What did they play at Violetta's party?
A recent reference question asked what is the instrumentation of the stage band in act 1 of Giuseppe Verdi's opera La Traviata. In case you've forgotten, the opera opens at the house of Violetta, who's giving a party to celebrate her recovery from illness. After Violetta and Alfredo sing the duet "Libiamo ne' lieti calici," the stage band (banda) begins to play, at which point the party guests exit to the next room to dance, leaving Violetta and Alfredo alone in order to fall in love. It seemed like a simple question. I pulled the authoritative Works of Giuseppe Verdi edition and found the spot to examine the instrumentation.
Surprise! The stage band was written on just two staves--like a piano score. No instruments were indicated at all. I looked at the back of the volume to see if it was included as a supplement. Not finding anything there, I went to the front matter. I found editor Fabrizio Della Seta's explanation in the introduction, where he states: "Following the practice of his time, Verdi wrote a guida banda (a short score on two staves), leaving its realization to the leader of the banda in theaters staging the work. This instrumentation could vary from theater to theater." Additional reading explained that the editorial practice of the Works of Giuseppe Verdi edition is to transmit what came from Verdi's hand. Other hands, while possibly significant to performance practice, were not incorporated into the edition.
Though a bit unexpected in this case, the notion of leaving details of performance up to the performer is not unusual. A cursory look at music history will show that many styles of performance expected performers to fill out what the composer only suggests or alludes to. Continuo realization (i.e. the art of improvising a keyboard part in vocal or chamber ensembles) is a frequent occurrence during the Baroque period, continuing into the latter part of the 18th century. Vocal ornamentation is expected in Handel operas as well as in many bel canto works, through the 19th century. Although it was nearly calcified by the 19th century, the act of improvising a cadenza at the climactic moment in a concerto was originally conceived as a point at which the soloist (be it pianist, violinist, cellist, or any number of instruments) could improvise, creating greater tension and the desire for resolution. Of course, this was a particularly significant area for vocal pyrotechnics in operas. Generally, however, the course of music history shows that as time progresses, most composers preferred to leave explicit directions and indications as to what and how their music was to be performed. Back to the original question. The critical notes for the Works of Giuseppe Verdi edition indicted a score published by Ricordi (the original publisher) had a banda orchestration, but as this source was inauthentic, it was excluded from serious consideration for the edition. I looked through a few more editions of La Traviata, but all had just the two-staff banda. Then I remembered our Harry G. Schumer collection. Harry G. Schumer (1893-1971) was a librarian at the Metropolitan Opera from about 1939 until 1969. Much of his career was spent at the company under the leadership of general manager Rudolf Bing (beginning in 1950), with whom Schumer apparently did not get along. When Schumer retired from the Met, he took with him his own materials - hundreds of slides that he took backstage and on tour, as well as many scores he had collected (or created) to supplement the Met's library. One part of his library (call number: JOB 89-24) contains miscellaneous parts and notes to various operas. Some were created for recordings, some for interpolations that may have been done by various singers or conductors, some contain lists of errors and corrections found in a host of opera scores (critical editions of most operas did not exist until well after World War II). It's a treasure trove of fascinating material. Sure enough, Schumer had dealt with La Traviata's banda problem. His collection offers two solutions. He had the banda sequence from the unauthorized Ricordi printed edition:
But Schumer had also prepared his own orchestration of the banda, more suitable for the Metropolitan Opera orchestra.
Where the Ricordi orchestration is somewhat unusual (1 piccolo, 4 clarinets, 2 horns, 1 flugelhorn, 3 trumpets, 2 trombones, double bass, and bass drum), for the Met, Schumer normalized the sound (flute, piccolo, 2 oboes, 2 clarinets, 2 bassoons, 4 horns, 2 trumpets, trombone, and double bass). Not only was I able to answer the question for our patron, but I was able to reveal the truth of the Verdi edition's editorial notes, showing how a leading opera company created its own orchestration to suit the orchestra at hand -- which, ironically, is exactly what the composer wanted.