TURANDOT AND ITS NOT-S0-HAPPY ENDINGS

No abandoned orphan draws the tears and the frustrations as does Turandot, Puccini’s final work, left incomplete at the composer’s death  in November, 1924 and rushed into completion by lesser hands soon afterward. True, the formidable Arturo Toscanini cut his performance short at the world premiere, at the point where the ailing Puccini himself had put down his pen. Doing so, he had carried the most grandiose of all Puccini’s operas to a crisis point but no further, relinquishing the stage to two larger-than-life monsters  facing each other with daggers drawn across an unbridgeable abyss,  but only fifteen minutes away from a happy ending. The intended love/hate duet that would have transported  these monsters across that abyss and into each other’s arms was never composed, at least by Puccini. On his manuscript sketch, however, he penciled indisputable evidence of the importance  he attached  to this climactic duet: the  infinitely revealing words: “qui Tristano.”

The story is well-known, give or take a vast array of contradictory  retellings. Toscanini, who held  Puccini in mingled adoration and contempt, visited the composer in the summer of 1924, heard  him, at the piano, perform the opera as it then stood, which may (or may not) have included his own projection of the final scene. Toscanini liked what he heard, and agreed to conduct the premiere  at La Scala scheduled for the following year. At Puccini’s death, Toscanini undertook to bring the work to completion in properly competent hands.  Puccini had completed his final opera in full score through the scene of Liù’s death midway in Act III. From then on to the end there  remained “a thirty-six-page draft” (says Joseph Kerman in Opera as Drama) or a pile of “twenty-three scarcely legible sketches” (says Julian Budden in OperaGrove and in his splendid recent Puccini biography) or “thirteen pages of sketches, which take the final scene only to the  crucial kiss” (writes Anthony Tommasini in The New York Times). Toscanini’s  first choice to complete the work (claims Budden) was Riccardo Zandonai; another considered possibility (claims conductor John Mauceri) was the young Viennese firebrand Erich Korngold. Both were rejected  by the Puccini family as being already too illustrious. The choice, with Toscanini’s grudging acquiescence,  fell upon Franco Alfano, still at the time not much more than a musical nonentity. (His one success, the 1904 Risurrezione, displayed a reasonable competence  if little more.)

Alfano delivered his commission; Toscanini rejected  it furiously, and ordered a rewrite. Toscanini conducted the 1926 premiere – postponed a year from the 1925 target date —  but drew the line at performing the Alfano score.. He conducted one single performance  (says William Ashbrook) and then turned the work over to his assistant Ettore Panizza, or he conducted three performances (says Harold Rosenthal), or “several,” (writes Andrew Porter). Considering Toscanini’s lifelong antipathy toward music he considered inferior, plus his famous on-again off-again regard for Puccini, it seems inconceivable that he would, even once, wave his baton over what he considered Alfano’s botched job. Despite his championing of numerous minor Italian composers,  in his complete  repertory list as compiled in Harvey Sachs’ eminently trustworthy biography, not a note of Alfano occurs. Alfano did recast and somewhat tighten  his completion – workmanlike if workaday, shaving his original 377 measures down to 268. Whether he accomplished this self-mutilation in time for the Metropolitan Opera premiere , seven months after La Scala, is buried in the dust of Thirty-ninth and Broadway..  What we hear today, invariably, is Alfano 2.

Alfano’s first version, clumsy as it may be over all, ends in a Technicolor blaze, with peals of brassy triumph and the lovers joining in the final music at top lung-power – the everybody-on-stage sing-along version of “Nessun dorma.” For the most part, Alfano 1 is the more tonsil-twisting of the two versions; the tessitura lies higher, most notably at the end when the lovers join in at the end of the “Nessun dorma” reprise with a pair of matched B-flats. (In Alfano 2, the chorus goes it alone.) Turandot’s aria, “Dal primo pianto” runs 79 bars in Alfano 1, pared down to 51 in Alfano 2. There is one recording – now deleted but worth the search: Josephine Barstow and Lando Bartolini on a Decca disc of operatic final scenes, conducted by Mauceri. In 1992, American  conductor Steven Mercurio created  his own conflation of Alfano 1 and 2 for the Opera Company of Philadelphia and led it to considerable  acclaim, with Alfano’s final brass spread high up around Philadelphia’s Academy of Music. Brilliant brass or no, it still makes for a sad contemplation,  that  325 years of grand Italian opera tradition should come to its sputtering end in the merely competent  hackwork of Franco Alfano. “Finita la poesia,” sings the crowd on the last completed page of Puccini’s manuscript, and they may have been right.

