Can I have the wrong music please? Revisiting music for ballet class

This article originally appeared in Dance Gazette (2) 2001. I’m not sure that I agree with myself any more on some aspects of f this, so perhaps it’s time for a review, 10 years on.

It would be a strange and unpopular tap-class where students learned tap to the Moonlight Sonata, or jazz dance to quadrilles.  Singers and dancers aspiring to be in Cats do not warm up for auditions with bel canto arias; b-boys do not practise headspins to Rameau. Music for training and practice in these genres is largely reflective of music for performance.

Yet when it comes to ballet, it is not uncommon to find a pianist (or a recording of one) playing a repertoire of mazurkas, waltzes, polkas, nocturnes and other light salon pieces from the 19th century in eight-bar phrases, while ballet companies are dancing in works which use a much wider variety of music:  Hans van Manen’s Five Tangos [1977] – Astor Piazzola,  new tango;  Michael Clark’s I Am Curious Orange [1988] The Fall, post-punk group; Balanchine’s Agon [1957] – Stravinsky at his most abstruse, polymetric and microrhythmic; William Forsythe’s In the Middle Somewhat Elevated [1987]  – Thom Willems hi-tech sampled percussion at extreme volume; Twyla Tharp’s “Junk Man Pas de Deux” from Known By Heart [1998] –  music played on “junk” instruments by Donald Knaack, or Balanchine’s Who Cares? [1970] lush, Gershwin songs orchestrated by Hershy Kay.

I have deliberately chosen examples from the 20th century repertoire to argue the case for using 20th music for class.  However, even the nineteenth-century repertoire is ill-served by the format of class exercises themselves. Think of the introductions to the Mazurka or Csárdás in Coppélia [1870], or the long preambles to the “Waltz of the Flowers” from The Nutcracker [1892. These have a sense of grandeur and expectation, they are portents of exciting things to come, not just four bars of “oom pah pah” to give the dancer an idea of the tempo. Very often, these introductions are at a different tempo to the music they introduce anyway.

Apart from the four bar introduction, the regular eight bar phrase is another passion-killer. Tchaikovsky’s waltzes in particular have skillfully crafted endings built-up over long time-periods, which keep the listener in a state of musical tension until the music finally – and noisily – resolves. These are the sections which end-up on the cutting-room floor because they “don’t quite fit” the exercise.  As I will argue later, doing things that don’t fit is a musical skill which needs to be trained.

The nineteenth-century ballet repertoire is not well-served either by being played on the piano. Playing a single sustained note on the piano (which is, after all, a percussion instrument)  is like throwing a stone into a pond – after an initial splash, all that is left are ripples fading into nothing. The percussive nature of the piano renders music which was written to played by stringed instruments lifeless, it cannot empathize with movement as the violin or cello does.

Nor does the piano do a very good job of representing the orchestra at its loudest and most exciting.  The music for the coda of Tchaikovsky’s “Black Swan” pas de deux from Swan Lake, for example, is very loud, with a drum beat in quality and prominence not dissimilar to a lot of rock music.  It is fast, loud and vulgar in the very best sense of the word.   However hard you hit a piano, you cannot make the sound of bass drums, cymbal crashes and triangles.

How much does the way that music is used in a ballet class really train students in the musical skills that dancers need?  Teachers and pianists expend time and energy on finding the “right” music for class, by which is understood eight-bar phrases, four bar introductions, suitable rhythm, appropriate tempo,  music for balancing, music for stretching, music for révérences,  music that ‘makes’ you land on one, jump on one, plié on six and so on.  Many of these musical choices, however, are predicated on assumptions about the relationship between dance and music which are simply not borne out by practice.

In performance, for example, a dancer frequently needs to be able to hear, but not listen to the music. What may be musically a quiet, peaceful ending may well be choreographically the cue for a dancer in the wings to run on stage and begin a virtuoso solo. Similarly, in a choreographic canon -  as indeed with a musical one -  it is important to be aware of one’s position in time relative to the previous person or group, not in relation to the absolute passage of musical time.  And of course, if a murderer arrives on stage six counts late, only the most obsessive dancer would insist on dying before he had been shot for the sake of musical  correctness.

Close inspection of the relationship between choreography and music does not always indicate the felicitous correspondence between, say, melody and movement that is implied by the exhortation “listen to the music!”.  The opening of the “Kingdom of the Shades” from La Bayadère [1877], for example, requires the dancer to move independently of and at counterpoint to the rhythm and shape of the melody.  Hans van Manen, coaching the Fifth Tango in his Five Tangos, told dancers at City Ballet of London “don’t move until you have to”[1]: to achieve the musical effect he wanted, dancers had to effectively be ‘late’.  It doesn’t look late – in fact, it looks extremely musical – but this musicality is gained by ignoring the music, not listening to it.

