Tannhäuser: Wagner's "problem" opera
Well, I'm working up to the Ring. But it's important to remember that, throughout most of the 19th century, Tannhäuser was Wagner's most popular work. With Lohengrin, Der Fliegende Holländer, and Rienzi—early works all, it was one of the few Wagner operas that most operagoers would have heard. Not only that, a good case can be made that the opera's unprecedented sensuality, its convincing musical argument for a carpe diem philosophy, and Tannhäuser's righteous attack on bluestockinged hypocrisy contributed, more than any other single work outside of Wagner's own prose writings, to the founding of the French Symbolist movement at the 1861 Paris premiere (16 years after the Dresden premiere). It could also be said that Gautier and Baudelaire—one of the few Parisians to support Wagner in 1861—and Mallarmé never recovered. (As a contemporary described this last poet at Wagner performances: "The orchestra dictated and Mallarmé wrote.")
Of course, Tannhäuser is worth listening to today for other reasons, not least because it contains a great deal of absolutely gorgeous and thrilling music. But it's important to remember that, with all its flaws, as of 1846 it remained the most revolutionary opera that had been seen on any stage, and had the lushest orchestrations. It is also fascinating as a laboratory in which Wagner, the working musician who covered his compositional tracks better than anyone since Mozart and Bach (in his other works, hardly a seam ever shows), is, for the only time, revealed uncomfortably straddling several stylistic stiles before going on to Lohengrin, the Ring, and the artwork of the future. A unique opportunity, therefore, to study the revealed craft of a composer who later became the very embodiment of the art of transition.
But in addition to the opera's historical, musical, and even literary importance, a buyer's guide to Tannhäuser is of unique value among Wagner's works because no fewer than three substantially differing performing versions of the opera have been used for the six recordings under discussion. Conveniently for our purposes, there are two recordings of each of the three versions, which came about as follows:
Immediately after the very first opening night in Dresden, in 1845, there came a set of hasty rewrites, mostly of Act III; the work was again revised between 1847 and 1852; and again, most substantially, in 1861 for the Paris premiere, which has gone down in musical history as a debacle rivaled only by the Paris premiere of Stravinsky's Le Sacre du Printemps. The members of Paris's notoriously conservative Jockey Club were used to skipping the first acts of all operas and arriving, usually drunk, just in time to ogle the legs of the corps de ballet in the (therefore) de rigueur second-act ballet. The management of the Paris Opera thus demanded such a ballet from Wagner, who, after some initial (and quite understandable) balking, agreed.
But Wagner endeared himself to no one by not only breaking all house records for rehearsals (163!), but also by inserting the ballet—since known as the "Venusberg Music"—at the very beginning of the first act. Nor did it help that Wagner's patroness of the nonce was the less-than-popular Princess Metternich, wife of the Austrian ambassador. On opening night the Jockey Club arrived on time, for once, but brandishing silver whistles, which they proceeded to blow throughout the opera. Fistfights broke out in the audience, and the production closed after three performances.
For Paris, Wagner also deleted Walther von der Vogelweide's Act II contest song, greatly expanded the role of Venus in Act I,i, and substantially rewrote the Act III prelude and the opera's final pages. Uncharacteristically discontented with even this many alterations, he then revised the opera a final time for an 1875 Vienna production. Still, in the last year of his life, Wagner admitted to his wife, Cosima, that he yet owed the world a Tannhäuser.
Most would agree; Wagner's many revisions of the work seem to have created as many problems, mostly of consistency, as they ever solved, though few would now wish to do without all the new music Wagner wrote for the Paris production, much of it now this "problem" opera's best passages. However, in his own defense, Wagner wrote to Mathilde Wesendonck in 1860, "Now when I had written Isolde's final transfiguration (footnote 1), 1 I was at last able to find the right end for the Holländer overture—as well as for the horror of the Venusberg."
The trouble is, that's exactly what it sounds like: Tristan out-takes spliced onto a much earlier work written in a wholly different style. Wagner did not hesitate to use his newly matured chromaticism, even quoting Tristan's four-note upward-sliding chromatic "magic" theme (though adding a fifth semitone). But the stylistic clash was even more glaring to contemporary audiences than it is to us today—as of 1861, no one had yet heard Tristan. These disparate musical languages call undue attention to themselves under all but the most diligent conductors, though I'm hardly the first to suggest that the opera's split musical personality mirrors, almost too explicitly, Tannhäuser's dual desires as embodied in the two female roles.
With all the substitutions and deletions, then, there's no simple choice between the Dresden and Paris versions: though the latter has by far the lusher, more commanding, unequivocally erotic music, and the crucially expanded role for Venus, the former's overall structure is more balanced, dramatically more sound. Bernard Haitink makes an excellent case for the Dresden version in his very fine 1985 recording (EMI CDCC 47295). Yet the Dresden Venus, with her many fewer lines, is a one-dimensional character compared to the Paris Venus's rich though still concisely drawn complexities.
Footnote 1: Wagner himself never called it a "Liebestod."