History

Following William Thompson’s retirement in 1896, his successor Richard Hague had the idea to start documenting and notating the ballets to preserve his predecessor’s legacy. This effort began in earnest in 1899, when Hague partnered with Ernst Meisenburg to create the first notations of The Wayward Daughter. Hague and Meisenburg (with Thompson’s occasional involvement) continued to notate ballets up until 1914, at which point the last surviving notations were made. 

Called Chorégraphie Codifiée by Meisenburg himself, the notation system is more widely known today Meisenburg Notation. However, some older 20th century sources refer to the system as English Notation, creating initial confusion as to whether different systems were being referred to.

Though the notations were occasionally used to stage ballets after the First World War, subsequent balletmasters soon began to add their own touches to the ballets and revise parts of choreography to better suit the advancement of taste and technique. This resulted in the so-called “traditional” versions which we are familiar with today. However, in the 21st century as ballet reconstructions became more of a fashion, the notations were used to present as close to Thompson’s choreography as could be managed.

Notations

Of the 29 ballets that were presented on the stage of Covent Garden between the first notations in 1899 and the last notations in 1914, 22 of those ballets were notated in some capacity. Of those 22 that were somewhat notated, only 16 were notated enough to be reasonably reconstructable. The other 6 ballets consisted of extracts that constituted only a fraction of the full ballet, did not include enough notated passages or did not include passages notated in enough detail to support a full length reconstruction. These so-called ‘unreconstructable’ ballets vary between more notated productions (e.g. Sylvia, whose first act was completely notated, second act sparely noted and third act completely omitted save the pizzicato variation) and less notated productions (e.g. Tristan and Yseult, whose notations only consist of ground plans for a waltz in the first scene of the second act).

However, the musical scores of all 28 ballets are mostly intact, whether in violin répétiteur, full score or piano reduction. The notations also include an essay by Meisenburg on how he developed his system of notation and how it ought to be read, aiding its decipherment.

Ballets Included in the Meisenburg Notations (with year(s) of notation)

Reconstructable

• The Wayward Daughter, ballet comique in three acts and four scenes (1899, 1907)
• The Beauty of Ghent, ballet pantomime in three acts and nine scenes (1900)
• Ondine, ballet pantomime in three acts and five scenes (1901)
• Esmeralda, ballet pantomime in three acts and five scenes (1901)
• The Devil to Pay, ballet pantomime in three acts and four scenes  (1902)
• Alice’s Adventures in Wonderland, ballet feriée in three acts and seven scenes (1902, 1911)
• The Swords of Toledo, grand ballet in four acts (1902)
• The Peri, ballet fantastique in two acts and three scenes (1903)
• The Corsair, ballet pantomime in three acts and five scenes (1904)
• The Vivandiere, ballet comique in one act (1904)
• The Amphitrite, ballet comique in one act (1905)
• The Buccaneers, ballet pantomime in three acts (1906)
• Coppelia, ballet comique in three acts (1908)
• Alcandra or The Amazons, ballet pantomime in three acts and five scenes (1909)
• Pygmalion, ballet pantomime in three acts (1910)
• Giselle, ballet pantomime in two acts (1913)

Unreconstructable

• Diavolina, ballet pantomime in one act (1901, 1902)
• The Two Peasant Girls, ballet comique in one act (1905, 1912)
• Sylvia, ballet pantomime in three acts (1906)
• Élodie or The Fairy of The Forest, ballet pantomime in two acts and three scenes (1908)
• The Fairy Doll, ballet fantastique in one act (1912)
• Tristan and Yseult, ballet pantomime in three acts and four scenes (1914)

Post 1914 and the Modern Day

Not all the reconstructible ballets survived the First World War into the modern day, and of those that survived all deviate to some degree from the notated versions.

The Wayward Daughter, Ondine, Sylvia and Coppelia survived well past the war into the modern day and are still performed by the Royal Ballet, though none in Thompson’s versions. Sylvia, Ondine and The Wayward Daughter were revised (or rather, rechoreographed) by Frederick Ashton in 1952, 1958 and 1960 respectively and Coppelia was rechoreographed by Dame Ninette de Valois in 1954. Whereas Ashton’s Sylvia remained relatively faithful to Delibes’ (and thus Thompson’s) score, the score of Ondine was revised heavily (and somewhat controversially) by Hans Werner Henze. Ashton and John Lanchbery made so many revisions to Hertel’s score for The Wayward Daughter (which Ashton renamed back to the original French title of La Fille mal gardée) to align it with their conception that they eventually abandoned Hertel’s score altogether, instead reviving Ferdinand Hérold’s 1828 adaptation of Dauberval’s original 1789 pastiche score. 

