Written by Jennifer Walden
Director Kathryn Bigelow won two Oscars for her film The Hurt Locker (2008) — one for Motion Picture of the Year and one for Best Achievement in Directing. On her follow up feature film Zero Dark Thirty (2012), she earned an Oscar nomination for Motion Picture of the Year. Will her latest feature Detroit continue that Oscar streak? We’ll find out on January 23rd. In the meantime, if you haven’t seen Detroit you can stream it on Vudu or buy it on Blu-Ray. The film is based on the events that transpired during the 1967 Detroit riots in which a group of police officers harassed and murdered unarmed guests at a hotel. It’s a story of racial prejudice and failed justice that unfortunately still holds relevance today.
Bigelow chose three-time Oscar winning sound designer and re-recording mixer Paul N.J. Ottosson to lead her sound team on Detroit. They’ve worked together on The Hurt Locker and Zero Dark Thirty — on both films Ottosson won Oscars for sound. Ottosson, who worked on Detroit at Sony Pictures Post Production in Culver City, CA, talks about building the film’s increasingly aggressive crowd sounds, mixing the Motown music-filled backgrounds, and how they spent weeks editing/cleaning the production dialogue because ADR just doesn’t work in Bigelow’s films. The result of the sound team’s effort is a cohesive soundtrack that supports the story and helps to fuel the tension on-screen.
What were director Kathryn Bigelow’s goals for sound on Detroit?
Paul Ottosson (PO): We didn’t try to make the sound a showcase for any one element. We just thought about the totality of the movie. It wasn’t like we were trying to make the score really cool or make the sound effects really cool. We just wanted to be one with the movie in all aspects.
We also tried to stay true to the locations and the era that we’re in, and be supportive of the story. Director Bigelow doesn’t lean on music a lot in her movies so we have to figure out how we’re going to fill up this entire space for 2 ½ hours. We had to build dynamics into the soundtrack using effects because that’s something that music would typically give you. So these were our discussions.
Where did she want to start with sound on this film? Was there a specific scene or sound that she wanted to tackle first?
PO: Getting the crowds right was very important. There’s also the space in the Annex (of the hotel) where we spend a lot of time. Even though we’re in that small space for a long time we can’t forget that the rest of Detroit was on fire and there was major rioting going on. We had to maintain the feeling of chaos happening in the whole city even though we’re focusing on just one event of the many that happened there. The story of what happened in the hotel was just the one that Kathryn [Bigelow] chose to focus on but it was not an isolated event.
Can you tell me about your approach to the crowd sounds? Were they all done with loop group or were there library effects too?
PO: A lot of it was from the set because Kathryn likes to make things real. A lot of the crowd was captured on-set by the production sound mixer Ray Beckett, who did a great job of covering the crowds and getting those recordings. Many of the crowds we were stuck with because they were on the production dialogue track.
Generally, riots don’t begin as super angry riots right out of the gate. They start with taunting and a bit of aggression and then turn vicious. You need to go through that progression.
I also recorded crowd sounds to fill out the track and to get the linearity and the growth of the crowds to sound correct. I recorded loop group outside and I scripted the scenes because we weren’t shooting to picture. So I timed out how long the scenes needed to be and what needed to happen in those scenes. Generally, riots don’t begin as super angry riots right out of the gate. They start with taunting and a bit of aggression and then turn vicious. You need to go through that progression.
I recorded loop group outside here at Sony. We had a big section of the Sony lot to shoot on, and we had a loop group of about 25-30 people. We recorded them in the alleyways, moving closer and closer to the mics I set up, and it builds to a singular point where the anger gets unified toward the police.
I haven’t ever shot group to this extent. I did a little bit of shooting loop outside on Fury, but those were more specific lines whereas on Detroit we needed these long scenes that were scripted out.
What was your mic setup like for the outdoor loop recordings?
PO: I had stereo mics where I was sitting, and then I had another six mics spread out at different distances from me. I wanted the group directed towards me but I needed that space around them. I had numerous mono mics put up in different configurations. I didn’t want an even pattern; I wanted it to feel real — sometimes louder on the left side or louder on the right, and then I cut it to where I needed it to be.
The important thing was to get long takes. I had 20 minute takes of all these people. I had two actors playing the cops and then the other 25 or so group actors being the rioters. They would talk back and forth and we would do this for 20 minutes. A lot of our group actors were African-American and many had been the target of racism one way or another in their lives, and at one point during our recording session they tapped into that and became truly angry. It was no longer acting. It was more like people letting go of a lot of anger that they had in them. I was just looking for really good actors, but by the end of our recording session I think we got the real thing, the real emotion. You can hear it in their voices in the film that the anger is very real.
