Notes Made in the Theatre and in the Dark: An exploration of the methods used in ballet documentation and reconstruction by Adelaide Robinson

Notes Made in the Theatre and in the Dark:
An exploration of the methods used in ballet documentation and reconstruction

Adelaide Robinson, Department of Library & Information Science, City, University of London

This text was written by Adelaide as a student assignment, in fulfilment of the requirements set for the Independent Study module at CityLIS, 2016/17.

Follow Adelaide Robinson on Twitter: @adafrobinson



The documentation of dance is a complicated field. As with any kind of live performance, there is a strong divide between what is technically possible to record, and what is possible to be reconstructed from those records. Not only is there a technical difficulty in recording a live performance – with video, written accounts, notation, or otherwise – but some artists believe that to try and document something in an inherently ephemeral medium goes against the point of their art. Therefore dance and theatre historians face challenges on many levels when trying to find documents and recreate performances for a new audience. Theatrical researchers usually have the luxury of scripts, scores, and written accounts from directors and actors, but those looking for ballet documentation have to work around the difficulty that is writing down a narrative with no words.

While documentation can be read and studied satisfactorily by those who simply have an interest in ballet, the job of reconstructing a past performance relies on these often fragmented and conflicting records. Methods of documentation in balletic dance history differ wildly from ballet to ballet: records can vary from informal sketches of dancers to highly complex sheets of notation. (Note; for the purposes of this essay, “dance” will refer to the European balletic tradition.) For some ballets we have no record of the production whatsoever, other than a name. In the modern, digital age, video cameras and new computer software – capable of recognising, recording, and storing movement – may or may not usher in a new era of ballet that will remain perfectly preserved for future generations. But whether or not even this will allow for a genuine reconstruction of any performance from the past remains to be seen. Using the various reconstruction and recreation efforts associated with Nijinsky’s Le Sacre du Printemps as a case study, this paper aims to explore the most common methods used throughout the history of dance documentation for ballet, and evaluate their effectiveness when it comes to reconstruction while keeping the following questions in mind; how and why has dance documentation evolved? And if the complete reconstruction of a ballet is impossible, then what is the point of documentation?

Dance Notation

In a 1986 review of recent developments in European dance history, Meredith Little presented the following argument.

“Studies in dance history may be boring to read because the material presented refers to nothing in our own experience; we must invent our own examples in order to make any sense out of it at all. How interesting would studies of Shakespeare be if, at best, only fragments of plays had survived?”[1]

While more standardised methods of notation and the increasing accessibility of filmed performances have proliferated in the ballet world in recent years, making reconstruction an easier task for future directors, reconstructors at present still have to work with these older “fragments” that the previous generation of choreographers left behind.

Dance notation is older than one might expect; European dance notation is generally agreed to have started with Pierre Beauchamp-Feuillet’s system of recording Baroque dance, (1700, in his work Choréographie), which was commissioned by Louis XIV. (It should be noted that other systems of dance notation were used before this publication, but in much the same way that Shakespeare is credited with inventing words as he recorded contemporary colloquial speech, Beauchamp codified existing systems into the first standardised notation system. It is also seen as especially relevant for being authorised by the French monarch.)[2] Dance notation then evolved through various forms and off-shoots devised by choreographers and dancers with different needs to fulfil. This diversity, while fascinating to study, does not lend itself to easy reconstruction of pre-modern ballet.

The most common and popular form of notation used today is Benesh, devised by Rudolph and Joan Benesh (1950s), with Labanotation (1928) a close second. Many directors now employ a notation expert, or “notator”, to work alongside choreographers and ensure that new productions are recorded to the full extent of their ability to be so. Notation experts are now often credited in programmes, as well as the original authors of notation scores, showing that notation has become something well-used and of value in the modern ballet world.

Some dance history scholars have expressed concerns about the growing dependence on notation scores in the current ballet world. Judy van Zile – professor of dance at the University of Hawai’i – argues that depending on the services of a professional notator diminishes the score’s accuracy, as the dance is being recorded by someone other than the person who created (or is dancing) it. This brings up several questions; most importantly, can a separate party accurately record a dance?[3] For the companies putting on popular ballets, this is not a problem; there is always someone available to teach who has danced Juliet before, for example, and this first-hand experience improves the process. For reconstructors of older and less well-documented ballets, trusting one person’s notes – a person who they may never be able to meet or ask questions of – can be a difficult, almost impossible experience. Dance notation is also not easy to read. As Meredith Little puts it, “researchers are finding that each notation system is a language with its own syntax; symbols have different meanings in different contact, and one must translate phrases, sections and whole pieces as well as individual steps and step units.”[4] If a company is without a skilled notation expert, possessing complex scores may be of little value to them.

