Don Carlos Rediscovered

Don Carlo.jpg

Erich Auerbach / Stringer

Bulgarian bass-baritone Boris Christoff (1914–1993), in costume as Philip II in Verdi’s opera Don Carlos at Covent Garden.

After months of negotiations, during which Giuseppe Verdi rejected King Lear, Judith and Holofernes, and Cleopatra as subjects, it was finally agreed: he would compose a new work based on Friedrich Schiller’s play Don Carlos as the Paris Opera’s showpiece production for Napoleon III’s Universal Exhibition of 1867. It was conceived as a grand opera in the French style: a five-act visual and musical spectacle with a large cast and orchestra, crowd scenes (not least an auto-da-fé), and—to satisfy French tastes—a ballet. Schiller’s sprawling drama was well suited to such a treatment, with its six major characters locked in personal and political struggles affecting three nations: King Philip II of Spain, who suspects his French wife, Elisabeth de Valois, of adultery with his distrusted son, Don Carlos, who was once betrothed to her and still loves her; the Marquis of Posa, Carlos’s boyhood friend, now the idealistic defender of freedom of thought and of the Spanish Netherlands languishing under Philip’s tyrannical rule, who becomes the king’s trusted confidant; the Princess Eboli, who loves Carlos and betrays him and Elisabeth when he spurns her; and looming over all of them, the Grand Inquisitor—the thundering voice of reactionary dogmatism.

Giacomo Meyerbeer had given Parisians grands opéras full of elaborate pageantry: a moonlit ballet for ghostly debauched nuns in Robert le Diable, a huge explosion to end Le Prophète. After twenty-four operas, however, Verdi aspired to more than theatrical display. With Don Carlos he intended, as the English critic and musicologist Andrew Porter wrote, “to give a new nobility and purpose to grand opera; to unite what was best about the Meyerbeer manner, with its enormous sonic and visual paraphernalia [what Richard Wagner derided as “effects without causes”], with his own ideas about character and dramatic impact and what he called warmth.” The result, according to Porter, is Verdi’s “most ambitious opera, one that combines his theatrical flair for portraying human beings in extreme plights, rent by conflicting emotional and moral imperatives, with lofty grand-opéra pageantry used…to demonstrate a belief in political and personal freedom.”

But such a grandiose vision proved unwieldy. At the rehearsal on February 24, only two weeks before its premiere, Don Carlos was too long: it began at 7:11 and ended at 12:23. The last trains for the suburbs departed at 12:35, and even though Verdi had already made numerous cuts, more were needed. Among the last passages to go were some ten minutes of music at the very beginning of act one, in which, after a brief orchestral prelude, a chorus of French woodsmen and their wives in the forest of Fontainebleau laments the deprivations of the war with Spain, and Elisabeth de Valois, the daughter of King Henri II, reassures them that peace negotiations are underway and it will soon be over.

The passage had an important dramatic purpose, setting the characters in a political world of misery and anticipation. Shorn of it, the opera opens in a private world of romance, as Carlos, the heir to the Spanish throne, steps out of the forest to sing of coming to France in the hope of seeing Elisabeth, whom he expects to marry to seal the peace treaty. He meets her and they fall in love, but when emissaries arrive to announce that she is to marry not Carlos but Philip, her people’s suffering—now expressed in only a few bars of music—means that she must sacrifice her own happiness and accept.

That opening scene was omitted not only from the first Paris performances of Don Carlos but from the printed score, and was not heard again until it was rediscovered a century later by Andrew Porter in the archives of the Paris Opera. As he wrote, “In the original orchestra parts…the cut passages had simply been stitched or pinned together, or pasted down; the same had happened in the singers’ scores. Line by line, it was possible to reconstruct a good deal of totally unknown music by the mature Verdi.” The discovery of eight passages in all—by the Verdi scholars David Rosen and Ursula Günther as well by Porter—made it possible for Günther to publish in 1980 an edition of all the variant versions of Don Carlos. This edition laid out for the first time the work in full as Verdi originally conceived it for the Paris Opera, without the pre-premiere cuts and without the substantial revisions that he made to the score fifteen years later.

Porter, who died on April 3 at the age of eighty-six, was surely the finest music critic of the late twentieth century. He began writing for the Financial Times in 1953; in 1972 he was recruited by The New Yorker, where he remained until 1992, after which he continued to write for The Observer and the TLS. His reviews, if one can give them such an ephemeral-sounding name, provide a veritable education in music history and appreciation: they are erudite, elegantly written, and expansive, and range so widely beyond the particulars of long-past performances that they still reward rereading. (All five volumes of his collected New Yorker reviews from 1972 to 1983 are now out of print, though used copies can still be found; most of his writings remain scattered in newspaper and periodical archives.)


Though he wrote on every period and genre of classical music as well as dance, Porter always had a particular affinity for opera. A longtime advocate of performances in translation, he prepared English versions of thirty-seven works, including Wagner’sRing of the Nibelung. He was particularly drawn to Don Carlos, the Verdi opera, he declared in 1978, of which he had seen the most productions. He wrote a number of articles on the lengthy process of composing it, as well as reviews of many performances in Europe and the US, all of which illuminate what has come to be seen as one of Verdi’s greatest works.

As The New Yorker’s music critic Porter was in Boston in 1973 for the first production—“in a sense,” he wrote, “the world premiere”—of “the opera that Verdi originally conceived and composed.” Despite the limitations of the borrowed scenery, a less than first-rate cast, and the absence of the ballet—he found Verdi’s full 1867 score, to the reconstruction of which he had made such significant contributions, revealed as never before:

“I thought it was tremendous…. The extensive, abundant five-act Carlos of 1867, before surgery, is a masterpiece of its kind…. In the revised Carlos, sentiments are less generously expressed; the lyrical appeal to our emotions is less open…. There was no reason to conclude that any of the composer’s pre-performance cuts had been prompted by artistic considerations, or for any purpose other than the need to shorten the running time.”

