The Ontology of the Recorded Sound

Andre Bazin's essay "The Ontology of the Photographic Image," maybe his most important, is now 63 years old, almost a quarter-century older than Bazin ever was. A gentle knife driven into the mind anyone who's ever read it and plenty of people you haven't. The blueprint for the third floor of the constantly growing tower of film theory, always almost reaching Heaven, but never quite getting there (once a floor is finished, we discover that there's one more to go).

Barely eight pages long in the University of California Press English edition of What is Cinema?, the one with the mauve color, "The Ontology of the Photographic Image" is followed a few pages later in that edition by a composite essay called "The Evolution of the Language of Cinema," edited from three pieces published in different magazines during the 1950s. "The Ontology of the Photographic Image" is important to Bazin's ideas like a seed is important to the tree it grows into. A fruit-bearing tree. "Ontology" and "Evolution" are masterworks of thinking. In these few pages, Bazin establishes a world of possibilities circling around a single idea: ontology. Realism. Reliability. What is the "reliable aesthetic," the one that doesn't try to manipulate its audience? Here is the idea of cinema as something other than a manipulation of reality; cinema can be a window to something real. Where is the truth? "Depth of focus brings the spectator into a relation with the image closer to that which he enjoys in reality." The truth can be discerned in the long take. But while he talks about a deep focus image, he never talks about a deep focus sound. Welles, Bazin's great ontological poet, chose his sounds very selectively, making a radio play on the soundtrack and a shadow play on the screen. The great post-syncer who would sometimes dub in the lines of his actors himself, imitating their voices. Yes, sound made the deep focus long take possible, but it's seen as only that: a means to an end. Why?

The primacy of the image in Bazin's writing--and in most writing on film--is understandable: the moving image was our first way to express the notion of cinema. The relationship between sounds and images in films is like the relationship between two siblings born a few years apart. When cinema was young (or middle-aged) a difference of 30 years meant a lot; sound was an upstart kid. But now that cinema is over a century old, the age difference isn't that much of a factor. So why do we still treat sound like the image's kid sister? Why do we write so much about images and the ways they are edited together and so little about sounds and their editing?

Yes, for a long time moving images were exclusive to movies. But now they've become a part of so many other media that there is no longer a direction correlation between the two. The Godardian notion of the Death of Cinema is the death of the orthodox cinema (a movie as a moving image) and the beginning of a more ambiguous notion of cinema, the birth of Cinema as a pure idea, like Art (which, let's not forget, was once inseparable from specific forms and subjects). When cinema dies (if it hasn't already), it will continue to haunt the world as a ghost, imbuing every action with its spirit. We are coming to a point where a moving image is no longer necessary to make a "film." Sounds and images have become equals, and so we should begin thinking of them that way. And, in order to do this, we've got to start at square one: to begin rethinking old, established theories and how they work in this context. The placement of the microphones should be just as important as the placement of the camera. Bazin's ontology is the foundation of many of the last half-century's ideas about cinema. If we believe in a "realist" (ontological) image, then we must believe in a "realist" sound. So what defines an ontological sound recording?

In 1930, Erich von Stroheim was given the opportunity to remake Blind Husbands as a sound film. The film was written and cast, but production was shut down a week before shooting was supposed to start; the problem was with Stroheim and his ideas about making a sound film. Stroheim wanted to record all of the sound directly. He believed that sound effects recorded in a studio were not the same as the actual sounds. He wanted to hear the actor's footsteps, and not the footsteps of a sound engineer putting on a pair of prop shoes and stomping on plywood. He wanted to record coach bells jangling by a lake because he felt it would not be the same sound as the one a foley artist's bells would make. He wanted a waltz band to perform live during a shot instead of having the musicians mime so that a pristine recording of the music could be dubbed in later. Stroheim's unrealized idea represents the two most important facets of what could be called an ontological sound: its importance in terms of the filmmaker's relationship to the viewer (that they are not "manipulating" but trying to present the microphone's view of "reality) and its importance as a gesture--that even if a post-synced sound and a directly recorded sound are identical, there is a moral distinction, an action, being made by using the original.

