Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps

Editing has very simple origins: a camera couldn’t hold much film. Even if it could, a given “good” take isn’t good for very long. If you want to make a film that lasts more than the length of a take or a camera roll, you have to put two pieces together.

You cut the section of the take that you want to use, then attach it with glue to another piece of film—and you’re editing. You put two opposing actions together and you’re intercutting—what Griffith called the switchback.

The technology of editing is simple. Unlike the mechanics of camera movement or the technology of celluloid or lenses, there haven’t been that many changes in the history of editing. There have been five phases. And the first phase was not really technological at all, when “editors” used to view the printed footage to make their cuts. They actually held up the film and eyeballed it. There are wonderful stories about the early days when films would actually be cut by the foot. If you needed three feet of a reaction, boom, you’d put in all three feet and drop the splice. That way you didn’t even have to bother to look at the film.

Editing became more sophisticated with the advent of the light box, which enabled editors to look closely at their footage and choose the precise frame they wanted to cut on. Glue was a problem, however. It was necessary to ruin one of four sprockets to apply it. If you missed the correct splice by one frame, it wasn’t possible to retrace your steps and take one frame off. You were literally stuck, unless you paid the lab to reprint the footage. And so splicing in early films was not as exact as it is now. Today we can study each frame, analyze continuity, and get it right. It’s harder when you’re using a light box and glue.

The Moviola appeared in 1924 and remained the standard editing system until the Sixties. The workhorse of the editing world, it was difficult to use. There were foot pedals and a hand squeeze and the film ran through a small gate.

The fact that it was difficult to operate suited the studio mentality rather well. Starting in the Twenties, the studios increasingly wanted to get the director out of the editing room. The more the director is in the cutting room, the less control the studio has. The studio hires a director, he shoots two, three films a year. Then the studio assigns editors. The director would already be at work on something else as his film was being edited. The studio stood between the director and the editor. The director would see the film he directed after it was cut. In response many directors came up with the practice of “cutting in the camera.” Directors like Ford and Hitchcock (when working for Selznick) would choose not to do a shot if they didn’t want it used, because they didn’t have control over how it was going to be edited.

And of course the editors would never teach you how to use the Moviola, because that’s their job and they want to keep it. The history of Hollywood used to be a history of secrets. There were “secrets” to lighting, “secrets” to cutting, and only the masters knew them. Or so the masters would have you believe. But the well-guarded tricks to editing were not esoteric at all. Established editors claimed that there were secrets in order to intimidate directors, producers, and studio executives and protect their power and their jobs.

Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps

This changed with the arrival of the flatbed, aka the Steenbeck or KEM (an acronym for Keller-Elektro-Mechanik), named after the companies that made them. The first flatbed editing tables were invented in Germany in the Thirties, and after more advanced models were developed in the Fifties, they came into widespread use in the Sixties. The flatbed was much easier to use than the Moviola, particularly because you could see the image more clearly. Anybody with basic manual aptitude could operate it. It consisted of an editing table with six plates, allowing the operator to play two picture reels and one sound, or two sound and one picture. You stop and start it with buttons. Very simple.

The flatbed was a very timely piece of technology, coming into widespread use right around when my generation started directing. We were the first non-studio generation—the film-school generation. The machine was perfect for us, because we could edit our own films to a degree. We wouldn’t be at the mercy of professional editors, who would tell you: “That cut will never work.” On my first film I had a splice that was six frames long, and the editor said: “You can’t do it—the eye can’t see six frames.” I said: “Let’s just do it anyway.” And by this time, tape had replaced glue. Glue was another encumbrance to the do-it-yourself editor—it required a certain degree of craft, it was time-consuming and frustrating. After tape became the standard, it was possible for a director to come into the editing room to try out an idea, make the splices, look at it, and say, “Hey, that’s pretty good,” and then the next day, you’d bring it up to the editor.

All of those changes in practice were due to technology. But in the history of editing, only one technology has represented a quantum leap: the Avid. Nonlinear electronic editing totally changed filmmaking.

Only a few films have had a transformative impact on the history of editing, and Sergei Eisenstein’s Battleship Potemkin in 1925 was one of them. During the silent era, artists in Hollywood watched films from all around the world, much more frequently than they do today. And after people in Hollywood saw that film, they couldn’t go back to editing the way they had before.

Eisenstein’s theories about the rhythm of editing (by time or action or tone), as set out, for instance, in “Methods of Montage,” seem obtuse nowadays—to my mind, sometimes unnecessarily so. But the Soviets understood that editing could work through assimilation. Faced with a string of images, the brain will work hard to make sense of it—even if there is no sense to be found. Primitive man looked at the night sky and saw animals, gods and goddesses. In movies it began with a famous experiment. The Russian filmmaker Lev Kuleshov intercut a matinee idol’s face with, alternately, a little girl in a coffin, a seductive, reclining woman, and a bowl of soup. It’s the same shot of the actor each time, but he appeared to be reacting to what was displayed before and after his close-up.

