Opera has always had an affinity to adultery. Secret meetings, betrayals, sudden discovery - this is the erotic stuff that unleashes shameless musical display. But, like all passions, adultery has its own history, and music has had to follow changes in the meaning of illicit love.
Up to the industrial era, adultery had an impersonal, economic side. Bastards (and, much more rarely, divorce) muddied the clear transmission of family property from generation to generation. Yet, when marriages were arranged for gain by parents, the hope that newlyweds might actually love each other proved just that - a fond hope. In reality, young couples were more likely to find love for themselves elsewhere. Practising the art of adultery in everyday life meant learning to juggle; the guilty couple needed to manage appearances. Adultery also provoked the complicity of silence within marriages. Although a spouse might well have known the other was unfaithful, an affair could be accepted only so long as it remained unspoken. Madame de Sevigne's letters reveal the arts of adultery practised sensibly among the upper classes; with equal finesse, the working class do so in Defoe's Moll Flanders.
By the end of the 18th century, however, this juggling act had started to break down; people wanted to marry for love, only love. In the world of money, inheritance meant ever less, and bastards could get work. Indeed, work was now seen as an effort of self-creation - as was family life. The Victorians easily used the phrase "founding a family", which would have sounded strangely egoistic a century before.
This sea change in society transformed the meaning of adultery: if family love is a form of self-expression, then falling out of married love implies a cataclysmic upheaval at once personal and social. One hasn't really known oneself. Adultery opens the gates to a delayed personal truth, it becomes the remedy of a mistake, and this personal discovery had to out in society - continued deception of others would mean only more lying to oneself. That's the distance travelled in the century or so between Moll Flanders and Anna Karenina.
The opera composer, no less than the novelist, had to come to terms with adultery as a form of self-discovery. In music, the distance travelled lies between Mozart's Marriage of Figaro and Wagner's Tristan and Isolde. In Figaro, true love eventually wins out; and in Mozart's CosI fan tutte, the main characters discover new desires. But Mozart's genius flourished in dramatising acts of seduction and deception, while the librettos of his operas gravitate to the theme of social appearances. (The stunning music celebrating fidelity at the end of The Magic Flute is, perhaps, the exception that proves the rule.)
In Tristan, we are thrust unashamedly into adulterous love itself, as a transforming, consuming act. Wagner dug deep into the medieval past for his story, making use of newly unearthed manuscripts of the Tristan legend. Considered just as a libretto, Tristan is a bit ludicrous: there is a magic love potion, a ship that fails to arrive in the nick of time (although Wagnerian transport is far more reliable than our own beloved Railtrack) and neatly symmetrical killings. But the music makes these stage devices powerful to the ear in a way they could never be on the page.
In this opera, written midway between his four epics of the Ring, Wagner delves deep into the possibilities of chromaticism, the half-tone steps that are the fundamentals of western music. Long melodies pour forth from narrow half-tone movements. Wagner sets this chromatic thrust against both crude harmonic shifts, such as the plagal cadences of church music, and more recherche harmonics, such as elaborated "Neapolitan-sixth" cadences - the harmonic adventure that, later, Debussy liked to use. In Tristan, the aural result no only compresses melody but also extends it through harmonic rupture.
This chromatic thrust helps Wagner the musical dramatist to achieve the intensity that transforms a creaky libretto into a compelling stage work. The great love duet in Act Two carries forward on chromatic melodic waves that are dissolved then recomposed, again and again. Although the signposts of musical memory, Wagner's leitmotifs, are used carefully in this opera, as in all his mature work, they are less important; the yet-to-be, the thrust forward, puts a different mark on the music of Tristan.
All of Wagner is long, and orchestral players like myself do occasionally get bored, losing contact with the music. But not in Tristan. For us, as for the audience, the intensity of the score makes it seem almost too short.
