Lee Smith, ACE won an Oscar and ACE Eddie for cutting Dunkirk. He was also nominated for ACE Eddies and BAFTAs for Inception and The Dark Knight. And nominated for an Oscar and ACE Eddie for Master and Commander.
His other films include Spectre, Interstellar, The Dark Knight Rises, X-Men: First Class, and The Prestige.
I’ve interviewed Lee before for ART OF THE CUT for Dark Phoenix, Dunkirk, and Spectre.
This conversation is about his collaboration with director Sam Mendes on the film 1917, which was shot to appear as though it was filmed in real-time in a single take – without edits.
This interview is also available as a podcast.
(This interview was transcribed with SpeedScriber. Thanks to Martin Baker at Digital Heaven)
HULLFISH: This was your second project with Sam Mendes?
SMITH: Correct.
HULLFISH: How did that relationship begin?
SMITH: On Spectre, Sam contacted me. He asked me to do Skyfall and I wasn’t available. And he contacted me again on Spectre and I was available and we had a great time on that. Then he contacted me again for this one.
HULLFISH: Got it. This is a movie that seems to not need any editing. I’m sure with you at the helm of the cutting room, that is not true.
SMITH: I learned my lesson when I looked at it and it said it was supposedly in one continuous shot. We shot for three months up to 39 takes on certain setups. You still need an editor, because if it was shot in one shot, it would be a very long day.
HULLFISH: Yeah, that’s true.
SMITH: Editing is not just about joining shots together. It’s about, rhythms and performance and music and sound effects and everything in between. I ended up finding myself as busy as I’ve ever been on this particular film.
HULLFISH: You talked about rhythms and performance. Other than choosing a take, what can you do? How were you able to affect rhythms and performance?
SMITH: Well, a lot of it was watching the performance and rhythm as each take unfolded and having a continuous talk backward and forwards to Sam as we built the movie. Unlike conventional films where you have conventional coverage, you know you can adapt and make many, many changes in post, we knew this wouldn’t be the case.
Of primary importance on this was that every day we were both completely convinced that the rhythm and the speed and the camera were in all of the right positions at the right time before Sam would move on to the next set up. That involved not just take selection, but sometimes music selection for temp music and sound effects because we were effectively building the movie from the get-go because of the limitations in post on what you can do. You just had to be 100 percent sure that the film was working as you were shooting it, and if it wasn’t, then we had to keep shooting or shoot again, reshoot. You had to do whatever it took to make sure you nailed it. There was no room for error.
HULLFISH: Were you on set? Were you watching a videotap? Or how did that work?
SMITH: No, no. I was in Soho, so in London. So Sam would watch the footage on his videotap, but I would then see it the next morning on a much, much larger screen and in much higher definition. I could give him the sort of unbiased opinion of what they shot the day before in a more theatrical environment, which was important to the both of us because had I been on set, you’re basically privy to all of the ins and outs of whether the day’s going well, not well, equipment behaving itself, not behaving itself.
Sam didn’t want all of those things going through my mind. The only thing that he wanted going through my mind was what was the best version of what they shot?
SMITH: Mostly. There was a little bit of discontinuity at the beginning of the shoot, but that was because of weather cover and a few other issues. But it dropped into order pretty quickly. You really had to be extremely careful with what the choices were, simply based upon how you lined up the next day’s shoot. So, continuity of performance, continuity of everything came into play and there were upwards of thirty-nine takes on certain setups, so if you’re wrong it had a horrible knock-on effect.
HULLFISH: How quickly were you able to respond to him from the time when they were actually shooting or wrapping up one day’s shoot? Were you able to say, Hey, I really think take seven is the best take, so then he could use seven as a reference for shooting the next morning?
SMITH: Every morning of the shoot I would view the footage first thing in the morning and then we’d discuss take selections and take choices and then commit. And then, depending on the complexity of what we were doing, that’s how urgent it was.
A lot of times basically we’d communicate every morning before they shot the first shot of the morning to make sure that we were all in sync with what they were lining up.
HULLFISH: Did you stitch together multiple takes for the same “scene” or set-up?
SMITH: Sometimes, yeah. Sometimes I’d join on either different takes to see how they ran — using parts of takes if I could manage to figure out how to put them together. So, there was a lot of editorial work involved in it on a daily basis, and more pressurized simply because there was no going back. Because of the way we set up each shot that continued into the next shot, the pressure was completely different to a normal shoot. But, exciting and kind of thrilling because you’re also making educated guesses as well, every day on what can and can’t work. And that was pretty exciting.