MPSE’s Q&A for “Detroit” Panelists are: Paul Ottosson – Sound Designer and Re-Recording Mixer; William Goldenberg – Film Editor; Harry Yoon – Film Editor; Jeremy Hindle – Production Designer; Moderated by Bruce Carse
The film opens up with a party scene and you hear the party crowd having a good time. Then the police storm in and break it up. The party turns tense and the aggression spills out onto the street as the police arrest party-goers. The scene eventually becomes a full-on riot. You did a great job of building that arc from a party crowd into a rioting crowd. How did you do that?
PO: I started by recording the loop group outside and we also did group recordings inside for the party goers for the second half of the day. We did long takes from the first frame of picture that goes to the end of the riot. I wrote it out and directed it and timed it so I would have this escalation. We did multiple takes of that and then we cut those together and layered it to get the right build. First, I needed the crowd to feel like they’re leaving but in the film as they get onto the street, there’s confusion and from there the aggression escalates.
Then, we recorded the party crowd during the interior takes. To that I added some exterior crowd of the people living in this neighborhood. One person says something like, “Get the hell off our turf.” That’s when the party crowd goes from a good mood to being upset. So we had those two different crowds that we worked with. Of course, we also had the crowd from the production takes.
It was a lot of work creating that space. There were some hard-panned sounds, with the exterior crowds (of people who live in the neighborhood) all on the left side and then the party people on the right side. In that overlap, there is some delay and echoes through the street.
I wanted to create these two sources for how this riot happened. The party crowd wasn’t rioting; they were more upset. Then we have the exterior crowd saying, “This is our turf,” and the cops are there harassing them.
When Kathryn shoots a scene with everyone mouthing, she doesn’t feel like she gets the right facial expressions. So that makes it challenging in terms of the dialogue because we need to hear it in order for the movie to make sense.
The crowds were very tricky and having the crowds married to the production dialogue was a challenge. When Kathryn shoots a scene with everyone mouthing, she doesn’t feel like she gets the right facial expressions. So that makes it challenging in terms of the dialogue because we need to hear it in order for the movie to make sense. But we had to maintain the intensity of a rioting crowd. So it was a lot of editing and a lot of pre-dubbing.
Did you have to do any technical ADR to get the dialogue clearer in those scenes?
PO: There are some lines in there that we don’t hear clearly but it didn’t matter so much because we understand the intention. In a situation like that, it’s ok to not understand every line because you still can understand the whole scene. Sometimes a word isn’t as important as getting the feeling of the scene right. Sometimes a line popping out during a riot can feel unrealistic.
ADR is really hard because Kathryn pushes the actors into a space that they’re not comfortable with while they’re acting on-set. Then when they get comfortable, it’s really hard to get that same feeling from them.
But, we did shoot some ADR there and more than not I was just using a word or half of a word to fix what was getting lost. With Kathryn’s movies, it’s hard to shoot ADR because she pushes the actors into situations that become to feel real to them. Then, when we get the actors on the ADR stage, they’re relaxed, like they had a massage that morning or a latte as they’re coming onto the stage. So I ask them to pretend that they’re super furious and they’re up against a wall, and it just doesn’t happen. Like, on The Hurt Locker, we did about 18 or 20 takes with the first actor we had in. Kathryn looked over and asked, “Is this going to work?” And I said, “No, I don’t think it’s going to work.” So we all took a break and Kathryn and I crossed out about 90% of the ADR lines. Then we went in again and had the actor put on a big heavy vest because the actors in the film wore nearly a hundred pounds of gear. So we piled some gear on him and Kathryn was wrestling with him and I was wrestling with him — we beat him up and then had him try the lines again. So, ADR is really hard because Kathryn pushes the actors into a space that they’re not comfortable with while they’re acting on-set. Then when they get comfortable, it’s really hard to get that same feeling from them. It’s a physical experience and she does so much to get that performance out of them on-set. So on her movies we always struggle with ADR and we know that going in so we spend a lot of time cleaning up the production tracks.
On Detroit, we did an additional eight weeks of dialogue editing because there were so much crowd on the dialogue as well as source music on the dialogue through most of the Annex scenes. It was a very challenging dialogue editing job. I think we did a total of 16 weeks of dialogue editing. It was important to me that the work was very consistent as far as noise reduction and track layout.
I know I can’t get the movie to work without good dialogue and I can’t ADR it because the audience won’t buy it. So we had to make the best of it. There were scenes in Detroit where Kathryn was playing source music and there was actually real live music on those dialogue tracks. But we couldn’t use that music because some of the music wasn’t cleared. Also, since they use different takes to make the picture cut, the source music is playing different parts of the songs and that wouldn’t make any sense. With tools and a lot of time, we managed to clean up all that music and then add in approved music during post sound.