The Devil to Pay, The Swords of Toledo and Pygmalion survived in revised versions of Thompson’s originals, though The Devil to Pay has since been retired due to its coming under criticism for its onstage portrayal of domestic abuse. The Swords of Toledo and Pygmalion are less often seen and are revived but rarely, though The Swords of Toledo has come under criticism for its dated depictions of Muslim and Moorish culture.

The influence of Marius Petipa on the world of ballet cannot be overstated, and following the Second World War the “London” versions of ballets started to be replaced by the “Petipan” versions. This change is most clearly seen in relation to Giselle, The Corsair and Esmeralda. All three survived both world wars, but following the Second World War they gradually began to lose their Thompson specific revisions, aligning their choreography with the international Russian standard. By the middle of the 20th century the pas added by Thompson in his revisions had been expunged, replaced with the pas added by Petipa in St Petersburg. 

However, three ballets that fell out of the repertory after the First World War have since been revived and are occasionally performed by the Royal Ballet. The first to be revived was Coppelia in 1954 by Dame  Ninette de Valois, though she based her revival on Petipa’s version, not on Thompson’s version as per the 1908 notations. Sir Peter Wright also created his version of the ballet in 1995. Revivals of The Buccaneers and The Beauty of Ghent followed in 1986 and 2005 respectively, but these two productions were but loose adaptations of the notations, with many liberties taken, parts of the score being rearranged and several sections of the ballet being completely re-choreographed.

The first full modern reconstruction of the notations was the 2012 reconstruction of the 1905 production of The Amphitrite for the Royal Ballet School, which did not become part of the repertory (though the choreography of the Pas de Deux was modernised and occasionally resurfaces as a graduation piece). There was a plan was to revive the 1902/1911 production of Alice’s Adventures in Wonderland in 2011, but the Royal Ballet instead commissioned Christopher Wheeldon to create a new ballet on the subject to a brand new score by Joby Talbot.

With the modern resurgence of ‘historically informed’ reconstructions of Petipa’s ballets from the Stepanov notations, there have been ventures into reconstructing Thompson’s ballets from the Meisenburg notations. In addition to the 2012 reconstruction of The Amphitrite for the Royal ballet School, extracts from The Corsair (the Pas des Éventails and the Pas de Distraction) were staged for the Royal Ballet in 2019 and a full reconstruction of the 1904 production of The Vivandiere was staged for the English National Ballet in 2024. Additionally, a reconstruction of 1903 production of The Peri is planned to be staged for the Birmingham Royal Ballet sometime in the late 2020s.

Meisenburg’s Essay

Although Meisenburg never published an official explanation of his system, the notations in the Covent Garden archives include an essay by Meisenburg on the subject of his notation. Dated to c. 1910, the essay includes information about how Meisenburg came up with his system of notation and how it ought to be read.

However, the essay is incomplete, surviving as draft copy which can be pieced together from the surviving fragments. While the essay is clearly unfinished, the surviving draft is a source which has proved invaluable for modern scholars with deciphering the system of notation. A critical edition of Meisenburg’s essay was edited and published in the 1980s, sanctioned by the Director of the Royal Ballet at the time.

The draft does not seem to have been ready for publishing or indeed consumption, given its state. Concepts are referred to before being fully explained later in the draft, explanations are repeated with slight variations and some concepts are only understood when comparing passages from different parts of the draft. Much of the “why” is not recorded, either why specific symbols were chosen or why similar movements were in different ways across the notations. In one of the fragments, Meisenburg begins to discuss “simplifications” that he doesn’t seem to have implemented in the surviving notation and in another fragment he begins to discuss “chains of movement” that similarly do not appear in the surviving notation. Additionally, there is the implicit assumption of sufficient knowledge of ballet technique, as Meisenburg freely refers to ballet steps without explanation. 

Thus, reconstructing the notations is, by definition, interpretative. The surviving draft (and by edition the published critical edition) gives enough information to read the notations to a decent degree of accuracy, but it is by no means a perfect notation system nor is the draft comprehensive enough to be didactic. 

As for the labelling of notations, Hague and Meisenburg seem to have been fairly rigorous. Notations were grouped by ballet, with indications that, alongside comparison with the score, allow choreography and music to be reliably matched.  Additionally, the notations are dated, though some contain more than one hand. Despite the rigour of Hague and Meisenburg, there are still some notations (c. 10-15% of the collection) that are ambiguous as to which ballet or which passage they correspond to.