Learning Space dedicated to
the Art and Analyses of Film Sound Design
What's new?
Site Map
Site Search
Sound Article List
New Books

Walter Murch in Conversation with Joy Katz
from PARNASSUS Poetry in Review: The Movie Issue (vol 22, 1997, p 124-153)

J.K.      The English Patient is a lyrical, non‑linear novel with an elaborate flashback structure. Can you talk about the freedom this allows in adapting it to the screen? Are people less susceptible to holding the visual images accountable to the story as they know it? Is there more room to interpret?

W.M.       In any film with a flashback structure you do have an extra degree of freedom in the ways the "beads" of the story can be strung together. The connections from scene to scene, particularly the transitions into the past and back to the present, are more allusive than they are in linear material, where one scene seems to trigger the next, like billiard balls colliding.

On the other hand, that freedom exacts a price. The filmmakers must have a strong, intuitive feel for the rightness and the "ripeness" of those transition moments; there are fewer objective criteria for what will work or not - if the transitions feel awkward, premature, or intrusive for whatever reason, the film can quickly become confusing or tedious.

Anthony Minghella, the writer-director of The English Patient thoroughly reworked the time transitions in adapting Michael Ondaatje's novel. But in editing the film, Anthony and I in turn revised Anthony's revisions, such that only seven of the screenplay's original forty transitions made it into the finished film. The other thirty-three were reinvented according to which scenes now found themselves adjacent to each other, and what worked in the language of film rather than on the page. Some things that look great in print fall flat when you see them up on the screen, and vice versa: Things that seem inconsequential on the page sometimes become luminous on the screen.

J.K.       Can you talk about that process?

W.M.       Even when you are shooting a film based on a "linear" screenplay, the challenges in structuring the material for the screen are similar to the challenges facing the translator of a text from one language to another. In the case of a film adaptation, though, it is the translation from the language of text into the new language of image and time.

There's a different weight that a moment (of picture and sound) carries in film, compared to the same moment conveyed by the written word. Everything in film is specific: this person with this color hair, saying these lines in this way, dressed in these clothes, lit by this light slanting at this angle, with these sounds in the background, etc. These details are always on screen. Every time you see a certain character, you are reminded of his haircut, his gait, the color of his eyes. A novelist need mention eye color only once. The mass of all those details, and therefore the amount of processing that your brain is obliged to do, gives a heft to film that text, which is suggestive and allusive, does not command. So you can often "take corners" in text - make sudden leaps and transitions - at speeds that would wreck the film. On the other hand, sometimes the opposite is true: The old saw that "a picture is worth a thousand words" is quite valid under the right circumstances.

In the screenplay of The English Patient, for example, there was a flashback to the desert quite soon after Hana and the Patient arrive at the monastery. It seemed fine in the text of the screenplay, but when we assembled the film it was clear that we needed to stay in the monastery longer before departing into the Patient's memory - to get our sea legs, so to speak, and familiarize ourselves with this new location and these two people suddenly alone together. But changing the placement of that transition meant that there were consequences down the line. We had to alter subsequent transitions to compensate for delaying the first.

But later there's a momentary transition back to the Patient during the sandstorm, just a single shot of him, with a dissolve of Katharine's hand seeming to caress his face. This wasn't in the screenplay. If you were to try to convey the complex feeling of that image in words alone, it might take more effort than it was worth.

In addition, there was the simple question of length. The first assembly of The English Patient was four and a quarter hours, so more than one third of that material had to be trimmed away to get to the present length of two hours forty minutes. As a result, many scenes were eliminated, bringing the survivors into closer, unintended proximity. This was sometimes serendipitous. When it wasn't, we had to discover other ways to structure the material, which in turn led to different interpretations, and so on.

Von Clausewitz, when asked to define war, said that "War is diplomacy carried on by other means." Taking his lead, I would say that film editing is writing carried on by other means.

J.K.       What are the pitfalls of making a literal adaptation of a work? Is there an artlessness to scripting and shooting the scenes in the same order as in the book?

W.M.       That depends on the book. It's the equivalent of making a literal translation from one language to another. Sometimes that literalness is appropriate, sometimes not. The Czech word "litost" may mean, literally, "compassion" in English, but because of how Kundera uses it in The Book of Laughter and Forgetting, "compassion" doesn't feel right. "Self-pity" would be closer, but elements of remorse and longing, and the urge to abase oneself, are in "litost" that are not in "self-pity." In fact, there's a specific word for self-pity in Czech, but Kundera passed over it and chose "litost" instead - "compassion" put into a self-pitying context. So the translator must know not only the literal translation of the word, but also all the words that the author could have used, but didn't. A penumbra of unchosen options hovers around every word in the original text. The more poetic the text, the larger and more luminous the penumbra.

An extreme example of the error of literalness is the 1947 RKO film The Long Night, directed by Anatole Litvak and starring Henry Fonda, which was an American adaptation of the French film Le Jour se Lève directed by Marcel Carné in 1939. During preproduction, the Americans realized that they owned not only the rights to adapt the film but the rights to the film itself. It was a good film, it worked, so why spend all this money adapting it when it could simply be reproduced shot for shot? The film was then broken into pieces, like a jigsaw puzzle, and each piece exactly reproduced, except American actors were speaking English. The director would refer to the French film and issue the appropriate instructions: "The camera is here, the lighting is like this. Your character is upset, and your dialogue is 'Don't forget to close the door when you leave.' Action. Cut. Okay, that's piece 59 out of the 1000 pieces we have to shoot. What's next?" The film failed because it had been made on an assembly line; no one had an organic concept of the whole they were trying to reinvent. It had been too atomized; and there was no place for serendipity.

