Building an Opera

Conductor Matthew Waldren explains the mysteries of getting a show onto the stage.


Published in

Scenario: The magazine of Opera Holland Park 2013.


Not to be reproduced or edited without the permission of the Author, Matthew Waldren.


What does a conductor actually do? Aren’t we there just to keep everyone together during the performance? Well, that’s certainly part of the job; and not always easy to do, particularly in opera where the orchestra, more often than not, cannot hear what is being sung by singers on stage who also, more often than not, cannot hear the orchestra in the pit… add in off-stage playing and singing, and the conductor is certainly important for maintaining musical discipline. But if traffic control were the whole job it would soon get pretty tiresome.


Performance is the culmination of the creative process, though since we always discover more detail during performance that creative process never really ends. In performance the conductor has the privilege of being the music, its conduit for the musicians, facilitating the music from the whole ensemble. In opera we also get the chance to channel a story and drama and draw the audience into this world we are creating. An opera begins months in advance with casting, direction, design, choreography and lighting design, and when this cocktail is combined with the live music and story we have the potential to create the greatest of artforms. This is a hugely difficult cocktail to mix, but it’s a wonderful concoction – and good to drink too, we hope.


To reach the stage where the cocktail is ready to be poured the conductor must, of course, learn all the ingredients – notes, rests, markings, words. Beyond this he must forge a real, detailed relationship with the work, get inside the piece and know it and all its characters absolutely. Each piece is different, and this relationship will develop in different ways. Some operas allow you to take the relationship at face value. Puccini’s Gianni Schicchi, for example, an example of genuinely laugh-out-loud opera which I conducted for Opera Holland Park’s Christine Collins Young Artists’ Scheme in 2012, leaves the conductor no place to hide. It is technically fiendish and the pacing (and casting) must be flawless for it to work. But Gianni Schicchi is Gianni Schicchi: 50 minutes long, set in one room, unaltered since it was written – by Puccini or anyone else – it needs no need cuts or revisions.


That actually makes it the exception. Among Puccini’s other operas, Madama Butterfly is one that was never set in stone. Originally written in two acts, it failed on its premiere in Milan in 1904 and was withdrawn. Between 1904 and 1907 no fewer than five versions of the piece were performed – all revised by Puccini himself. The last revision is usually presented today, but the original version gets an occasional outing, and sometimes a conductor decides to include fragments of one version or another. Another example: in Paris in 1858, five years before Les pêcheurs de perles was written, Jacques Offenbach produced the two-act comedy Orphée aux enfers. In 1874 the piece was revised and doubled in length, becoming a four-act work. Nowadays, the shorter piece is generally preferred, but often with additions from the grander 1874 version. It is the conductor’s job to establish if adding additional material makes musical sense, and (along with the director) if it makes dramatic sense for that particular production.


If we look even further back to Monteverdi’s L’incoronazione di Poppea then we really do find a jigsaw. There are two sources for the composition from the 1650s; one in Venice, copied by the wife of the composer Cavalli which includes his alterations, the other – which differs considerably from its Venetian counterpart – in Naples. There are also two libretti and an original scenario. Furthermore, the opera cannot be entirely attributed to Monteverdi, and it is hard to be sure how much of the work he actually composed. Problems also arise with the orchestration, and one must look to contemporaneous sources to assist in ascertaining what forces may be required. Performance history will also have a bearing on the decisions the conductor and creative team will come to on how best to mount a coherent production of the piece.


Les pêcheurs de perles is a bit like that. It is a wonderful piece, but it has been much overshadowed by Carmen and much criticised over the years. It also has no fewer than three versions. The conductor must, in collaboration with his colleagues, realise a version that works best for each new production.

Bizet wrote Les pêcheurs de perles in 1863, setting the music to a libretto by Michel Carré and Eugène Cormon. It received 18 performances that year, and no more in Bizet’s lifetime. The libretto – originally set in Mexico and entitled Leila – has generally been regarded as slight, with little background given to the characters and with more than a few convenient coincidences. But this is hardly unique in opera (or elsewhere). As an audience we suspend our disbelief, and Bizet’s sound-world and skill as musical dramatist helps us considerably in that regard.


