A Conversation with Walter Murch – Part 2

Her Name was Moviola, continued

In Part 1 of my recent discussion with Oscar-winning editor Walter Murch, he explained about a documentary film with which he’s currently involved. In that film he takes a look at some of the processes traditional film editors went through. We continue that conversation here in Part 2.


What was the experience like to go back in time, so to speak – working with a Moviola again?

I hadn’t cut any dailies on a Moviola since 1977, 45 years ago. Dan had not done anything on a Moviola since 1994. But it all came back instantly. There was not the slightest hesitation about what any of this stuff was or how we made it work. Interestingly, that’s very different from my experience with digital platforms.

In what way?

Let’s say I’m cutting a film using Avid and finish the work. Then three months later I get another job using the same Avid. In those three months the muscle memory of my fingers has somewhat evaporated. I have to ask my assistant questions similar to, “How do I tie my shoelaces?” [laugh] Of course, it comes back – it takes about three or four days to get rid of the rust. Then in about a week I’m fully back.

Murch editing Coup 53 with Premiere Pro

So that’s an interesting neurological question. Why does editing on the Moviola not have the slightest evaporation in 45 years, whereas editing on a digital platform that you are very familiar with start to evaporate if you are away from it for a few months? I think it’s because every skill in Moviola editing is a completely different set of physical muscular moves: splicing is different from rewinding is different from braking is different from threading up the Moviola, etc. etc. And each of them makes a different sound. Whereas the difference between ‘splicing’ and ‘rewinding’ in digital editing is simply a different keystroke. 

In our emails we had talked a little bit about the differences between an upright Moviola and flatbeds like KEM and Steenbeck. Would you expand upon that a bit?

Ironically, the outliers in this are the flatbeds. In a sense, both the Moviola and nonlinear digital are random access machines. With the Moviola, everything is broken down into individual shots, which are rolled up and put into boxes. There might be two or three or sometimes six or seven shots in a box. When you want to see a shot, you ask your assistant, “Can you give me 357, take two?” That’s kind of what happens digitally, too, except you are making the selection by typing or mouse-clicking. Digital is much more random access: you can select internally within the shot.

A KEM or a Steenbeck on the other hand is linear. Everything is kept in the dailies rolls as they came from the lab. If you want to see a particular shot, you have to find it in its ten minute dailies roll. What I would do is thread it through the sprocketed prism, without going through the drive sprockets. Then I’d just spool down at very high speed with my hands on both the take up and the feed and watch for 30 seconds or so while it was winding until I got to the shot.

Next, I would put it on the screening side screen and lock it into the sprocketed motor drives, to figure out a place to edit it into the cut. On the KEM the picture module would be on my left and the sound for that on my right. The center would be what’s coming into the film. That’s my way of working, but everyone has a different way of working.

Murch and Taghi Amirani, director, Coup 53

When George [Lucas] cut on the Steenbeck, he was using a one-screen Steenbeck, so that option of having two screens was not available to him. And so he would make select rolls. He would go through the dailies roll, cut out the good bits and then either hang them on hooks or build them into a separate selects roll. For philosophical reasons, I don’t like working that way, but it’s certainly a valid way of working.

I liked the ‘dailies roll’ method, because as I would be hi-speed scrolling for the shot I wanted, I often would find what I needed instead. As I would be spooling, I would glimpse alternate takes, things that I had initially rejected, which proved to be valuable, because as the film evolves, maybe they would now be helpful. Even when editing digitally, I still also construct what I call ‘KEM rolls’ of everything shot for a scene strung together.

We humans have this fascination with vintage analog gear, whether it’s film or audio. Is it just that touch of nostalgia or something different?

The basic idea behind the Moviola film is that if we don’t do it now it will disappear from history. It was difficult enough now in 2022, which coincidentally is the 100th birthday of the Moviola. The first one was built in 1922. If we don’t do it now, it will become exponentially more difficult.

Not the machine itself. I think they’ll always hang around, because they’re iconic, like ancient sculptures. But ancillary equipment like mag film was really hard to get. Also hard was the pressure-sensitive thermal tape for the Acmade printer. We eventually found it from Pixar in California. But if we hadn’t found those tapes, we couldn’t have made the film. It was as simple as that. Without that specific tape, all of this inverted pyramid would have just collapsed. 

Every film in Hollywood from about 1925 until 1968, let’s say, was cut on a Moviola. Before 1925 cutting was done without any machine. The editors were just cutting by hand, assembling shots together and then screening the assembly. So the screening room was in essence their Moviola. They would take notes during the screening and then go back to the bench and trimmed and transposed shots or whatever. Ultimately all of the classic films from Hollywood after about 1925 were forged on the Moviola. That was my experience standing all day for 12 hours a day at the Moviola, with all this [winding noise] stuff.