The crafting of Tristan's sound particularly suits its subject, adultery. The chromatic thrusting conveys to the ear the sense of transfiguration that adulterous passion displays on the stage. Some critics maintain that this opera is Wagner's version of Romeo and Juliet, which would make it a story of doomed love, but that's not quite right. Just as the music keeps evolving, so does the love between Tristan and Isolde. The philosopher Jerry Fodor aptly remarks that, by Act Three, the dying Tristan no longer needs assurances of Isolde's love - "what he wants is her permission to die". And, by the end of the opera, Isolde's love-madness has changed from the abandoned lust that came upon her when she first drank the love potion: it has become sweeter, maternal. So, too, the chromatic shifts of her "Liebestod" aria have a new harmonic context.
What marks Tristan as a work of its time is the crime that the lovers commit; the intensity of illicit love is intense because illicit - it is not "normal" to feel so much. Hans Sachs - perhaps Wagner's only truly sympathetic character, and just a man - will mock boundless love in Die Meistersinger. But in Tristan, Wagner sought to express Eros without restraint in an era when people believed in moral discipline. No transforming love without transgression, then: this was the conundrum written out in the pages of Anna Karenina, sounded out in this amazing opera.
It may be that the passions of adultery today can no longer fuel opera as an art form. At least, that possibility occurred to me when I attended the recent production of Tristan at the Royal Opera House in Covent Garden. The director, Herbert Wernicke, announced in the press his desire to make something modern of this work. On stage, he has stripped away the element of transgression by the simplest of dramatic means: Tristan and Isolde never touch. They are isolated from each other, moving up and down two ramps that are like separate bedrooms. The reason for this, Wernicke said, is that he conceives the opera to be a story of individual longing, of Sehnsucht. Desire has replaced passion, and pent-up desire commits no crime. So it is, indeed, quite modern: adultery has lost its bite.
Much modern opera production, particularly in Germany, pits director against composer - directors such as Wernicke arguing with the music, rather than appearing as the composer's servant. There is nothing wrong, in principle, with converting Tristan into an opera about masturbation rather than coition. But even with a production worthy of his gifts, Wernicke could not, I believe, have brought off this particular argument.
Wernicke, who has done brilliant work in Frankfurt and Salzburg, was poorly served by his two principal singers. Gabriele Schnaut, the Isolde, shouted throughout; Jon Frederic West, the Tristan, bawled. At least Schnaut has a good ear, and so delivered on Wagner's chromatic distinctions, while West wandered uncertainly from tone to tone. The conductor Bernard Haitink, however, performed beautifully - the sound coming from the pit was ravishing.
So was the look of the stage. Wernicke designed it like a Malevich painting, using simple planes and strong colours. He has sensed, as has the American director Robert Wilson, how closely the experience of colour complements that of hearing sound; musicians constantly talk about colouring a sound, something we intuitively understand, although have difficulty explaining. However, just in the look of the set, Wernicke began to lose his argument.
Whereas the music becomes ever more gripping, the stage, after the first shock of its visual beauty, becomes quite boring. The bold, stark colour planes remain static, while the colours of the music shimmer, dissolve and recombine. Similarly in the dramaturgy, Tristan and Isolde on their separate ramps desire each other in Act One, Act Two and Act Three. Then the curtain falls. Desire has no narrative.
My sociological self tells me that there is something socially significant about the failure of Wernicke's argument. Transgression and violation give value to the everyday world, as well as to the forbidden. Take the illicit out of social life, and you wind up with what Herbert Marcuse called a one-dimensional society; provocation becomes just a little prick of sensation, nothing more. In love, as in politics, our desires have to strain against the rules to carry weight.
I don't mean that the Victorians got it right. Adultery is no crime, although marriage ought to be more than just another affair. But it does seem to me that artists - writers, as well as composers - have trouble fashioning dramas from love, as Wernicke does, when love itself is reduced to desire. Moreover, love carries its weight because it can be dangerous, to oneself and to others. This is the element of adultery that opera has exploited, through all the twists and turns in the meaning of adultery itself. And that's why, I think, Wagner won the argument at Covent Garden.
Richard Sennett is a musician and writer. His latest book, The Corrosion of Character: the personal consequences of work in the new capitalism, is published by W W Norton (£9.95)