Bizet’s harmony and scoring was striking, and it certainly wasn’t to everyone’s taste. Hector Berlioz, the 19th-century musical enfant terrible of Paris, was one of the few critics to praise the opera; the general opinion was that Bizet had made a valiant youthful attempt but that his compositional style was too disparate. In fact, Bizet himself may have had something to do with perpetuating this received opinion; in 1867 he wrote in a letter:



My musical baggage is pretty slender. An opera [Pêcheurs] very much discussed, attacked and defended – in fact a failure, honourable, brilliant if you allow me the expression, but nonetheless a failure.



It is a shame that many commentators since have focused on the word “failure” rather than the word “brilliant”: you have only to listen to the prelude to realise the brilliance. Perhaps the failure Bizet spoke of was the box-office failure, or the failure of his critics and audience to grasp the piece? In any event, the conductor must free himself from Bizet’s own doubts and focus on his depiction of character, drama and place through music.


Bizet cannot really be criticised for not creating an authentic musical sound-world for the Ceylon of the opera; it was not a requirement of the epoch, and no composer would have been well-versed in that culture. Bizet’s writing of course relies mainly on prevailing French, German and Italian musical styles. But it is simplistic to argue that he relies only on these idioms, and he uses them with great imagination to invoke a different air, a different perfume, a different climate. Although he was certainly influenced by pieces such as Félicien David’s 1844 composition Le désert, he takes us to a sound-world and a Ceylon of his own imagining. Three examples to listen out for include Nadir’s Act 1 aria, his off-stage song in Act 2 and Leila’s incantation to Brahma in Act 1. Each is strikingly beautiful and sensual, and none are ‘of’ 19th-century Paris.


Through a daring use of harmony and orchestration (particularly cor anglais, harp, oboe, and flute), and a deceptive use of metre and sinuous melody, these pieces transport us to another place and give us glimpses into the hidden lives of the characters. Surely Leila’s second verse of coloratura after her incantation is more about her own personal desires than a religious experience? This shows Bizet’s remarkable talent. In 1863 music wasn’t as global an art as it is now, or even as it had become by the early 1900s when Puccini wrote Butterfly. By then, there was real interest in other musical cultures, and – although it was still brave – it was far more accepted to write in pentatonic scales and create a Japanese sound world. In Carmen, too, being set closer to home, it was perhaps easier for Bizet to create the Spanish and gypsy flavours of the composition, even though these are not really authentic. Nonetheless, it seems that the music and libretto of Les pêcheurs did not thrill the Parisian audience and critics, and the piece dropped out of the repertory.


And it doesn’t really feature again until Bizet’s death in 1875, when music from the opera was played at his funeral. It was then that Carmen­-fever swept the world. Carmen had been mauled by the critics during its premiere run in 1875; now it was a classic example of death making a career. A production in Vienna later that year sparked Carmen’s success, though the opera wasn’t revived in Paris itself until 1883. Les pêcheurs de perles made its own reappearance ten years later in 1893.


And when Les pêcheurs did return it was in a souped-up new version which had basically been Carmenised. The order of Act 3 was altered, a trio by Benjamin Godard added, and Zurga – instead of being left a lonely figure unsure of his fate – was stabbed to death. A Verdian cabaletta-ending of Zurga and Nadir’s Act 1 duet was also dropped in favour of a swelling reprise of the main tune. This has become the duet ‘Au fond du temple saint, performed as a stand-alone number in a thousand classical concerts, that we all know and love. The changes to Act 3 arguably heighten the drama, but Zurga’s death certainly verges on the melodramatic. That said, the revival breathed new life into the opera and won it a place in the standard repertoire. Bizet would almost certainly have wanted to make changes to the score for any revival, in any case. He wouldn’t have expected a revival to be successful in its original form, as his 1867 letter suggests. Composers are very prone to make changes to their work and it’s a mistake to imagine that compositions are set in stone.