J.K.      There are infinite ways to translate novelistic language and scenes into image. You could take a whole conversation in a novel and decide that, filmically, you'd present these two characters in a certain spectral light; or it may be raining and they exchange only three words. In effect, each could translate an entire chapter of a novel.

W.M.      Yes, occasionally we find that a line of dialogue in the script, written to evoke a certain feeling when read, may inspire an expression on the actor's face which makes the line itself superfluous. At that point we can (and should) remove the dialogue. In retrospect, we can see that the function of the dialogue was simply to act as the sculptural mold of an emotion; but once the emotion has been provoked, the words may become redundant and even occasionally damaging to the very emotion they were intended to convey. This redundancy happens all the time in filmmaking, it is part of the process itself, and you have to be constantly alert to it. Films which seem overly long are usually suffering from precisely this problem.

Time, in film, doesn't correspond to time as we perceive it in real life, where someone getting up and crossing the room and going out the door apparently follows a sequence of causally linked events; if the chain were broken, he wouldn't get where he was going. In the theater, time is also (mostly) causal since stage actors are living people tied irrevocably to the nature of real time. But in film, because of the nature and power of editing, people can pop up in the most unlikely places and it seems perfectly natural. In fact, if you were to represent movement in film in real time, everything would seem unendurably long‑winded, probably because time in film is more as we experience it rather than as we perceive it: That man crossing the room was not consciously aware of every step he actually took, so in film we don't show it all, unless we intend to make a point of it. We try instead to represent movement in a more economical, suggestive way. It's our version of poetic compression.

Because of his knowledge of film, the screenwriter is already confronting these issues of time, duration, rhythm, weight, momentum, contrasts between light and dark, etc. But he is grappling with them on the page, in terms dictated largely by the acts of writing and reading. Then the director (the same person, in the case of The English Patient) annotates the written page with more detailed and thematically elaborate ideas for the dramatic, visual, and sonic treatment of the material. Then all of us (actors and the heads of various departments: costume, art direction, camera, editing, sound, etc.) in turn interpret the director's interpretation. The director's job is to see that these different interpretations mesh in an interesting way. It is rather like the relationship among composer, conductor, and performers in an orchestra.

  Anthony was keen, for example, to distinguish between the world of the monastery, where action was comparatively static, and the desert, where the action was dynamic. His idea, which he passed on to the director of photography, the production designer, and makeup and costume, was to make the monastery look like a watercolor, and the desert bolder and more graphic. Which is the antithesis of what conventionally happens in film: The present, as the dominant reality, is sharply prominent, and the images of the past become rippled or sepia or washed out. In The English Patient, memory is bolder than the present.

J.K.      That seems more accurate, because memory and dreams are incredibly sharp and specific; they're not fuzzy at all.

W.M.      It's certainly true of The English Patient, since the person doing the remembering is a dying amnesiac immobilized in bed, and much of the "present" action takes place in his sickroom. By contrast, the past teems with adventure and sexuality: planes flying, bombs exploding, sandstorms, handsome people making love, all kinds of emotion and trauma. So Anthony had to take the clues from Ondaatje's text and transpose them into another language, which infuses chronology and rhythm with images and sound. That process of transposition continued through the shooting and the editing of the film.

One of my favorite observations on this subject comes from the French filmmaker Robert Bresson: "Your film: three lives and two deaths. It is born in your head, it dies on paper; it is brought to life again during shooting, where it is killed on film; and then resurrected in the editing, where it opens up like flowers in water." I think what he meant by "kill" is not so much destroy but rather trap, although as every writer knows, a certain amount of destruction occurs in getting something from your head to the page; the idea must first undergo an imprisonment in words on paper if it is to have a separate existence in the reader's imagination. The text is then interpreted and brought to life by the actors; the camera, in turn, "shoots" it, trapping it onto film. Then, in the editing room, the footage is dismembered, carefully rearranged, and transubstantiated into a third life. This astonishing, sequential death and resurrection must happen for the film to be truly alive; if you think you can get from the original concept to the finished film without destroying something, you're mistaken. But there's hopefully a value gained at each stage that more than compensates for the losses. Finally, I would add a fourth "life" to Bresson's list: The film is resurrected again when shown to audiences. They frequently bring things to the film-emotions, reactions, experience, insights - which the filmmakers would never have expected, and the film itself changes as a result.

Bresson spoke in more pungent terms, death and resurrection, about what I'm calling translation, but his images are particularly appropriate in the case of The English Patient because the novel did not follow a typical novelistic, let alone cinematic, chronology, the process of adaptation was correspondingly wrenching. When you read a book by, say, John Grisham, you can see the film it will probably be; he has already begun the process of translation. I'm sure Ondaatje never thought his novel would be turned into a film. In fact, when you finish reading The English Patient, you probably think, "Well, they'll never make a film out of this." Even though the images are powerfully visual, it doesn't immediately seem like there's a film there: no conventionally likable main character struggling in sequential time against a hostile force and ultimately prevailing. Just the opposite: The main character (Almasy) is spiky and "difficult"; there's no one to hate, really; and the ending resolves around the breakup of the relationship between Hana and Kip, the discovery of Katharine's dead body, and Almasy's death by lethal injection. It took Anthony's particular genius to see the filmic potential in all of that. More to the point, then, was this process of destruction and reinterpretation. So much so that when Anthony went off to Devon to write the screenplay, he took lots of research material but not his copy of the novel. He wrote the screenplay without any direct reference to the book, but because it had so deeply impressed him, whole lines of dialogue and scenes inevitably wound up in the screenplay, although in different combinations than in the original.

J.K.      What, in the process of the death and rebirth of The English Patient, were the links that carried through? What were the things Anthony latched onto, the touchstones in the translation process?