In the police station scenes, it’s the same thing. When we have two actors in a room there’s a huge group outside the room. That group was active because Kathryn doesn’t want the actors to feel like they’re on the set pretending something. So there were no clean dialogue tracks for this entire movie.
When they’re in the Annex and there are people yelling outside, there are really people yelling outside and you can hear that on the dialogue tracks as well. So it’s a huge challenge to get it to be clean, but also the ADR doesn’t work out well on these movies.
Who was your dialogue editor?
PO: Robert Troy. He truly is one of the best — perhaps the best dialogue editor out there.
In most scenes in Detroit, there’s music playing in the background, or TV broadcasts playing in the background, or a combination of both. Talking about the music tracks first, were those tracks changing all throughout editorial?
PO: In every movie I’ve ever worked on, they start out very ambitiously with the source music. They pick songs that will be $1,000,000 to clear. For Detroit, there was an insane amount of source music, something like 50 tracks. Some of the songs you hear clearly and others you don’t but they’re all part of this fabric we’re creating. The tracks changed but the tone of the music generally was the same. It might be that some artist had a huge hit song and that will be a very expensive song to clear. It might be that the artist had a similar song that never became a hit and that will be more reasonable to clear — maybe $20,000 instead of $100,000.
That is a process that Kathryn and the picture editor go through and change the songs. Also, there’s a music supervisor named George Drakoulias and he found all these great tracks from that era, songs that would be playing on the radio that night. So they chose the music and I mixed it in.
There was a great scene in the Annex where everyone was gathered in one room talking about music. The Motown song on the radio gives way to Coltrane on vinyl, which then gives way to a TV broadcast. Can you talk about the design and mix on that scene?
PO: In that scene, there’s also a piece of source music playing from outside at the party that they just left. So that’s all in there trying to find its own space. Plus, you have people talking, with overlapping dialogue at times.
It’s such a delicate balance to hear all of that music and then hear as one piece of music takes over another.
I’m really happy about how that scene came out. If you don’t work with sound, you might not know how complicated that scene is. I could mix a shootout scene two or three times over more easily than I could mix that scene. It’s such a delicate balance to hear all of that music and then hear as one piece of music takes over another. We have to hear the dialogue there too. So, do we raise the dialogue or does that make it all get out of balance? Do we lower one piece of music, or all of the music, and then do we have to lower the TV? If we lower the TV then now all of the Foley sticks out. It’s such a super fragile scene. If you go in there and change just one little thing than the scene is at risk for being remixed again. I think a big shootout scene is more forgiving than that scene. A shootout is synchronized sound for things that we see but there’s enough chaos in it where it’s a much more forgiving situation. It’s not that action scenes are easy to mix; it’s just that there are more things you can buy into and accept. Whereas here, we want to play all these sounds, plus there’s sirens going off outside. There are an endless amount of sounds in that scene. The goal is to be able to hear it all without being so disruptive that it takes over the dialogue.
We tried to play with perspective as well, because sometimes that helps. For example, I can place the TV on the left side and place the record player slightly on the right. But then we move around the room and now those sounds need to follow. Like, when the one actor walks up to the record player then that sound becomes more important because it’s center. Then there’s a moment where he goes into his own mental space about Coltrane. Up to that point we were playing all of the different music sources as mono, but when he touches the record player I start opening up the sound a little bit just to give it more presence in the room. It goes into stereo for a while but then it comes back in to a mono sound. There are so many moving pieces in that scene. I don’t know how long it took me to mix it, but it was a long time.
Then you have the processing on the radio music which sounds different from the processing on the vinyl, and then the processing on the TV broadcast sounds different too. Can you discuss your processing on those?
PO: All of those source music tracks start out pristine and I had to find different futzes for them, one that’s suitable for the radio, another that’s suitable for the TV, and one for the record player. Also, don’t forget about the music that’s coming from outside that’s echoing around the whole space.
There’s one compressor I like from McDSP that has a great filter. I like the hard-core cutoff filters in it. They are super brutal. I would use that plug-in to manually compress the tracks so that I could control it and make it as tight as I needed it to be. I also control how I send it out to the room itself. If the radio is coming from the left side, well it still plays across the whole room because of the reverb reflections in the room. So after I compress and filter the music tracks, I send that to a reverb.
The futz setting for the vinyl was maybe less hard than the radio because the speaker was a bit bigger than the radio’s speaker. They all sound different, as if they’re coming out of different size speakers. For the record, there’s a bit of additional noise added. Sometimes I’ll achieve that by compressing a copy of the track really hard to destroy the sound and then I’ll mix that layer in on top to give a little taste of that noise to dirty it up.
The sounds of sirens are also common in the backgrounds because the riots are on-going. Can you talk about the use of siren sounds? Were there guidelines to when and where they should be used?
