Wednesday, October 26, 2005

The triumph (?) of Ariane

Ariane et Barbe-Bleue, Paul Dukas' sole opera, is a prequel of sorts to Debussy's Pelléas et Mélisande. But while the latter is (despite legions of detractors) widely loved and performed, the City Opera run of Ariane that ended last Saturday is notable for its having happened at all.

Why the difference? We might ask Leon Botstein, conductor of the City Opera run and a 1999 concert performance of the piece. Back then, he wrote:
We always want to believe that the standard repertory reflects the enduring best of music. If something is not standard and popular, we often assume that there must be a good reason. But that is frequently not the case. The truth is that in the performing arts, particularly music, what remains in the standard repertoire is the result of habits and tastes that have as much to do with convenience and prejudice as with anything we might call quality. If we listen to Ariane, we might have difficulty in finding enough fault with either the music or the libretto of this masterpiece to warrant its disappearance from the stage.
This is, I think, half-right as to the Dukas. The music is certainly interesting -- a more straightforward take on the post-Wagnerian elements in Pelléas, one which nevertheless reaches some depth of mystery in, e.g., the offstage chorus of the first act. Were it easier on the voice, this music alone might've earned Ariane some fringe status in the operatic canon.

But it's hard to grasp how the author of Pelléas, one of the great literary peaks in opera, came up with this clunker. The joke about Maeterlinck & Debussy's masterpiece is that nothing happens (...and then Mélisande dies). Yet at every moment something is struggling to happen. Characters reach towards each other as strongly as those of Don Carlos; their failure, as the Schiller/Verdi characters' failure, is the drama and pathos of the piece. And more than that: that the space between these people can't be overtly pierced by them makes it the mystery of the world, a space in which the opera's flowery "symbolist" text and subtle musical shadings float as longed-for explanation, not obfuscation. It's an ideal marriage of schema and style.

Ariane et Barbe-Bleue is the opposite. The action goes straight forward, but it's the lack of impediment that makes the evening dramatically inert. Ariane neither suffers nor is threatened nor even really doubts -- she simply marches down as she came to do, brushing off token opposition from the Nurse and Bluebeard, and frees the wives (including "Mélisande"). Soon after, the wives decide they want their bondage back and Ariane, apparently unperturbed, walks off. Her only struggle is with the score.

Whether Maeterlinck's post-Pelleas concoction works as allegory (of women's lib perhaps, or wedding-night insanity?) or perfumed poetry, this dramatic nullity's been pretty much sufficient to keep Dukas' work off the stage. Nonetheless NYCO failed to make the best case for Ariane, using an overly literal, poorly lit, and too-often silly production in which serious and mysterious elements could find little foothold. Paul-Émile Fourny of Opéra de Nice did little with this City Opera directing debut.

I'd enjoy listening to the musical portion of the opera at home, in a good recording. But what first-rate soprano would bother with the killer title part? Vaness replacement Renate Behle tried gamely but was overstrained and overmatched. Ursula Ferri showed a stronger voice as the Nurse, but couldn't cope with the high parts of her big door-opening sequence. The wives sang what little they had well, as did Ethan Herschenfeld as Bluebeard. Botstein advocated impressively from the pit.

But if all these forces had been arrayed for an Ariane-as-black-comedy, set by the young Hindemith... That would've been something.


  1. I can't take issue with you over the "nothing happening" problem with this opera - that's true enough, certainly in canonic dramatic terms. Yet I think that you're underestimating this opera, and particularly doing it a disservice in comparing it to "Pelléas". It is much closer in concept, both musically and dramatically, to Bartok's "Duke Bluebeard's Castle".

    All three opera (Debussy, Dukas and Bartok, composed in that order) share the common point that *none* of the characters are real. They all represent states of mind, not actual people. Debussy is the most humanist, the most subtly layered, but Dukas and Bartok share a curious intransigence that is a clear reflection of a certain period view. There are, actually, quite a few operas (and particularly quite a few French operas) of this immediate pre-WWI period that deal with characters that are more iconic than realistic. Which is, perhaps, one definition of Symbolism.