W.M.      In the broadest sense, the interweaving of the personal and the historical. Anthony was interested in how these two forces interact, how what we think of as history seems, in part, to control our lives, and yet if we examine it we think, "Well, what is history but the sum of individual decisions made by people like me, with their own concerns?"

To a degree, the events on which the WWII North African campaign turned are traceable, in this film, to the doomed love one man harbored for another man's wife, even to his retrieving her dead body from a cave in the desert. To do that, which on the face of it seems absurd, he was willing to give the Germans access to information they wouldn't otherwise have gathered. As a result, the whole movement of men and machines across North Africa took a different complexion than it would have had this adulterous affair not flared up. Yet ultimately, all of the above would not have taken place without the forces of history focusing on North Africa, where war was being prepared.

How do you distinguish between the large forces causing individual events to happen and the individual events that in turn cause large things to happen? The drop of water is carried along by the wave, but what is the wave without the many drops of water? That's one of the questions The English Patient poses, as it explores the various forms of love and friendship that bind people together and tear them apart, and the good and disastrous things that emerge as a result. As much as we want everyone to love everyone else, it's impossible without someone getting hurt. History is the writing of that intimate truth on a grand scale.

J.K      All translation is, in a way, profoundly unsatisfactory. One could argue there's no such thing as translation: In poetry, you're composing a new poem. You take what you can, you find the urge, the life, of the poem in Spanish and you make a poem in English from them.

W.M.       Yes. But when you get right down to it, even the original poem is a translation from another mysterious language, what Stephen Spender called the base ground of poetry, "a rhythm, a dance, a fury, a passion that is not yet filled with words." The challenge of writing poetry is this struggle to express the previously ineffable, taking ordinary "grocery-list" words and forcing them, by juxtaposition and compression, to convey things they weren't meant to. Or simply to create new ones: Shakespeare invented over 1500 words in the process of writing his plays and poems.

J.K      During editing, does the character of a film change? Have you ever worked on a film whose screenwriter thought, for instance, "We're going to get at the relationship between personal history and larger history," but by the time the film was in the can it had metamorphosed into something else?

W.M.      The form is always changing, and certain aspects of it wind up being illuminated differently; discoveries are made. But at its heart, if it is to be successful, a film has to be faithful to the original conception, the central truths and interests that gave birth to the project. This was particularly so in the case of The English Patient. The form was changing ceaselessly all through the writing of the screenplay and then through the film editing, but we were always trying to find ways to more clearly and emotionally convey what had driven Michael to write the novel. Which were three things, in my view: the interweaving of the personal and historical; the pursuit, in that interweaving, of the different kinds of love, friendship, and conflicts that arise; and how those conflicts work across different time frames. I don't think you could turn The English Patient into a comedy or straightforward action-adventure story.

But it's also one of the editor's responsibilities, since what leaves his hands is the finished film, to be aware of and exploit any new possibilities that may arise. He must monitor the strengths and weaknesses that develop during the shooting of the film, and make judgment calls about which strengths to enhance and which weaknesses to suppress, remembering that if you completely suppress the weaknesses, that becomes a weakness in itself.

J.K.      What do you mean?

W.M.      When John Ford cast John Wayne in Stagecoach, he was taking a chance on a relatively unknown actor; up to that point, Wayne had starred in many Republic pictures, formulaic westerns known as "oaters." Appearing in a film by John Ford was the equivalent of moving from television to the big screen. This was Wayne's big break and he wanted to impress Ford. As a result, he overacted in his first scene. Ford beckoned him over and gave him a tip: "If you want to be great, be great in three scenes and be adequate in everything else." Wayne took Ford's advice to heart, and they went on to forge one of the most enduring actor-director relationships in film. But Ford's note seems slightly chilling at first. "What does he mean, be 'adequate?' Shouldn't I be as great as possible all the time?"

First of all, you should understand that for Ford, "adequate" meant "appropriate to the situation." Definitely not mediocre. To be "great," on the other hand, meant to be otherworldly, so that the performance is sublime, beyond analysis by conventional means. If you were to follow Ford's recommendation, you would first have to ask, "Which three scenes am I going to choose to be great in?" To answer that, you must really, deeply understand your part, and know when and how to invoke some kind of supernatural acting force. If you succeed in doing that, the proportion between "great" and "adequate" allows your performance to be seen in a larger context. Whereas if you are trying to be great all the time, chances are you won't achieve it. And even if you could, the monotony of greatness would spoil your performance: The audience would lack any comparative criteria by which to judge it. Also, by hogging the screen all the time, you wouldn't allow the other actors their chance to be great, which means that you'd kill the film pursuing this selfish and abstract goal of perfection.

J.K.      So we're talking about the dynamics of intensity. There has to be contrast, otherwise there's no texture in the performance.

W.M.      Yes. And what's "bad" is always up for discussion. Something bad in one film can be good in another. In Apocalypse Now, for instance, Francis [Ford Coppola] broke one of the cardinal rules of filmmaking: Don't let the actors look at the camera. Yet he did it in a consistent way that increases the subjective identification of Willard (Martin Sheen) with the audience: Other actors, talking to Willard, will look right at the camera, but Willard, looking back at them, will look to the camera's left or right, conventionally. When Willard looks at the camera, however, it's read as a conspiratorial glance at the audience. You can see this most clearly in the scene at Nha Trang, where Willard is given the mission to terminate Kurtz's command. It's not noticed as a mistake because the technique is right for this particular film; the opening reel, which features Willard trapped in his hotel room, is intensely subjective. So another responsibility of the filmmaker is to decide what, for a given film, is good and what's bad. There are no fixed rules, though rules of some sort have to be observed consistently. Of course, technical considerations enter here: Is the shot out of focus? Did the actor remember his lines? Though sometimes the dismay and confusion of an actor trying to remember his lines, put in another context, can be wonderful.