HULLFISH: What was the schedule like that? Did you do any prep work?
SMITH: I was on for some prep work where we were just discussing some of where the joins would have to be. Really it was just theoretical. We had sort of sketches of what they would be and how they would work. Some of them appeared simple. Some of them appeared incredibly complex. But, until you shoot it, you don’t really know what the conditions are.
Part of the fun of it was — you make a lot of plans and then you hope that it all works on the day. Sometimes we had to modify what we were thinking, and other days it was smooth as silk. So you never knew. What appeared to be complex things turned out to be not-so-complex. Simple things turned out to be complex. It’s a tricky way of making a movie. Not for the faint-hearted.
HULLFISH: Do you know how many shots there were? Not how many edits necessarily, but how many shots there actually were in the movie?
SMITH: Yeah, but I can’t tell. I’m under the cone of silence. I know all and everything.
HULLFISH: Of course you do.
SMITH: And under pain of death you will never know.
HULLFISH: Oh, come on.
SMITH: The gentleman who edited Birdman (Douglas Crise and Stephen Mirrione) was quoted as saying, “Whatever you think it was, it was a lot more.”.
HULLFISH: I believe that.
SMITH: It’s true.
It required every skill I think I’ve ever had over a long career to figure out how to do everything. You can make assumptions when you’ve been editing for a long time about what you think you can make work and what you patently think is not going to work. That was all of our discussions and where you raised red flags about something and discuss the next day’s shoot and talk about where the projected join would be as opposed to now you’re actually looking at it. It’s for real. You can sort of make an educated guess and say, “yeah, maybe not quite like that.” Then you adjust accordingly.
There are no match moves using robotic cameras, that kind of thing. This was all shot with many, many different rigs, in a reasonably organic fashion. It’s not like you’ve got computers running cameras that are matching exactly — which I have worked on with other films.
This is more shooting something in the reality of those trenches and the mud and the weather and the sun cover. It was all very real. It’s not a film that was shot on a set or a stage where a rig could even be controlled, which makes it even more of a wild endeavor.
HULLFISH: Obviously, you have the choice of take to affect pace. And maybe the choice of putting two takes together to make one, but what else are you using? Music, sound effects, ADR?
SMITH: We used a few speed ramps — accelerating the actual native camera speed. Also, between the choice of sound effects and music rhythms, they were our other two weapons to adjust pace.
Sound effects and music can very much change your perceived rhythm and that affects an audience quite dramatically. And we were very aware of that.
While I was putting the film together, I was doing a lot of advanced sound work. I brought the sound crew on a lot earlier than you ever would in a normal shooting environment simply because we were always considering that we had to post the movie as you were shooting it, so you had to know that the sound was going to work. You had to know that the music, even being temporary music had the right rhythms and all of that to either wind up the tension or decelerate the tension or where to release the audience.
It just really had to be figured out as we were going. I remember about halfway through the shoot we took half the movie and showed it to the crew — the heads of departments — and they were all kind of blown away because it played like you were watching a finished movie.
We could only do so much in the visual effects as far as crane removal and rig removal and things that were just patently distracting in 1917. But we could do enough — fast enough — to just keep proving that the film was working. And it was working with rhythm and pace and performance. All of that came into play.
And of course, you can adjust performance because I’ve got 39 takes of people talking, so that’s a lot of dialogue.
HULLFISH: What was the shooting schedule and then how much time did you have before you got into the mix?
SMITH: I think it shot for about three months. I think it was 65 days of actual shooting. And then I was on the job for about nine months all up. It ended about a week ago for me. I was still checking the last of the visual effects the morning of the British premiere, which was literally a week ago.
A massive amount of environmental visual effects work that was done. It’s not showy visual effects. It’s just things to just solve problems in the background of shots. Like no-man’s-land and places like that. They were very complicated and painstaking to complete. So, a lot of our time was spent doing visual effects and visual effects reviews. And then, when Tom Newman came on there was a lot of time put into the music and we were doing a kind of a rolling mix with temps and the audience test screenings and all of the above. We traveled to New York to do a test screening that went very well. Then you get into the color timing and the finalizing and the race to the finish. Nine months all up is pretty short for a film of this size. Normally, I’d be on it for more like 12 months.
HULLFISH: You mentioned the screenings. What were you able to do? After a typical screening you’d maybe say, let’s delete that scene or intercut two scenes or something.