Kuleshov’s discovery was that, through editing, a filmmaker could control not only what images the viewer encountered, but how those images were received. Eisenstein’s theories were based on the logical combination of images. Bowl of soup, man’s face—he’s hungry, that’s logical. But he also realized that the brain was capable of doing more in putting together related images. In Strike he crosscut between a slaughterhouse and workers being executed. That’s an idea, not an emotion. The brain doesn’t naturally connect that slaughterhouse equals dead workers. Editing could not only arouse emotions, it could create ideas. Eisenstein became one of the first theoreticians of filmmaking. When I was in film school, he was as well known for his didactic books, Film Form and The Film Sense, as he was for his films.

And yet when you study his films, you realize that, for the most part, Eisenstein edits the way everybody does. In his writings he describes editing through theory, but in practice he edits instinctively—which is why his films work. The one theoretical principle every artist knows is that theories have to be broken. Usually, the first ones to break with the theory are the ones who invented it. The imitators are stuck trying to make the theory work; the artist is already on to something else.

An artist is going to use what works, and Eisenstein is no exception. A quick reading of “A Dialectic Approach to Film Form” might suggest that Battleship Potemkin’s Odessa Steps sequence is all cuts, all dialectic. But as much as Eisenstein is known for promoting editing as the effective tool, he was still smart enough to know that, when push comes to shove, move that camera. He himself writes in “Methods of Montage” that what gives the Odessa Steps sequence its tension is the violation of his formal rhythmic editing scheme. Eisenstein cited the counter-rhythms on screen; for me, it’s the camera movements that, in the end, give the passage its real emotional impact—not the cutting.

The Odessa Steps scene was a game changer. When they saw Battleship Potemkin in Los Angeles, they were stunned by the way Eisenstein was able to create momentum, to pull out an emotion, to stretch out time (how many steps were there?). And this was before the Moviola. The most famous scene in the history of editing was made on a light box.

Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps
Potemkin Odessa Steps

It’s not that different from what Tony Scott did decades later. Eisenstein planned everything in advance down to the millisecond whereas Scott would shoot enormous quantities of footage and assemble the scenes later. But they both use multiple shots with intensive changes of angle and perspective. I like Domino (05) for the simple reason that it’s an example of something taken to the extreme. There are so many edits in that film that hardly anybody gets to say more than five or six words before a splice comes in. Scott would shoot action with multiple cameras from the same angle, running at different speeds, with different lenses. You can find production stills showing his rig: six cameras on top of each other. One camera is even hand-cranked so he can speed up and slow down the film speed, making the image flash and pulsate.

We generally shoot much more footage today than filmmakers did in the past. There’s a lot of fast cutting in silent cinema, but the average number of cuts in features has doubled twice since film history began—and they’re about to double again. Which means you need more raw material, which means you shoot every setup with multiple cameras. You can only light properly for one camera, and so when you shoot this way the lighting suffers. But technology has an answer for that too. Now we “light” in postproduction as well.

Why are there more cuts? Because film styles change, because our ability to process images has accelerated, but also because we can. The technology permits it. Each advance in editing makes it possible to edit more rapidly and with greater complexity. Now we need all that footage, because we’ve retrained viewers to expect all those cuts and multiple points of view.

One of the reasons directors like Ridley Scott and David Fincher keep shooting commercials is because they get to try out new techniques. You could pick any two Scott movies, place them side by side, then check which commercials he did in between to see where the new tricks come from. Because commercials have large budgets, a director can try a hundred wrong things as long as he gets one right. Whereas in a film, you’d better get one of two right. But the interesting thing is how new film technology rewires our brains.

Eisenstein used his montage theory to political ends. He depicted much the same thing as Griffith had—the powerful against the powerless. And because he was a political man, heart and soul, and was much rewarded by the Communist system, the assumption was that his kind of editing was intrinsically political.

The Surrealists were among the first to realize that what Eisenstein and Kuleshov were talking about wasn’t politics. It was about how the mind operates on images, and how it assimilates them and makes sense of them. Man Ray, Buñuel in Un Chien Andalou (29), and Cocteau in Blood of a Poet (32) understood the power of editing. They grabbed Russian thinking and ran with it.

They made better use of the Russians’ ideas than Hollywood did. We would use montage for action sequences, but we really didn’t believe in the mashing of images. Eisenstein was talking about a Hegelian dialectic. A + B collide and C is created. I don’t know if it quite works that way, if it’s that intellectual. It’s more as if A + B + C coexist and create a D, which may not even be visible on the screen.

