Al Pacino and Andy Garcia in The Godfather Part III (1990).

One part painter, two parts prestidigitator, Walter Murch is, quite simply, one of our greatest living motion picture editors. He has a long and storied career as both a picture and sound editor, and has authored one of the most insightful little volumes on film editing ever published. "Blink of an Eye" should be required reading for every film student.

Murch's credits include sound mixing on The Conversation, The Godfather Part II, and American Graffiti; editing on: Ghost, The Godfather III, The Unbearable Lightness of Being and First Knight. He also collaborated on the screenplay for THX-1138 (directed by George Lucas) and directed and co-wrote the 1985 Disney release Return to Oz. Murch's work has been recognized with many international awards, including an Oscar for sound editing on Apocalypse Now. In 1997 he was honored with an unprecedented double Oscar for film editing and sound mixing on The English Patient-the first digitally edited motion picture to receive an Oscar. Just before this interview, he supervised the re-edit of the Orson Welles classic Touch of Evil, based on Welles' own 54-page memo. The "new" Touch of Evil was released this fall to much critical and popular acclaim.

Gene Hackman in The Conversation (1974).
Victor Wishalla interviewed Walter for MovieMaker at the sold-out 4th Annual Sedona International Film Festival in northern Arizona, where Walter was on the faculty of the festival's filmmaking workshop.

Victor Wishalla (MM): Walter, how can filmmakers with limited budgets get the most out of the editing process today?

Walter Murch (WM): Well, the most recent thing I've worked on was The Apostle, a $5 million film directed by Robert Duvall, using his own money. Basically, he took over the same crew that shot Slingblade, a $1 million film. There was considerably more production value in the The Apostle, but they used the same methodology. They videotaped directly from the negative, looked at dailies on tape and digitized into an Avid from the videotapes. Then they cut everything on the Avid and cut the negatives, using the Avid output. They never printed work prints at all. Without seeing it on film, of course, you never know quite what you're going to get, but it turned out very well for them.

Working like this, you have to rely more on the lab reports for what may or may not be in focus.

Somebody has to look at it critically very early on. Once it's on the Avid, if something is wildly out of focus you see it, but if something is lightly out of focus, it's hard to tell-especially at the numbers you'd be resolving at as an independent. We did The Apostle and the The English Patient at [a digitized resolution level of] four. Thelma [Schoonmaker] did Kundun at 60, but they had a much bigger budget than we had.

Jane Fonda in Julia (1978).
MM: While working on a small monitor, how do you visualize the final picture on the big screen?

WM: One of the main risks of using the Avid without work prints is not seeing it on the big screen before the negative is cut. You have to find ways to work around this problem.You'll find that many of the mistakes you make as an editor come from the problem of looking at a television screen rather than into the image of a motion picture. I set up a 27" Sony monitor to look at either the Avid, as I'm doing it , or videotapes. Out of 3x5 cards, I cut out two little people, full-body silhouettes of a boy and a girl.

I place these so that it seems like they're looking at the screen. I calculate how big they are to make sure that if this screen was 30" wide, they have to be proportionate to that. #That's a reminder to me that I'm not making something for TV, but for the big screen.

Once you get used to this it becomes second nature. On the big screen, you're not aware of the frame as such, you're more deeply sucked into the story. The rhythms with which you tell the story will be different. Sometimes you'll stay longer on shots, sometimes you'll cut away sooner. A lot of editing is instinctive. You feel when you are ready to cut away from a shot.

MM: How do you apply your feel for the film's rhythm to transitions and dialogue?

WM: Ultimately, I want a dialogue scene between two actors to have the feel of a natural ebb and flow of exchanges of information, threats, love or laughter between two people.

Robert Duvall in Apocalypse Now (1979).
Watch two people talking. As they talk, they reach a point where they've made their main point, but still continue. For instance, if I say: "It's very hot out here today, don't you think?" The portion "I think it's very hot out here today" is really the key line. I'm just being polite by adding an extra phrase. You'll find yourself looking at one person until they've made their essential point. Then you'll find yourself looking at the other person, wondering what their response is.

The first person's dialogue will overlap into the reaction shot of the second. Then you look at him until he's had what he has to say and cut back to the first person reacting. So there's this ebb and flow, the dialogue dances with the issues/31/images. This is a wonderful, often unnoticed-but critical-part of what makes a scene come alive.

In contrast to that would be holding on a person while they talk, then cut to the other person. You're on them until they're finished, and so forth. That produces a staccato; I call it "Dragnet Style," after the old TV series, which used it very effectively.

Like any style, it can be overused and you have to find what's appropriate. Under normal circumstances, your reaction to what's being said has a much more fluid feel to it. We try to capture that fluidity in how we manage overlaps with dialogue.

MM: How do you use sound overlaps in transitions?

