Perhaps you saw the article in the New York Times last March under the provocative headline “Have We Been Playing Gershwin Wrong for 70 Years?” It’s about his beloved An American in Paris, a symphonic tone poem that features the distinctive sounds of French taxi horns and which will be performed on The Philadelphia Orchestra’s Opening Night Concert September 30. The inimitable honkers are just one of the ways Gershwin evokes the bustle and atmosphere of the French capital as it sounded in 1928. He reportedly spent a lot of time tooting on actual taxi horns—the old-fashioned kind with the squeeze bulb—before choosing the four he wanted.
Then, for unknown reasons, Gershwin did something that created a long-running musical mystery: In his score for An American in Paris he indicated which taxi horn should play which part with a set of circled letters: A, B, C, and D.
Over time those designations came to be seen as the musical pitches A, B, C, and D. But new scholarly research now claims that those letters were just labels, not pitches. And the researchers point to a 1929 RCA Victor recording of the work credited to the Victor Symphony Orchestra, with Gershwin himself, as proof. Sure enough the taxi horns on that recording sound very different from the ones we’ve been hearing for decades in concert halls, on recordings, in movie soundtracks, and in the current hit Broadway show (which is about to close). The pitches are more like A-flat and B-flat above middle C, a D above that, plus an A-flat below middle C. Producing that lower note would require a longer horn than the ones orchestras have been using for years.
Enter Philadelphia Orchestra Principal Percussion Christopher Deviney; those taxi horns are his responsibility. The Orchestra has its own set of American in Paris taxi horns? Mais oui! Complete with travel case.
While Chris found the Times story intriguing, he didn’t think a single recording justified changing the horns. For one thing, the taxi horns we’ve grown used to are pretty much in tune with what the rest of the orchestra is playing.
But then Chris heard from an old friend, a fellow percussionist in the Chicago Symphony. His father had played percussion with the Cincinnati Symphony, including on a performance of An American in Paris shortly after it was written. Gershwin himself had brought the taxi horns to Cincinnati for the occasion. Chris’s friend has the photo to prove it:
That’s Gershwin on the left, seemingly about to honk a taxi horn that’s a whole bunch longer than the other ones. Long enough to produce that low A-flat.
That was all it took to convince Chris that the “wrong horns” story was legitimate, and he should find some new horns. Chris called Steve Weiss Music in Willow Grove, PA, one of the world’s leading suppliers of percussion instruments. They could order him taxi horns in A-flat, B-flat, and D above middle C. But the long one in a low A-flat below middle C? Sorry; nobody makes it.
Undeterred Chris told them to do their best. And a few days later he went to Willow Grove to check on progress. Three of the horns were on order. And just for grins, a store employee had rummaged through a drawer of miscellaneous horns, and voila! There was a bicycle horn in the missing pitch.
As for audiences, he doesn’t think they’re going to mind the new taxi horns, no matter how unfamiliar (and more dissonant) they may sound. After all, they’re supposed to represent the cacophony of traffic in the middle of Paris. “It’s not like we’re taking taxi horns and trying to get melodies out of them.”
In fact the new/old taxi horns should feel right at home with The Philadelphia Orchestra. That Victor recording from way back in 1929, the one the scholars used to back up their theory about the horns? The label reads “The Victor Symphony Orchestra with George Gershwin,” but at least one source says, it was actually members of The Philadelphia Orchestra.
Plus ça change …