western Europe were eagerly imported, and in which the influence of Wagner was matched by that of the symbolist movement, with Alexander Skryabin (1872–1915) a prime musical mover, albeit one who wrote no operas. When Debussy’s arch-symbolist Pelléas was first performed in Russia in 1907, it seemed to some like a homecoming. Indeed, the next twenty-five years, up to around 1935, were heady times for Russia, with the children of the Silver Age (many of them leaving the country permanently or for prolonged periods) proving influential in an emerging, pan-European musical aesthetic. At the same time, though, the tumultuous political events then unfolding inexorably led to a new period of cultural isolation for the country.

The political upheavals of these decades – the Revolutions of 1917; the ensuing civil war and Lenin’s consolidation of power; Lenin’s death in 1924; Stalin’s take-over and gradual assumption of dictatorial control during the late 1920s and early 1930s – were accompanied by constant debates about the proper place of art in the new, revolutionary society that Russia had proclaimed. Although the Mariinsky and Bolshoy were almost immediately brought under state control, with free tickets supplied to the workers, the position of opera – as ever, commonly seen as the elitist art form – was delicate. Lenin personally disapproved of an entertainment he found reeking of upper-class culture. He was also suspicious of what he saw as music’s general tendency to embellish the world in beautiful sounds, and thus to paper over the injustices and inequalities of the society from which it came.

But there was a problem. Opera proved tremendously popular in the new Russia. More embarrassing still, the crowd-pullers remained those classic revenants of the despised, Tsarist nineteenth century. Out of the immediate confusion of the Revolution emerged two competing camps, both with some measure of official support. At one extreme, those collected under the group called RAPM (the Russian Association of Proletarian Musicians) wanted genuine proletarian music, an art fit to reflect and articulate the new state. This position was constantly weakened by the fact that the proletariat continued to have stubbornly bourgeois taste, and because attempts to create revolutionary music ex novo (usually with liberal recourse to folk idioms) often sounded suspiciously like the nationalist styles of the previous century. At the other extreme was a group called ASM (the Association of Contemporary Music), who were what we would now call progressives: they wanted to reject the old, bourgeois art of the nineteenth century and enthusiastically embrace the new musical ideas coming from Western Europe. For a time, particularly in the 1920s, the progressives seemed to gain the upper hand. Their most successful composer was initially Nikolai Yakovlevich Myaskovsky (1881–1950), who was famous above all for instrumental music; in opera the key figure was Vsevolod Meyerhold (1874–1940), a theatre director who grew up within the symbolist movement and then, in the early days of the Revolution, developed radical new expressive techniques of anti-naturalism and artificiality, many of which later became associated with Brecht. At the same time, modernist operas such as Wozzeck and Jonny spielt auf were performed, and Russian film of the 1920s became internationally recognized as an aesthetic powerhouse with significant avant-garde cachet. Emerging directly out of this milieu was the precocious talent of a young composer called Dmitri Shostakovich (1906–75).

Shostakovich’s first opera, The Nose (1930), is based on an absurdist short story by Gogol and concerns a civil servant who misplaces his nose, pursues it around town in farcical episodes and is finally reunited with it. The work owes a great deal to Meyerhold, with whom Shostakovich was much involved at the time, not least in its decision to put on stage such a patently unrealistic, not to say unrealizable fantasy. Musically, it is a wild compendium of the most radical Western styles, with Stravinskian neo-classicism, a heavy dose of Wozzeck-like extreme vocal and orchestral effects, and much play with popular music idioms, particularly dance tunes. In this last regard, the debt to Weimar Zeitoper is clear. And, as in Jonny spielt auf, or indeed in Ravel’s L’Heure espagnole, among a salad of iconoclastic effects the word-setting returns to naturalistic declamation. It is a good indication that speech-opera was by this stage virtually the default mode amongst those who wished to bring the art form into the orbit of twentieth-century radicalism. In this sense, The Nose marks the closing of an historical circle. Musorgsky had been one of the first to hold to natural speech as a model for operatic vocal writing, and Debussy learned the lesson from that source. French composers writing opera after Debussy kept to the programme in a new way, adding sarcasms and exoticism, and Weimar Germany ran in tandem; and then, with Shostakovich, speech-opera returned to its country of birth.