J.K.      You say, in your book In the Blink of an Eye, that as editor you are the ombudsman for the audience. A higher percentage of the audience will not have read Ondaatje's novel than, say, a Grisham novel. Still, are you concerned that readers of The English Patient will object to its adaptation?

W.M.      It's an impossible thing to predict. What images did this particular reader bear in mind when he read the book? Chances are the actor chosen to play a certain character won't quite match the virtual actor that the reader had conjured up. Mario Puzo's description of Michael Corleone in The Godfather would never make the average reader in 1971 think of Al Pacino, particularly at that stage in his career. And yet now it's hard to think of Michael Corleone without thinking of Al Pacino; they just seem made for each other. In fact, Pacino's evocation of Michael has overwhelmed the character in the novel.

We tend to think of good casting as finding an actor who has a strong physical and emotional relationship with the part he is being asked to play. This, what I would call vertical casting, is true as far as it goes. But a film can have good vertical casting and still fail; everyone may be individually right for their part, but there's no interactive chemistry. The missing ingredient, then, is something you might call horizontal casting: the need to find an ensemble of actors who, among themselves, have real-life interrelationships that in some way mirror the fictional relationships of the characters they're asked to play. It's the weaving of the horizontal and vertical, as in a fabric, that gives flexibility and strength -which is to say balance-to the casting of a film. Marlon Brando was the acting Godfather to the four youngsters who played the Corleone Brothers. When each of them - Pacino, Duvall, Caan, Cazale - made the decision to be an actor, I'm sure Brando loomed large as an inspiration. Now their dream had come true: They were acting with him, competing with each other, trying to impress "dad" in different ways. So when the camera stopped, they all were still in character. No matter what happened in a scene, it couldn't go wrong.

The goal of good casting is this kind of internally self-correcting balance: not an abstractly perfect Michael, who exactly fits the description in the book, but one who, despite imperfections, takes part in a larger whole and thus helps to achieve in film the same kind of balance that Puzo achieved in his book.

Similar considerations apply in the decisions about what material to include in an adaptation. There are scenes in The English Patient not in the screenplay, and vice versa. But what you're striving for is the novel's proportion and balance, achieved by means appropriate to the new medium of film. Sticking too respectfully with the details of the original work is generally fatal. In the novel, for instance, the story ends with the reaction of the characters in the monastery to the news of the bombing of Hiroshima. This doesn't happen in the film. The material was actually in the screenplay, and it was shot and assembled, but the film just didn't seem to want to accommodate it; Hiroshima intruded as an extraneous political event - the outside world crashing in and destroying the microcosm of the monastery - which was exactly the point in the novel. But for some reason, probably related to that question of weight, this didn't seem to work in the film, so we decided to take it out. That, however, jeopardized the ending of the film, since in the screenplay Hiroshima triggered the final collapse of the relationship between Quip and Hanna. We had to find an alternative structure, using scenes that had been shot for other purposes, to resolve the ending in a satisfying way, not only for the film in its own terms, but for people who had read and liked the novel. To advocates of the novel, we may or may not have succeeded.

J.K.      What is the legacy, in film editing, of Eisenstein, Pudovkin, and the clash of images in montage cinema? Taking two disparate images, showing them in juxtaposition, and expecting a certain additive reaction ...

W.M.       It's a huge legacy that has evolved over time, from a rudimentary grammar based on Hegelian dialectic (this was the Soviet Union, remember!) to something that is more developed, in a musical sense, and less theoretical. Perhaps we don't do it as boldly, crudely, or self-consciously as it originally was done, but at its core, that's what editing is all about: finding a method to splice the text and subtext, and produce a value added from the collision of the two. This quest extends throughout the filmmaking process.

When adapting a film, you orchestrate the images and ideas of a novel. In music, the orchestrator cannot permit all the instruments to play the same note together for any length of time. He can get away with it on two or three occasions, but no more than that. The rest of the time, he states a theme and then its opposite, which harmonizes with it, all expressed by different instruments, and gets something "additive" as a result, bigger than any of the elements that went into it. Then yet another group of instruments develops its theme, which eventually supplants and merges with the first. And so on.

The interaction of these themes in a film would be, to take a simple example, an actor saying, "I love you." Those words by themselves convey ardent feeling. But the actor can deliver them so they hint, "I don't love you." That's an interesting ambiguity, a tension between the words and the attitude of the speech. But then the director can stage the scene to insinuate, "Yes, but he does love her, really." And the editor can assemble the scene to hint, "But he really doesn't." All of a sudden, you've got four levels, and others can be added, through production design and costumes, to get a layering of reversals around "I love you / I don't love you / I do love you / I really don't / But I do, I do." How those ideas cohabit will decide whether it's a compelling film. Again it resembles orchestration in how you state a theme, develop it, contradict it, undercut it, support it, resolve it. If all of the instruments always played the same note, the orchestra would sound like a child's xylophone. You can barely listen to that stuff for fifteen seconds. To sustain two and a half hours of concert or cinema, you need more development and elaboration.

J.K.      What's the visual equivalent of the orchestra playing the same note at the same time? Is it when the actor says, "I love you," and the scene is shot, acted, lit, and edited that way, so there are no emotional harmonics?

W.M.      Exactly. That's bad filmmaking. You find examples of it all over.

J.K.      I want to ask you about artifice, both in image and sound. In Apocalypse Now, the opening sequence, with its four layers of images, gets at Willard's emotional state. Flames overlay his hair, which works as a metaphor for his raging restlessness; he would consume himself if he didn't get back to the jungle. Can you talk about the editing of this sequence? You say, in In the Blink of an Eye, that editing "should look almost self-evidently simple and effortless," yet that sequence is anything but.

