The Conversations: Walter Murch and the Art of Editing Film

by Michael Ondaatje

  This is the most significant thing that I think I do. If I had to abstract one element from the way I work, I'd say that no matter how you work as an editor, this is a good thing to do. You can have completely different approaches to everything else, but do this.

  The wonderful advantage—and it is miraculous to me—is that by doing it you quickly develop instructive feedback. Not only about this particular shot, or film, but about your own talents as an editor. When choosing a cut point, an editor is like somebody playing a violin…. There is a bowing technique that you have, that you want to develop and have mastery over, whatever kind of music you're playing. You want to have a touch or a tone.

  When I mark frame 17 and the next time frame 19, I have a feeling that goes with each. When I mark frame 19, I feel, Oh, it was a little longer that time—I can feel it. Then, looking at the counter, I realize, That was two frames. In this context, that's what two frames feels like: one-twelfth of a second. But I now have an emotional feeling in my gut about what a twelfth of a second feels like, with these shots in this context, and that's teaching me something.

  Throughout all this, you're working with the rhythms of the actors, and the rhythms of the camera moves. You are internalizing everything—the rate of the speech of the actors, how they deliver their lines, how they are physically moving in the space, how the camera is or is not moving in the space. You are taking all this into consideration, and that is what, over a period of time, allows you to begin to assimilate and learn the particular language of this film. What's the rhythmic signature of this scene? And then, of the whole film?

  Every time conductors confront a piece of music with a new orchestra, they have to determine the rhythmic signature. An editor is doing that with the film.

  O: If you've cut the film at this rhythm, and then you go into the mix, doesn't the work you do in the mix alter the pacing of the scene—if you add, say, a musical touch or the sound of rain, during the mix?

  M: Yes. And that's one of the peculiarities of the way I work. When I assemble a scene for the first time, I turn off the sound. Even if it's a dialogue scene. I look at the people's faces and imagine what they're saying and read their body language. Significantly, this envelope of silence allows me to imagine the mix the way it will finally be. I'm allowing the space for these sounds in advance. Even though I'm not sure exactly what they will be.

  I find this method essential, because the only sound that's recorded at the time of filming is dialogue and it's sometimes quite rough. You can become mesmerized by the particularities of that sound, which is not the way it's going to be when it's all cleaned up and it has music and sound effects running along with it. It's important for me, when I first assemble a scene, to imagine the music and the sound and the dialogue working together in some ideal dynamic form.

  O: So this editing by hitting the seventeenth frame is done in silence?

  M: I look at it solely as a piece of silent film, imagining the music and the sound as much as I can. I construct the whole scene silent, run it back silent, and make revisions in silence. Does it work? I turn on the soundtrack and confront the reality of what is now added by the dialogue. Sometimes it's exactly the way I imagined it. Other times, fortuitous things have happened that are much more interesting than what I could have ever achieved intentionally had I been listening to the sound.

  Of course, there can be mistakes. I might select a take that is good visually, but without knowing it I was imagining the reading from another take, which is smoother. So I make a correction: I'll use the good sound from that other take and superimpose it over the good visual, so the actor is saying one thing, visually, but the sound is coming from another take. Because that's what I heard in my head when I was putting the scene together. Jean Renoir would have had me burned at the stake! But you try to seize these opportunities, which sometimes come by chance and sometimes from an impression you've gleaned that's deeper than the reality actually in front of you.

  This method allows you to superimpose the rhythmic signature of the film on shots that have no internal dynamic at all, which are simply held for length.

  In The Talented Mr. Ripley, for example, Ripley is sitting on the beach and looks out to sea. There's a shot of the sea. How long do you hold it? You hold it for as long as the thoughts you imagine Ripley is thinking can be held while you are looking at that shot. As I run the sea shot, we're looking at his point of view and we're thinking what Ripley is thinking. When those thoughts have danced in my head, with the image, to this point of fruition, I mark the frame. And then go back and do it all over and hope to hit the same frame. It's still amazing to me that this happens, even though I've been doing it for thirty years. But because the thoughts have their own internal dynamic, which is miraculously invariable, the shot can last for fifteen seconds or more—360 frames—and you can still repeatedly hit exactly the same frame, the 361st.

  O: And this mental decision has been set up by your prior knowledge of the way Ripley has been thinking before?

  M: Yes. I have internalized the rhythms that the actor has given me, and internalized the rhythms that the camera operator has given me. The camera operator has been internalizing the rhythms that the actor is giving him and the rhythms that the director is giving the camera operator. So all of that is coming into play, and I am now metabolizing all those things and getting to the point where I can superimpose those rhythms even on shots that have no internal dynamics at all—no dialogue, no camera movement, no actors moving.

  O: Like the learned habit of a guy who shoots clay pigeons—knowing how long the target needs to be in the air before he raises his gun and points …

  M: It's very similar to gunslinging. That's the reason I stand when I edit—I'm fully engaged in my body. It's possible but hard to imagine somebody shooting clay pigeons while sitting down. There's something about the engagement of the entire body in the rhythmic motion that allows you to hit the clay pigeon. Similarly, when gunslingers faced one another on D Street in Kansas City, they stood, they didn't sit on chairs and shoot at each other.