Or maybe not. The latest solution of Turandot’s finale problem comes from what might seem unlikely at first view, but which actually makes strong musical sense. Luciano Berio, has now tried his hand at a Turandot completion worthy of the score and its creator. It seem an unusual project  for Berio, one-time avatar of Schoenbergian atonality and longtime kindred spirit to Pierre Boulez, to take on. But we should remember  that Berio is not only a prolific opera composer in his own right but a  passionate defender of Italy’s musical heritage all the way back to Monteverdi and before, His new version of Turandot has been sanctioned by the Puccini estate. After trial concert-performance  runs in the Canary Islands and Amsterdam,   led by Riccardo Chailly, it was staged last June by Los Angeles Opera under Kent Nagano and, two months later, at the Salzburg Festival conducted by Valery Gergiev, received – if a broadcast tape can be believed – with the mix of puzzlement and ecstasy attendant   on any major premiere of new and controversial musical substance.

Something else among Berio’s  credentials stamps him as the proper agent to bring Puccini’s near-masterpiece to a fitting conclusion: his prowess as a highly skilled tamperer.

The 1968 Sinfonia, his best-known work, includes one movement  in which a gathering of familiar repertory tunes (Beethoven, Berlioz, Debussy) moves in a stream-of-consciousness progression while another part of the orchestra plays Mahler and a chorus declaims activist graffiti.  A recent score called Rendering subjects a folio of Schubert’s deathbed  sketches – for a symphony that would have been No. 10 if completed —  to a reworking that builds handsome and flexible bridges between Schubertian romanticism and Berio’s own love of that language.

Berio’s “tampering” with Puccini’s expressed wishes and actual sketches runs to 307 bars, midway in length between Alfano 1 and 2; the performance under Gergiev ran a few seconds over fifteen minutes, almost exactly the same length as the recorded  Callas version (of Alfano 2, naturally). The Adami/Simoni text undergoes two cuts: about half of Turandot’s “Dal primo pianto” has gone,  and so has the choral finale with the “Nessun dorma” reprise.

What Berio has done, actually, has been to recast the entire time-span of this final scene, to the point where his own musical fabrications impart a far more naturalistic flow to the events themselves. Small glints of Puccini’s  music speed the process; Calàf delivers his crucial kiss to the music (“gli enigme sono tre”) of his first response to Turandot in Act II. Rather than the echoes of “Tristano” that Puccini might have evoked for the ensuing duet, there comes next what amounts to a small tone-poem,  nearly three minutes’ worth of purely orchestral music, skidding through harmonies dense and disturbing, dealing wordlessly what no ice-bound soprano need verbalize upon the tenor’s first kiss. The harmonic density may raise hackles among  purists; remember, however, that  Puccini’s own last years were largely spent in discovering the musical world around him – Stravinsky’s Petrouchka and, more remarkable, Schoenberg’s Pierrot Lunaire.. There’s good reason to suspect that he might have welcomed Berio’s tampering far more heartily than Alfano’s flattening.

Turandot finirà pianissimo”; wrote Puccini to librettist Adami. “This new Turandot,” said Berio to a radio interviewer, “will end exactly so.” Prince Calaf (“il principe priapico,” says Berio, the “horny Prince”) reveals his name – in a close rewrite of Alfano’s frantic crescendo on a four-note figure – and Turandot drags him off to meet his fate. “Amore!!!” sing they both on his-and-her B-flats (as in Alfano 1) as faint echoes of “Nessun dorma” percolate through the orchestra. But then the music subsides; little by little there is darkness both visible and audible. An audience, awaiting the customary “happily ever after” choral outburst of the Turandot they’ve all known and loved, sits stunned. You hear it on the tape: a long moment  in which the silence is further prolonged, then the cheers.

Does it work? As music drama created by Luciano Berio – composer of the magnificent Un Re in Ascolto revived not long ago in Chicago and eminently deserving – it works quite well, a “rendering” of thematic fragments by Puccini to stand beside his Schubert piece. The diehards will surely have trouble with the new Turandot. There were dark mutterings after the Kent Nagano-led Los Angeles performances last June – brought on by the somber new ending and by Giancarlo del Monaco’s  murky staging that obliged the antagonists  — neither alternating pair a vocal experience to sing about — to bellow at one another over the abandoned corpse of hapless Liù and to perpretate their final music in what looked like somebody’s abandoned attic.  But even  Gergiev, who delivered a stupendous performance at Salzburg, by the way, has confessed that he will probably revert to Alfano 2 when circumstances so ordain. As an example of a great composer’s respect for a venerable colleague and countryman, however, Puccini/Berio is definitely win/win.