Of all the New Wave filmmakers, Jacques Rivette was the most Bazinian, and his films, full of creaking floors and camera noise, represent ontology not only in their long takes but in their use of direct sound. I think of The Nun and its noisy footsteps, and The Duchess of Langeias and its loud furniture. Hou Hsiao-Hsien made the first Taiwanese film to use exclusively direct sound, City of Sadness, shooting (and recording) in long takes not only as a way to bring out the truth of a past world but to capture the noise of a passed time, the little irregularities of dialects and accents that would probably be replaced with an easier-to-understand, general Taiwanese Mandarin if he went to post-sync it at a studio. Here is the ontological sound inseparable from the ontological image. A dual "realism." A completed circle.

The ontological image is not superior to the manipulated one; both are equally "cinema." But the idea of its importance led to so many ideas and counter-ideas, just like the auteur theory did. It's a statement made to provoke responses. So if we attach importance to the ontological sound, what is the place of sound-montage, or of the mixture of ontological and artificial sounds that makes up the bulk of most movie soundtracks? There is more, of course. There's no such thing as a complete answer to a question--only more questions.

The Color Green

An Affair to Remember and Police Beat

Would anyone remember Lee Marvin's scarf in Seven Men from Now so well if it was the usual red or yellow instead of that fertile green that stands out against the dusty brown of Lone Pine?

Why does the color green always stand out so vividly in movies? Is it because we see it so rarely? It's like the George Harrison of the movie palett: an essential (and underrated) element, and it's always a pleasure to see it take the lead.

It seems like filmmakers' passions always lie with red or blue, never with green. Maybe Technicolor is to blame: it was so good and vivid with those colors and green was more the domain of Agfacolor (also known as Ansocolor). Jacques Tourneur's Stranger on Horseback has some of the sharpest greens in cinema, thanks to the Anscocolor, but they say the director was never happy with it.

Blue was fashionable for a while. I hope green will be all the rage some time soon.

Small Head of E.O.W., by Frank Auerbach

Oil on board, 1957-58

Some things about what I imagine the painting to look like make me uncomfortable. I think of the feces-stained walls of Hunger's prison cells, and how this vaguely fecal color forms a face. Small Head of E.O.W. doesn't look pretty, but it seems beautiful. The basic outline of a head, neck and hair interrupted by a single Egyptian eye that seems to be pointed at something behind the sitter. The suggestion of shoulders that are both delicate and hulking. This human shape being something that exists on the surface of an undulating mass, like sea foam forming on waves. A persona shaping. Auerbach, successor to Joyce.

The Trouble with Hats

Those Awful Hats (1909), directed by D.W.Griffith


Fantasmagorie (1908), by Émile Cohl

Movies as windows into the past. Both D.W.Griffith's Those Awful Hats and Émile Cohl's early animation Fantasmagorie feature jokes about something that was apparently a big problem for filmgoers a century ago: women's hats blocking the screen.

I think of the countless clever shorts played in American movie theaters (especially AMC) nowadays that tell the audience to turn off their cell phones--the Indians sneaking up on a herd of buffalo only to have it scared away by an obnoxious ringtone, the submarine whose presence is given away to the enemy--and I hope that someone will preserve them so that people will watch them a hundred years from now and laugh that we fussed so much about phone calls.

The Missing Passage

The recent Region 1 boxed set of David Lynch's work--"The Lime Green Set," as it's called--includes, amongst deluxe versions of his short films, Eraserhead, The Elephant Man, Blue Velvet and Wild at Heart, a DVD that's referred to as the "Mystery Disc" on the packaging. Its contents are fascinating: the Rabbits online show, a large collection of shorts and, most interestingly, deleted scenes from the Wild at Heart work print.

It seems like the most accurate term for describing any moments ("scenes") from Lynch's films is to call them passages, a term that's also handy with Godard. So these aren't so much deleted scenes as omitted passages, sentences cut from the ends of paragraphs, the Gypsies that were trimmed out and painted over on the canvas of Manet's Boy with a Pitcher.

Importantly, these missing passages aren't included on the Wild at Heart disc--by being placed o the Mystery Disc, they serve as an afterword, a collection of numbered moments that we can insert into the Wild at Heart of our imagination.