Hollywood stole what it could from the Russians, and perfected the classical style of editing—by which I mean the “invisible” edit. The Hollywood definition of good editing was, from the Thirties to the Sixties, the invisible cut. If you didn’t see the cut, that was a good splice. The idea was that the filmmaking should be invisible. To put the viewer in the space and hide filmmaking’s presence.

So we spent 40 years creating that, and we’ve spent the past 40 taking it apart. The shower sequence in Psycho (60) is at once the acme of the Hollywood style and its fracturing. Psycho was made in the Moviola era, but Hitchcock had worked with the editor, George Tomasini, on a half-dozen earlier films and had total control over the editing process. The director and editor were working from the same side of the table. And Hitchcock had a freedom in Psycho because it was a low-budget film and he was a big-budget director.

Psycho shower scenePsycho shower scene
Psycho shower scene
Psycho shower scenePsycho shower scene

The cutting of the shower sequence wasn’t easy. It was a time-consuming experiment due to the limitations of the Moviola—it was difficult to run a sequence forward and backward. Once the flatbed became the norm, more and more editors cut scenes in the same style. But Psycho was 10 years ahead of the KEM.     

The biggest technological change in the history of editing came with the Avid in the early Nineties—it might just be the biggest game changer in film history. Before digital editing, you would shoot film, select the desired takes, print positive dailies from the negative, splice them together, and review them on the flatbed and in the screening room. The only way you could watch more than two or three minutes of footage at a time in the Moviola era was in a screening room. The flatbed stretched this out to 10 minutes. The more you see your film, the more you try out new ideas.

But you can only try so much when you’re editing with a workprint. Every time you cut it, it stays cut. Every time you tape it, it leaves a mark. Every time you project it, it gets worn. Every time you try something, you’re damaging it. And if you damage it enough, you have to pay for a new workprint. And that’s money—and the assumption was that if you were any good as an editor you would have figured out how to cut it by now, rather than having to print all this new footage. There was a disincentive to experiment. Another limitation of linear editing is that you can only save one cut at a time. If you made changes after you screened a cut of a scene or of an entire film, that cut was lost and only existed in your memory.

With the advent of the Avid and the nonlinear editing process, anything was possible. A director today can be simultaneously working with multiple editors on multiple cuts with opposing editorial strategies. You can put an experimental cut together in the morning and discard it after lunch. Before the Avid, you screened your film every 10 weeks. Now you screen it every two days. The editor keeps every version of every cut and can access it instantly. It’s possible to experiment with multiple and overlapping effects without waiting or paying for the lab or having to use internegative, which results in a drop in quality. Visual experiments that would not have been often undertaken before with linear editing are now standard operating procedure. Montage made viewers experience movies differently. So does digital editing.

JFLJFK Oliver Stone
JFK Oliver StoneJFK Oliver Stone

Just as Psycho anticipated the flatbed, two films anticipated electronic nonlinear editing. One was Performance (70). The other was JFK (91). I was dazzled when I saw it. I knew Oliver Stone and I felt I had underestimated him. I couldn’t believe how he held all those possibilities in his head, all those different ways to jigsaw that cut together—the multiplicity of different material seamlessly integrated. You can’t plan that. It’s trial and error.

For this reason I assumed for many years that Stone’s style had been changed by nonlinear editing. I assumed the Avid remade him. Only much later did I discover that the Avid began being used shortly after JFK. Because Stone and his editors were using so many formats—35mm, 16mm, 8mm, photos—they transferred their material to and edited it on videotape, anticipating electronic editing.

My contention has been that technologies begin as toys and evolve into aesthetic tools, but the case of JFK demonstrates that sometimes artists anticipate the toys. Whichever is the case, Stone never looked back. JFK was his last non-digital edit.

There is an artistic downside to technology. Faster editing results in shorter editing schedules, leaving less time for the director and editors to “live” with the film. Artistic possibilities can be intimidating. A certain lack of freedom can be quite comforting. When I made Mishima in 1985, because of the peculiarities of the situation by the time I started shooting, the people who had financed it in Japan had withdrawn, ensuring that it wouldn’t be shown in Japan—in effect, I was making a film for nobody. This made the burden of making it much heavier. It’s easy to say, we gotta sell tickets, we have to make compromises. When you realize that the only criterion of excellence is your own, it weighs heavily.

In another 10 years it may be possible to direct a film in postproduction. Directing, redirecting, and editing every scene in every way possible. Now that’s a paralyzing thought.

Next: camera movement

1 | 2 | 3 | 4