Pictured: a section from the Sunday 18 March 2017 programme for “The Human Seasons/After The Rain/Flight Pattern” at the Royal Opera House. “Flight Pattern” premiered on 16 March 2017, and the Benesh notator for the production was Gregory Mislin.[5]

Documentation in the Digital Age

As documentation in the library and information sector in the digital age has become increasingly computerised, and many documents are now ‘digital-born’, so has the world of dance documentation made tentative steps towards the same. As Judith Gray put it: “Of all the arts, dance would seem the least likely to accede to the vagaries of rapid change and the relentless advances of this modern technology,” yet computer software has become far more prevalent in the field of dance notation recently.[6] Janos Fügedi’s 1998 review of the application available for recording and preserving dance, while dated, still gives us an insightful look at the potential for these types of software. Computer applications for dance generally include the following outputs; computer animations based on dancers’ movements, born-digital dance notation, and databases of scores. Fügedi has a positive outlook on computer applications, stating that dance research can gain “great benefits” from computer applications in the field and ends his review on the hopeful note that “the connection of local dance notation collections to the Internet (…) will open a new horizon for factual, documented dance based comparative research.”[7] Unfortunately, the majority of dance notation collections are still only available in physical libraries, with computer applications reserved for those with a higher knowledge level of programming and better resources for the more intense processes such as motion capture.

Putting a greater focus on dance documentation in the digital era has already changed how ballet is produced and shared. For example, the copyrighting of a live performance has been especially difficult for choreographers in the past, as original choreography needs to be “fixed” before it can be put under copyright: the performance is not enough. With the proliferation of better notation systems and the introduction of video cameras to most theatres and dance studios, choreographers now have much better access to enforcing copyright of their original works.[8] Dance documentation also looks to become even more digitally-minded in the future. When computer software for dance notation was new, the hope was that it would lead to a program that would deliver an ‘enhanced’ score, with video and music to accompany the notation. As described by Wilmer and Resende, “in this way a student would simultaneously (1) hear the music of the ballet, (2) have access to the music score, (3) watch a performance of the ballet, (4) see the dance score, (5) see illustrations of each position or movement of the dancers body and (6) find out the name of each position or movement.”[9] Although there is no such definitive program yet, dancers and choreographers do now have much easier access to filmed performances, and the advance of motion capture and animation in documentation software enables a greater understanding of recorded movement than a simple, printed dance score.

Case Study: Le Sacre du Printemps (The Rite of Spring)

Le Sacre occupies a special place in the history of dance reconstruction and recreation. There are over a hundred different productions associated with Stravinsky’s original score.[10] While most of these are completely different dancers, some have tried to recreate Nijinsky’s original choreography; using sketches, eyewitness accounts, and reviews. The lure of the Rite of Spring most likely stems from the fantastical stories of its first performance on May 29, 1913 in Paris. The combination of Stravinsky’s intense, overwhelming score, and Nijinsky’s modern choreography produced a hostile environment for the audience and led to a riot. The performance ran for only six more nights in its original run. The scandal of “the riot of spring” has been a key ingredient of its popularity ever since.

The difficulty of reconstructing Nijinsky’s Sacre has been part of an enduring myth of “the lost masterpiece”. Years after the ballet was dropped from the repertoire, by the director of the Ballet Russes, Serge Diaghilev, it was put up for revival by the same man. According to reports at the time, “apparently no one could remember the movement, even though some of the original cast were still performing with the company.”[11] A revival was in fact staged, but was unsuccessful and used practically new choreography. The most true to life reconstruction of the Sacre was researched and written by Millicent Hodson and Kenneth Archer, and staged by the Joffrey Ballet in 1987. It is the methods of dance documentation used for this reconstruction, and the accounts of Hodson and Archer themselves on the process, which will make up the largest portion of this case study. The fragments available to them demonstrated many different varieties of dance documentation; from sketches, to notation, to eyewitness accounts written down in interviews and diaries.

There are almost no records of any notation for Le Sacre. As the production was unlike any ballet that had come before it, there was a difficulty in fitting the expressive new movements in Nijinsky’s choreography into established notation systems. Nijinsky himself seemed to be frustrated with notation, having been quoted as saying:

“I am forced to cry for a ‘partition of movements’ where to place my instruments – which are the human bodies – in a manner that is in absolute accordance with a white canvas for Bakst or group of islands for Debussy. My composition is even less simple because the human body does not possess just 4 strings but an infinite multitude of sensitive and expressive elements.”[12] (Quoted in Hector Cahusac’s, “Debussy et Nijinsky,” 14 May 1913.)