WM: We use that a lot. Look at how we use sound in The English Patient. Many times the sound for the scene that's about to happen starts to bleed into the end of the earlier one. You are aware of something happening, but you don't quite know what is is. Then, when you cut to the second scene, you find out.

It's like what happens when the alarm clock goes off and you don't wake up, but incorporate the sound of the alarm clock into your dream. Then you wake up and say "Oh, it's just the alarm clock." We used that technique a lot in how we moved from one scene to another.

Ron Howard and Cindy Williams in American Graffiti (1973).
MM: How does a film's genre influence your editing style?

WM: I take my cues from the director, from the script, from what the actors are doing and the rhythms they're doing it in. I try to find the one or two key actors who are setting the tone for the film-usually the stars. I find the rhythm that they're doing things in and use that to influence my own rhythms. Once you begin to really feel these rhythms, you can extend them into areas that may have few or no actors in them. For example, say you have a shot of a landscape, how long should that shot be? It depends on what the rhythms of the film have been up until then.

MM: What happens before you accept a project?

WM: Typically I read the script, and I'm like an actor, wondering if he should take the part. An actor wants something that both reveals his strength and, hopefully, also asks him to do something new. Asking the actor to do something completely new is occasionally interesting and challenging, but it can be risky, because you're out of your depth. And we've all seen films in which actors have been cast who are out of their depth.

On the other hand, a film that only asks the actor to do what he's already done before, that's a limiting thing. You want a certain degree of familiarity and also expansion.

When I read the script, I ask myself: "Do I respond to this material? Are there certain things about this that I've never done before that might be interesting to see?

That's the basis. Then I'll type up a series of six to 10 pages of notes about the screenplay-what I thought of it, what I thought its strengths and weaknesses were, where I thought they might get into trouble, any ideas that might be good to incorporate in the screenplay-sort of a free association on that level. I then send this to them with a little note, 'this is what I thought, let's see if we're on the same wavelength.' Sometimes people will read that and respond with 'we don't want any criticism from you at all.' That's fine. I'm glad to know that soon because I don't want that job.

On the other hand, if they hear what I have to say and respond something like "well, we have a lot of interesting things here. I disagree with this this and this, but these other things are good"-that's the beginning of a good relationship. Already I'm beginning to commit to the project in an interesting way. They're hearing what I'm saying and are beginning to incorporate what I think about the project.

MM: Let's say you've agreed to act as a project's editor. What comes next?

The English Patient (1996)

WM: At that point I start timing the script. I sit down with a stopwatch and try to visualize everything in it and the amount of time it will take per scene. I put a number next to every scene: this one is 30 seconds long, this scene is two minutes and 55 seconds long. I do this three or four times. An involved process, it takes a couple of days. You go into sort of a dream state to do it. Then I list all these figures and compare them. If every time I've thought that scene #37 was 30 seconds long, that's a good indication that there's something solid about that scene.

On the other hand, if I timed a scene three times and each time it comes out at different lengths, that's an indication that there's something hard to visualize about that scene. I then send those timings to the director and the producer so they can begin to thinking about it.

For an independent filmmaker that would be absolutely crucial to do, because you're trying to save money wherever you can.

If the timings of the script come out to be two and one half hours, and you know you want this film to be one hour and 45 minutes, it's clear that you'd be better off cutting your 45 minutes now, rather than spend all that money shooting it.

Film editors are very good at doing this; after all, that's the world we live in-the world of timing.If you, as an independent filmmaker, hire an editor, one of the things I would ask is "are you good at timing films?" If an editor is good at timing, it's an indication that they're a good editor. It's not always true, because it's a knack and you have to visualize something that isn't there yet, and be able to read with imagination. Once you have those timings, you can talk about things in a much more informed way with the director.

MM: What else is important to remember when it comes to stretching production dollars

WM: Also in terms of timing, a good number to keep in mind is 30 percent. Say you time a scene. It lasts 120 seconds, you estimate. Inevitably, when you shoot the scene, entrances, exits and all, it'll come out slightly longer-that's good. You want it to be slightly longer than it will finally be. But you don't want it to be too much longer. If it's more than 30 percent longer, that's a red flag.

If it's under 30 percent, with good editing, you'll find ways to compress that scene, while keeping everything essential in it. If it's over 30 percent, if all of your scenes are over 30 percent-and this is just my personal experience, -you won't be able to get the film down to it's length by compressing. You'll have to lose whole scenes. You'll have to lose chunks of flesh, sometimes whole characters. Elements of the story will have to be abandoned.

MM: As further explained in your book, "Blink of an Eye"?

WM: Yes, although "Blink of an Eye" is not a technical book. It gets into the artistic-and some technical-reasons for why I do certain things and how to get the maximum out of the process of creative editing, based on my experience. MM