Nijinsky and the ballet detectives

A husband-and-wife team are recreating the legendary dancer's lost work

Actors and musicians don't know how lucky they are. For them, reviving the creations of the past is a matter of breathing life on to a page. Dancers rarely have that luxury. Until as recently as 40 years ago, once a ballet dropped out of a company's repertory, it simply ceased to exist.

Yet now, the Royal Ballet is about to present a work which became a ticket-stub of history after just five performances in 1913. The ballet is Vaslav Nijinsky's Jeux, one of four he made for the legendary Ballets Russes in Paris. And its recreation is the result of several years' graft by a husband-and-wife team whose working method is part detective work, part inspired guess.

Before Millicent Hodson and Kenneth Archer turned their cultural spyglass on this 18-minute piece, it amounted to little more than a rather fine Debussy score and a biographical footnote. None of its choreography had been written down, and almost everyone who had ever seen it was dead. "Never was so much artistic fame based on so little artistic evidence," wrote the American critic Joan Acocella in her introduction to Nijinsky's diaries published last year.

Vaslav Nijinsky was ballet's first male superstar. Women were said to swoon over the very thought of his famous jump; his sexual appetites were the talk of Edwardian dinner tables; a top racehorse borrowed his name. More arguably, he was also the 20th century's first great choreographer.

Yet ballet lovers today have little idea how his ground-breaking dances looked. The exception is L'Aprÿs-Midi d'un Faune, which remained in fairly constant currency after the demise of the Ballets Russes, thanks to its famous Debussy score, and which reappears alongside Jeux on the Royal Ballet's bill. The Rite of Spring, which faded from view not long after its notoriously violent reception two weeks after the premiÿre of Jeux, got the Hodson and Archer treatment in 1987.

On the face of it, there wasn't a lot to go on with Jeux. Yet it seems to have been the very lack of obvious signposts that spurred the pair to put it back on the map. "It's a process of excavation," explains Hodson, snatching a break from rehearsals. "Or like assembling a complex machine with no instruction manual," puts in Archer, who is given to prowling about the studio with a video camera.

"We start by gathering as much information as we can," says Hodson - production photographs, sketches by contemporary artists, sketches by the ballets' designers, first-night reviews ..." She prods a bulky ring-binder of pencilled diagrams and cuttings - the mobile reference library she and Archer keep by them at all times. They have also spoken at length to relatives and protégés of the people involved "though that part gets harder as the years go by". It was easier, she says, back in the 1980s, when they were still able to speak to veterans like Marie Rambert, once directly involved in the Ballets Russes.

They also managed to track down the original bassoonist of the orchestra, a nonagenarian who, though vague about what happened on stage - "since he was always looking in the wrong direction" - was able to give a vivid idea of the atmosphere.

But a scrapbook, a musical score and the right ambience do not make a stageable dance, even one as fragmentedly modernist as Jeux is believed to have been. Nijinsky called it a poÿme dansée; others ballet cinématographique, detecting in its splicing of free images the techniques of early film.

This absence of continuous development was clearly a boon to the reconstruction. But there must have been times, surely, when Hodson and Archer found gaps in their material, and had to make bits up? "Rather than invention, we call it intervention," says Hodson. "Having studied all four of Nijinsky's ballets now, as well as those made by his sister, Bronislava, who was very close to him, we feel we know what means he was likely to use to get from A to B to C." Sometimes, she says, it's a case of turning the principles of design into movement. "In Jeux, it's a triangle, that keeps bringing the dancers into three-way formations and three-way divisions of the body: the head facing one way, the hips another, the feet changing. Other times the ground pattern of the stage design - six circular flowerbeds making a triangular labyrinth - makes a decision for you."

The scenario was easy to establish, since Nijinsky had his own poem, in French, printed over the piano score. Three tennis players in white are in a park at dusk, and ostensibly the theme is the sensuality of sport. But there is a mind-game being played, too. Hodson and Archer believe that the characters were based on Duncan Grant and the two Stephen sisters (later Vanessa Bell and Virginia Woolf), whom Nijinsky once saw playing tennis together in Bloomsbury. He clearly detected the three-way sexual attraction that was going on there, they say, and fed those hints of yearning and jealousy into this ballet.

In brute terms of audience value-for-money, is it worth it? Why not take the music and invent a new ballet? Hodson and Archer have clearly chewed over this question many times. "We believe," she says with mission-statement firmness, "that this is a great part of the 20th century's aesthetic. It's like Picasso's Les Demoiselles d'Avignon. For years that picture was thought insignificant, and kept turned against the wall. Now it's seen as the turning point of 20th-century art. What if it hadn't survived? Nijinsky's choreography is the same."