The story is well-known: Giselle discovers that the man she loves is in reality a prince betrothed to another woman. Devastated by grief, the young peasant girl succumbs to madness and dies. She joins the Wilis, young brides to be who have died before their nuptials and who condemn men to dance themselves to death. If this ballet, first performed in 1841, has lost nothing of its fascination over the centuries, it is particularly thanks to those bewitching winged creatures, the Wilis, dressed in tulle and on points. Anne-Marie Legrand, in charge of the Soft Dressmaking Workshop at the Palais Garnier, confides the secrets of the making of the emblematic tutus from the “white act” of Giselle.

The Soft Dressmaking Workshop (in French Atelier Flou, “flou” meaning blurred or indistinct) is dedicated to the conception of the female costumes, unlike the Tailoring Workshop, which makes the male costumes. Why these names? I couldn’t give you the exact reason. To my mind, when you look at a male costume made by the Tailoring Workshop, you notice that it has a more structured look, with fabric cut on a flat surface. For the female costumes, however, a large part of the work is done on the tailor's dummy because a pattern is not enough to work from. The fabrics are all-important and each one requires a particular approach. We have to be very reactive in our work, moulding and sculpting the fabric, particularly for the drapes. I think that’s where the term “flou” comes from, because we sculpt diaphanous fabric for women whose curves can be infinitely varied and subtle.

As head of the Soft Dressmaking Workshop, I prepare the models of the costumes. The decorators arrive at the workshops with designs that I make up in three dimensions. The designs are more or less flexible, depending on the decorators. I have to reconcile the vision of the artistic team with what we can do and especially with the constraints and particularities of dance costumes, which is our speciality. We make suggestions to the decorator and eventually the design is finalised. Then, I create a pattern which I pass on to my two workshop assistants who do the cutting out. Then they pass on the job to the nine dressmakers. We also use temporary staff when the workload is really heavy. At the moment, we’re working on a revival of the ballet Giselle as well as on two new productions so there are twenty-seven of us in the workshop!   

The costumes for Giselle are redone regularly for several reasons. Firstly, because it’s a ballet that occupies an important place in the company’s repertoire and which is often performed, in particular on foreign tours. The costumes get a lot of wear and are stocked in containers: the dancers barely have time to take them off before they are packed away, sometimes still slightly damp. Silk yellows very quickly so we have no choice but to renew the costumes.

Once the skirts and bodices have been cut out, the dressmakers get them ready for fitting. There are always two fitting sessions. At the first, the costume is not finished. Between the first and second fitting it takes five days' work to carry out the considerable job of pleating the organdy silk used for the Wilis. After the second fitting, we make the final adjustments to the bodice before we assemble it with the skirt. It is painstaking work, all done by hand, in order to fit it perfectly to the dancer's body.

There are various sorts of skirts and tutus. The type used in Giselle is what we call a “romantic tutu”. At the end of the 18^th century, with grand ballets like La Sylphide, the long skirt with several underskirts became the emblematic costume of the ballerinas. It is also known as the “Degas tutu” in reference to the painter Edgar Degas, who often took dancers as a subject for his paintings. But at the dawn of the 20^th century, the tutu was shortened, became rigid and began to be worn above the hips: the pancake tutu or English tutu was now the order of the day. This is the tutu used in Swan Lake, for example, and therefore the emblematic ballerina’s costume in the collective unconscious today.

Making the bodice and the tutu requires a considerable amount of work. One single tutu in Giselle takes 23 metres of tulle, cut into seven layers placed one on top of another. We use different types of tulle with different characteristics for each layer: first comes a stiffer tulle to structure the skirt then come layers of increasingly fine, supple tulle. The layers are gathered, pinned and stitched by hand, one by one, onto a yoke. Then we do what we call “points de bagage” : large, loose stitches that keep the layers together during performance. To make a complete costume, it takes at least sixty hours.

In the second act of Giselle, the dancers all wear romantic tutus and points, which is why it is called the “white act”. It’s the most enchanting and it’s when the plot moves into the realms of the supernatural. We are in the kingdom of the Wilis, ghosts of young women who died before their weddings. I think the tutus make an essential contribution to this unearthly atmosphere. Their whiteness seems to reflect the light of the moon, - it’s extremely beautiful. And the “unreal dance” with which they ensnare men would really lose something of its hypnotic power without the effects created by the fabric. The diaphanous quality of the tutu gives the Wilis' movements an ethereal and floating quality. In spite of the twenty metres of fabric, on stage it appears infinitely light. The romantic tutu has become an integral part of the ballet Giselle.

interviewed by Milena Mc Closkey

In the entire history of ballet, I know of nothing more perfect, more beautiful or greater than Giselle”, wrote Serge Lifar with alacrity. It is true that the popularity of Giselle has never wavered, and neither has its place among the most important creations in the history of ballet, and indeed dance in general. For whilst remaining the incarnation of a certain era, Giselle is timeless; the epitome of romantic ballet, modern re-readings of it have often sought only to render it yet more romantic.