However, although no score from Nijinsky has been found, there was a rumour that he intended to produce one. In Hodson’s preface to the reconstructed score, she quotes a contemporary of Nijinsky, as having heard the choreographer say:

“I have invented symbols to represent the dancers. The note on the stave represents the head, their gestures are indicated by stylised attitudes. I have transcribed the Sacre and intend to transcribe all of (my ballets)… And in ten, twenty, a hundred years, they will be able to dance these ballets as they danced them today.”[13]

The reconstruction process would certainly have been easier if Nijinsky had followed through on this statement.

Rarely does a reconstructor have to rely on a score (whole or otherwise) alone; the most successful projects have had access to a multitude of different, less official records to help the process. While notation systems are useful and complex enough to record the more complicated aspects of movement, it is important to not discount other, more informal forms of records; such as sketches, written accounts from company members and other involved peoples, reviews, programs, costumes, and other such souvenirs and mementoes associated with attending a dance. Much of the information we have on ballets lost to history is due to reviews. As documents, reviews can provide dance historians and reconstructors with a wealth of important information. An example of this can be seen below:

 Pictured: an illustration accompanying a review of the 1921 run of Le Sacre du Printemps in ‘The Sketch’. [14]

This illustration, although meant to provide humorous insights for the review, actually contains a wealth of useful information about the 1921 revival. Firstly, it gives us the name of the principal dancer, Lydia Sokolova, which is useful as the names of individual dancers were not always recorded in early 20th century programmes. There are the sketched out costumes, which provide valuable reference points for costume reconstructors; showing clothing, accessories, the weight and hang of the material, and the fact that the dancers appear not to wear pointe shoes. Most importantly, in the illustrations and in the jokes in the captions, we find useful descriptions of the actual movements of the dance. While ‘oranges and lemons’ is meant to be a humorous commentary, it and the accompanying illustration tells a reconstructor that a movement occurs in the dance where a dancer moves through two lines of parallel dancers with their hands linked. The series of illustrations above the main picture provides an even clearer representation of a particular movement, sketched out in stages. While this document was not intended to be a record of this kind, documents such as this can be invaluable to a reconstructor. Written accounts in reviews are also useful. However, reviews are by their very nature biased, and do not present an objective record of a performance. Reviews in papers can be sensationalised to sell more copies of a publication. This may be a problem in the case of “the riot of spring”, where different accounts of Le Sacre’s first performance may be unreliable due to being repeated and embellished in a ‘Chinese whispers’ style.

In a place where performance studies and fan studies intersect, documentation of any live performance owes much to the devotion of its audience. Many sketches of the original 1913 Sacre have been found and used in reconstruction attempts. The most influential were drawn by Valentine Hugo (née Gross); an illustrator, painter, and ballet enthusiast whose works can now be found in the Valentine Gross Archive in the Victoria and Albert museum’s Department of Theatre and Performance.[15] Valentine Hugo, a devoted follower of the Ballet Russes, made many sketches of their rehearsals and performances between 1910 to 1914. Her motivations and process were described by Richard Buckle in the following quote, presented in the introduction of a book of Hugo’s sketches:

“The motive, of course, behind the tireless jottings of Valentine Gross, was to record a series of uplifting theatrical experiences (…). Her notes made in the theatre and in the dark could be caught unconscious, because she did not know what or how she was drawing. Together with the scribbled name of a dancer or the collar of a costume they were an aide-mémoire which might turn out to be legible and helpful, or as happened in a number of cases, might not. She would not consider these notes as drawings and would probably shrink from showing them to anyone else.”[16]

Pictured: Page of a sketch book showing blue crayon preliminary sketch made in Théâtre de Champs-Elysées during rehearsal or performance of Le Sacre du Printemps, Diaghilev Ballets Russes, 1913. Sketch by Valentine Gross.[17]

These informal forms of documentation should not be discounted; in many ways, Valentine Hugo’s “scribbled” sketches may have been of more use to reconstructors than any attempts at notation the choreographer tried to make. Drawings from a passionate fan can record more emotional context than even the most complex of notation systems. Hodson and Archer’s reconstruction of Le Sacre du Printemps hinged on two important documents; Stravinsky’s musical score, on which he had written descriptions of some movements above the stage, and Nijinsky’s contemporary Marie Rambert’s recreation of the score based on Dalcroze eurythmics, which was recovered after her death in 1982.[18] Despite this, it may have been the ‘unofficial’ documents that were the most useful to the reconstruction efforts.