It is no surprise that proletarian groups took extreme exception to The Nose and succeeded in getting it removed from the stage after a brief first run. Shostakovich withdrew into ballet and instrumental music, which – being non-verbal – were a lot harder to criticize. But he soon embarked on what was probably intended as an operatic compromise: an attempt to answer his critics and write a Russian opera suited to the times. The result was Lady Macbeth of Mtsensk (1934), one of the most famous operas of the twentieth century, albeit often celebrated for strange reasons. A sign of Shostakovich’s change of heart can be seen in an article he wrote in 1933. He declared there a distinctly un-Nose-like operatic aesthetic, one that could almost have come from Bellini a hundred years before: ‘In opera people don’t talk, they sing. Consequently, the text must be a singing one, it must give the composer maximum possibility for freely flowing song.’²⁵ Such pronouncements might have sounded comforting to those who were opposed to avant-garde attitudes and styles, but they sit – to say the least – oddly with the tale on which Shostakovich decided to base his opera.

Written by the composer with the help of the librettist Alexander Preys, Lady Macbeth’s libretto is based on a 1865 horror story by Nicolay Leskov. Katerina (soprano) is in a loveless, childless marriage with a rich merchant called Zinovy (tenor). They live with Zinovy’s father, Boris (bass). Zinovy departs on a business trip and Katerina starts an affair with one of the firm’s employees, Sergey (tenor). Boris discovers the affair and beats Sergey brutally. Katerina retaliates by murdering Boris, lacing his mushrooms with rat poison. Sergey moves in, but late one night, after Katerina has had a frightening vision of the ghost of Boris, Zinovy unexpectedly returns. He suspects adultery and begins to beat Katerina. Sergey, who has been hiding, intervenes; together they overcome Zinovy, Katerina strangling him and Sergey performing the coup de grâce with a heavy candlestick. They hide the body in the cellar. On Katerina’s and Sergey’s wedding day a drunkard accidentally discovers Zinovy’s body and the lovers are arrested. The final scene takes place near a bridge across a river; Katerina and Sergey are convicts en route to Siberia. Sergey has transferred his attentions to a fellow prisoner, Sonetka (contralto). He manages to trick Katerina out of her woollen stockings and gives them to his new mistress. As they all cross the river, Katerina pushes Sonetka into the freezing stream and jumps in after her. Leskov’s final image describes how Katerina ‘threw herself on Sonetka like a strong pike on a soft little perch, and neither appeared again’.²⁶

Surprising as it may seem, this operatic version had toned down Leskov’s original, in which Katerina and Sergey also finish off a young boy who turns up claiming to be Zinovy’s heir. The alteration turns out to be part of a larger shift of focus: while Leskov’s Katerina is indeed a kind of Lady Macbeth – constantly encouraging violence, driven by her sexual needs, staunchly unrepentant – the operatic Katerina is granted unique musical status, so much so that the audience can identify with her. An example comes at the end of Act 2, leading to Zinovy’s murder. The scene is Katerina’s bedroom, where she and Sergey are sleeping. The introductory music involves three tiers of string sound – a sinister low bass, a ‘halo’ of upper strings and an ascending melody emerging in the middle register. Katerina wakes Sergey and asks him to embrace her, the music building to a huge, Straussian cadence and lyrical climax as she utters her lover’s name. We seem set for an ecstatic love duet, but instead Sergey launches into a petulant rant about his lack of status as her secret lover. He too has plenty of passion and high notes, but the orchestral behaviour is markedly different. Instead of underpinning the singer’s sentiments (as it had with Katerina), the accompaniment seems blithely indifferent, with spiky rhythms on chattering woodwind, a tuba-driven bass and obsessive, mechanical ostinati. Katerina tries to calm Sergey, and the moment she sings the orchestra moves back into sympathetic mode. As she kisses him again, a beautiful, ethereally long, Mahlerian string theme takes over, the harp accompaniment emphasizing the kinship with the famous Adagietto from Mahler’s Symphony No. 5.