  From The Talented Mr. Ripley: How long do you hold the shot of Ripley looking out to sea— “as long as the thoughts you imagine he is thinking can be held.”

  “Waiting for Provocation”by Anthony Minghella

  Walter has become inextricably bound up with my ideas about film, with my plans as a filmmaker. His rigour in the cutting room; the standards he has set for himself and expects of his collaborators; the tacit understanding that he is a fellow filmmaker, a peer, not a servant of the director; his profound grasp of every aspect of the filmmaking process—these make him a partner, and an exhilarating one. He has been a writer, he has been a director, he has done pioneering work on the way films sound, and the way they look. Among the many things Walter has taught me is the necessity for every element in the film to work, and to be working in concert. His technique of lining the cutting room with stills extracted from each of the movie's setups is a constant reminder that each cut affects the entire movie, that each sequence lives inside a gestalt, that internal rhymes in a scene have to relate to the larger rhymes.

  He has an extraordinary grasp of how music works in a movie, and unusually for Walter, it is not a theoretical strategy. He seems to throw music at the film, carving up cues, subverting their intended placement: a savant with the score. Watching him at work with Gabriel Yared's painstakingly choreographed sketches is, for someone who prides himself on being a musician and possessing a musician's ear, entirely destabilizing. I remember having devised with Gabriel a series of rules to organize the composition of the English Patient score, with a particular orchestration delineating scenes at the monastery and in the desert. Walter listened to these cues with a certain detachment, while I explained their intended destinations. He then stood, as he always does, at his editing lectern and laid in cues, apparently randomly, using the Avid to stretch and contract lengths, often not listening to the entire piece, and certainly paying no attention
to the map I had outlined. The results were often startling, always provocative.

  This was one of several occasions where one or both of us have left the cutting room with emotions running at danger point. And yet the finished version of the movie reflects as much of Walter's sense of how the score should sound as it does mine or Gabriel's.

  And I've learned to relax into this, to wait for these provocations, just as I expect them with the picture. I have no interest, watching dailies, to indicate to Walter how I've imagined a scene will cut, even though I have always planned a cutting sequence. I've come to understand that the joy of film is in letting everybody play and empowering the play. Walter knows who the director is; the director knows he's got an enormous resource and should exploit it.

  Ripley has an opening title treatment where “talented” evolves as the final adjective in a rosary of descriptions that have preceded Ripley's name: mysterious, unhappy, fragile, and so so. An attempt to identify the contradictions surrounding this complex character. So: The brilliant, baffling, brittle, loyal, tender, curmudgeonly, impenetrable, wise, wonderful, cheerful, stern, obsessive, loving, abrupt, professorial, encyclopedic, patient, impatient, and essential Mr. Murch.

  O: What about the other crucial consideration in editing—where to begin a shot?

  M: I'd discovered very early, during The Conversation, that I was allergic to a certain style of editing, which is called “matching action.” That's the standard kind of Hollywood editing, where you cut from one shot to another in the middle of a character's movement. It was considered significant when it was invented, because it seemed to mask the moment of the cut. Somehow the audience was supposed to think this was all continuous action.

  When I started out, I obediently tried to work that way, and discovered that I didn't like it. I wound up instead finding places where the movement of the incoming shot was just beginning, like a flower opening up. So my cuts generally are made not at the midpoint but at the beginning of the gesture.

  I prefer to initiate the motion in the incoming shot. I'll take a shot right to the point where a character is about to move his head, then cut and initiate that motion in the next shot. There are times when I do not do this, mainly in fight scenes, action scenes.

  But on the whole, I'm taking into consideration, at the point of the cut, where the audience's eye is and in what direction it's moving, and with what speed. The editor has to imagine the audience's point of attention when the film is projected, and has to be able to predict where ninety-nine percent of the audience is looking at any moment. You've said you always look at the person typing—the court stenographer in courtroom scenes. You may be the one percent I can't account for… ! But by and large I have to be able to say with some certainty that at such-and-such a moment ninety-nine percent of the audience will be looking at this point on the screen, and in the next second they will be looking here. That means that their eye is travelling, say, left to right, to the upper corner of the frame, at a certain speed. If I choose to cut at this point— at frame 17—I know that at that moment their eye is right here, in the Cartesian grid of the screen.

  That's a very valuable piece of information. When I select the next shot, I choose a frame that has an interesting visual at exactly that point, where the audience's eye is at the moment of the cut, to catch and redirect their attention somewhere else. Every shot has its own dynamic. One of the editor's obligations is to carry, like a sacred vessel, the focus of attention of the audience and move it in interesting ways around the surface of the screen.

  However, in the middle of a fight scene, you want to abuse an audience's expectations. You want to send their eye off in one direction, then cut with something that is going completely in the opposite direction, somewhere else. That induces in the audience the sense of visual disorientation you get when you're really physically in a fight with somebody. You don't know where the next punch is coming from. We try to duplicate that feeling visually.