W.M.      Yes, although it stands somewhat apart from the film as a whole. It functions as an overture; the story itself doesn't really begin until the two soldiers come up the stairs to get Willard and take him to Nha Trang. So you're looking at a character whose head is enveloped in flames, and then at slow-motion helicopter blades slicing through his body, superimposed upon a whirling ceiling fan, and strange sounds and music intermingling from different sources; you're probably aware you're watching a film, not an imitation of real life. Even dreams, despite their odd surreality, don't look quite like that. Inevitably, the superimposed images in Apocalypse Now betray a self-consciousness because they come at the very beginning and are intended to expose and explore Willard's inner state of mind. If there had been no resonance between that scene and the film as a whole, the opening would have been a meaningless exercise, empty virtuosity.

J.K.      When did you decide to get at his inner state that way? Were those sequences planned from the beginning?

W.M.      They weren't in the original screenplay. They derived from an early sequence shot for the film, in which Kilgore (Robert Duvall) orders a napalm strike because he wants to suppress the mortar fire preventing his soldiers from surfing. That napalm explosion was photographed by six or eight cameras. One camera, far removed from the action, had a long telephoto lens and was running at high speed. When Francis watched the dailies of the material, something about that camera angle - the calm jungle, flattened by the long lens, suddenly, in slow motion, erupting in flames, and those hallucinogenic pterodactyls of helicopters passing in front at odd angles - captivated him and made him think of it as an opening image for the film.

He'd also shot, as an exercise in getting at the character, a scene with Marty Sheen (Willard) trapped in his hotel room, pacing, fuming, drinking, smashing the mirror in which he had been studying himself. It had been shot with two cameras even though it was not intended for the film, and yet turned out to be so powerful that Francis wondered if there was a way to bring that unintended shot of the jungle bursting into flames together with the unintended scene of Willard drunk in his hotel room. Those two images don't quite go together; you couldn't simply cut from one to the other. So some "connective tissue" was shot: the upside-down shots of Willard, the ceiling fan in the room, etc. This whole overture emerged organically out of the filming. It then became my responsibility to execute it, to take those diverse elements and devise something that serviced the film in an interesting way.

J.K.      It was your eye which merged the fire with his head.

W.M.      Yes. There were no storyboards, at least none that I ever saw. It occurred to me that if a certain section of the burning jungle shot was superimposed with a certain section of the angle on Willard, it would appear that his head was on fire. I also took advantage of the visual and sonic parallels between the helicopter blades and the blades of the fan. And I browsed around in the footage from the end of the film to see if there was something I could steal to give hints about the end at the beginning - the massive stone head of the Cambodian Buddha, for instance, which occupies the right third of the screen, staring implacably at Willard's upside-down head on the left-hand side.

J.K.      Do you think Hollywood has changed much since the time Apocalypse Now was made?

W.M.      Hollywood swings back and forth from exploration to exploitation. We are in, and perhaps emerging from, a period where exploitation is dominant, or at least preponderant. The balance was to the other side in the early 70s, but only because Hollywood at that time was in the final agony of the old studio system as it had been organized in the 30s, 40s, and 50s. For various reasons - the impact of television, creative exhaustion - it was a period where the compass was spinning, true north had been lost. Where was the industry going to go next? Would it survive at all? A lot of the maverick films from the early 70s are simply the result of people getting the chance and trying, instinctively, for the most part, to find new compass points in the wilderness: Sam Peckinpah's films, road films like Easy Rider, the films of Coppola, DePalma, Scorsese, Lucas. Then by the mid-70s, with the successes of The Godfather, Jaws, and Star Wars, a pattern began to emerge. Money was pouring in, the concept of the "summer film" took hold, and these maverick films provided a model to be copied. But coloring in the squares of pre-existing pattern can only take you so far. It seems to work for a while, and then it suddenly stops; you've wrung as much interest out of it as you can get.

J.K.      We're swinging out of this era?

W.M.      That's my intuition. We're not out of it yet, and some people are beginning to despair (see David Thomson's essay in the December 1996 issue of Esquire), but too much is changing right now, technically and economically. And my generation, which was young in the 60s, is now in the ambiguous position of being the establishment they fought against, having more power than they ever thought possible, but poised to be overthrown in their turn.

There's an inherent tension in film between its economic function as a mass medium and its ability to "get at" things that other art forms cannot. This is nothing new. I'm sure artists in the past, in music, painting, and other media, enjoyed the same kind of tension. But the economics of film are such that it magnifies the problem. On the one hand, film has to pay its way, and it's so expensive that it has to pay handsomely. To do so means that, generally speaking, you must appeal to a broad audience in an easily marketable way. The issues you're dealing with have to be common currency; they have to be known from the outset, with the film only giving them a topical spin. This is the exploitative side of the equation.

On the other hand, contradicting all of the above, what keeps film alive is its unpredictability: If its function became exclusively intestinal - merely to find the best way to efficiently metabolize pre-existing truths into the social bloodstream - it would eventually wither away. This is what happened to social realist art in the Soviet Union. Everyone liked it, or said they liked it; it was comprehensible, predictable; it trumpeted generally acknowledged "truths" to the world at large; but it was dead at the root, sustained only by the largesse of the commissars. This is also why, with a few exceptions, film sequels are disappointing: The territory is too well-known; the food has already been partly digested before we eat it.