The scene thus far has followed what we might call an ABA form, with the extreme orchestral contrasts between A and B drawing all the sympathy to one character. The remainder of the scene, although it broadens the musical horizons, repeats this effect. Sergey goes back to sleep, and the ghost of Boris appears before Katerina. There are broad hints of ancient operatic ghosts, not least of the Commendatore of Mozart’s Don Giovanni in the enunciation of Katerina’s name to a descending octave; but the orchestra is again in overdrive. Noises are heard outside; they rouse Sergey and he hides. Zinovy enters to discover signs of dual occupation, and he and Katerina indulge in a huge shouting match (one of the great marital rows in opera: Wotan and Fricka are restrained and courteous by comparison). The orchestra returns to that mood of indifferent energy that underpinned Sergey’s rant. Zinovy, the least heroic character imaginable, is introduced by an ironic trumpet fanfare and even a hint of Rossini’s Guillaume Tell overture. The relentlessly jaunty, driving rhythms continue unabated until Sergey returns and the lovers strangle and club Zinovy to death. The scene then closes, wonderfully poised between its two contrasting musical worlds. As Sergey drags Zinovy’s body down to the cellar, a strange Mahlerian funeral march sounds forth, a strumming low bass, with solo clarinets and then bassoons (prime movers in the preceding bouts of woodwind chatter) now subdued but once or twice edging perilously close to joviality. There is a brief, shimmering pause and a string melody as Katerina asks Sergey to embrace her and tells him ‘Now you are my husband’. To close the scene, the funeral march reappears, now played on the trombone, this unlikeliest of instruments keeping the sheer uncanniness of the ending intact to the last moment. It is laughter in the dark made musical.

In Meyerholdian terms, which would later resonate in Brecht’s theories, the audience in Lady Macbeth is compelled by the music into an alternation between identification and estrangement. We identify with Katerina, but we are alienated by the uniformly hideous men in her life. This constant back-and-forth continues throughout the opera. The sense of orchestra-encouraged identification is at its boldest in Katerina’s solos, particularly her central monologue in Act 1, ‘Zherebyonok k kobïlke toropitsya’ (The foal and the filly), in which she pours out her sexual longings. The strings here sometimes come close to a conventional accompanying role, and the clarinets and other woodwind, elsewhere so cheery and impertinent, assume a melancholy, hyper-expressive function. As in Katerina’s other solos, the harmonic idiom is also stable, even though she has the unnerving habit of launching angular lines into the upper extremes of her register, sometimes painfully isolated from the orchestral stability beneath. On the other hand, the seduction scene (in many ways it is closer to rape) that ensues between Katerina and Sergey is again underpinned by farcical, manically over-energetic music, even including a series of graphic, descending trombone glissandos to depict what befalls Sergey after his energetic sexual climax.

So extreme are these contrasts that questions about musical meaning inevitably crowded in. The opera was at first wildly successful in Russia, but many in the West found the mixture of violent events and trivial music inexplicable. As late as 1960, the US composer Elliott Carter declared that ‘the relation of the music to the action is unaccountable’.²⁷ One of Russian music’s acutest contemporary critics has suggested a dark interpretation. A clue is offered by Shostakovich himself in a programme note he wrote for the first performance: ‘It would be fairest of all to say that [Katerina’s] crimes are a protest against the tenor of the life she is forced to live, against the dark and suffocating atmosphere of the merchant class in the country.’²⁸ From this it would be simple to understand the opera as a reflection of what was then Soviet orthodoxy (and critics at the time indeed saw it so): in other words, Shostakovich’s trivial music does a magnificent job of dehumanizing the ‘merchant class’ (represented by the vile Zinovy), and the heroine Katerina does her job by exterminating them.

