Exterminating the Recording Angel

If the film that you are about to see seems to you enigmatic or incongruous, that is how life is also. It is repetitive like life, and, like life again, subject to many interpretations. The author declares that it was not his intention to play with symbols, at least not consciously. Perhaps the best explanation for The Exterminating Angel is that, rationally, there is none.¹

In The Recording Angel: Explorations in Phonography, Evan Eisenberg claimed: "It seems to me that recorded opera is a less artificial act than live opera."² He was responding, in part, to a spat between opera critic Conrad L. Osborne and producer John Culshaw, over the latter's 1967 recording of Strauss's Elektra; such were Culshaw's studio interventions, Osborne complained, "the recording stands in relation to a live presentation much as a movie stands in relation to a stage original: it … must be conceived on its own terms." Culshaw's view was that producers who attempted to capture the sound of an opera house (Osborne's preferred mode), rather than acknowledging that recordings were devised for consumption in the home, were deceiving listeners to a greater extent than those who used the studio to realize the score's complexities so that it "hurt" or "involved" them, as they thought the composer intended. It is difficult to imagine a similarly heated debate taking place between critics and producers today; in the era of the MP3, recordings are more likely to be discussed in the diminishing terms of compression rather than hifidelity.³ Yet the relationship between live and recorded performance continues to be tested within opera studies, through discussions of DVDs, simulcasts, films, and productions that take opera as a starting point to explore transmedial relationships.⁴ What happens, though, when a composer reverses the usual—or at least, classical—sequence of live rendition to recording and creates an opera from a film, as is the case in British composer Thomas Adès's The Exterminating Angel? Must it be conceived "on its own terms," as Osborne claimed of a movie version of a play? Does recasting cinematic characters in live human form threaten to destroy the "recording angel" Eisenberg celebrates?

Thomas Adès's The Exterminating Angel was premiered in Salzburg in 2016 and has since been performed at Covent Garden and the Metropolitan Opera House (from where it was broadcast to cinemas around the world in November 2017). The libretto, by director Tom Cairns, derives from a 1962 film by Luis Buñuel, El, angel exterminador. Buñuel explained the plot very simply: it is "the story of a group of friends who have dinner together after seeing a play, but when they go into the living room after dinner, they find that for some inexplicable reason they can't leave."⁵ Adès apparently saw the film as a teenager and was struck by its operatic potential. There are numerous reasons why El ángel exterminador would attract a composer: the "play" the dinner guests have been to see is in fact Donizetti's Lucia di Lammermoor and the singer of the title role is among them, along with the conductor and his piano-playing wife. It is a moment of musical potential—a performance evaded—that traps and then, by its eventual realization, liberates the guests. Beyond those possibilities for diegetic music, there are opportunities to see reflected in the guest's situation something of the operatic spectator's experience. Cairns claimed: "Opera isn't so much like this anymore, but it has a reputation for being a glittering elitist activity. [Buñuel's characters] have just all been to the opera, so there are all sorts of connections with the opera house. The main thrust of what we want to do is to let people see that it could be them."⁶

Or, as Adès put it, somewhat ambiguously: "every opera is about getting out of a particular situation."⁷ By "situation," Adès may have meant simply a plot's dénouement: comic opera that ends in marriage (Cos'ı fan tutte, La Cenerentola, Der Rosenkavalier), failed romantic...