SMITH: We laughed about that when we were showing it to the studio. I said, “I’m not quite sure what the notes are going to be.” But, you know, intelligibility. It was quite a complicated soundtrack. We wanted to make sure that when we screened it, that everyone could understand the film. We needed feedback on the music. Like with a lot of audience test screenings, you just need to know where you stand.
Thankfully, we had a fabulous screening with an audience that just loved it. So it really just then left us to complete the film and perfect everything, On a schedule like this, even when you’re test screening, there are many, many visual effects shots that aren’t complete and things that need a lot of tender loving care. But at least you’re not worrying about your movie not being liked. You’re just worrying about making it as perfect as you possibly can.
HULLFISH: Tell me a little bit about your ability to be a storyteller in this instance. What were you able to do that affected story?
SMITH: Well, I think a lot of times it was just continuous talk with Sam about each day and what works and what didn’t. The great thing with Sam is if I ever mentioned that I was struggling with anything or something wasn’t quite how I imagined it would be, he had no problem in picking it up on the next day and shooting again.
Likewise, if he sees something, he could call me up on the weekend and say, “Look, I’m just not comfortable with this. It seems to be okay, but it’s not great.” And then we’d talk it through and talk about ways of modifying it. Then, if needs be, he’d go back and have another go at shooting it. I’ve got to say, every time a modification was made, it was for the better. That’s the mark of a great filmmaker. We’re always striving for — not just okay — it’s got to be great. Everything had to be perfect.
He was always asking whether I ever had the feeling that we should be cutting, which was something that I was carrying with me the whole way through the shoot, as if I was ever watching the footage from the day before and in my mind saying, “Cut.” Or “turn the camera around” or “we’re too long on their backs” or “we’re too long before we look across.” “Why aren’t we looking at what they’re talking about?”
A few times we had to go back and modify shots because of that.
HULLFISH: But there were discussions of actual making cuts? Was that ever considered?
SMITH: We actually did. There was a little bit of advanced editorial work where we’d talk about whether something had run its course and we needed to modify the next section based upon how we were feeling about the section before and a couple of times he modified the script and what they were shooting to increase the speed of events that were happening because we had the luxury basically of continuously watching the movie and then going to the next day.
Everything that happened before was basically giving you a guide as to what the next day would need to be in terms of the shoot. There were a million and one micro-adjustments happening on a daily basis. So it wasn’t like figuring this film out and then just going out and shooting it. It was nothing like that.
HULLFISH: You mentioned that you were trying to choose takes and that shooting next days were sometimes based on the take that you chose. How close were you in your first assembly before Sam came into the editing room to keeping the same takes? And then, as he screened with you were takes changed and how much?
SMITH: We didn’t change a single take.
HULLFISH: Wow! No, take changes from the assembly!
SMITH: Correct.
HULLFISH: (laughs).
SMITH: You’re the first person who’s asked that question. That was a cool question.
HULLFISH: And I liked the way you answered it. I’m thinking about calling the Guinness Book of World Records and seeing if we can get you in there for something. That should be a world record of some type. That’s pretty darn impressive.
SMITH: It was pretty impressive, I thought. We could have changed takes, but this was a combination of both of us being incredibly diligent as it was going along because we just didn’t want to get caught. And honestly, to change too many takes would have caused just a massive knock-on effect that probably might have destabilized making it on time. So I knew what the stakes were when I came into this.
I don’t think I’ve ever concentrated so much to get something so it was unassailably correct. Well, at least you try. But it was pleasing. If we were doing any take selection, it was while we were shooting, we went backward and forwards a couple of times on different takes and then thinking about it for a couple of days, but in general terms, Sam and I were either in sync or very close on take selection and then it would be discussed a few more times. And then we would just go with whatever was deemed to be the perfect take.
And of course, with takes this long, there is no perfect take. There’s the BEST take. Takes are full of human moments. We also had to sort of guess that some of those moments might be correctable. And I had to try and correct them with my team as quickly as possible to prove that they were correctable. That was what made the entire shoot pretty thrilling, I’ve got to say.
SMITH: Like a camera bump, or poor stabilization, where you had a beautiful, perfect take but because you’re in the real world with people carrying physical cameras, the camera operator might be turning to get out of the way might have been knocked by someone. Then we’d look at it, work out whether we could reposition the frame and just do some tricks in the digital world to solve this momentary problem because you wanted to save that take.