It felt like blacksmithing in comparison to what we do digitally. And obviously you are physically cutting the film. I used the Inviso film splicer on this, which was invented in the mid-1970s. So it’s a little bit of a cheat to say in 1972, because I invented the Inviso in 1976. We also used my other invention, which was to make the hooks on the trim bin out of 1.75mm crochet hooks. These have a barb at the end of them, so that prevented several pieces of film from falling off when you hung them on a hook.

I mean, this is inside baseball, but one of the fascinating things about analog editing is that it is open to physical tinkering. For instance: the Acmade numbers were not printing boldly enough for some reason. What to do? Howard’s solution was to wrap a piece of adhesive tape around the sprocket wheel, forcing the film closer to the print head. You could fix something, just like working on a car from 1956. If you had a problem with the carburetor, take a screwdriver and bang away at it. Right? Today with digital fuel injection or now with electric motors, it’s hopeless for an ordinary person to have any access to the engine. To a certain extent there’s a similarity with digital film editing. Of course, if you know how to code and you know what a database does, you can do very sophisticated things in the digital realm, but that’s the requirement.

When Dan was syncing the film up, there was a sound recordist on the shoot, who was a young film student – maybe 24 years old. At the end of one session, he asked Dan confidentially, “Did you do this on every film?” [laugh] It was just incomprehensible to him that we had to do all this very physical work.

John Gregory Dunne wrote an article around 2004 about film craft. He said, a director friend of his called the old way, which is what we were doing in Moviola, “surgery without anesthetic.” That’s interesting, because how did you do it? Well, first of all, we had to do it. There was no other way. How did Michelangelo carve David? Did he sharpen his chisels after every ten bangs? How many assistants did he have? Just those really ephemeral things that were necessary to do well, which I don’t think we have any record of. So that was another reason to do the film. It was just to say, this is what you had to do.

I read Michael Rubin’s book Droidmaker about the start of Lucasfilm. You are heavily featured in it. He referred to a picnic event you hosted called the Droid Olympics, which was directly related to the film editing techniques we’ve been talking about. Please tell me a bit about that.

Droid Olympics, 1978 – Worst Hand Anyway

The first event was held in the summer of 1978. I was working on Apocalypse Now and had been for a year. I was editing on a KEM. Richie Marks and Jerry Greenberg were working on Moviolas. There was an army of assistants re-constituting the dailies after we had cut a scene. I would cut stuff out and hang it on a bin. Then at the end of the day everything would have to be put back together again on the dailies rolls. This was re-constituting the dailies, which was very tedious work.

Marcia Lucas speed splicing. (L-R) Duwayne Dunham, Richard Hymns, and Dale Strumpell look on.

Steve, my assistant, and I were working in the same room. He had an action figure called Stretch [Armstrong]. It was a rubbery creature who could reach across the room with arms that would stretch. He’d taken Stretch and manacled him with wire to the rewinds and put Stretch’s body inside the sync machine. Stretch was being stretched on the rack of the sync machine. I said, “Steve what are doing?” And he said, “Stretch has to suffer!” [laugh] It was a way of blowing off steam from all of the semi-mindless, but crucially exacting work. And so I thought, they need a break.

My wife, Aggie, had put on a horse show earlier that summer for the local kids where they could do various horseback riding skills and get blue ribbons. So I thought, well, we’ll have one of those for a decathlon of skills in film. How fast can you splice? How quickly can you rewind a thousand feet? How accurately can you guess how many feet are in an arbitrary-sized reel of film? Those kind of things.

Murch winning rack stacking

All of the Apocalypse Now editors and Lucasfilm editors were invited. I think The Black Stallion was editing at the time, so they were invited. I think anyone in the Bay Area working in film was invited and it was just a wonderful afternoon. We staged it again two more times – 1983, I think, and then also in 1987. By the time we thought about doing it again, everything had become digital.

There’s probably a digital equivalent of that. But, I guess it wouldn’t be as much fun physically.

No, it wouldn’t be as much fun to look at. There were all kinds of ridiculous things that happened. Carroll Ballard was rewinding and it got out of control and the loops went way up, probably six feet on either side of the sync machine. He didn’t know what to do and panicked. And then, suddenly the loops collapsed and the sync machine flew up into the air and the film got torn to shreds.

Those are the things that you wanted to see! Much more exciting than watching the beach-ball spin around.


This conversation continues in Part 3.

A Conversation with Walter Murch – Part 1

©2023 Oliver Peters