Film is permanent (hopefully); theatre is ephemeral. The inevitable result: directors who split their time between theatre and film wind up being judged for posterity only by their film work. That's just the way things have to be, since we can't actually know for ourselves what it was like to experience their theatre productions in the theatre. (We can guess, from descriptions, stills, and -- hopefully -- video evidence. But all of that just enables us to imagine what the production might have been like.) Some directors' careers look much worse than they really were when they're discussed solely as film directors. Rouben Mamoulian, for example. Film buffs often talk about him as someone who never fulfilled his early promise and got fired from a lot of movie projects, like Laura. But while he was making disappointing movies in the '30s and '40s, he was revolutionizing the theatre with productions like Porgy and Bess and Oklahoma! After he got fired from Laura, he went back to New York and did Carousel. We can't see those productions, so half his career -- maybe the more important half -- is not something we can evaluate well.
All that is a way of saying that we can't fully judge a director's theatre career, even if we have video clips or complete videotaped productions in front of us. (Watching a videotape of a production is not watching a production. I guess you could say that watching a film on TV isn't really watching the film, but at least we're not seeing it from a distance.) But, that said, it's always fun to come across more evidence of a famous film director's theatre career, or to better understand the place theatre had in his career. One superstar movie director who made theatre a huge part of his career was Luchino Visconti, who alternated movies with stage productions, and especially opera productions.
Most of Visconti's productions aren't available in video form; I don't think I've ever seen any video of his work with Maria Callas. They were the most famous director-star opera team of their era, creating (reportedly creating, I mean; I wasn't there) a new level of psychological credibility for so-called "frivolous" Italian operas. Perhaps his most famous non-Callas production was Verdi's Don Carlo at Covent Garden in 1958, one of several '50s productions that turned this opera from a rarity into one of Verdi's most-performed works.
Anyway, in 1964, Visconti re-teamed with designer Filippo Sanjust and conductor Carlo Maria Giulini for a Covent Garden Il Trovatore, and this production was taped (I think by the BBC). Clips of this production have appeared on YouTube, and I've been greatly enjoying watching them. I don't think they bring me that much closer to understanding what Visconti brought to opera production, especially since Trovatore is not a piece that calls for a very interventionist directorial approach.
The plot of Trovatore takes a lot of crap, though it's actually a good libretto with a strong, solid melodramatic story; unlike many confused, convoluted plots, it's quite easy to follow and the characters' motivations are always clear. But there's nothing very complex about most of the characters. They mostly have very simple, unadorned motivations of jealousy, love and duty, and they express these feelings in equally simple, direct ways. That means the director needs to keep things simple as well, not get in the way of the characters. That's the way Visconti seemed to see it; in Il Mio Teatro he's quoted as saying that the characters of Trovatore are all preoccupied with their own emotions, that they express themselves to the fullest when they sing, and that the things they sing rarely demand an answer from anyone else. (In other words, the characters are not trying to communicate with each other. They're just thinking out loud.)
Anyway, what that all adds up to is that Visconti is not trying to re-invent Trovatore or even make a statement about what he thinks it means; he just wants to give the characters a chance to express themselves to the fullest. So his staging of Azucena's entrance song, sung by the great Italian singer Giulietta Simionato, isn't that much different from the way it's done in A Night At the Opera. But that's because there aren't many ways to do this number without getting in its way. It's a crazy old gypsy woman remembering the time her mother was burned at the stake; the director's options are limited, and it's up to the singer to sell it -- which she does. Incidentally, Simionato is still alive as of this writing, going on 100. That's almost as awesome as she is.
This is the number in two parts: the song, followed by the recitative (with tenor Bruno Prevedi) and then in part 2, the follow-up musical narrative where she fills us in on the backstory: she kidnapped the child of the man who put her mother to death, but in her insanity and delirium she murdered her own child instead, and raised the other child (the tenor) as her own. Yes, that pretty much is the inspiration for every Gilbert and Sullivan opera ever, but these baby-switching melodramas were very common in the 19th century, and this is one of the more plausible examples, because the character is just crazy enough to have done something like this.
The famous thing about this production was that Leontyne Price, originally set to star, had to pull out at the last minute and was replaced by one of Covent Garden's resident sopranos, Gwyneth Jones. She did a fine job, probably better than Price would have. (Price sang a lot of this role beautifully, but she wasn't good at the coloratura that other parts demanded. Jones even did the trills.) It made her an international star, but unfortunately, she didn't live up to the promise she displayed here; she took on a lot of heavy parts and her voice developed a heavy wobble. She was a fine singer with a long career, but a different kind of career than this would have suggested.
Here she is in the Act IV number followed by another number featured in A Night At the Opera, the "Miserere." (The tenor, Prevedi, was one of a number of good-but-not-great Italian tenors in the '60s; it was probably the last era when there were a lot of good Italian tenors. By the '70s, Pavarotti was practically the only one.) Again, I wouldn't say there's anything striking about the staging except that it doesn't get in the singer's way with a lot of extraneous detail; there isn't even anybody on stage except her during the "Miserere" number. (The director has to decide whether the tenor and the chorus should be offstage throughout or if they should be visible, like the tenor is visible at the window in A Night At the Opera.)
Visconti felt that Jones and Prevedi didn't quite get what he was going for, the idea that the characters were emotionally isolated and not listening to each other or waiting for a response, but that the singers who did get it were Simionato and (as the villain) the British baritone Peter Glossop. Here they are in the scene where Simionato is captured, and Glossop discovers that she is (he thinks) the person who murdered his brother and the mother of his romantic rival. That's fairly clear; what Visconti and Sanjust were thinking in putting the villain in a hat like that is less clear.
And here's the finale, which ends as follows:
- Soprano has taken poison, dies.
- Tenor is hauled offstage to be put to death.
- Baritone gleefully tells mezzo that her son is dead.
- Mezzo tells baritone that, no, the tenor was only her adopted son and he was actually the baritone's brother.
Verdi, as he frequently does, wraps up all this plot stuff in about one minute near the end, so the director's job is to make it all as emotionally clear as possible; every action has to "read" in only a few seconds.