So this second aspect of film, its explorative side, is ultimately just as necessary as the exploitative. But it's terrain that is, by definition, less well-known. Milan Kundera calls it going behind: "The mission of poetry is to discover an essential truth but one not known in advance." Two decades earlier, Robert Bresson wrote a note to himself that touches on the same thing: "The filmmaker is making a voyage of discovery on an unknown planet. Be as ignorant of what you are going to catch as is a fisherman of what is at the end of his line." And ten years ago, Coppola said the same thing using a different analogy: "The director is the ringmaster of a circus that is inventing itself."

This ignorance of what is wriggling at the end of the line is risky, of course. It makes the investors rightly nervous, because sometimes an old boot gets pulled up. But when it's successful, it can hook and reel in a truth smarter than the fishermen who caught it. And this ennobles the people involved, the medium that made it possible, and the audience that is witness to it, since all know they're in the presence of something greater than the sum of its parts. "Great novels are always a little more intelligent than their authors," Kundera wrote in Art of the Novel. "Novelists who are more intelligent than their books should go into another line of work." A purely exploitative cinema would, similarly, make filmmakers smarter than their films, smarter than their audiences, smarter even than the medium itself, and this would eventually produce a corrosion that spreads from the author to his audience via the medium itself, and eats away at the connections between all three until they fall apart.

J.K.       Is the self-conscious use of sound, as in The Conversation, a technique to be used cautiously? In the beginning of the movie, we watch Cindy Williams and Fred Forrest talking and we hear these ambient, otherworldly sounds; it disarms us at the beginning because it's so self-conscious.

W.M.       It's a boldly stated version of the goal of any film: to present compelling contradictions and resolve them inventively. In dramatic terms, you want to pick a character and a thematic situation that seem completely at odds with each other, yet coerce them to coexist.

J.K.       Like two halves of a metaphor. They have to be at that point of tension ...

W.M.      They have to be as far apart as possible, and yet not so far apart that there's no way to fuse them. If they're too close to begin with, they produce a flat, uninteresting metaphor; if they're so far apart that you can never resolve them, the whole thing doesn't make any sense.

In that scene from The Conversation, the soundtrack has, up to a certain point, evoked a kind of urban idyll: The traffic is lullingly monotonous, you hear distant music, people taking their lunchtime break. Then suddenly into the middle of all that come these invasive, distorted, crazy sounds, which pose a question: What's that sound doing with these images? And then, as if to underline this, you start to notice mysterious men with rifle-like devices on rooftops, funny things sticking out of windows looking down at the people in the square. Ultimately, the answer to the question posed by the sound turns out to be what the film is about.

J.K.       Francis says in his journals that in the writing phase of a film, you're operating in two universes. In the first you're getting back the sensation that something is "there" behind the concept you're working on; in the second, you translate that into perceivable language, actable work. What is "actable work?"

W.M.       On a primary level, we make motion pictures, so there has to be motion of some kind, kinetic or dramatic. Secondarily, we are making e-motion pictures. You need motion or emotion, and preferably both. A film composed of static images can work if it somehow arouses emotion in the audience. Carl Dreyer's Joan of Arc, for example, is relatively static. It's hard for audiences to get into it nowadays - our sensibility has changed - but if you do fall under its spell, it can be very emotional, even though it is not "motional."

On the other hand, many people go to movies simply to see people and things move through space in surprising and exciting ways; they bring all the emotion they need with them, like a box lunch. This approach goes back to the films made by the Lumière Brothers before the turn of the century, when simply seeing a train arrive at a station was thrilling. An action-adventure film has to find ever more exciting ways to get characters moving through space, with objects exploding or collapsing on them, and just enough story to keep everything hanging together. Rather like the relationship between songs and plot in early twentieth century Broadway musicals - there was just enough plot to keep people on stage, but the audience really came for the songs. Sometimes that's enough, though a steady diet of cotton candy isn't nourishing.

Ideally, you want motion and emotion, but neither comes about unless there's conflict. A filmable character must have a conflict, as deep within himself as possible. If a character in a film does one thing and says something else that completely contradicts it in an interesting way, then the audience is hooked. Or, in an action-type film, the characters may be flat, but they get trapped in an unlikely situation. How will they get out? In both cases, the resolution must involve some kind of motion. It may simply be kinetic motion; but it can also be dramatic motion: the transformation of character. "Actable work" involves, I believe, the statement and resolution of forces or principles that are initially as far apart as possible.

J.K.      The way Metaphysical poetry yokes opposites together.

W.M.      Yes. You want an audience to be just aware, not too aware but just aware, of how far apart those things are, to wonder, "How can this be? How can this resolve itself?" In the end it does, somehow. Michael Corleone doesn't want to be part of the family, but we know the Mafia doesn't tolerate defectors. You're either with us or against us. That a favorite son doesn't want to be a part of the family automatically creates a conflict. Events conspired to make him, the most unlikely son, the Don. He had to take a kind of essential femininity in his personality and crush it, to suppress that side of himself for what he saw as the good of the family. And that conflict fueled the events of three films; that contradiction became "actable work."

J.K.       Why do you cut without sound?

W.M.      I do so only in the early stages of assembling a scene. Eventually I add the dialogue, of course, but initially I pretend that I'm working on a silent film. It allows me to focus more clearly on the emotions playing on people's faces, on their movements and body language, and to try to interpret the scene using just those elements. Once I have the basic structure of the scene, and it interests me without words, I restore the dialogue and then make further changes based on the new chemistry of image and sound. But initially I put myself in the state of mind of aphasics who can't speak or understand their native (or any other) language. As a result, they become adept at interpreting their interlocutor's body language and facial expressions. If you go to a foreign country, you're effectively aphasic. If you stay there long enough, you may become adept at knowing instinctually not only if someone is mad at you but perhaps why. I exile myself to the foreign country of the film, to try to invoke the images and emotions I'm after purely through posture, expression, and rhythm of movement. I do it for somewhat the same reason that Anthony went to Devon without a copy of The English Patient. I know that if the dialogue is present from the beginning, it will dominate the scene and inhibit the visual reinterpretation of the material.

