The Edit Finds the Film

The edit is not where you finish the film. It's where you find out what film you have.

What you do once you open the footage, defending the movie you meant to make or building the one you actually shot, is the entire job. I've edited a lot of films, including The Killing of Kenneth Chamberlain, which earned me an Independent Spirit Award nomination for editing. Here's what three of those edits taught me.

The Killing of Kenneth ChamberlainStop protecting things

One apartment, one continuous event, and an audience that already knows the ending from the title. Every usual tool is gone. All you have is when you cut and how long you stay. Early on we stayed on shots longer and lived with Kenneth. As the tension built, so did the cutting, until by the end I was cutting every second to second and a half. Duration is a performance choice, and you make it in the edit.

I'm also in that film, as one of the officers. When you cut yourself, you'll protect your own takes without noticing. My rule: you don't get a vote. Pick the take that's right for the scene, then ask whether you'd pick it if it were someone else's face. And it goes further than acting. Your favorite line, the shot that took four hours to light: the footage doesn't know what anything cost.

The RitualLet the editor beat the writer

I have a story credit on The Ritual and I edited it, which sounds efficient and is actually a trap. The writer has reasons for every line. The editor only asks whether the scene is alive. If the actor already did it with their face forty seconds ago, the line goes. And with actors at the level of Al Pacino, Dan Stevens and Abigail Cowen, takes aren't better or worse. They're different. Choosing between them isn't quality control. It's authorship. Stop watching takes in isolation: you're choosing a scene, not a take.

SolitaryThe assembly is supposed to be bad

The first assembly is always long and hard to watch, on every film. It isn't a bad version of the movie. It's the raw material laid out in order. You wouldn't look at a pile of lumber and call the house bad. So watch the whole thing once, touch nothing, and write down two things: where you got bored and where you leaned in. Your instinct about where is almost always right. Your instinct about why, in week one, is almost always wrong.

How I set up a cut

1. Watch and log everything before you cut anything. If your best take is one you never watched, it doesn't exist.
2. String-outs before scenes. Every take of every setup in order, so building a scene is choosing, not hunting.
3. Name things like a stranger has to find them. In four months, you're that stranger.
4. Never overwrite a version. Duplicate and date every meaningful pass.
5. Passes have jobs. Assembly, structure, scene work, rhythm, finishing, in that order.

The edit is the last rewrite, and the only one where you're working with what exists instead of what you meant. Do it honestly and the movie gets better than the script.

Try this this week

1. Watch your footage all the way through. A short, a scene, anything. No pausing, no fixing.
2. Write down where you got bored and where you leaned in. Not why. Just where.
3. Then go make the film. That list is the most honest note you'll ever get on your own work.

Questions, answered

How do you start editing a feature film?

Watch and log all the footage first, build string-outs for each scene, and work in passes: assembly, structure, scene work, rhythm, finishing.

Why does my first assembly look so bad?

Every first assembly is long and hard to watch. It's raw material in order, not a verdict on the film.

How do you edit a film you also wrote or acted in?

Let the editor beat the writer, and give yourself no vote on your own takes. Choose what's right for the scene.