Editor John Axelrad, ACE, has been in post-production since 1991, including a stint as Oscar® winner Anne Coates’ assistant on Out of Sight and Erin Brockovich (I interviewed Coates previously here). He’s edited dozens of features including Slither, We Own the Night, Two Lovers, Crazy Heart, Something Borrowed, Miles Ahead, and Krampus. Editor Lee Haugen started as an apprentice under Axelrad and then cut several films including Repentance, Dope, and Miss Stevens before joining Axelrad again to cut Amazon Studios’ new theatrical release, The Lost City of Z.
HULLFISH: Tell me a little bit about working on a movie with two editors. How did you guys work that collaboration out between the two of you?
AXELRAD: Well it was kind of a natural process to have the two of us cutting. I was hired as the editor and I had been co-editing with another one of my former assistants, Kayla M. Emter, who’s now cutting on her own. It was such a liberating experience for me to be co-editing with somebody. We did Miles Ahead together and we did James Gray’s previous movie, The Immigrant, and I just found that experience to be creatively fluid. It allowed us to really explore the film from different points of view and to really enhance the creative process of putting the film together, because when you’re cutting with someone and you’re bouncing ideas off each other, I think it is to the benefit of the movie. It’s not just one perspective of how something can be done. And my favorite part of the process personally is working with the director and trying out ideas: trying to look at the entire structure of the movie; making sure that character arcs are working; that the film itself is well paced; that the characters are alive and vibrant. Lee was also a former assistant of mine, and I saw an opportunity when Kayla was unable to work with us on The Lost City of Z. Lee actually started the film first, because I was finishing up Krampus, so I couldn’t start on the dailies right away. So Lee actually went to Northern Ireland and started the assembly process.
HAUGEN: It was my first time working in the two editor system as well. John has been so gracious to allow creative contributions from all positions. He was the first person to have given me an opportunity in feature films in 2008 on James Gray’s Two Lovers. Our paths just didn’t line up after that. Two Lovers was such an eye opening experience of how to make a film. Working with James and seeing his process and working with John and how they interact together. It was the best learning experience I could ever have and it was fun to be reunited again eight years later.
AXELRAD: It definitely was.
HAUGEN: To have the whole crew, except for Tom… Tom was busy.
AXELRAD: Tom Cross was another assistant of mine. We did five films together. I was not in a career position yet to share editing credit with him, although he did get additional editing credit on everything. But I really wanted to make sure Tom got the exposure working with directors and being able to sit in the chair and collaborate. I like having a team. I’ve got three guys around me that we can bounce ideas off of. It just speeds up the process of editing because different ideas are flying back and forth at a much faster rate. Lee edited the movie Dope, which won him the Best Editing award at Sundance. That was two years ago?
HAUGEN: Yeah, two years ago.
HAUGEN: Thank you
AXELRAD: So it wasn’t that Lee just fell off a truck. He established himself as a successful editor.
HULLFISH: I didn’t think that was the case. So I’m assuming you’re cutting in AVID. Are you using a Nexis, or how are you collaborating?
AXELRAD: Yeah, Avid shared storage on Nexis. We have a good shorthand we developed together. If one of us does changes or does something, we use markers in Avid so I know what he’s done and he knows what I’ve done. And we just kind of develop a process as we go. So much obviously depends on how the director likes to work.
HULLFISH: Tell me a little bit about versioning. I talk to editors all the time about organization. There’s all the organization of doing versions which is compounded when you have two editors. Do you have naming conventions that you’re using? Or bin conventions to keep that stuff straight?
AXELRAD: If it’s dailies we simply divide up the scenes. Lee would be editing certain scenes and I would be editing certain scenes. Sometimes, I’ll take Lee’s scene that he’s edited and I’ll work on it after he’s done. I’ll give him a scene that I did and let him play with it and come up with something else. Then we cut into the reel the version of the scene that we feel is best. We’re obviously editing out of order, so it’s always a guessing game when you’re cutting along with dailies.
Avid works very well with shared projects and shared storage. So only one of us has access to a bin at one time, and if we’re dividing a show into reels in the case of The Lost City of Z or in the case in the film we’re editing now, which is Papillon. At one point we had seven reels. We would just call back and forth to each other, like “Hey can I get reel 1? Are you working on it?” It really just comes down to communication. Before we attack something, we get a series of notes from the director and we do a little pow-wow. And we come up with a gameplan of “okay you’ll work on this and I’ll work on this.” Then after that we show it to each other and give each other notes on the edits.
HAUGEN: John is probably the most organized editor I’ve ever met.
AXELRAD: I’m a little anal retentive. I think that is what he’s trying to say.
HAUGEN: Organized is a nicer word.
HULLFISH: Do you both organize your bins differently so you had to use different bins? Or did you come up with a mutually agreed bin layout and organization?
AXELRAD: I confess that I made him conform to how I like to do it.
HAUGEN: That’s the way I learned to organize the bins from John back on Two Lovers. I adapted that into my own workflow as an editor after that. So coming back to it wasn’t really that much of a stretch. We organize each dailies bin by scene, in script order and frame view. And when we do the first cut, we leave that editor’s version in the bin. And if we continue to do multiple editor versions, we leave them in the scene bins and then we’ll also add them to the reel as we start to build the show. That way we always have a back up of what we originally did so we can reference them.
AXELRAD: That’s an advantage of co-editing with people that have worked with me before in an assistant capacity. I can imprint a certain system so we already have a shorthand going in. I imagine it would be different for me if I joined a show – like I know Tom Cross was editing on Joy with multiple editors. These are people he hadn’t worked with before and they’re all established and I’m sure they have different methods. So I would say that that’s much more complicated if they have different ways of working.
HAUGEN: It’s a good system. I changed things I as I grew, but I think we worked out a good solution.
AXELRAD: Communication is key.
HULLFISH: Are either one of you guys selects reel guys? Or is it always working straight out of the bins – straight off the dailies of the clips themselves?
AXELRAD: It depends on the scene. I’m a huge, huge fan of ScriptSync. For dialogue scenes it is invaluable to me. Often times it is very time consuming for the assistants to do the ScriptSyncing. So if I need to edit a scene right away I’ll just dive in, but ideally I could be working on something while they’re ScriptSyncing a scene. I like to watch the dailies first, mark everything up, mark the takes within the takes, and mark moments and things that I like. Making sure the system, which is kind of a color-coded system of markers, will help my assistant do the ScriptSyncing. And once that’s done, we dive into it. Then we’re able to edit and compare performances. If it’s an action scene with action beats, then I do like to do a selects reel. I’ll go through and mark everything up and maybe break the action scene down into beats, and just cut together the beats that I think are the best moments. And then, from that selects reel, I’ll do the assembly.
HULLFISH: I will pull a selects reel but I will do the selects reel in sections, beats in other words. I don’t do every line or every exact moment. A lot of times if there’s a blocking change in a scene, I’ll say, “Okay here’s where they’re standing up against the window” and then it’s all the shots when they move to the other side of the room. And then one character moves over to the couch and sits down and then I’ll break all that stuff down. Is that kind of what you’re talking about?
AXELRAD: That’s what I mean when I say in terms of beats, and it depends if it’s an action sequence. What I usually will do is make select reels based on character or camera angle. So I’ll say, “Okay, these are all the selects from this character” and they’d be – for example – in the order of a fight scene. Therefore when I’m editing, I instinctually say, “Okay, now I feel like I need to go to this other character” and then I have all the selects for that part of the scene in a separate sequence. It’s different for a scene with dialogue, which may have less “beats,” but I think the method you’re suggesting of how you organize is the same.
HULLFISH: Then I do my bins the same way: organizing bins the way that I imagine the edit might go. So if I know that there’s a big jib shot or something that begins a scene, the director probably wants that to be the first thing. So even if that’s setup I’ll put that at the top of my bin. And then normally wideshots. Then all the two-shots together. I do them visually, not necessarily in the order that they were shot; but in the order of if I’m “looking for a specific visual where can I find it the easiest?”.
AXELRAD: And that’s exactly what we do. We do it in frame mode, and kind of in the visual order of where they occur in the scene. I think it’s easy to scroll down and say “well I’m done with the stuff up top because we’ve already been through that and the coverage doesn’t continue”.
HULLFISH: And then if you’ve got pickups or something you know that those pickups are not going to be at the top but further down or wherever they belong.
Axelrad: I feel like I can’t really get into a scene until I’ve really gone through all the footage and marked it up. It’s like I can’t make my bed until I’ve cleaned and organized everything in my bedroom first.
HULLFISH: So tell me your process, Lee.
HAUGEN: I do sit down and watch the dailies. But I first start with the wide shots so I get a layout of what the scene is going do and where it’s going. Mentally I can prepare as I watch the rest of the dailies. I can envision how the scene is going to playout in my mind. Then, as I go through and watch all the dailies, I will sometimes pick selects of things that I really like and I’ll just put them into the sequence. And then I assemble the first rough pass very quickly, because I do want to see it as a whole. That gives me more of a grasp on where the scene is going to go and what the director is really going for. A lot of the time I will set that scene aside and revisit it the next day, because I do like to walk away, especially if I’m frustrated with a scene or it’s just not fitting together easily. Seeing it with fresh eyes is a good solution for me. That way, I don’t waste too much time fighting with myself during dailies.
AXELRAD: On The Lost City of Z, I did not participate in the assembly since I was finishing up Krampus, so Lee was in Northern Ireland during all of principle photography. A 50 day shoot or something?
HAUGEN: 55 I think.
AXELRAD: 55 day shoot, and Lee put together a 4 hour assembly, and he was caught up to camera. And I don’t know how he did that. I mean it was a good assembly, the problem is that it was 4 hours long. That’s something I struggle with the most is going through dailies. There’s the perfectionist thing in me that I can’t even move on to the next scene until I feel like I polished it. I think that can be a handicap, because in dailies, the purpose is to make sure that there’s not a major problem that your director needs to know about or damaged film or that they don’t have a scene they think they have. So it is a handicap being a perfectionist in that phase. That’s why, for me, I’ve found so much creative energy working with another editor because we can stay efficient together. It’s a very symbiotic thing that we have going.
HULLFISH: So it was a 55 day shoot. Then what happened? How long did you guys go from there?
AXELRAD: Well, for James Gray that’s when it all begins. Every director is different, but James is very much, “Okay, let’s start with scene one, and perfect scene one before we go on to scene two.” So, I learned very quickly, having done three films previous with James, you can put all this sweat and blood and tears into the assembly and it doesn’t matter. Because James wants to rediscover all the footage for himself with you in the room. And to his credit, he’s there with you twelve hours a day and you’re reviewing and he’s watching the paint dry as we edit. But that helps his process and he’s seeing how things go together. That’s where Avid ScriptSync comes in so handy, especially for someone like James, who’s very, very performance-based and wants to be able to compare performances. In the case of The Lost City of Z that’s where the work begins. What Lee did with his assembly was a wonderful blueprint. We were able to watch the whole thing together and kind of know ahead of time, “Okay we’re gonna have some trouble when we get to the second act.” It just puts it in the back of your head. Some directors just like to work on the overall structure first, but working with James he likes to paint the fence one panel at a time before he steps back and looks at the whole fence.
HULLFISH: Martin Scorsese has the quote you may have heard that “No movie is ever as good as the dailies, or ever as bad as the first cut.” What’s the value of doing that first cut, then? What’s the value of putting it together at four hours and saying, ”Okay now we don’t have a movie”?
HAUGEN: I think it’s very important to watch the editor’s assembly because it does lay out the blueprint. It also gives you a rough idea of things to look out for on the first pass of the director’s cut. I know in The Lost City of Z, the jungle is a large portion of the film. And it became Charlie Hunnam and Robert Pattinson for a long period of time in that four hour cut. After viewing we realized we needed to keep Nina, Percy’s wife, more present. So when we were in the jungle we used her voice to read some poetry. which kept her character alive.
AXELRAD: There’s a reason they call it the ‘assembly.’ At first I was offended when I started out. I said, “What do you mean assembly? This is the editor’s cut.” But it really is an assembly because your responsibility is to put things together, which you have to do quickly. You have to do it out of order, based on the order they shoot the scenes. As the editor, I would say during the shooting process it’s 50% politics and 50% editing. Because it is your responsibility to let people know if there’s a problem. And you have to really dance a very fine line: knowing to speak with the right people and not get people panicked. But oftentimes there’s a problem with the film. In the case of The Lost City of Z, they shot on 35mm film and there was a lot of film damage from the humidity and the jungle heat and things like that. So these are the things that – if they simply just shot it and didn’t look at the footage until shooting was over – there would be some unpleasant surprises. So it’s kind of the purpose of an assembly to make sure they have what they need to start to work when shooting is over.
HULLFISH: Great point. The other thing with an assembly: calling it the editor’s cut is kinda funny, because it’s a cut where you have to kinda set your ego aside a little bit. Because I look at a scene and I’m think, “This scene’s never gonna make it into the movie, so I might as well just cut it out.” or “These three lines that start this scene are never going to make it, so let’s just start here”. But in an assembly you can’t, right? You have to do it by the script. So it’s not really the “editor’s cut” because if it was the editor’s cut, you would have cut those scenes out and you would’ve cut those lines out.
AXELRAD: Exactly. If editors had a more time, I’d first deliver the “assembly,” and then I’d trim it down and deliver an “editor’s cut.” It really depends on who you’re working for, because I always felt obligated to include everything. And when I started working with James a few times he said, “You know, you don’t have to include everything I shoot.” So it depends on the needs of your director. But I’d say the general consensus among editors is that you don’t purposely remove dialogue or scenes in an assembly just because you don’t think they’ll make the final cut. You have to swallow your pride at this stage, because the intent of the assembly is not to show what you would do, but rather to have the whole film laid out for you and the director to have a starting point from which to work.
HAUGEN: It’s true even if one dialogue scene is ten minutes long, which we had a couple of those, and I thought, “Wow, that’s a long scene.” But there’s really great moments in it and you can just tell it’s going to be really great once we tighten it up.
AXELRAD: It’s also helpful to reference back to the assembly. Working with James, we’ll edit on the scene for weeks and weeks and change it every which way. And then he’ll say “you know, let’s look at the assembly for that scene again,” which basically has everything in it and the kitchen sink. And it’s just a way to hit “restart” if we’re working a scene and beating it down and it’s become too contorted to a way that’s not working. Watching the full assembly of that scene again may make a few lightbulbs go off to how to approach it differently.
HULLFISH: Tell me a little bit about your feelings on temp music?
HAUGEN: That’s an easy question for a James Gray film, because he has very specific music that he wants to put in his films. And he does so much research ahead of time that we do not temp during the assembly. And we work the scene with James first before we put in music. We want to make sure that it is working and that it is functioning without the help of music. Then as we start to build the show, we add music.
AXELRAD: Personally I love editing temp music and sound effects. I love doing that sound design as a picture editor. I always joke that I’m a frustrated music and sound editor working as a picture editor. But working with James, it’s absolutely true, sometimes music during early stages of a cut can be a band-aid to the point where it is detrimental to the structure of the scene. Among the directors I’ve worked with, I’d say maybe a third of them say absolutely no temp music at the start, because they really want to see if a scene is true to itself and not being masked by music. But it also depends on the genre. When I edited Krampus, the expectation was to thoroughly do a temporary sound design and work with temp music. For a horror-comedy it’s so critical to the genre. Obviously we’re editing first before we include the music, but I think you can really know if something is working after you’ve cut it and watched it and then to have experiment with music. Sometimes I have found the structure of the music perfectly fits the edit that you did. Or sometimes it’s a little off and you might adjust the picture edit to the music, because there is a rhythm to editing. If you find the right piece of temp music, you kind of know when you’ve got that connection when the two are working together in harmony.
HULLFISH: David Woo was telling me he cut a motorboat chase scene to Steppenwolf’s “Born to Be Wild… you know… ”Get your motor runnin’, head out on the highway…” Then he stripped that completely away and the final audio track was just the sound of the motorboats. But the song gave him his rhythm and impetus.
AXELRAD: It depends on what you’re editing. With an action sequence I would say “why not edit with music?” When you’re doing a big drama scene, I always say dialogue first– dialogue should be pacing it. For something like “Crazy Heart” and “Rudderless,” I often let the music guide the edit.
HAUGEN: When I did Dope it was very music heavy. The main characters are in a band and they had four original songs. So those songs were what I pulled from to find the rhythm and the pace of the film: that energy and that pulse within the music they were creating.
HULLFISH: Talk to me a little bit about sound design on Lost City. You were just saying that you’re a frustrated sound editor dressed as a picture editor. Tell me about what you use. Is it all production tracks? Do you have a big sound effects library you like to use? Are you using a lot of like pad room tone and stuff to make things smooth over?
AXELRAD: To start with, there’s the production sound. I always like to talk with the sound recordist before the shoot starts. Make sure that we’re on the same page. I think it’s imperative that we work with the full eight tracks of the WAV file. Lee can attest that I’m very fussy with dialogue editing. And you know the first step is making sure the dialogue is smooth. And I like being able to access all eight tracks of the poly WAV file. Not that I’m using all eight tracks, but I like to be able to choose between them to find the best track for dialogue, whether it’s the lav or the boom. Or if I need ambience or fill from what he or she records. So the first step is to make sure that the production tracks, technically, are smooth and synced and no drop-outs. And then we talk start talking sound design.
HAUGEN: And we mostly pull from a large library that I received from John back in 2008.
AXELRAD: It’s many Terabytes big and many, many sound libraries and custom designs.
HAUGEN: We monitor LCR (left-center-right, as opposed to plain stereo). We fill out everything, every scene has stereo even if it’s just room tone. In The Lost City of Z it was a lot of jungle, and we tried to change it up to help with change of location. Give you a different feel from scene to scene as they were traveling throughout the jungle.
HULLFISH: In Avid, how are you achieving left, center, right? As direct outs? How are you monitoring that?
AXELRAD: We do direct out. I know some people that use a digital interface to do that, I think Tom has a digital interface to do that. But we’re still accustomed to the Mackie Mixer and so we have all 8-channel direct output. So it’s analog but we designate the first four channels as mono (to the center speaker) and the second four channels as stereo (left and right speakers). Editing 3.1 is just a fuller experience, especially when you preview the movie in a theater, to have the dialogue coming out from the center speaker and then you’ve got your stereo music and effects coming out left-right. It’s funny because when I go to New York to edit they just don’t do that. They edit stereo. I think LCR is maybe more of an L.A. thing. And I even know a lot of people that are editing in 5.1 out of the Avid, which it’s very capable of doing.
HULLFISH: I just talked to somebody who was doing that.
AXELRAD: I think like on a big show, if you’re doing a Marvel movie, you want to be editing in 5.1 or 7.1. For a James Gray movie I think editing 5.1 would be overkill, at least at this stage. We’re expected to do a preliminary sound design, but we don’t have time to mix 5.1. They usually don’t hire the sound supervisor until later in the process – usually in the middle the director’s cut. You’re expected to have some sort of sound design blueprint in the movie and I do enjoy working on that—including temp music. I know it’s a sensitive union issue because a lot of music editors get upset: why is a picture editor doing music editing? And I do know a lot of picture editors who respectfully say “that’s not my job.” They don’t want to do it, because they don’t want to take away work from a music editor or from the sound designer. I do it because I enjoy doing it, and that’s always my defense when I get criticized for doing it. Once you know the edit is working, sound design is really the polish you need to really sell it.
HULLFISH: I agree. Let’s talk about the overall or story pacing of the movie.
HAUGEN: This is an epic film, and starting with a four hour assembly, it was very challenging to get it down to the run time that it is, which I think is 2:20 with credits.
AXELRAD: We first started by cutting out every other frame to get it down to two hours.
HAUGEN: For some reason it didn’t work.
AXELRAD: Nah, that didn’t work.
HAUGEN: But as we kept cutting it down and focusing the story more we found the most difficult part was the beginning of the film. Trying to get to the jungle in an efficient amount of time, because that is where the film really takes off.
AXELRAD: And it really depends on the director you’re working with. We’re editing Papillon right now with director Michael Noer, and he comes from more of a documentary background. So his approach is to look at the whole structure. So we constantly are watching the whole movie. We’re just working first on overall structure and arc and looking at overall pacing before we start getting into scene by scene micro-pacing. With James Gray it’s the opposite. He’s a very linear thinker and he wants to go scene by scene by scene. So each scene, with James, has its own internal pacing that works. And then once we get through that process – which takes many, many weeks – then we sit back and look at the whole movie and say “wow okay, now we have an overall pacing issue.” These individual scenes work by themselves, but now let’s look at the big picture and fix that. So, as editors, we have to be adaptable depending on how a director likes to work. For us we can’t rest until the pacing of the entire film is working, and that’s when the tough choices come about: what to cut? what to shorten? what to omit and what to put back in?
HULLFISH: That’s an interesting point: you take some stuff out and it seems to work for a while, and then you decide, “I think we need that scene back.”
AXELRAD: That happens a lot.
HULLFISH: Were there any scenes like that in this movie? Can you describe why that might have happened? Why a scene that seemed like a good idea to take out ended up going back in? What was the purpose of that?
AXELRAD: There’s two different levels to that. You’ve got experimentation where we say, “Hey, we don’t need this let’s take it out” and we kind of retroactively change the story through loop (ADR) lines and through editorial sleight of hand. And so we try to do that a lot to help streamline things, to help change the flow of the story where we feel that there’s a problem. Sometimes it doesn’t work or sometimes the narrative gap is too big and then we’ll abandon that idea and restore something.
HAUGEN: There’s one sequence that we did take out scenes and put scenes back a couple of times. It was the World War I sequence, there were three of four main scenes. And at one point we wanted to make it all about his character, Percy Fawcett, and about him being a great leader in the war. And when we sat back and watched it, we didn’t think it flowed with his story. He’s striving to get back to the jungle – to get back to what he’s really deep down inside looking for. And so we did swap out some scenes to make sure his obsession came through.
AXELRAD: I can remember over the course of the films I’ve cut, and even what we’re editing right now, we’ve taken scenes out, deciding, “We don’t need this, we can tell the story more efficiently doing it this way.” And then we screen it for people and things bump for them. Then we say, “Okay well maybe what we thought would be a better way to tell the story is causing some character interpretation problems.” Even though it’s narratively efficient, character-wise it’s confusing. So we undo something that we thought we were very clever to do, and we restored a scene that we cut out. Even though, narrative-wise, it drags a little bit, it’s essential for understanding the characters and their relationships with each other.
HULLFISH: Let’s talk about crafting performance from actors. I’m sure you’re working with very talented actors, but they’re not working in context sometimes.
HAUGEN: James is great at getting different levels of performance from the actors which is a blessing for editors. That way we can help tell the story as best we can. There’s one scene where an actor goes from being this stern person that keeps all his emotions inside and ends up getting up and yelling at another character. We experimented with many different ways to see which performance fit better: whether he should yell and storm out of the room, or be calm and cool and just see in his anger through his expression. We stuck with the internal performance.
AXELRAD: I think it’s the adage of “less is more.” Sometimes outward emotion or yelling and screaming maybe a tendency for an actor. But I think you can craft a performance about what is not said, the unspoken word, and just through nuance and facial expression. And in the case of The Lost City of Z, there are a couple of instances where we can say much more by saying less. And that really was by removing dialogue from a scene. I know in Two Lovers, which we worked on together, there were whole sequences that just weren’t resonating with people on an emotional level. So we simply removed the dialogue and made this montage out of it, set to the right music, and it really brought sympathy to the characters. A famous case of this is in Raging Bull, where there’s a scene where they’re not speaking — but if you look at it, Scorsese just simply removed their dialogue. Their mouths are still moving, yet you don’t notice it because you are emotionally invested in the scene. You could imagine that scene with dialogue being a far different scene emotionally than without. And so those are some of the things we do to craft and hone performance.
This was fun to construct, editorially speaking. The images from the church of the baptism were from a deleted scene that was to open the movie. During the editing process, we wanted the aftermath of Percy stopping the arrow with his book to be more spiritual, so we cannibalized footage from this other scene. This was a great example of building an emotional moment in the editing process that was not intended from the original shoot.
HULLFISH: Thank you so much for your time guys. Good luck on your film.
Axelrad: Thanks so much, Steve. It was a pleasure speaking with you. We love to talk about the craft of film editing.
Haugen: It was great speaking with you, Steve. Thank you so much.
The first 50 Art of the Cut interviews have been curated into a book, “Art of the Cut: Conversations with Film and TV editors.” The book is not merely a collection of interviews, but was edited into topics that read like a massive, virtual roundtable discussion of some of the most important topics to editors everywhere: storytelling, pacing, rhythm, collaboration with directors, approach to a scene and more.
Thanks to Moviola’s Todd Peterson and Evan O’Connor for transcribing this interview.