Cuban Overture (urtext)

The trip to Havana, which Gershwin undertook with several friends in 1932, was actually the decisive musical drop. Gershwin stayed in the luxurious „Almendares“ hotel. His stay was not secret for long: „The siren call of Cuban music with its strange rhythms and plaintive melodies lured George Gershwin to Havana. He’s looking for new inspiration“, wrote The Havana Post. Gershwin, meanwhile, was initially looking for relaxation and entertainment. You met him while sunbathing, golfing and in night clubs. „I spent two hysterical weeks in Havana, where no sleep was had“, he later wrote. And of course it was the music that got Gershwin infected. „Cuba was most interesting to me, especially for its small dance orchestras, who play most intricate rhythms most naturally“, he was enthusiastic about the music – and the musicians.

Lewisohn Stadium in New York was an important place for Gershwin. In an „All-American program“ in the summer of 1931, his „Rhapsody in Blue“ was performed. An „All-Gershwin program“ was now planned for the summer of 1932. Gershwin had already composed a number of orchestral pieces by then, such as the „American in Paris“, the Piano Concerto and the „Second Rhapsody“. But his ambition was great to release a new symphonic piece for the major event. That is why he began the „Cuban Overture“ in July 1932. He started the orchestration on August 1, it took him eight days – after that there was little time to do all the editorial work, the production of the performance materials and of course the rehearsals until the premiere on August 16.

The piece was first premiered under the title „Rumba“. The premiere was a success. The audience at the Lewisohn Stadium reacted enthusiastically, as did the second performance in the Metropolitan Opera („Met“) shortly after the premiere. The critics wrote mixed reviews, but rather positive. Some found a „merely old Gershwin in recognizable form“, the others „rich sonority“ and „fascinating rhythms“, the others saw „Rumba“ as a competition to Ravel’s „Bolero“, others were amazed: „The work’s materials have been treated in ways which depart from anything Gershwin has done in earlier scores“.

The recently quoted critic, Olin Downes, probably had the clearest view. In fact, the „Cuban Overture“ is full of fascinating details, especially for those who know Gershwin’s orchestral work well. The middle part is in Gershwin’s oeuvre, apart from some passages in „Porgy and Bess“, unique. It is designed as a polytonal canon that is harmoniously highly complex. Even experienced analysts would have to take a close look at this part in order to understand the harmonic mode of action.

In terms of sound, Gershwin is perhaps the closest to the sphere to which he had already committed himself when composing the „American in Paris“. Here Gershwin sounds really French, like Gershwin’s revered tonal language of the „Groupe des Six” around Darius Milhaud and Francis Poulenc. It is very typical for the unbiased George Gershwin that he places this expressionistic part – which is quite demanding for the listener – between the irresistible Cuban rhythms with quotes and allusions from „La Paloma“, „The Peanut Vendor“ and the Cuban national hit „Échale Salsita“.

The handwriting of the score shows a certain agitation. In many places Gershwin does not write out passages, but refers to the same parts written out elsewhere. He uses measure repeat signs more than usual in an almost stenographic manner. He notes “3 TIMES” above the penultimate bar instead of writing out the bar three times. Gershwin is also less reliable than in most of his other scores in the uniform designation of the parts with expressive, articulation and dynamic signs. Nevertheless, there are no important questions left unanswered, the autograph is self-explanatory almost everywhere and, in addition to manual and editorial corrections, only requires the unification of the voices and the addition of obviously incorrect names or accidental omissions.

Gershwin later arranged the „Cuban Overture” for piano four hands. The basis for this was the Particell, which he had created together with Oscar Levant on the two pianos of his New York apartment. He later refined it to a real transcription. With this engraved and printed piano version there is an authentic reference source that can answer some open questions. Some of these questions arise when you look at the version of „Cuban Overture“ that is popular today.
After Gershwin’s death in 1937, all five concert works orchestrated by Gershwin himself were initially performed in their original form, but from around 1942 they were replaced by versions that were considerably alienated. The name Frank Campbell-Watson, who is called „editor“ in several scores, is the central figure in this inglorious „Gershwin affair“. The Max Reger student worked as a composer, church musician and publisher at the New York music publisher Harms, which published Gershwin’s pieces at the time. He worked on at least two works himself („An American in Paris“, „Concerto in F“), while on two others („Variations on ’I got Rhythm’“ and „Second Rhapsody“) he let contract arrangers edit the pieces.

Most of these „transformations” go far beyond meaningful editorial interventions – in some cases they reverse the intentions precisely defined by the composer or simply remove particularly sophisticated effects. The revisions were probably more about establishing a certain marketability, perhaps also about extending the copyright protection periods. The changes were never authorized by the Gershwins, and Ira Gershwin has even repeatedly campaigned to keep his brother’s original versions in distribution: He had used them himself throughout his life and was always satisfied with it, Ira said.

The „Cuban Overture“ was reasonably lucky compared to the other works. It has remained largely unchanged, even though Frank Campbell-Watson said about her a sentence which is really alarming for pure Gershwin fans: the overture needs „more housecleaning than anything else“. He handed over the job of „cleaning up” to an editor who is not known by name. The result of this „cleaning“ was even dirtier: „Because the published score is laden with errors, I recommend the conductors and historians that it no longer be used“, writes the American musicologist Jeff Sultanof, who dealt with the original version of the „Cuban Overture“.
The present Urtext edition is to be regarded as a diplomatic edition based on the final own-hand manuscript score. Changes or additions that clearly resulted from the musical content or that were returned to the score from Gershwin’s piano version are denoted in brackets (or in dotted lines for crescendo forks). The most important differences compared to the Campbell-Watson version edited by a third hand are:

In bars 289 and 295 the clarinets in the edited version play together with the flutes and oboes. In the original version, the clarinets are only the answer to the other woodwinds.

In the original score, the coda (bar 355) begins with seven bars with percussion only. These measures are also missing in Gershwin’s piano version, but they would hardly be performable on the piano. These bars are missing in the edited score.

According to the manuscript score, the penultimate bar has to be played three times („3 TIMES“). In the piano version, this is confirmed in written notes. Campbell-Watson hat the bar once only.

Gershwin’s score does not contain a wood block. In the edited score it is played unisono with the snare drum from bar 43ff.

In countless points in the score known today, dynamics have been changed and articulation signs have been added unprovided by the original sources and not deriveable from the musical or dramaturgical logic.

Boris Hellmers-Spethmann
September 2015