The sequence labeled Scene 78 is the most fascinating, as potent and doomed as the famous Winkie's scenario in Mulholland Dr. or the phone call in Lost Highway. Like those sequences, it's in some ways completely self-contained. Viewed outside of the context of the movie, it could be a short film. On the other hand, if imagined as part of the whole of Wild at Heart--or, more importantly, once viewed, remembered as being part of Wild at Heart (and with Lynch the memory of a film is as important as the act of watching it)--it becomes significantly more ambiguous. It's both completely fathomable and a total mystery.




The Tradition of Consistency

Theophrastus gives us 30 characters. His intention isn't dramaturgy or social observation. He wants to be a guide (like a guidebook), to create a text that would help describe (you could say characterize) people in the same way a map describes a city (incompletely, that is, creating a set of names and coordinates to navigate through a whole with infinite social possibilities). So we have The Boor in the same way we have Logan Square and The Coward in the same way we have Lakeshore Drive. R.C.Jebb's 1870 translation has the old Greek writing:

"I will describe to you, class by class, the several kinds of conduct which characterise [people] and the mode in which they administer their affairs; for I conceive, Polycles, that our sons will be the better if such memorials are bequeathed to them, using which as examples they shall choose to live and consort with men of the fairest lives, in order that they may not fall short of them."

We've invented many more systems--as many systems as there are people, probably even more. Advertising is a system for dividing a population into consumer groups. Politicians create a system that clearly delineates people into us and them. A system is always a knife. It's for carving up a whole into pieces so that it can be more easily digested. People are not characters; character, characteristics are a system we use to more easily analyze something. When we think of any person as a character (and, in our memories and our experiences, we frequently do) we cut off part of them, and we can keep cutting. Perhaps this is the great Utopian aspect of cinema: that it does not include a system for dividing up its audience.

So why is it that film analysis -- whether it's published criticism or two people talking in a lobby -- wants to tether itself to a notion of character -- a very clearly delineated, orthodox concept of character, whether in respect to the subjects or the filmmakers (in auteurism, themselves characters)? There is nothing wrong with the character system -- many films have been made with it in mind. But it is never a good idea to use the same system for everything. Every film calls for a new approach. Though literature is our favorite scapegoat when it comes to attacking lazy criticism, adherence to the character system, the character tradition, is more likely the fault of the guide we frequently use to describe what we think constitutes "good" acting: consistency. A person has to be like an equation ("X would never do that!" when we should instead accept what X has done and try to understand it, and that involves more systems--morality, ethics, etc--illustrating that the mind is a sort of infinitely regressing trap).

The art of creating an "original character" is like the art of cooking: a new combination of old ingredients. Any original idea can be subdivided into older ideas. This does not discredit originality--but we should see things for what they are. Behind every new idea is a transposition of an old idea, a previous assumption. There will be new ideas as long as there are old ones. The older the better.

People who dislike David Lynch's work, or at least certain aspects of it, will occasionally mention the "poor characterizations"--especially in Wild at Heart. But with Lynch a person is never a character--instead, a character is ascribed to them. They are given names, but those names aren't theirs to keep. Sometimes a single set of characteristics can travel between actors (Mulholland Dr., Lost Highway) or the same actor can be given multiple distinct sets of characteristics and names (Inland Empire). Or, as in Twin Peaks, a notion (whether it's a character, a feeling of guilt) can spread out like an umbrella and cover multiple persons at different times.

There are no "characters" in Ozu's late films. Instead, familiar faces are given names and roles, but they're never characters in the orthodox sense. There are no "characters" in late Hitchcock (certainly not in Topaz): instead, images of people serve as magnets for the metal filings of our emotions. And the people of Cassavetes and Pialat are like quotes with ellipses on either side: there is a acknowledgment that we are seeing something out of context, removed from its reality, but it is important that this removed passage be quoted.

Lynch, along with Abel Ferrara, presents the most intelligent portrayal of identity of any contemporary American filmmaker: in every one of us, there is a million, and sometimes a million of us are one. It depends on the system.