If a perfect reproduction of a live performance is impossible, then what is the point of dance documentation? Many artists shy away from documenting, (specifically filming) their work, arguing that to document a live performance is to go against its inherently ephemeral nature. However, Renée Conroy argues that the purpose of documentation is vital in order “to enable today’s dancers to have a more robust kinaesthetic understanding of the works that pave the way for contemporary choreographic masters”.[19] By having more access to documents that record dances and dancers in their historical context, choreographers will be able to study the technical aspects of dance in a way that they may have not previously explored. Having a stronger focus on historical records of dance; including notation scores, photographs and pictures, and notes, could improve the choreographic process for many a company, and bringing dance documentation out of shoeboxes and old journals could improve the overall standing of dance history. Had the sketches of Valentine Gross been ignored, for example, a wealth of essential information on the Sacre, and several others of Nijinsky’s ballets, would have been lost. In his introduction to her works, Buckle praises her efforts: “dancers, choreographers and historians must for ever be grateful to her for the pains she took.”[20] While most methods of dance documentation – even film – are arguably unsuccessful in completely capturing a performance, documentation should still be treated as a vital part of the production process. Without it, many productions would be lost to history.

There is also the argument, as Sarah Rubidge puts it, that “works are artistically valuable in themselves, and should be made available to contemporary audiences for their intrinsic, rather than solely for their historical, value.”[21] Ballet enthusiasts should be able to relive the productions of the past in as much detail as possible; not just to learn, but to enjoy. If dance documentation continues to evolve in the digital age, those wanting to re-experience a production should not have to analyse sketches and score fragments, but instead should be able to watch the performance from their own homes or their local theatre and performance collections. With virtual reality making its way into the arts, balletomanes may even have the chance to relive ballets in a totally immersive experience. While many companies have put up 360 degree videos of rehearsals and behind-the-scenes featurettes before, the Dutch National Ballet were the first company to premiere a ballet that was created and produced for virtual reality. The show, ‘Night Fall’ was created in conjunction with the Samsung virtual reality department, premiered with free access online on World Ballet Day in 2016, and was also available to watch at the VR cinema in Amsterdam on the weekend of the 27th of August. This ground-breaking event highlights the third reason that dance should be documented; for posterity, for the dance itself, and also for the wider opening of access to the arts. Richard Heideman, press manager for the Dutch National Opera & Ballet, told Digital Trends that “it is also meant to see if we can reach a new audience with this project. We intended to reach out to people who would normally not buy a ticket to a theater or ballet performance, but are willing to try the VR project — and we hope it gets them inspired and excited to also try the live on stage experience one day.”[22]

Not only is documentation essential for reconstructing the ballets of the past, but it should also be used to introduce more people to the ballet of the present. Valentine Gross’s sketches were sold to tourists and ballet lovers in programmes as Le Sacre du Printemps played in Paris. These pictures were no doubt regarded fondly by those who never made it to the original, seven-show run, as they are regarded fondly by dance historians and enthusiasts today. As dance documentation evolves further into the digital age, it should continue make wider access to live performance art a priority.


Anderson, Jack. “THE JOFFREY BALLET RESTORES NIJINSKY’S ‘RITE OF SPRING.’” The New York Times. October 25, 1987.

Conroy, Renee. “Dancework Reconstruction: Kinesthetic Preservation or Danceworld Kitsch?” American Society for Aesthetics, 2016, 5–9.

David, Irving. “Choreography and Copyright – Make the Right Moves.” Dance UK, 2012.

Dormehl, Luke. “The World’s First Virtual Reality Ballet Experience.” Digital Trends, 2016.

Fügedi, János. “Computer Applications in the Field of Dance Notation.” Studia Musicologica Academiae Scientiarum Hungaricae 2/4, no. 39 (1998): 421–41.

Gray, Judith A. “Dance Technology: Current Applications and Future Trends.” In The Evolution of Dance Technology. The American Alliance for Health, Physical Educations, Recreation and Dance, 1989.

Hodson, Millicent. Nijisnky’s Crime Against Grace: Reconstruction Score of the Original Choreography for Le Sacre Du Printemps. Pendragon Press, 1996.

Hugo, Valentine. Nijinsky On Stage. Edited by Jean Hugo and Richard Buckle. First. London: Studio Vista Publishers, 1971.

