New classics for old. (Attitudes).

ONE OF THE BEAUTIES OF DANCE IS ITS STRANGE IMPERMANENCE--IT MOVES LIKE THE WIND, LEAVING RESULTS BUT NO LASTING TRACES OF ITSELF. YOU GO TO A CONCERT AND HEAR A SYMPHONY. ON YOUR RETURN HOME you can, if musically adept, read a score, and certainly you can play a recording of that same symphony. The interpretation almost certainly will differ, but the main text of the symphony will remain the same. And your ears will function in the same way.

OK, now you go to see Swan Lake. You come home. Even a professional dance notator is unlikely to have a score of that particular production. And you put on a video, although again the actual production, and I'm not talking about the interpretation here, is unlikely to be the same. Finally, your eyes will not function in the same way--the video is smaller and two-dimensional. Moreover, the impact of actuality is much more potent in the theater than in the concert hall.

So how can we even talk about ballet classics? It's not easy. There is only one eighteenth-century ballet extant, Vincenzo Galeotti's The Whims of Cupid and the Ballet Master from 1786 and still officially in the repertoire of the Royal Danish Ballet. When we look at the nineteenth century, apart from a nice handful of Bournonville ballets, also preserved in something like aspic by the loyal and lonely Danes, the situation is murky.

We can be reasonably certain--through Stepanov notation and living tradition--of the original choreography of Petipa's The Sleeping Beauty (1890) and the Petipa/Ivanov Swan Lake (1895). Large chunks of Petipa's La Bayadere (1877) are still with us, although the production we know is an amendment by Agrippina Vaganova in 1932. All of the male dancing was later rechoreographed, gloriously as it happened, by Vakhtang Chabukiani for Vladimir Ponomarev's 1941 Kirov production, with the Golden Idol dance then added by Nikolai Zubkovsky.

The Coppelia we know best is not the 1870 Arthur Saint-Leon choreography (although the Paris Opera maintained something approximating it until the late 1940s), but stagings often based on Petipa's St. Petersburg version of 1884. As for Giselle, that touchstone of Romantic ballet, the renditions we know today all stem from Petipa's final St. Petersburg production of 1884, which did, there is no doubt, retain strong elements of the 1841 Paris original with its choreography by Jean Coralli and Jules Perrot.

That is pretty much our classic (pre-twentieth-century) heritage. Yes, probably a fair part of Le Corsaire is still Petipa, from his 1899 production; parts of the 1898 Raymonda, one act of his 1881 recension of Paquita, and something of his 1866 Esmeralda all survive--as must, here and there, the Ivanov Nutcracker, which was given in Britain (at least the second act) as late as 1951. The early Danish repertoire is undoubtedly the best preserved, although even here a certain erosion has occurred in the last fifty years.

At the beginning of the twentieth century, largely with the choreographer Mikhail Fokine and the impresario Serge Diaghilev, a sea change took place, and choreography was more or less regarded as sacrosanct, linked almost inextricably to its score. Other versions may exist--but as completely new choreography, such as Balanchine's The Firebird or Robbins's Afternoon of a Faun or Les Noces. The other day I even saw a version of Copland's score for Appalachian Spring without Graham's choreography, but that was just a sad aberration.

So our main concern in keeping ballet classics must be with the tattered fragments we have fitfully secured from the nineteenth century. This is our tangible past in theatrical dance, our heritage and antiquity. It's not much. So should we play around with it? Surprisingly, I think, with reservations, the answer is probably yes. We should obviously preserve most zealously what we have, but this does not prevent concurrent attempts at renovation.

The issue was brought home to me by Sylvie Guillem's production of Giselle, first produced by the Finnish National Ballet, and presented in Los Angeles and New York by La Scala Ballet of Milan. Now this was not the kind of total revisionism shown by the likes of Mats Ek, whose production ends with Giselle and Albrecht stark naked in a lunatic asylum. However, it was still a far cry from the Giselle we know and love, which has been changed very little since the beginning of the twentieth century.

Now, Guillem has retained, more or less, the original scenario, and throughout has been scrupulous to include what you could call movement motifs from the accepted choreography, so all the time you are seeing dance fragments that seem vaguely familiar if often differently assembled. Where Guillem has been particular is in emphasizing the story's dramatic aspects and trying to accent them more realistically.

Indeed, her first scene has the air of Pietro Mascagni's opera Cavalleria Rusticana in its verismo account of peasant life, while her second act (far more successful to my mind) with its disc-like playing area, druidic stones, and swirling mist suggested something out of Wagner. But the care for details--as a typical touch, each of those betrayed brides, the Wilis, had a slightly different wedding dress to haunt in--and the overall attention to Romantic style made this a remarkably effective Giselle. Indeed, I loved it, but not, I caution, for everyday use.

Anna Kisselgoff made a shrewd comment when she likened Guillem's approach to that of Peter Brook in his own revolutionary productions of Shakespeare's A Midsummer Night's Dream and Bizet's opera Carmen. But Brook's radical methods did not change the crucial text of either the play or the abbreviated opera. And if you agree, as I think we should, that the basic text of a ballet is its choreography, then Guillem has gone a lot further than Brook ever did. Yet, if we accept it for what it is, this new window on Giselle is not only enjoyable, but could have a salutary effect, in its dramatic relevance and resonance, on all those traditional productions of Giselle that I trust and presume we will continue to hug to the bosom of our classic repertoire.

Senior Consulting Editor Clive Barnes, who covers dance and theater for the New York Post, has contributed to Dance Magazine since 1956.