If Giselle is rightly considered one of the summits of romantic dance, it is not only because this ballet, created around 1840 by Théophile Gautier and Jules-Henry Vernoy de Saint-Georges, is a product of its time; it is also because the many performances of it that have been given right up to our own time have revisited, accentuated and streamlined some of the elements that constitute its romanticism.

The story of Giselle was entirely inspired by German romanticism: the idea for it came to Gautier in response to a passage from the manifesto Über Deutschland (On Germany) by Heinrich Heine on the subject of Vile, creatures from German and Slavonic folklore, the ghosts of young fiancées dead before their nuptials and haunting the woods to carry off imprudent wanderers with them into the afterlife. The ballet’s plot, situated in a medieval, bucolic Germany, begins in the first act with a scene featuring folk dances whose strong “local colour” is reminiscent of certain works by Victor Hugo or Musset; and the second act, in which the Vile appear is dominated by a dreamlike, fantasmagorial atmosphere, heightened in the first production of the ballet by the decors of Charles Ciceri, “a great specialist of lighting effects, sunrises, moonlight and evocations from beyond the grave”, wrote Serge Lifar.

After a resoundingly successful period, in France and elsewhere, that continued up until the 1860s, Giselle seems to have gone out of fashion and disappeared from the bill. But the ballet was later to enjoy a renaissance in Russia, at the Mariinski Theatre in Saint Petersburg where, in 1884, 1887 and 1899, the French ballet master, Marius Petipa, presented new version of Giselle. In the course of these performances, the original libretto and the choreography were modified, notably, those elements judged to be purely decorative and not necessary to the drama were cut out.

It was in this new mould that, in 1910 and 1924, Giselle was re-exported back to France by the Russians, with memorable new productions. Deliberately modernised, the ballet radicalised certain elements that had been present initially and, as a result, it could be considered as even more romantic than the original. In Act II, for example, all the elements of daily reality, whose appearance contrasted with the ghostly presence of the Vile, were cut: the halt of the hunters at the beginning of the tableau, the confrontation between the peasants and the Vile that follows, and the arrival of Princess Bathilde at Albrecht’s side at the end (the curtain now falls on a despairing and solitary Prince). Thus the Act now belongs entirely to the Vile, nothing more disturbs the dreamlike and sepulchral atmosphere created by their presence on stage. In Act I, the “Madness Scene”, in which Giselle discovers that her love for Albrecht is impossible, was also modified: less danced, more mimed, it offered a vision of madness that was both more realistic and more dramatic.

As Lifar said, these changes contributed to a more “poetic” conception of ballet, in keeping, in his opinion, with what Gautier had wanted (Gautier had had to make a few concessions to his co-librettist Vernoy de Saint-Georges, a confirmed author of ballets orientated more towards entertainment and bourgeois drama) and closer to the spirit of German romanticism that had inspired him in the first place.

The interpreters of the role of Giselle also changed, and with them, the way the role was conceived. After Carlotta Grisi, the first Giselle, a blue-eyed blond whose appeal as a young peasant lay in her freshness and vivacity, the role was reinvented by the great Russian ballerinas who then appropriated it: Anna Pavlova, Olga Spessivtseva, mysterious brunettes who embodied a more tragic, ethereal Giselle, perfect when they mimed her madness or took on the aspect of ghosts clad in the winding sheets of the Vile.

Yesterday and today, Giselle embodies the apotheosis of Romantic ballet; but it is clear that, from one period to another, one is not speaking of entirely the same romanticism. The ballet that was performed in 1841 was of a prosaic, bucolic romanticism, still close to light entertainment, relying to a large extent on effects of local colour. In the 20^th century, it became more poetic, more absolute, with a romanticism dominated by themes of dreaming and death that, as a result, is timeless.

Inextricably bound up with the history of costume and women’s fashion, the chignon has become one of the ballerina’s characteristic features. From the technical necessities intrinsic to the evolution of dance to its elevation to the status of myth, we look back over the history of this emblematic hairstyle from the world of dance.