  O: And also probably in a love scene.

  M: Yes. There is no stage line to love. In a passionate love scene there's actually an advantage to be gained by crossing the stage line as many times as possible. If you look at the dance scene in Ghost, after Sam and Molly have been playing with clay and start to dance to the music on the jukebox … that's full of cuts that cross the stage line. Each cut, once the dancing gets passionate, puts the characters on the “wrong” side of the frame.

  Visually, I'm taking care of the eye, it's rhythmically and sensuously done but— Wait a minute! Isn't she supposed to be on the left and he's supposed to be on the right? No, no, it's the opposite! What that does is put you in the state of mind of making passionate love to somebody—disorientation, spacelessness. You are physical beings, but you have gotten to a spaceless place. By fracturing the grammar of film, in that way, you induce in the audience a little of that same mentality.

  If I cut it the way it should be cut—according to classical film grammar—it feels kind of flatfooted: these people are doing passionate things and we're just standing outside them, watching them. So we're not as fully involved. But by breaking the rules, you can bring the audience into the madness that is passionate love.

  LAST CONVERSATION

  TORONTO

  In 1928, Walter's father left Toronto to move to the United States. Now, more than seventy years later, Walter was back in the city, accompanied by his son (also named Walter, and himself an assistant editor), to work for several months on the editing of K-19: The Widowmaker, a film starring Harrison Ford and Liam Neeson, about a hidden piece of Soviet history—the Chernobyl-type meltdown of Russia's first nuclear submarine. So our final conversation took place in Toronto, in June of 2001.

  On his last Sunday we met at my house. In a few days he would be returning to California to finish the editing. Walter, relaxed, spoke about his experiences as a director and scriptwriter. He reminisced about his childhood love for the Oz books; and about his use of the cult book Wisconsin Death Trip as a historical source for his Return to Oz.

  He went on to speculate whether one could perhaps design a notational system for film the way there is one for music, and as our conversations drew to a close, he ended by returning once more to the subject of dreams.

  Murch directing Fairuza Balk, who played Dorothy in his film Return to Oz, 1985.

  BLESSED UNREST

  O: I wanted to talk to you about working as a director on Return to Oz. Since then—1985—you've continued as an editor but have not gone back to directing. Do you want to direct again? Was that something you felt at ease and comfortable with?

  M: I learned something during the process, which is that I'm not temperamentally interested in directing for the sake of directing. Return to Oz was a project that was dear to my heart. I initiated it and I am very glad to have made the film, which I think is wonderful and strange in a way that is true to the spirit of the original Oz books. There are some people who love the process of directing for its own sake, of mobilizing large groups of people. Francis is a classic example of that. I'm just not—I'm a more solitary person.

  On a moment-by-moment basis, the state of mind that you're in when you're editing is probably very similar to the state of mind the writer is in when reorganizing material he's already written, and deciding what order to put it in and what to truncate.

  The advantage of writing and editing is that at any time you can stop what you're doing and walk around the block, or have lunch, or take a phone call, or go dig in the garden and think. At any particular moment, the editor has the freedom to interact or not interact with the material. You can always step away and let the subconscious do some work.

  In directing, once the rails have been laid down, you have to pretty much stick to them. If you're shooting a scene, you have to keep shooting it even if it's not working as well as you hoped, and you have to make the best of it no matter what the variables are. I'm exaggerating to make the point. If something really isn't working, then of course the dire
ctor has to stop and rethink. But you're stopping a huge machine when you do that, and there are very large knock-on costs, which reverberate for a long time afterwards.

  That leads me to another point. You, as a director, are the audience for whom the actors and the technicians are performing. They're looking to you to see how it's going. Emotionally, you are the litmus test. This is exacerbated by shooting out of sequence—which means an actor is sometimes asked, in a single day, to go from carefree exuberance to crushing grief. And to repeat that exuberance thirty times, then have lunch, and then do crushing grief.

  The usual “keel” of continuity that keeps a theatre actor's performance on track is gone. There is no natural keel, because of the fragmentary nature of the filming process. So the director becomes the keel. He has to have the whole film in his head and be able to respond, Yes, this grief is all very well, the actor is right to be grieving, but the grief I'm seeing is too blue, it needs to be a redder grief—whatever that means—because of where the scene is in the film. Because the director has a reference point that is not immediately accessible to everyone else, he knows it's the wrong colour grief, and now must ask the actor to change.

  So as a director you must be very much in the present, paying attention to all the minute things everyone is doing. At the same time, you're being surprised by what they're doing—usually in some proportion of good and bad. You try, in shepherding the scene through, to emphasize the good and eliminate or deemphasize the bad, either through something the actor does or through where you put the camera. You then have to balance that against the material that's already been shot and against what's going to be shot. Decisions that you make have to meld with what has gone before, and they will affect things in the future. To be truly effective you have to live simultaneously in the past, the present, and the future.