And a lot of times we would find a way to do it. Sometimes it was a hair-raising correction, but you just had to talk it through with a lot of people and just say, “We’re really confident we can make this.” We would do a rough version to make sure it works, but given a bit more time, we can make it perfect. That was another daily sort of thing where we’d have to be very sure that we knew what we were talking about. You wouldn’t want to get to the other end of it and have to say, “I can’t get from there to there.”
HULLFISH: In a normal editing situation — and I know you’ve done this in past films that we’ve talked about — an edit can be a weapon. A cut can be something that propels the story, propels the pace, propels the movie. You were not allowed to do any of that. Did that hamper you? Or was it just an interesting challenge?
SMITH: It’s just a different challenge. I could still do certain things. I said to Sam — I’m going to try and get the best of everything in the movie. And no matter what that meant, I had sort of free range to do that.
For example, if there were three sections that I wanted to figure out how to get the three best parts out of three different takes, that was up to me to figure out how to do that in a doable fashion, because there’s no point just saying, “I like these three sections, but there’s no possible way that they’ll be able to still look like one shot.” That would be pointless. But it was surprising how many ways I could figure out how to do it.
That was a learning curve. That’s not something you would normally do. You wouldn’t agonize over it so much. So, it was just something I’d never done before. It was a definite puzzling challenge that he threw out. I’m pretty confident he knew that’s why I’d do it — because he said, “It ain’t going to be easy!” And I said, “Yeah, I don’t like easy.” Easy’s no fun, I get bored.
This was a really interesting way to handle this particular story and immerse yourself with these two boys. I was all in for it. “It sounds ridiculously complicated. Let’s do it.” And hopefully, at the end, when people watch it, they think it looks simple.
I’m pretty convinced 99 percent of people will think it looks entirely simple and that’s exactly how it should be.
HULLFISH: About music: because you were trying to get so much of it done so early, did you try to find temp ahead of time by weeks and weeks and have a great library of stuff that you knew you were going to try to use or was a music editor doing that kind of stuff?
SMITH: Well, I was selecting all the temp music. Tom (Newman) did some demo work that was nonspecific that I managed to use, but the film was coming together so fast that Sam wanted to know it would work. So he really wanted working music the entire way.
What I was trying to do was — basically every night after I put the shots together — would be to improve the sound and pick music, so he’d be able to watch it and we’d be able to start to give it a vibe of music. Then we’d talk about the music and talk about the style and the rhythm of the music — which is why you use temporary music, to begin with. You still have to have an understanding of how music is gonna work and affect the film. That was a big part of my day.
When you end up having to screen for an audience, you’re still using temp score. All of that has to be working and working really well. Bad temp music will kill you maybe quicker than anything.
HULLFISH: On very long takes — in a normally edited movie — that long take has a certain truth to it because there are no edits. Do you feel like that’s something that this movie has, or is that an advantage of doing it this way?
SMITH: I’m not sure about that, because I think editing is an incredibly powerful way to tell a story. As you said, I can adjust rhythm and pace and performance basically at will. I think doing it in this one-shot version is probably the most dangerous way you can make a movie because it’s either going to work or it’s not. It’s kind of thrilling. I guess if there’s any editing to be done in a film like this, it has to be done — as we did — on a daily basis. No point me being quiet and getting into post and saying, “I didn’t really like that. I thought the camera should’ve turned around there” because that would be pointless.
Kind of editing just by dialogue and editing by camera moves and editing in your head. Everything that you’re seeing still has to tell the story. You just work within the constraints of this process. I’ve got to say, it was fascinating. I’ve been editing a very long time.
HULLFISH: So you’re going to send me a timeline right?
SMITH: God. If you saw it the timeline on the Avid it looked insanely complicated,
HULLFISH: I’m sure it did.
SMITH: It was nothing like what you’d imagine.
HULLFISH: I believe it. Thank you so much. Lee. It was great talking to you again. And as always, good luck on your next project.
SMITH: You too.
To read more interviews in the Art of the Cut series, check out THIS LINK and follow me on Twitter @stevehullfish or on Instagram @stevehullfish or on imdb.
The first 50 interviews in the series provided the material for the book, “Art of the Cut: Conversations with Film and TV Editors.” This is a unique book that breaks down interviews with many of the world’s best editors and organizes it into a virtual roundtable discussion centering on the topics editors care about. It is a powerful tool for experienced and aspiring editors alike. Cinemontage and CinemaEditor magazine both gave it rave reviews. No other book provides the breadth of opinion and experience. Combined, the editors featured in the book have edited for over 1,000 years on many of the most iconic, critically acclaimed and biggest box office hits in the history of cinema.