There is little doubt that Shostakovich attempted to present his opera along these lines (there is more in the programme essay to suggest it); but, as with virtually all such political readings of operas, there are loose ends and anomalies. Most pressing is the question of why, if such a message were intended, Shostakovich chose Leskov’s tale in the first place: why set to music a story whose negative portrayal of the heroine, albeit tempered, was so unmistakable? It seems as plausible to suggest that Shostakovich was attracted to the tale, at least as he adapted it, because of its possibilities for ambiguity: for a chance to indulge his lyrical vein (recall that remark about how an operatic text ‘must give the composer maximum possibility for freely flowing song’) without falling into pastiche or – worse still – seeming un-modern. In brief, Lady Macbeth allowed him to explore further that hinterland between comedy and tragedy – the chilly space inhabited by that Mahlerian funeral march at the end of Act 2.

Whatever its political intent, the most famous incident surrounding Lady Macbeth occurred on 26 January 1936, some two years after its first performance. A revival in Moscow was attended by Stalin, together with his political henchman Vyacheslav Molotov and his culture minister, Andrey Zhdanov. Two days later an unsigned denunciation of the opera appeared in the official Soviet newspaper Pravda under the title ‘Muddle Instead of Music’. The article principally addressed the opera’s musical style, which it overtly associated with Shostakovich’s mentor, Meyerhold:

From the first minute, the listener is shocked by deliberate dissonance, by a confused stream of sounds. Snatches of melody, the beginnings of a musical phrase, are drowned, emerge again, and disappear in a grinding and squealing roar. … This music is built on the basis of rejecting opera – the same basis on which leftist art in the theatre rejects simplicity, realism, clarity of image, and the unaffected spoken word – which carries into the theatre and into music the most negative features of ‘Meyerholdism’ infinitely multiplied. … The power of good music to infect the masses has been sacrificed to a petty-bourgeois, formalist attempt to create originality through cheap clowning. It is a game of clever ingenuity that may end very badly.²⁹

This attack on an artist of Shostakovich’s international standing was unprecedented. Indeed, perhaps his success in the West was in fact a prime motive. The penultimate paragraph reads:

Lady Macbeth is having great success with bourgeois audiences abroad. Is it not because the opera is absolutely unpolitical and confusing that they praise it? Is it not explained by the fact that it tickles the perverted tastes of the bourgeoisie with its fidgety, screaming, neurotic music?³⁰

The attack may have resulted simply from Stalin’s personal disapproval, but this Pravda article sent a signal to Soviet artists, one that they would disregard at their peril. The experimental, Western-influenced movement went into retreat. Those in official favour but with progressive sentiments, such as Myaskovsky and Khachaturian, avoided opera altogether, leaving it to be peopled with earnest, song-laden pieces about Soviet achievement. Meyerhold was arrested in 1939; he was tortured, ‘confessed’ to crimes of espionage and was executed early the next year. Shostakovich made patient attempts to rehabilitate himself, famously with the Fifth Symphony (1937).

Many celebrated works followed, but Shostakovich never again attempted an opera. Works involving ‘confused streams of sounds’, speech turned into music rather than conventional operatic singing and formal tidiness, had been officially identified as perilous, possibly even fatal to their creators. In André Gide’s Retour de l’URSS (Return from the USSR, 1937), one interlocutor, prodded by Gide to account for Shostakovich’s fall, put the matter in terms that have resounded for centuries, and might just as well have been said of Rameau:

‘You see’, explained X, ‘it wasn’t at all what the public asked for; not at all the kind of thing we want nowadays. Before this he had written a very remarkable ballet which had been greatly admired.’ (He was Shostakovich, whom I had heard praised in terms usually reserved for geniuses). ‘But what is the public to do with an opera that leaves them with no tunes to hum when they come out?’³¹

Decrying a loss of melody – tunes to hum – is the universal protest of the operatic hedonist. Yet in some times, in some places, this critique could have a dark side, and did not always represent benign hankering after opera’s older forms, or their pleasures. We are left wondering how to evaluate competing losses – how to judge those who make the laments. Not all unmelodious operas, operas with speech patterns instead of song, are heroic metaphors for contemporary political defiance, as proven by Capriccio. But by 1945, opera’s great undertow – the expressive power inherent in the melodic arc, as performed by a human voice – was demanding a faith that for many composers was beginning to look blind.