Although founded in 1661, the Royal Academy of Dance did not allow female dancers to appear on stage until 1681. These ballerinas, representatives of “la belle danse”, wore costumes that, from head to toe, were very similar to those of the court. Their hairstyles thus resembled those worn at Versailles: heavy powdered wigs, often adorned with crests of feathers, as can be seen in the engraving of Marie-Thérèse de Subligny in 1700 or in the sketches of costumes by Jean Bérain made between 1704 and 1726.
These cumbersome headdresses were rapidly pared down from the beginning of the 18^th century onwards. The way people danced changed, becoming, with the “narrative ballet” both more technical, requiring speed, virtuosity and ever greater freedom of movement. Marie Sallé and Marie-Anne de Camargo were the greatest exponents of this period, appearing bare-headed on stage. It was at this period that costumes became lighter and better adapted to dance.
Their aim was to avoid hampering the dancers. The fashion for Antiquity at the end of the 18^th century was to sacralise very simple costumes and hairstyles with loose hair, like that of Madeleine Guimard, shown here in an engraving.
In spite of the establishment of a wardrobe service in 1805, with an official hairdresser to ensure a form of harmony and uniformity amongst the dancers, the portraits of ballerinas that have been found from this period demonstrate that discipline was not the order of the day and each ballerina was free to arrange her hair in the manner that suited her.
Little by little, wigs disappeared entirely. At the beginning of the 19^th century, the fashion was for flat braids, both in town and on stage. Dancers, like the celebrated Fanny Elssler, helped to popularise them.
What began as a solution to a technical imperative gradually changed into the quest for an aesthetically pleasing line. In his writings on dance, Théophile Gautier, a great lover and chronicler of opera and ballet, mentions above all the dancers’ bodies, their legs and the gracefulness of their arms; the evocation and appreciation of their hair occurs only from this perspective: “Mlle Elssler should also arrange her hair to accentuate the back of her head; her hair, knotted lower, would soften the straight line of her shoulders and neck”, wrote T. Gautier in 1837 in Le Figaro¹. Dark haired dancers were also appreciated for the contrast their hair made against their white skin; again referring to Fanny Elssler, T. Gautier remarked that “her hair shines like the wing of a raven framing a mask of pale marble.²”
Thus the chignon was subsumed little by little into the attributes of the ballerina and contributed to the construction of her figure. During the 1830’s, this image crystallised completely around the character of the Sylphide interpreted by Maria Taglioni. The tutu of white gauze, the pink satin shoes and the chignon worn low, sometimes with a coronet of flowers, constituted the dancers’ uniform. The image of the ballerina includes all these elements, chignon included, and has persisted right up until today.
The romantic “white” ballets of the 19^th century thus sacralised this vision of the ballerina. Thanks to new rules in 1880, the Opera Ballet more rigidly enforced the wearing of what had become the ballerinas’ uniform, from the costume to the hairstyle. Thus, hair worn in a low bun for the now emblematic ballets like Giselle, Swan Lake and La Sylphide has been adopted by dancers from that day to this.
The model that inhabits the collective imagination was not however unanimously appreciated when it was first established. Spanish ballerinas and “bayadères” also met with great success with their long hair worn loose, a sign of sensuality that contrasted with the cold elegance of French ballerinas. Once again, Theophile Gautier’s account provides us with information as to the taste prevailing at the end of the 19^th century, notably through his eulogiums on Mlle Priora, a Spanish dancer: “Her noble and regular head forms an antique cameo. A forest of black hair (...) gives to her beauty a sort of savage accent that differs from the pale grace of ballerinas³.”
In spite of everything, this orientalist vein gave way to the queen of hairstyles that the chignon had become. If the fashion was for turbans (La Source by Arthur Saint-Léon, La Bayadère by Marius Petipa) or for ballets largely inspired by folklore (Coppélia), and even for improbable accessories (Fanny Elssler’s headdress of feathers in La Chatte métamorphosée en femme by Jean Coralli), all these elements used the chignon as a basis.
Firmly established among the accoutrements of the ballerina, the chignon has taken on more or less varied shapes over the years, above all following the rhythm of feminine fashion - the little crimped bun, placed higher or lower on the head - it has been deployed in an infinite range of possibilities without ever disappearing.
Contemporary choreographies still include this timeless hairstyle. Although ballerinas are sometimes called upon to wear their hair loose on stage, it is often to convey the intensity of a situation, as in The Rite of Spring, or the despair and abandon of a heroine killing herself as at the end of Romeo and Juliet.
Unchanged since the crystallisation of the portrait of the ballerina in the middle of the 19^th century, the chignon has imposed itself in both artistic representations of the ballerina and in the practices of dancers from their earliest age at the School of Dance.