For all its historical interest, the Don Carlos of 1867, cut or uncut, has seldom been performed. Verdi, determined to make any further cuts himself if theaters wanted a shorter version, revised the score again in 1882–1883, removing the entire first act and recomposing some of the remaining scenes in his later, tauter style that was to come fully to fruition in Otello and Falstaff, operas in which he pared everything down to “just what is necessary for the action.” In 1886 he approved another version, in which the first act from the published 1867 score was rejoined to the revised acts of 1883. Both used an Italian version of the libretto rather than the original French, with the Italian version of the title—Don Carlo—since it had fallen out of favor in France and was by then mostly being performed in Italy. For decades the opera was performed in one or the other of these two versions.

Once the lost music of 1867 was rediscovered, however, opera companies began to incorporate some of it in Verdi’s later versions of the score. The Metropolitan Opera, in a new production of the opera in 1979, added to the five-act version of 1886 the scene at the beginning of act one that Verdi had cut two weeks before the premiere in 1867 (the current production, which premiered in 2010, omits it). Claudio Abbado conducted a major new production for La Scala’s bicentennial in the 1977–1978 season that included the opening scene as well as the scene in which Eboli exchanges masks with Elisabeth at the beginning of act three, the scene in which Philip and Carlos mourn the death of the Marquis of Posa in act four, and the original, longer ending of act five, with its booming chorus of inquisitors replacing the slimmed-down ending of 1883 with only Philip, Carlos, Elisabeth, the Grand Inquisitor, and the mysterious monk who might or might not be Philip’s father, the retired emperor Charles V. Porter found it a deeply personal experience: “It was thrilling to me to hear Abbado and the Scala orchestra bring to life, so richly, music that had gradually taken shape, line by line, under my eyes as I prized open the stuck-down pages of the 1867 orchestra parts, one after another, and copied them to build up a full score.” The performance of January 8, 1978, was televised:

Yet Porter always had doubts about performing such “mixed” versions of the score. Already in his review of the Boston production of 1973, he argued that “the ‘unknown’ passages can be restored only to this earlier Paris score; they do not fit the revision.” In general he thought that companies should perform either the four-act 1883 version (“the most powerful, the most effective”—as Verdi himself put it, “greater conciseness, and greater strength [nerbo]”) or, given the time and resources, the five-act 1886 version without additions (or cuts), though he also believed that “mature music that Verdi composed in 1866–67 is too good to leave forever unperformed, and so there is a strong case for presenting from time to time, in all its splendid length, the grand opera” written for Paris. He sometimes argued that the original prelude and chorus was the only music from it that companies should consider adding to the revised score, though on occasion he had doubts even about that, feeling—not without reason—that it had weak passages. Abbado, when he made a studio recording of the opera in 1983–1984, returned to the 1886 version, respecting what he took to be Verdi’s final thoughts, though he used the French text and also recorded a number of the deleted 1867 passages in an appendix disc. That performance has wonderfully refined and passionate orchestral playing, though that can seem overcompensation for a cast uneven except in its uniformly poor French (see the first audio clip above).


Julian Budden, the author of the magisterial three-volume The Operas of Verdi, generally agreed with Porter’s preference for not mixing dropped passages from 1867 into later versions of the score but admitted that such a stricture could seem “rather pedantic,” noting that Abbado’s inclusion at La Scala of the lament of Philip over the corpse of Posa at the end of act four (music that Verdi reused in his Requiem) “made an unforgettable impression.” Porter, writing of that performance, complained that though this passage “begins magnificently,” it “bursts into the tonic major in a blatant way.” But he does not connect this dismissive musicological point to any larger musical or dramatic argument about the scene, which strongly establishes not only the significance of Posa’s death for Carlos and especially for Philip, to whom he had become a trusted friend precisely because of his independence of thought, but also the now-irreparable break between father and son.

Despite claiming to find mixed versions mostly unsatisfactory, Porter collaborated on one himself for the Théâtre du Châtelet in Paris in 1996 conducted by Antonio Pappano, which both incorporated some of the original1867 music (but not the prelude) and substituted intermediate versions of some other passages, though Verdi’s later revisions of those are swifter and finer. But at least it was sung in French. Porter frequently complained about companies using the flawed Italian translation, which, he wrote, “falsifies the sense. It blunts Verdi’s keen rhythms and doesn’t fit his phrasing. It makes fancy what is clear and direct [and] casts a film of rich, thick sludge over the clarity of the original.” The Châtelet production was recorded; its cast, particularly Jose van Dam as Philip (in the previous clip) and Roberto Alagna as Carlos (below), makes a persuasive case for the superiority of the French text.

As Porter understood, there is no “definitive version” of Don Carlos, “but, rather, a variety of possible alternatives to be decided on according to the circumstances of the individual performance that is being planned.” It is up to conductors and singers and directors to choose the music they wish to perform, though “this does not provide performers with an ‘anything-goes’ license.” One imagines that in the end Porter would have agreed with Budden’s broad-minded conclusion: “When performed with sufficient musical and dramatic understanding any combination of versions can be made to sound convincing.”

Subscribe and save 50%!

Get immediate access to the current issue and over 25,000 articles from the archives, plus the NYR App.

Already a subscriber? Sign in