    The choice of names, in itself, is terribly significant. The other wives; Mélisande, Bérengère, Sélisette, Ygraine - these have a sort of Pre-Raphaelite quality about them. You can just see these wispy, ethereal women. They're not instruments of destiny, they are pawns. Ariane, or Judith, on the other hand - these are powerful names, steeped in myth, but should not be mistaken with feminist symbols. Certainly Dukas was quite clear in *not* wanting his opera to be considered as a feminist allegory, however tempting it might be to portray it as such.

    The point with both of these epic characters is the sacrifice they make. Judith's search for light tears away all the protections her husband has put in place to hide his deepest secrets, to the point that he can no longer tolerate her presence save as a cherished memory. Ariane's search for truth takes precedence over everything else in her life, and notably over all social relations, for she sacrifices her marriage and her kinships (her "sisters") to her implacable need to understand. This is where Maeterlinck rejoins Schiller in a sense of the demands of duty over personal needs, whatever the artifices of his language. You say :

    Ariane neither suffers nor
    is threatened nor even really
    doubts -- she simply marches
    down as she came to do,
    brushing off token opposition
    from the Nurse and Bluebeard,
    and frees the wives

    I think that it's wrong to percieve this as dramatic ineptitude, but as an inherent tragedy. Ariane walks into this marriage with the fully-formed notion of rescuing the wives. She knows, in advance, what she is prepared to sacrifice in search of that knowledge. Bluebeard has lost from the moment he accepted her presence in his life, something of which she is fully aware, and this is precisely why she will not permit the peasants to harm him, and constantly assures them of her own safety. This is no hapless heroine in need of rescue, constantly nagging Sister Anne in the hope that the cavalry is coming. That was never her function in life. She manoeuvred herself into this marriage, it's clear from the very first words; she's conscious of her charms, or Bluebeard's weakness towards her. The interest is not in the action, but in the reaction, and that final scene, when the other wives choose not to face the full blaze of Ariane's diamond-bright light, but to remain in the dappled shade with their broken Bluebeard, constitutes a veritable dramatic climax.

    I will add that, as with "Pelléas", it becomes imperative to be able to hear the text clearly, or to have access to a first-rate translation - something very difficult to achieve with Maeterlinck - in order to grasp the nuances of the language. Dukas' score is beautifully structured, in my opinion, in that it lets the voices ride over the orchestral texture. Yes, it's a killer sing for the Ariane - she has precious little rest during the two hours of the opera, and it's a very difficult tessitura - but she doesn't have to fight an overly dense orchestration.

    In case you didn't know, "Ariane et Barbe-bleue" was not actually conceived as a stage play, unlike "Pelléas". but always as an opera libretto, and was written by Maeterlinck very much in reaction against "Pelléas" in its operatic setting, mainly because Debussy refused to cast Maeterlinck's mistress Georgette Leblanc as Mélisande. There's no point attributing high motives to creative artists - they're just human beings like the rest of us! :-) Just more talented.

  2. Thanks for your thoughts. I think I have to take issue with this, though:

    "All three opera (Debussy, Dukas and Bartok, composed in that order) share the common point that *none* of the characters are real. They all represent states of mind, not actual people."

    While the characters have an archetypal cast to them, I think all the players in both Pelleas and Bluebeard act towards each other in recognizable human terms -- that is, as actual if narrowly-viewed people.

    In the Bartok, for example, Bluebeard is the Self (well, of one sort anyway) and Judith is the Other... But they enact not a masque between two "states of mind" but the human drama underlying many a relationship.

    That's what I don't see in Ariane: the women, as you say, are pawns, so that she herself, whether you see her as human or not (I don't; she seems an irritatingly smug demi-god), doesn't have any recognizable interactions with any of them. Ditto Bluebeard.

    But maybe I'll change my mind upon living with a recording.


Absolutely no axe-grinding, please.