Cut-Ups

Laura Dern in Inland Empire is as many characters as the viewers wish to divide her into (the more adventurous can say that she is really only one). Inland Empire and Mulholland Dr. are films without scenes--a scene, after all, is a system by which a whole can be clearly delineated. Instead, by presenting the work as an interconnected whole, Lynch gives us the opportunity to create our own systems, to cut them up as many ways as we want. Mulholland Dr., especially early on, frequently coalesces into something akin to scenes. But Inland Empire is a long cinematic superimposition, posing the same question a double exposure poses: are we looking at a single image, two distinct images, something else? As behind every new idea there are old ideas, behind every recognizable image is complete abstraction.

In 1984, a young Austrian named Peter Tscherkassky, a former student of journalism (the system of shaping experience into "news") and philosophy (the system of systems!), laid three minutes worth of celluloid down in a darkroom in the shape of a film frame and exposed a single still image from the Lumiere's first "actuality" (really the second take) on it, each blown-up microscopic section of the frame creating a new image. Concreteness—"realism"—is something that only exists on the surface, and the moment you begin to divide you find abstraction.

As a painter, Lynch knows this well. The painter coalesces, combining something completely separate from the painting (the paint) to create an image. Even in a "realistic" image, if we look far enough, we will see that there are "abstract" elements: colors.

This is what makes the Bible one of the greatest works of art. With its numerous versions, translations and rewordings, it is a text that can be subdivided infinitely. It provides us--like the concept of characterization when applied to people--with a rudimentary system (chapters and verses) as a guide (like a map), but it can be broken up and re-contextualized any number of ways: by word, by sentence, by any number of pages, verses, words, sentences, chapters, and, of course, passages. The Bible is a machine capable of generating an infinite number of meanings, working as long as someone somewhere is even thinking of it. And so is the cinema.



The Intruder

The first image of Scene 78 is one of its two close-ups. Johnnie Farragut (Harry Dean Stanton) is playing with a novelty pen. A wide shot: two men--who will be named later as Reggie and Dropshadow (or, rather, they will have those names ascribed to them) arrive and introduce themselves. They're embodied (and supporting players in Lynch films are always embodying something), respectively, by Calvin Lockhart and David Patrick Kelly.

The preservation of the original splices in the work print gives every cut a jittering quality, like a snare hit. Even the most casual viewer becomes aware that the image consists of pieces attached to each other to form a chain. The soundtrack is finished, crisp, clear almost as if to contrast with the unfinished nature of the images. It is a pair of ominous arms trying to hold something together, trying to keep it from falling apart.

Stanton, with his beige suit and dapperly held cigarette, is doing his best Cary Grant, his best American cool to Lockhart's affable Honduran. Lockhart is playing friendly, talkative, but what comes through is a subliminal menace. His blue shirt seems tuned to the low hum on the soundtrack, its electrical blue seeming more and more intrusive as the dialogue progresses. Lockhart says he runs an appliance repair shop and then tells Stanton he works for the Honduran government. He has a license to kill. He tells Stanton that he and Kelly are going fishing and them offers to show him his combat pistol, holding up a metal briefcase. You could edit out half of his lines and make him into a friendly tourist. Or you could edit out the other half and make him into a dangerous rogue. As Stanton stands up to leave, there is a close-up of a tattoo, somewhat illegible, across Kelly's knuckles.

Lynch is a filmmaker who is sometimes criticized for his simplistic worldview. That observation is accurate, though the idea that it's a detriment is misplaced. His work consists of many conflicting simplistic wordviews operating side by side, several systems going at once. In Lockhart's character alone, there are two possible pieces we can cut out and make something easy to understand (he is a friendly foreigner with a thick accent on a trip to New Orleans / he is a killer working for a foreign government), both originating from archetypes. Simple ideas an ordinary American like Stanton might have about people from other countries. Lynch does not reject the notion of "consistent character"--he embraces it more than any of his contemporaries. The more "consistent characters" a person can be, the merrier.

But say we put those two halves together, like Scene 78 does. What we are left with is not two characters, but a sort of infinite whole. Something that by contradicting itself (as the placidity of the wide shots is contradicted by the menacing second close-up) in every direction creates an ambiguous image from which ideas can be carved by anyone willing to hold a knife. Lynch can be cut up forever.