Jarvinen, Hannah. “Kinesthesia, Synesthesia and Le Sacre Du Printemps: Responses to Dance Modernism.” The Senses and Society 1, no. 1 (2006).

Koegler, Horst. The Concise Oxford Dictionary of Ballet. London: Oxford University Press, 1977.

Little, Meredith. “Recent Research in European Dance, 1400-1800.” Early Music 14, no. 1 (1986): 4–14. doi:10.1093/earlyj/14.1.4.

Rubidge, Sarah. “Reconstruction and Its Problems.” Dance Journal 2, no. 1 (1995).

V&A Collection. “Le Sacre Du Printemps | Gross, Valentine | V&A Search the Collections.” Accessed April 20, 2017.

V&A Theatre and Performance Collection. “Le Sacre Du Printemps | Gross, Valentine.” V&A Collections. Accessed April 17, 2017.

Wilmer, Celso, Cristiana Lara, and Cristiana Lara Resende. “Illustrations and Nomenclature Stave for Dance Movements: What Visual Communication Can Do for Dance.” Resende Source: Leonardo 31, no. 2 (1998): 111–17.

Zile, Judy Van. “What Is the Dance? Implications for Dance Notation” 17, no. 2 (2009): 41–47.

[1] Meredith Little, “Recent Research in European Dance, 1400-1800,” Early Music 14, no. 1 (1986): p. 4

[2] Horst Koegler, The Concise Oxford Dictionary of Ballet (London: Oxford University Press, 1977). p. 60

[3] Judy Van Zile, “What Is the Dance? Implications for Dance Notation” 17, no. 2 (2009): 41–47.

[4] Little, “Recent Research in European Dance, 1400-1800.”

[5] Cast sheet: The Human Seasons/After the Rain/Flight Pattern, 18th March 2017, Royal Ballet, London. (London: Royal Opera House, 2017)

[6] Judith A Gray, “Dance Technology: Current Applications and Future Trends,” in The Evolution of Dance Technology (The American Alliance for Health, Physical Educations, Recreation and Dance, 1989).

[7] János Fügedi, “Computer Applications in the Field of Dance Notation,” Studia Musicologica Academiae Scientiarum Hungaricae 2/4, no. 39 (1998): 421–41.

[8] Irving David, “Choreography and Copyright – Make the Right Moves,” Dance UK, 2012,

[9] Celso Wilmer, Cristiana Lara, and Cristiana Lara Resende, “Illustrations and Nomenclature Stave for Dance Movements: What Visual Communication Can Do for Dance,” Resende Source: Leonardo 31, no. 2 (1998): 111–17,

[10] Millicent Hodson, Nijisnky’s Crime Against Grace: Reconstruction Score of the Original Choreography for Le Sacre Du Printemps (Pendragon Press, 1996). (Introduction).

[11] Ibid. p.7

[12] Hannah Jarvinen, “Kinesthesia, Synesthesia and Le Sacre Du Printemps: Responses to Dance Modernism,” The Senses and Society 1, no. 1 (2006). p. 78

[13] Hodson, Nijisnky’s Crime Against Grace: Reconstruction Score of the Original Choreography for Le Sacre Du Printemps., p. 23

[14] De Grineau, Brian. The Pagan “Shimmy Shake” at the Prince’s. July 6, 1921. V&A Theatre and Performance Collection, Blythe House, London.

[15] V&A Theatre and Performance Collection, “Le Sacre Du Printemps | Gross, Valentine,” V&A Collections, accessed April 17, 2017,

[16] Valentine Hugo, Nijinsky On Stage, ed. Jean Hugo and Richard Buckle, First (London: Studio Vista Publishers, 1971). p. 12

[17] V&A Collection, “Le Sacre Du Printemps | Gross, Valentine | V&A Search the Collections,” accessed April 20, 2017,

[18] Jack Anderson, “THE JOFFREY BALLET RESTORES NIJINSKY’S ‘RITE OF SPRING,’” The New York Times, October 25, 1987,

[19] Renee Conroy, “Dancework Reconstruction: Kinesthetic Preservation or Danceworld Kitsch?,” American Society for Aesthetics, 2016, 5–9. p. 2

[20] Richard Buckle in Hugo, Nijinsky On Stage. p. 14

[21] Sarah Rubidge, “Reconstruction and Its Problems,” Dance Journal 2, no. 1 (1995).

[22] Luke Dormehl, “The World’s First Virtual Reality Ballet Experience,” Digital Trends, 2016.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s