J.K.      Can you talk about the difference in thought process between shooting and editing?

W.M.      For the director, what characterizes the shooting, emotionally speaking, is the need to live in three time frames simultaneously. This is certainly unnatural and frequently painful. He must be absolutely attentive to the present, or the performances will start to wander. He must dwell in the past, bringing to bear an intense awareness of everything he's done up to this point. And he must anticipate the future, which is to say he must realize that present decisions will ramify unpredictably unless he is very careful. Emotionally, editing is much more like writing: I have time to consider things and balance them. If I change something in one scene, I can go back and make corresponding adjustments to an earlier one. I have the leisure to walk away from the material and think about it, sleep on it, deferring decision to the morning. Directors rarely have this luxury.

However, a director can defer certain problems to the future. For the editor, there is no future. If I can't solve a problem, the film is going to look bad, and so am I. It doesn't matter where the problem came from, I have to solve it, with the material and vocabulary that have been given to me. Sometimes, when there's no solution to the problem, the answer is to reframe the context in which the problem exists, to change the situation that gave rise to the problem in the first place.

J.K.       How do you do that?

W.M.      The Conversation is a film about a private detective uncharacteristically agonized over the potential death or harm that some of his work may have caused and might cause in the future. Up to that point, in previous films, characters in Gene Hackman's role were conventionally portrayed as being hard-boiled; Hackman himself had just played Popeye Doyle in The French Connection, and Francis had just directed The Godfather, in which many people were killed without a second thought.

So early on in editing The Conversation, we found that test audiences were bringing the wrong expectations to the film, and it was making them impatient: The last thing they expected was Gene Hackman to anguish over somebody who might have died because of some miscalculation he made. We couldn't "solve" the problem and make Harry Caul into something he wasn't, so we had to find a way to defuse people's preconceived notions, to get them to pay attention to small things, so that the death of a person loomed larger by contrast.

As a result, in the opening reels of the film, rather than acquiesce to the audience's initial impatience by accelerating the action, we instead did just the opposite: We dwelt an unusually long time on small things that would ordinarily have been cast aside in the interests of getting on with the story, in order to tune people's expectations back to normal, where life is precious.

J.K.      So you have the detailed, intimate sound, which played a huge role in the film, and the costuming - his disposable raincoat, which was distracting, peculiar, and creepy yet pathetic. Are these some of the ways to change the context?

W.M.      Yes. And just spending time with the details of this man's life: a kind of character study of a lonely, ordinary man who has this extraordinary occupation. When Francis was writing The Conversation, his goal was to fuse the forms of character study and suspense story: Hesse and Hitchcock, Steppenwolf and The Wrong Man. For those two to share the same film - talk about bringing contradictions together!

In Hesse's Steppenwoff, the central character's name is Harry Haller, so in The Conversation the Gene Hackman character's name was initially Harry Caller, an appropriate name for someone who spends his life bugging telephones; in a later draft it was shortened to Harry Call. But Francis's secretary made a mistake typing the screenplay and on one page spelled his name Caul. When Francis read that he knew it was even better. A caul is the shroud like membrane which sometimes surrounds the fetus at birth. Caesar was born with one, and it was supposed to portend greatness. On the one hand, the silent "u" in the name softened it; it still sounded like "call," but wasn't spelled that way.

On the other hand, the visual idea of the caul became the determinant for a whole stylistic treatment in the film, which was to wrap Harry in semi-transparent plastic as often as possible - the raincoat that he always wears, for example - so as to insulate him from the rest of humanity by this membrane. Even when he isn't wearing the raincoat, whenever he gets threatened, when something bad is going to happen, he moves behind some kind of transparent plastic, or glass, to throw up a barrier between himself and the threat, but a barrier through which he can see and still be "safe." So the translucent wrappings play a large part in the movie, all because of a slip of the finger in the retyping of the screenplay.

J.K.      What's your taste in poetry? Do you read much contemporary poetry? Do you ever find poetic techniques analogous to what you do in the editing room?

W.M.       I have to confess to being what you might call "poetically challenged." I love poetry in the abstract, but don't care for poems in particular, somewhat like those unfortunate people who love humanity but have a hard time with human beings. I wish this weren't the case, and am somewhat mystified by my tone-deafness. Thankfully, there are a few exceptions: Rilke and Edgar Lee Masters are poets I have found myself reading regularly, sometimes on a daily basis for months on end. But much poetry, particularly contemporary American poetry, I find too willfully obscure and/or self-absorbed. I do, however, love to read the work of several authors who write prose with a poetic sensibility, like Curzio Malaparte, Antoine de Saint-Exupery, Milan Kundera - and of course Michael Ondaatje. I'll read these favorite books over and over again, like listening to a piece of music, until the binding falls apart.

I think you could make a case for an analogy between the length of the individual line in a poem and the duration of the individual shot in a film. Formally, it is the "broken" structure of poems, particularly those written in the last hundred years, which first strikes the average reader; poems are now less often written in regular meter, nor do they "wrap around" and fill the space available, like prose, but instead distribute their words on the page in a seemingly arbitrary but secretly architectural way.

Why does a poet choose to end one line where he does, on a particular word, even though it may make no grammatical sense to do so? Perhaps because of what you might call a rhythmic and contextual "ripeness" in the line, and because he wants to draw subtle attention to the last word in the line and the first word in the next. Similarly, I will keep a shot on screen until it feels rhythmically and contextually "ripe." If I held it any longer, or shorter, the balance of the movements of forms and ideas within the shot would be spoiled, because I want to draw subtle (and sometimes not so subtle) attention to the final image of the outgoing shot and compare it with the first image of the next.

