There’s a great bit in Singin’ in the Rain in which Don Lockwood and Lina Lamont record a love scene for their next costume drama blockbuster, The Dueling Cavalier. While they act out their characters’ mutual passion, Don, sotto voce, insults Lina and talks of breaking every bone in her body; the director, seated a few feet in front yet apparently oblivious to the content of the dialogue, congratulates their efforts. In the silent cinema, what the actors actually said to each during filming other was of no importance, because none of it was heard by the audience; the visuals were all that mattered. Later on, when, following the success of The Jazz Singer, production of The Dueling Cavalier has switched to sound, the actors reshoot the same scene, only this time, they have a script to adhere to; what they say does matter. The technological necessity of speaking into the microphone is too much for Lina to contend with, and her dialogue is only patchily recorded. Don, meanwhile, makes his own error; departing from the script, he exclaims ‘I love you!’ over and over again, as he was wont to do as a silent movie actor in the days when such nonsense didn’t form part of the film. Nobody realizes how ridiculous this will sound to the audience; at the test screening, people jeer and laugh at Don’s effusive declarations of love just as they jeer and laugh at Lina’s squeaky voice and the erratic way it has been recorded. Unintended hilarity is further increased when the soundtrack starts playing out of sync, so that Lina’s voice is heard while the lips of male actors are seen moving. The sound finally slows down to the point that Lina’s voice sounds like a deep, distorted, inhuman roar. The Dueling Cavalier was supposed to be a high-toned drama, but thanks to the vagaries of the new technology, it provokes derisive laughter, soundtrack working against visuals – or, perhaps, revealing them as they ought to be seen, by exposing the genre’s outdated techniques and attitudes (although the best of silent cinema at the time was anything but outdated in technique, something Singin’ in the Rain fails to acknowledge). The rupture between sound and image in The Dueling Cavalier threatens to lead to humiliation for its stars, but the test audience’s reaction proves to be a blessing in disguise. Taking their cue from the laughter, the filmmakers adapt the movie into a musical comedy, changing the title to The Dancing Cavalier. Sound and image are successfully reintegrated, with one exception: there’s nothing that can be done to rescue Lina Lamont’s voice, and so it is dubbed by Kathy Selden. Finally, sound and image are at last fully integrated during a live performance, when the curtain is raised behind the lip synching Lina to reveal Kathy, the true singer.
Above: Don Lockwood romancing (or not) Lina Lamont. See this article for an extra sound/image in-joke on the part of the makers of Singin’ in the Rain.
I thought of Singin’ in the Rain when I watched My Case for the first time. I found myself wondering what the actors actually said during the performance/recording of the second and third repetitions, when the sound is first removed and then distorted. We have already heard Régio’s lines, and so, if we take it for granted that the actors are simply repeating the performance (as happens in the theatre), we know what they are saying even if we can’t hear it. But there’s no way to be sure the actors are sticking to the script; as with Don Lockwood and Lina Lamont, they had the freedom to say whatever they pleased, or to say nothing at all. Oliveira leaves the possibility open of another script, another story, that isn’t told, but left for us to try to construct. In the second repetition, we try (unsuccessfully) to connect the voiceover to the visuals; in the third, we try (unsuccessfully) to decipher the distorted dialogue, straining to catch anything that might be intelligible. The possibility of unknown plots other than Régio’s is acknowledged not only through the variations in staging and the obvious variations in sound, speed, action etc., but also through subtle variations in camera placement. In the second repetition, the camera is closer to the performers than it is in the first, making it slightly more conventionally ‘cinematic’, although there are no close-ups. The camera is pulled right back for the third repetition, so that it’s often difficult to see the actors’ faces (something heightened by the exaggerated high-contrast lighting). The camera is also, for the most part, more static, making the third section seem more like filmed theatre (in the literal sense) and also, perhaps, very early talkies, in which the camera rarely moved due to the demands of the primitive sound recording equipment (or so the story goes). Each time, the action is viewed afresh, heard afresh.
Above: Three views of Bulle Ogier, illustrating camera placement in the first, second and third repetitions. Note also the lurid colors and exaggerated lighting of the third image.
Oliveira’s repetitions, nodding to film history (the silent movie and the early talkie), inquire into the relationship between image and sound in the cinema. Film is a visual medium, goes the mantra; scripts, therefore, ought to aim for verbal economy and avoid wordiness (voiceovers especially). Visual expression is key. Thus, Panofsky writes that ‘[in] a film, that which we hear remains, for good or worse, inextricably fused with that which we see; the sound, articulate or not, cannot express any more than is expressed, at the same time, by visible movement; and in a good film it does not even attempt to do so’. I don’t know if Oliveira ever read Panofsky’s essay, but if he did, I’d imagine he’d have let out a great big ‘Hmmm… ’ at this point, or the Portuguese equivalent thereof. Sound and image become radically disassociated in My Case, especially in the second section, in which Régio and Beckett collide and do not reach a resolution. I mentioned in my previous post a shot (in the first section) of the Intruder, the Doorman and the Actress fighting among themselves while the Author, off-screen, delivers his monologue. In this instance, it isn’t difficult to process both the visual and the aural information simultaneously because, although different things are going on in both, the action we see is simple, and we have the reassuring cognisance of a connection between sound and image: we have already seen the speaker, we know he is in the same space as the other three (even though his portion of that space is now out of view), we have seen him interact with one of them and have every reason to believe that all four will interact some more. The connection between sound and image has been loosened, but not wholly severed. In the second section, however, there is an absence of any readily apparent connection between Régio’s play and Beckett’s text; this very absence might encourage us to supply our own connection, but such an attempt would, I think, be futile except at a somewhat tenuous thematic level (see previous post). There are points at which the Beckettian voiceover coincides with Régio’s characters speaking, but none of those characters can be identified with the speaker, and their various concerns do not mesh with his. Watching and re-watching this section, I found it impossible to devote sustained and equal attention to words and action/image (though there are periods of respite in the form of pauses in the voiceover). Beckett’s text is cryptic and elusive enough in its own right; it becomes even more so when our natural instinct to make sense of things leads us to try to connect it with Régio’s piece. Oliveira’s presentation of Régio, meanwhile, is unusual enough that it invites its own attempts to make sense of it. Why is it silent? Why is it black-and-white? Why has it been speeded-up? Is it a replay? If not, how does it differ from the first version? Was it recorded live? There’s too much going on here to process at the same time, with the result that my concentration went back and forth between sound and image, the inextricable fusion described by Panofsky well and truly undone.