The 1893 version of Les pêcheurs de perles was the basis of all further productions until the 1970s, since Bizet’s original orchestral score was lost. However, in the ’70s a speculative re-orchestration of Act 3 of Bizet’s 1863 version was commissioned, and this version has been an option for larger opera companies since. Then, in 2002, conductor Brad Cohen (well known to Holland Park audiences) orchestrated an ‘authoritative’ version of Bizet’s 1863 original following the discovery of Bizet’s conducting score (which contained only six staves but included annotations of the missing orchestration), offering yet another version complete with the original Act 1 duet. So now we have three versions. As with everything human, all music and drama is imperfect, so there cannot be a perfect ready-made version to pick off the shelf.


For a conductor, facilitating the music and drama requires flexibility. This happens as part of a team, collaborating to tell the story as effectively as possible. The approach will alter considerably for every production of the same piece, as creative teams and casts change. We are often initially critical of performing something that isn’t as close to the original composition as possible. But opera is – or must be, in order to thrive – a living, breathing artform and not a museum piece. Carmen, for example, can be performed with extended or curtailed dialogue, with the posthumous recitatives, or a mixture of both. Bizet himself, according to Massenet, was perhaps intending to insert recitatives for performances in Belgium and Germany in any case. Conductors also often decide to dramatically reduce much of Carmen’s off-stage music, and there have been productions where directors and conductors have decided to kill off the officer Zuniga at the end of Act 2 instead of having him come back to buy oranges in Act 4. Some of these choices are determined by the concept of a particular production, and the conductor’s job must be to make sure that it all works musically. Beyond this, any changes we do make can only be justified if they help us get to the heart of the story.


Opera is also a business, and there are many practical considerations when deciding how to tell a story and which version of a story to tell. Although the conductor is not, thank God, responsible for the economics, it is a very real consideration. The cost of dogmatically pursuing a particular version may mean you can’t afford the cast you want, or the chorus or orchestra numbers you need to do it justice. Or, if you do, you could make a loss that threatens future productions. This also raises the issue of the cost of performance rights in the UK, a subject for another entire article… The venue where you are performing is also important, as it may require an orchestral reduction or even a complete re-orchestration. Again, you have to be flexible and creative.


Beyond all this, however, the entire point is to entertain the audience and there will be musical decisions to make this happen. Les pêcheurs de perles has a clear example: that duet. It is so well-known and well-loved – but what we are used to hearing is the posthumous 1893 version. Should we always perform the 1863 original because we finally have the correct orchestrations? Neither version really alters the narrative, so there really is a straight choice. I rather like Bizet’s original 1863 cabaletta ‘Amitie sainte’, there are good arguments for performing it, and it’s wonderful that the option exists. But we should acknowledge that this duet, in its later 1893 form, is the one thing that most audience members will recognise in the opera, and it seems churlish to deny everyone the enjoyment of hearing it sung beautifully and in context.


So our planning process Les pêcheurs de perles has taken us through many practical obstacles, through discussions about the narrative and music, through Zurga’s death or survival, but has ultimately always been about how best to tell the story and to entertain. So this evening you will hear some music from 1863 and some music from 1893; however, if we do our job successfully, it shouldn’t make a blind bit of difference to you because your energies will hopefully be focused on the characters and story.


In about five weeks from the time of writing the singers and I will meet for the first time and music rehearsals will begin. A week later production rehearsals will commence. The singers will bring their own unique personalities, thoughts and ideas to the process, and together with the director we will all continue the creative process and uncover more layers of the lives of our protagonists. The chorus will arrive soon afterwards, bringing the lives of their characters with them, before the orchestra joins the fray. Before long the culmination of the process will be upon us as we attempt to draw you, the audience, into the world of Les pêcheurs de perles. The audience is at the heart of our creative process and if you can also go home whistling a tune, then that’s a bonus – and means that I’ll have done at least part of my job right.


Matthew Waldren read music at Cambridge and continued his studies at Guildhall School of Music and Drama. A conductor and coach, and formerly a singer, Matthew has primarily worked in opera over the past decade.