Because of the nature of text, however, the poet extracts meaning not only out of the juxtaposition of the last word in one line and the first word in the next, which are temporally close, but also between the last word of one line and the last word of the next, which are architecturally close on the page. We don't have that ability in film. All the impact and beauty of a cut from one shot to the next lies in the direct "retinal" comparison of the outgoing image with the first frame of the incoming image, and here there are certainly visual equivalents to poetic (and musical) techniques of comparison and contrast: closeup to wide, light to dark, rough to smooth, slow to fast, etc. These are the visual equivalents of rhyme, allusion, alliteration, metaphor, and probably many other poetic techniques if you cared to make a detailed study of them.

J.K.      Do you like variations of tempo and modulations in longer poems, and are they models for editing films? The Waste Land, for example.

W.M.      It's impossible to think of film without variations in tempo, just as it is for music and poetry. Music and film, however, are generally more repetitive than poetry. This probably has to do with the fact that they are performances whose progress, once started, cannot be stopped or reversed. In film, we use repetition (of similarly composed shots, of dialogue, of sound effects, etc.) not only for their rhythmic effect, but to alleviate ambiguity, whereas the eye, reading poetry, is constantly and automatically flicking backwards, often unconsciously, to re-examine the meaning of certain words. This re-examination, whose pattern is unique to every reader, is the hidden repetitive structure of poetry. To the extent that poetry can be read aloud, of course, it is like music and film. And I think you will find more repetition in ancient poetry precisely because it was meant to be recited to an audience rather than read silently as text by a single reader.

J.K       Is editing a book different from editing a film?

W.M.      Well, I've never edited a book, but I suspect that in many ways the editor of a book and the editor of a film share similar responsibilities: to offer counsel to the author (or director) about the largest structural issues and the smallest details of expression. Questions range from "Should this scene be in the film (or book)?" and "Shouldn't these two chapters be reversed?" to "Shouldn't you use another word instead of 'very?' " But with a book, at the end of the day it is the author who goes off and does the actual writing. In film, however, it is the editor who selects and assembles the images in a certain order with a distinct phrasing, and the director who suggests revisions based on what has been presented to him. The give and take between who is "editing" and who is "performing" is constantly shifting in film, the numerator becoming the denominator and vice versa.

J.K.      Do your editing philosophies and methods change depending on the cinematic vision of the director you're working with?

W.M.      My deepest responsibility to the work remains the same: to tell the story in the most appropriate, engaging, economical way. Certain technical things also remain the same: how I sharpen my pencils, so to speak. But each film is a universe unto itself, and everyone who works on it, director included, is trying to discover the laws by which this universe works: what kind of character, action, color, tempo, sense of reality, sense of humor suits the film. They're particular to each film - what works on one film won't necessarily work on another. And as they evolve, you absorb these laws so completely that they become unconscious reflexes, almost like being infected with a benign virus. The difficulty then is the amount of time off you find between films, to "cure" yourself of that virus. If it's too short, less than six weeks in my case, you inadvertently infect the next film with the rules of the previous one, which may be inappropriate. And this is good neither for you nor the films you're working on.

J.K.       The English Patient is an exquisitely beautiful film. Did you consciously edit the composition and images for the sake of the gorgeous effects (not that they're divorced from the plot)?

W.M.      That would be similar to a composer who, knowing his work was going to be performed by a skilled instrumentalist, writes so as to take maximum advantage of the particular performer's strengths. (Mozart's relationship with the clarinetist Anton Stadler comes to mind.) He shouldn't distort the structure of the composition simply to show off virtuoso technique; rather, the virtuosity should arise stealthily and organically out of the melodic and harmonic needs of the composition. Similarly, when I know there are beautiful shots waiting for me in the dailies, I want to use them to their best advantage, but quickly become ruthless if the shots turn out to be superfluous to the story. Their beauty then almost turns into a liability, like handsome but empty-headed people. I much prefer a necessary shot to a beautiful shot. If it is necessary and beautiful, so much the better.

J.K.      What was the intent with the ending? It seemed almost Wagnerian in its grandiose sweep and highly wrought climax.

W.M.      It depends on what you mean by "ending." The last reel of the film - the death of the Patient - always struck me as "Bachian" rather than Wagnerian: powerful but formally restrained and somewhat contemplative, like a hard lump in the throat which won't go away. The Wagnerian part, for me, came earlier, when Almasy carries the wounded Katharine to the cave and she tells him that she always loved him. There is a conjunction of dialogue, emotion, music, and scenery at that moment which does seem as close to Wagner as you could safely get.

J.K.      Did Minghella decide consciously to keep the film close to the rules of classic cinema, and not parallel the novel's experimental form? Was this a concession to popular taste?

W.M.       Film is a popular medium, and the audience is never far from our thoughts, the way the ocean is never far from the thoughts of a ship-builder. But that shouldn't prevent, in fact it should even encourage, beautiful ships being built.

November 8. 1996 Fantasy Studios Berkeley, California

© Joy Katz & PARNASSUS Poetry in Review 1997

This article is reprinted and reproduced on with permission of the copyright holder.

To Walter Murch Articles

         Your Learning Space for Film Sound



Star Wars Sounds Film Sound Clichés Film Sound History Movie Sound Articles Bibliography
Questions & Answers Game Audio Animation Sound Glossaries Randy Thom Articles
Walter Murch Articles Foley Artistry Sci-Fi Film Sound Film Music Home Theatre Sound
Theoretical Texts Sound Effects Libraries Miscellaneous