PO: The sound design is like a score, so sometimes you need to create tension to build up to an emotional peak. So sometimes, I’d use the sirens to tell time. I’d come in hard on sounds or cut out hard to indicate a time change in the same scene.
I’d use them to help communicate that we’re in a different location, because it’s dark sometimes and the audience might not know where they are. So maybe I would play the siren pretty big and then we cut to somewhere else and I completely take out the siren, like in an abrupt way. By cutting out the siren I can indicate that we’re in a different location. Or, sometimes I would use the siren to come into a scene, to help communicate that we are in a different location.
Everything that you hear in the movie was 100% what I and the crew wanted to have in the movie.
Sometimes an actor would say something in the movie and I’d hang a siren over that to create some space afterwards. There’s a rhythm in the soundtrack that we’re trying to find, just as you would with music. There’s a rhythm between the sound and the silence. Maybe you’re expecting to hear a sound and then you don’t and that makes you hold your breath a little bit. I think about the sound in a musical way. There’s nothing that was thrown in at random in the movie. Everything that you hear in the movie was 100% what I and the crew wanted to have in the movie. We thought about the sound a lot.
What was your approach to the gunfire in the film? How did you want the gunfire to sound?
PO: I didn’t want the gunfire to be impressive at all, or oppressive, or cool. I didn’t want the audience to think about the gunfire so much as someone might be dying after it. I never wanted the audience to think, “Oh, hey that’s a cool gunshot.” It just needed to be something that was raw and ended in a very finite way, especially inside the Annex.
For the gunshots outside, there is a little more mechanics in them because they weren’t shooting at people so much as objects, like when they shot out the streetlamps.
All the sounds were taken from real guns, either recordings that I did or from one of my sound effects editor Hamilton Sterling.
The main point is that I never wanted the guns to sound like they’re cool ass guns. This is just the wrong movie for that.
When you’re shooting inside of a room, it doesn’t feel like a single source anymore. It sounds like everything exploded around you because guns are so loud, especially when you’re in a tight space.
I have a pretty extensive military background. I was in it since I was pretty young, from when I was like 13 years old, so I’m very familiar with what guns sound like. The sound of a gunshot has this feeling like it hits your body. And when you’re shooting inside of a room, it doesn’t feel like a single source anymore. It sounds like everything exploded around you because guns are so loud, especially when you’re in a tight space. There’s almost no directionality. The sound is all around you. So that was my approach for inside the Annex.
Although the courtroom scenes near the end of the film are much quieter than the riot scenes (or the scenes at the Annex), the tension there is still as strong. How do you use sound to help achieve that?
PO: When we don’t have a score, I have to use sound to do what the score typically does. To get the courtroom to stay incredibly quiet and tense, I needed to have much bigger sounding scenes around it. And the opposite is true for big sounding scenes — I need to have quieter scenes around it.
For the courtroom, I chose reverbs that sounded a little bit colder with a little more high-end and a bit of echo. Dramatically, there’s a lot of tension in those scenes. Good sounding movies are always helped by a good, engaging story. The sound is really supportive of that story. This is a well-written story, and well-directed with fantastic acting in it. All of that is a part of the puzzle. All of the pieces work well together when we all do our best job at it.
In your opinion, what’s the most stand-out scene for sound on this film? Why? What would you want other sound people to know about that scene?
PO: We talked about two of them — the riot crowd scenes were really challenging because all of the things I said before. Then, that scene up in the Annex stands out to me because I never before had to mix a scene that had three pieces of music playing, plus a TV broadcast. If you compare that scene to a big spaceship battle sequence, in the spaceship battle it’s all on the screen and you see it as well as hear it. For this Annex scene, I spent an immense amount of time to get it to be what it is but it’s not a show piece. Like, you wouldn’t put it on your reel. I would probably pick a scene out of Fury or Zero Dark Thirty.
But I’m super proud of the editing that I and my team did on this film. And I’m super happy with the mix that I ended up doing on it. With a movie like Detroit, I try to look at the whole movie more so than individual scenes. From a sound design perspective, it’s more about the arc of the entire movie. In totality, what we ended up with was a very coherent movie in all aspects. The sound evolved over the course of post production so that by the time I got to reel five, I realized that something in reel one wasn’t going to work anymore. Even the sound followed a long story arc. The sound design was never about a shot or a single scene; it had to do more with the accumulative totally of it all. We thought a lot about the sounds that went into the movie. It wasn’t about one singular event; it was about the totality of the soundtrack and how supportive we are of the story. Through sound, we wanted to make the story feel more coherent and I think we succeeded in making the audience feel like they are part of the movie.
A big thanks to Paul Ottosson for giving us a behind-the-scenes look at the incredibly balanced sound of Detroit – and to Jennifer Walden for the interview!
Please share this:
+ free sounds with every issue: