Editing Life of Pi
At the start of my interview with Tim Squyres, the editor of most of Ang Lee’s films, including his latest, Life of Pi, I tell him how much I like the movie. I say that I know I like it because its images, its ingeniously affecting conclusion, and, most of all, the headspace it created for me have lingered for days. Upon waking each morning, scenes have come flooding back. And the subtleties of the film’s ending, which contains a rich meditation on the role stories play in our lives, have resonated in my mind in unexpected ways.
“I get a lot of emails from people who saw it at the friends and family screenings,” Squyres replies, “and they say, ‘We’ve been thinking about it all the time.’ It is that kind of film.”
If the Hollywood movie business’s migration to 3D is to reach a satisfactory aesthetic end, then it’s to prompt just this response in its viewers. In addition to telling a great story with compelling performances, a 3D movie must do more with its extra dimension than popping objects out of the screen. Working in 3D, an artist must create an aesthetic philosophy around screen space. This philosophy doesn’t need to be the same for each director. But just as Andrei Tarkovsky claimed that a director’s signature was his handling of time, perhaps in the future a director will be judged for the singularity of his or her concept of the three dimensions.
Based on Yann Martel’s Booker Prize-winning best-seller, Life of Pi tells the story of the young son of a zoo owner stranded after a shipwreck on a life raft with a live and very scary Bengal tiger. It’s a tale of strategy, survival and, as mentioned above, storytelling, and it takes its place among such ruminatively beautiful Lee films as Crouching Tiger, Hidden Dragon; The Ice Storm; and Brokeback Mountain. Editing a film in 3D is just one of the topics Squyres and I discussed only a few days after the film premiered at the New York Film Festival. We also talk about the Avid, the challenge of portraying tedium without being tedious and the state of things for editors today.
Filmmaker: Having worked with Ang for so many years, what was different about the editing process on this film?
Squyres: Shooting in 3D is hard, shooting on water is hard, and then there were special effects, so on this film you had to plan a lot. Shooting in 3D on water, we wound up being able to get far less coverage than we would normally have gotten just because it was so hard getting the shots. Many of the scenes on water, you had to previs them; you couldn’t just go out and shoot the scenes and then figure out [the editing] later. So I had fewer options than I normally have because [Ang and the production] were so constrained in shooting. I didn’t have the ability to rethink completely how I was going to structure scenes because I just didn’t have that many choices.
Filmmaker: Were you involved in the previs process?
Squyres: Yeah, not a lot, but I got them and commented on them.
Filmmaker: At the New York Film Festival Q&A, Ang said that a big section of the film exists as a cut film in previs. Was he directing those sequences to match that previs, and did you have to cut them that way?
Squyres: Well, when I was cutting, I knew what the previs was, but I didn’t care much. My job is to look at the footage and see what I can do with it, not to just plug in pieces to fit a plan. Now sometimes there weren’t really a lot of other ways to do it besides what was planned. But, no, I never saw my job as matching the previs. That wouldn’t be a very interesting job. And also, they didn’t always shoot the previs. Often once [they] got out there with the actors and the waves going, [they] realized there was a better way to do it. So just matching the previs wouldn’t have been possible in a lot of places.
Filmmaker: In terms of working with Ang on the dramatic arc of the movie, what were the concerns you faced as an editor, whether it was with performance or story or pacing?
Squyres: Well, a lot of that stuff you work out in the script stage, and I was very involved in many drafts of the script. But then, once we shot, it was the thing you always have to deal with in movies, which is, it’s too long and you have to start taking things out. The real difficulty was that for a large chunk of this film there is a person who’s drifting in the ocean and who is not in control of events. Normally on a film, you want a strong narrative through-line. But this is not like a caper film, where one scene leads to the next. Things just happen when they happen and [Pi] reacts to them. So trying to make it feel like there’s clear development from one scene to the next, when narratively there really isn’t, was the real challenge in this film. And cutting it down, trying to focus and make character development points about [Pi’s] relationship with [the tiger] and his growing ability to survive, and trying to make those feel like you’re watching a narrative, when in fact, there’s really not any kind of strong narrative drive through that section, that was the hardest part. Because we had established that [Pi] writes in a journal, we had an excuse for voiceover [and we used it when] something needed to be explained.
Filmmaker: When was that?
Squyres: That was when [audiences] were puzzled. Why is [Pi] feeding the tiger? In the book, it is very clear why he felt he had to feed the tiger, but in the movie, you know, if the tiger’s a problem, why not just let the tiger starve? If he’s going to be feeding the tiger, it’s really important that we make sure everybody knows why. But you want [that voiceover] to feel worthwhile and entertaining in its own right, and not to feel like you’re just sticking something in to explain something. It has to feel integral to the film.
Filmmaker: I imagine you also had to combat, in some ways, the essence of the story, which is a tale of being stranded on a boat for more than 200 days. There’s a tedium associated with that.
Squyres: Yes, there was always [the question], how do you portray tedium without being tedious? We wanted to portray that he’s there for a long time and it takes a toll on him. There were quite a few more scenes, some of which I think will wind up on the DVD, of him learning to survive and then deteriorating. We’re not necessarily 100 percent sure that those were all good to take out. But some people in the [test] audiences were feeling the tedium maybe a little bit more than they wanted to, certainly in a mainstream movie. You know, any time you take a scene out, if your script is well written, that scene was there for a reason. It wasn’t there just to bore people. It was there to tell you something and [we needed] to figure out if we could do without it and still make our points. Those [are the] kinds of decisions we had to face a lot of. They were quite a challenge on this film.
Filmmaker: And when you mention being involved in the script process, what kind of feedback did you give at the script stage?
Squyres: I would give very detailed notes about all kinds of things in the script, anything from individual words to overall scene structure. And then Ang and David Magee, the writer, and I sat down a bunch of times and talked about our various concerns. You know, the one thing you have to remember when you’re an editor is that at some point, anything that’s a problem is going to be your problem because everyone else will be gone. So if there was an issue that concerned me, it was good to air that out beforehand.
Filmmaker: Tell me a little bit about your initiation into 3D and the things that you had to suddenly integrate into your editing process?
Squyres: Well, you know, whenever you’re editing you’re [mentally] translating the image on the [monitor] to what it’s going to look like on a big screen. In this case, I’d never done 3D before, and I didn’t want to cut the film in 2D and imagine it in 3D the whole time. So we cut the film in 3D. In fact, I’ve still never seen the movie in 2D. I wore glasses all day, and in our editing room in New York we had a Christie projector and an 11-foot screen. My cutting monitor was like a good small theater so I could experience 3D the whole time. Sometimes I worked on a plasma screen, but when I worked with Ang, [we worked] almost exclusively on the big screen. Our goal with 3D was to make a really immersive experience. With 3D, you have to think a lot about it being comfortable [for the audience]. When you watch a movie in 3D, you’re asking your eyes to do things that they weren’t designed to do. So [as an editor] you have to be very aware of what depth people are looking at and to try not to snap them back and forth too much while, at the same time, accomplishing the things you want to accomplish. In 3D postproduction, there is “convergence” and “interaxial” — or, I guess what’s often referred to as “interocular,” or IO. That’s the distance between the lenses, and it determines the amount of depth in a shot. Convergence is what’s in front of and behind the screen in terms of the image. In postproduction, I can’t change the IO, but I can change convergence. Probably a third of the shots in the film we changed convergence. And that became part of my job. I sort of became the postproduction stereographer. You can [debate] whether that’s part of the editor’s job, or if you should have a separate stereographer. But in our case, I took that job on. Figuring out 3D is really fascinating, and now that I’ve done it, there are a lot of shots where I wish the IO was a little different.
Filmmaker: You’re referring to where in the screen space objects sit and where the audience is directed to look?
Squyres: There are a lot of things that tell you where to look — like who’s talking, for example. But convergence sets what’s playing on the screen, what’s behind it and what’s in front of it and how far. If I have a big landscape shot, I can choose to put [an object] either on the screen, or eight — or 20 — pixels behind the screen, and that will affect how the shot feels. You have to think about it in terms of the shot and also [the shot] you’re coming from and what you’re cutting to. For example, there’s one place where we do a fadeout at night and then a fade-up in the morning. At night, the stars are about 20 pixels back, 20 pixels behind the screen. In the morning, the clouds are about 32 pixels behind the screen. The way that we wanted to do the fade-up is that the sun comes up first, and then everything else follows. If you just fadeout, where infinity’s at 20 pixels and then you fade-up the sun at 32 pixels, your eyes are still thinking infinity’s at 20 pixels, so the sun feels like it’s in a hole. It feels very strange. So what we did is this: when the sun comes up first, it comes up at 20 pixels back, and then, as the rest of the scene comes in, the sun drops back to 32 pixels. You don’t feel it [when you watch the movie], but if we did it differently, you would feel it, and it would feel strange. For every dissolve you have to think about how the shots interact. For some of our transitions, the outgoing shot has to drop back as the dissolve is starting, or even before the dissolve starts. There’s one dissolve where we cut from a wide shot of the boat at night to a close shot of Pi writing. If you just do the dissolve, this little tiny boat looks like it’s floating right on Pi’s nose; it just looks dumb. Leading into that cut, we had to drop the boat back, so [coming in] it feels like it’s behind his head. [In 3D] you have to carefully consider every dissolve, every transition. So [the stereography] really is part of the editor’s job.
Filmmaker: That’s fascinating. And I’m sorry, but can you explain how pixels are the unit of measurement here?
Squyres: Yes. That’s the offset between the left eye image and the right eye image. Let’s say you have an image of a star. If you look at it on the screen without the 3D glasses, you’ll see two stars. When you put the glasses on, each eye only sees one and they automatically reconverge. If the right eye image is to the right and the left eye image is to the left, [the star] feels like it’s behind the screen. If it’s the opposite, it feels like it’s in front of the screen. And so, by repositioning the images left and right, you can control what’s behind the screen and what’s in front of the screen, and you can do that in postproduction.
Filmmaker: What did you cut the film on?
Squyres: Avid Media Composer 5, which doesn’t have 3D support. Media Composer 6, which I was a beta tester for, has fantastic 3D support, but it didn’t come out in time for this movie.
Filmmaker: How did you add that 3D functionality?
Squyres: It was cumbersome, especially because in early versions we did a ton of editorial comps ourselves. And then, there are also some more complicated sorts of things besides fades and dissolves and cuts in this movie. There’s a lot of other fancier transitions, and we did all those in the Avid too. It was really hard and very time consuming, but it’s much, much easier nowadays [in Media Composer 6]. But we found all kinds of different ways to work around those things.
Filmmaker: You were doing the comps yourselves? You weren’t sending them out and getting footage back?
Squyres: It was a combination. We did a lot ourselves. We had a company called Halon that was doing a lot for us. These are just temp comps, but we put the tiger in and everything. Halon did the previs, and they gave us all the tiger and animal animation separately, just on green, so we could comp it in ourselves. At every level of the process, we wanted people to be able to watch it like a movie. And you can’t do that if the tiger’s not there.
Filmmaker: Given that it’s the 20th anniversary of the magazine, I’d like to ask you about the craft of editing and how it’s changing. How do you think you’d be approaching the craft and the business of editing if you were starting out today?
Squyres: It’s tough. Now there’s so much technical stuff you need to learn in order to do a job that shouldn’t be a technical job. It’s more a creative job. I hesitate to say that because it implies a distinction between the technical and the creative, which is misleading because the two complement each other. They’re not in opposition. But our job has become much more technical than it was. What has really struck me is that a lot of people now go into editing from the technical side. They learn the equipment, the software. I remember how [I realized] on Sense and Sensibility that my job, fundamentally, was to pass judgment on great actors’ great performances. Not only was I allowed to say, “This take from Emma Thompson isn’t very good,” it was my job. That’s what was required of me. And that had nothing to do with the technical. That’s artistry on an entirely different level. To be a good editor, there’s an awful lot of technical nowadays that you have to master that you didn’t have to in the past, but you still have to be just as good at the artistic side of it. Hopefully people will be able to cultivate both, but I think it’s harder now than it used to be. It’s also harder because of the way the assistants work now, which is a little more distant from the creative process that the editor and the director go through. You’re not in the same room as much, and I think that’s a drawback for assistants [wanting to move up the ladder]. And I don’t know how you overcome that.
Filmmaker: How has your work as an editor evolved alongside Ang’s as a director over the years?
Squyres: Well, I kind of learned to edit by editing Ang’s footage, and Ang learned to direct with me cutting his footage. I cut his first movie, so we both grew up in this together. We have similar sensibilities about things. Our relationship, actually, was greatly aided by the Avid. His first film I cut on film, and he had never worked with an editor before. His English was a bit limited, and he didn’t really know how to talk about getting a movie made. You know, as a director, sitting in the editing room, that’s all you can do — talk. So his second film, The Wedding Banquet in 1992, we cut it on Avid. It was one of the very first films cut that way. And it was great because you didn’t have to talk so much. You’d just do stuff quickly and not worry about the consequences. And then you had something to talk about: Do you like that or not? Now, I think [Ang and I] understand how each other thinks very well, and I can anticipate things very thoroughly with him. We barely talk during production. We talked a little bit more on this film, partly just because I was in Taiwan. But when I wasn’t in Taiwan, we hardly talked at all. During Crouching Tiger, which was a 129-day shoot, we spoke twice during that entire shoot. And he never gives me any instructions. I cut scenes and show him just so he can be confident that he got the scene. He gives me a lot of leeway to explore, and I generally know his sensibilities and from experience how to approach a scene that he has shot and what kinds of things he’ll like and what kinds things he doesn’t. We’ve been at this a while and have really learned how to work together.