Ravel once called Debussy's Prelude to the Afternoon of a Faun "one of the few miracles in the history of music," but surely his own Daphnis and Chloe is no less miraculous. Written for Diaghilev's Ballet Russes and premiered in 1912 (the year before the same company gave us The Rite of Spring), Ravel scored the work for a huge orchestra (employing both a wind machine and a wordless chorus) that he then used for an incredibly wide variety of expressive and sonic effects. Later he called it "a choreographic symphony," "a vast musical fresco," but in fact there are no words to properly describe Daphnis and Chloe: almost a hundred years after its premiere, it remains a unique creation, truly one of a kind.
The two performances of this music that Charles Munch and the Boston Symphony Orchestra recorded for RCA in 1955 and 1962 have so completely dominated the competition that it's easy to forget how many other distinguished performances were recorded during that same period and how few have been recorded since. Though Munch early and late both deserve their reputations, it would take a less sybaritic disposition than my own to have resisted the diverse pleasures of Pierre Monteux (who conducted the premiere), Leonard Bernstein, Andre Cluytens, Ernest Ansermet, and (a little later) Jean Martinon.
Since that "golden age," it's been pretty much a drought, and those in search of a modern recording of the score have been forced to choose between performances that are well wrought, but unfeeling (Abbado) and those that are altogether clueless (Rattle). At the time, Charles Dutoit's Montreal recording was highly regarded, but it seems clear now that was mostly because of the sound--it was one of the first digital releases one could listen to without cringing. The performance is, like so much of Dutoit's work, suave, sophisticated, and entirely unmagical. Of the two Boulez recordings, I prefer his first, mostly for the virtuosic playing of the New York Philharmonic. The remake with the Berlin Philharmonic might indeed be an improvement, but it's so brightly recorded that it's hard to listen to without squinting. In this score, as in so many others, Boulez is perfect at rendering correct pitch, timbre, and intonation; at the same time, his literal mindedness pretty much bleaches out any trace of spontaneity or genuine feeling. If one is looking for a recent version in state of the art sound, I favorably reviewed Myung-Whun Chung's recording with DGG last year and I would still stand by that recommendation. But I'm now happy to welcome these two new recordings, both of which are competitive with the best in terms of both performance and sound.
Both Haitink and Levine have recorded this work before (Levine with the Vienna Philharmonic, Haitink with the Boston Symphony), but neither of those performances achieves the same level of excellence as these new ones. It was certainly a fascinating and instructive experience listening to the two, for they couldn't be more different in every way. The choice here is between Levine's youthful Romantic approach and Haitink's cooler, more refined classicism. If I could only have one of these recordings, I'd probably go with the Levine. His version is more theatrical, athletic, and (like Chung) he shapes the individual episodes with greater character. There's also more juice in the faster music.
But the Chicago performance also has its virtues, and they are considerable. Haitink's ear for balances gives Ravel's lush textures an utter transparency, a luminous, shimmering clarity, in which every note registers its full expressive effect. Though unhurried, Haitink's approach is surprisingly sensual and atmospheric for an octogenarian, and he does an even better job than Levine in evoking the dream-like languor of Ravel's Grecian idyll. A specialist in the music of Bruckner and Mahler, Haitink is used to organizing large orchestral works. Here he takes the symphonic nature of the score seriously, and follows a single line of development from its barely audible first notes to an orgiastic climax that is every bit as exciting as Levine's (and maybe even more so for having been so carefully anticipated). Besides, Poulenc's Gloria, perhaps the composer's greatest orchestral work, is a substantial bonus, and Haitink deftly balances the mix of wise guy insouciance and profound reverence that gives this music its essential and very distinct character. In the final "Quid Sedes," soprano Jessica Rivera sings with a pure tone and a reverential tenderness that are just what the music calls for.
The Boston and Chicago are two of our greatest orchestras, and they realize their respective conductors' visions of the ballet with a combination of breathtaking virtuosity and unforced naturalness that is all the more impressive given that both performances were recorded live. The choruses also contribute brilliantly, but (not surprisingly) are used to different effect. For Levine, the chorus is more prominent, assertive: there's always a sense of dialogue between chorus and orchestra. For Haitink, the chorus is recessed, otherworldly, more a part of the orchestral fabric. The recordings are both spectacular, but also very different: with Levine, you have a seat in the orchestra, and the sound is immediate and forceful. With Haitink, you're in the upper balcony, a more distant perspective that conjures up a very wide and convincing soundstage.
Given how few acceptable modern recordings we have of this music, it seems an embarrassment of riches to have two such good ones released almost simultaneously.
Though I have showed a slight preference for Levine, in effect you can't go wrong with either of these superb performances. And if you love this irresistible music as much as I do, you'll be perfectly happy to live with both.