If dancers sometimes have to ignore the music, there are also occasions when the music ignores them.  In a 1958 film[2] of Le Corsaire [1856], Alla Sizova ends the female variation from with the following:

Grand fouetté relevé into attitude derrière en croisé,
Relevé à la seconde
into double pirouette en dedans [arms 5th position] (repeat 3 times)
Grand fouetté relevé into attitude derrière en croisé Relevé à la seconde, plié in 5th position, relevé in 5th position, arms 5th position[3].

- and all this to Anton Simon’s dreamy, sentimental waltz Souvenir du Ball. There is almost nothing in this music which supports or gives impetus to the movements, nothing which hints at the effort required: it is soft, lyrical and practically without accent.  This is not to claim that the music is inappropriate – on the contrary, it is the ‘effortless’ music which helps to make Sizova’s performance look effortless too. Les Sylphides [1909] is also full of examples of “big” jumps done to “small” music.

An extreme instance, perhaps, of “not listening” is to be found in Neumeier’s Ondine [1994], in which dancers initially learn one section to a recording of Thomas Hampson singing Begin the Beguine, but only later in the rehearsal period perform it to Hans Werner-Henze’s dense, atonal, modernist score of 1958.  Neumeier’s aim was for the audience to see the dancers embodying the lush, warm elegance of Cole Porter that they had rehearsed to, while they heard Henze in full angst.

And what about the skill of not moving to music?  As any corps de ballet member who has danced in the second act of Swan Lake will attest, there is a lot more to standing still to music than just standing still.  Similarly, during the pas de deux of Les Sylphides, the transition from stillness to movement in the changes of the corps de ballet require a dancer not only to know precisely when to move, but also how to move.  While some may argue that this is simply stagecraft or choreography rather than a musical skill, “when” and “how” are fundamentally musical questions.  Coming in at the wrong time with an inappropriate quality is the essence of musical slapstick; it is the basis some of the funniest sequences in Jerome Robbins’ The Concert [1956].

Then there is the skill of moving to no music at all. Many pianists, faced with the task of accompanying pas courus or running steps, will hear the word “run”, and play lots of fast notes high up on the keyboard. Yet dancers frequently have to run in silence – at the beginning or end of a solo for example.

When running is accompanied by music in ballets, there is rarely a like-for-like correspondence between “running notes” and “running feet”.  If the music can be assumed to bear any relationship at all to running, it may perhaps highlight the motivation of running  – anger, despair, madness, love, fear, for example, or hint at the goal of running  – a lover, a murder victim, an elusive sylph.  In perhaps the most famous instance of pas courus – Myrtha’s entrance in Act II of Giselle [1841]the music yields almost no sense of tempo, rhythm or metre at all: nothing but ethereal high strings accompanied by the harp in an extremely slow six that almost defies recognition.

Perhaps the most important musical skill of all, though, is to be able to adjust to differences in tempo.  Andria Hall, ballet mistress at the Deutsche Oper Berlin, would frequently ask the pianist “can we have it at the wrong speed now please?”   Although in an ideal world, conductors would attend rehearsals frequently, would get to know their dancers well and remain in constructive dialogue with them during the performance period, the reality is all but ideal, and dancers should be prepared for the just-tolerable, if not the worst.

Always finding the “right music” for class can be like giving an alcoholic another drink – it keeps one problem at bay while masking other more important ones. That is to say, at the very point at which the dancer’s musical skills might be developed by being forced to deal with a musical problem (such as the wrong tempo, music which seems inappropriate to the movement, music which is not in eight-bar phrases – or, indeed, not having any music at all), teachers and pianists are often tempted to try and make the problem disappear by “correcting” the music.

When the purpose of class is to prepare for a rehearsal, or to warm-up for a performance, ‘four-in, four phrases of eight’ is a very valid and useful shared model which helps to make class run smoothly.  Within this structure, there is the possibility of drawing students’ attention to music, by working against the music rather than with it.  Suki Schorer’s account of Balanchine’s use of music in class in this way gives some excellent examples of how this can be achieved. [4]

If ballet classes are to distinguish themselves significantly from aerobics and educate students musically, though, there is no better place to start than with occasionally asking for the ‘wrong’ music – too fast, too slow, too loud, too quiet, too irregular.  It is deviation from the norm that makes music interesting;  it is how dancers and  choreographers deal with music that makes ballet interesting.  Having the “wrong” music for class is not a mistake, it is a place where the art of music and ballet meet.


[1] reported by Christopher Hampson, ballet master of City Ballet of London, circa April 2000.

[2] Matheson, J. Simons, C. (Prod.) The Glory of the Kirov. London: NVC Arts, 1996.

[3] My thanks to Lynn Wallis for transcribing these steps.

[4] Schorer, S. (1999) Suki Schorer on Balanchine Technique. New York: Knopf

© 2001 Jonathan Still