Randy Thom

Designing a Film for Sound

Published in Cinesonic - Cinema & the Sound of Music - AFTRS Publishing, Sydney © 2000
Transcribed & edited by Philip Brophy from the talk delivered at the conference in 1999
Contact © 1997

Introduction

Randy Thom occupies a unique place in the unwritten history of sound design in the cinema. Firmly grounded in the fundamental mechanics of narrative, he is possibly the most knowledgable sound designer currently working within the Hollywood system. His awareness of narrative conventions, dramatic form and acoustic logic is impressive - but his ability to interrelate all three in his designs is formidable. Years of experience working with a wide range of American directors and producers have allowed him to develop a flexible methodology so as to blend his work into an equally wide range of projects. Sometimes his work is deliciously noticeable on the soundtrack; other times it is judiciously subsumed into the film's story. Among them are landmarks in the contemporary field: Colors, Wild At Heart, Forrest Gump, Mars Attacks and Contact. By virtue of having experience not only as a supervising sound editor, but in everything from being a sound effects recordist to a final mixer, Thom stands as a practitioner with a finely tuned sense of craft and a flair for imaginative orchestrations of sound.

Philip Brophy


Seeing film versus hearing film

It's often said that when you have a problem finishing a film - that is, a problem figuring out what the best ending is - it means that you've had a problem beginning it. This is because the beginning and the end always need to be connected in some way. And so it is similarly difficult for me when I speak about sound to think of where I'm going to start.

I guess I'll start with what call ridiculous idea number 1: "Film is a visual medium."

The ridiculous idea that film is a visual medium continues to be promulgated by people who have no idea what film storytelling is about. I'm angry about it because it implies that what I do as a sound designer is a secondary activity in film. I know it is not, and I know many very wonderful film-makers who agree. You may have noticed, as I have, that when you're in a movie theatre and the sound stops working for some technical reason, funnily enough, you never hear anybody in the theatre say "it's okay, film is a visual medium anyway."

It's clear that what we see inhabits our consciousness in a way that what we hear does not, and it's certainly true that in some ways our visual apparatus dominates, sometimes subverts, our oral apparatus. But, that doesn't mean that seeing is more important than hearing, and it certainly doesn't mean that visual images in a film are more important than sound. It's just that since we do tend to be more aware of what we're looking at than what we're hearing, that it's easy to jump to the conclusion that film is a visual medium and that the visual is more important.

In fact, since we're not usually so aware of what we're hearing, or of how it is affecting us, sound is all the more effective a tool for storytelling. You can put messages into the sound that the audience isn't going to be aware of in the same way that they tend to be aware of visual images. And, of course, in an ideal situation the audience is not thinking that they're being fiddled with at all, at least in typical narrative film storytelling.

People often ask me if I'm able to go see a film and not think about the sound. It's easy for me to go to a film and not think about the sound, unless I'm bored by the film, and then I begin to think, how did they do that, or why did do they do that. But one of the magical and wonderful things about movies is that if they're at all well made it's very easy to lose yourself in the storytelling. So, I'm going to talk about the role sound can play in films and discuss some films that I've worked on and other films that I haven't worked on.

My background in film sound

I do sound for movies. I do a variety of jobs in sound for movies. Sometimes I only mix the sound on a film. That is, I show up pretty much at the end of the process and push faders up and down which determine how loud any particular sound is going to be at any particular time, and make other adjustments and decisions. On other films, all I do is make sound effects, fabricate sounds, and hand them to other people and they edit them into the film and still other people mix them into the film. On some films, I am in charge of all the sound, except the music, and sometimes I'm working very closely with the composer who's in charge of the music. I typically do that on Bob Zemeckis films like Forrest Gump and Contact and I wouldn't want to do that all the time because it's too big a job, especially if I'm trying to be in charge of all the sound and sound design and mix the film. It's too much responsibility for one person to have on one film after another. I would go crazy or become an alcoholic or something.

So it's good for me to do different things on different films and take a vacation from certain kinds of work. I love to go out and just collect sounds, and that clears my head and makes way for a certain kind of serendipity. I work most at Skywalker Ranch, which is George Lucas' post production facility in Marin County which is just north of San Francisco. I've been associated with George Lucas and his company since the second Star Wars film, The Empire Strikes Back, which was the second film I ever worked on. The first film that I worked on was Apocalypse Now. I had done a lot of sound work before that, mostly in public radio (actually more like Communist radio), very small community radio stations, and some in music recording studios.

In the radio stations I produced and/or recorded radio plays. I never thought I was going to be a sound person until I wandered into a radio station on a college campus one day and they were looking for people to be DJs, and I did that for a while, and then got tired of it and started looking around the radio station for other things to do and so got involved in radio drama and producing little documentaries, etc. Then, in the mid 70s I started thinking about working in movies and banged my head against the walls that are set up to keep you out of the movie business - and there are plenty of them, though I think there are not quite as many now as there were 20 years ago - and finally made a very lucky phone call to this guy Walter Murch who went to the University of Southern California with George Lucas and who came along with Lucas when Lucas joined Coppola to start American Zoetrope in San Francisco.

The first film they did was THX1138 which was an elaboration on a student film that George had done at USC and the film didn't make very much money. They figured after they'd convinced some people in Hollywood to give them some money to make this movie that they had sort of screwed up and blown their wad and that was it, they weren't going to get another chance and they were sitting around wondering what to do next. Paramount Pictures called Coppola and said would you like to direct this pot-boiler Italian novel called The Godfather. They added, you are Italian after all. These young renegade film-makers debated quite a bit about whether they wanted to sell out and stop making art movies and work on The Godfather instead.

Francis, having the good business sense that he has, decided he should probably sell out, at least temporarily. So he agreed to direct The Godfather and then the fortune that he made from The Godfather was used to fund American Graffiti which was the next project that Lucas directed. Apocalypse Now and other films followed. That's a little history-in-a-nutshell of the Northern California film-making bunch which I'm a part of. I'm proud to say that I almost never work in Los Angeles, so when it's said that I work on Hollywood films, it's not literally true. I do go there mostly for meetings with people to try to get jobs.

Designing sound for film

As I said, Apocalypse Now was the first movie I ever worked on, and no film could have been a better training ground for film sound because everything that could go right or wrong did on Apocalypse Now, and it was quite a wonderful mix of people. At least a third of the people who were hired to work on that film had never worked on a film before, partly because Francis and Walter and the others didn't want to import a lot of people from Los Angeles up to the San Francisco Bay area, which is where they wanted to edit the film and do the sound for the film, because they were afraid that they would import L.A. work attitudes: it's only a job, I'll just collect my pay cheque and hang out here everyday until they'll let me leave. So, they went out of their way to recruit young, enthusiastic wild-eyed people like me with what they thought were interesting ideas. But they brought in a core of people who had a lot of experience, too. There were a couple of older English sound editors, Les Hodgson and Les Biggins who had worked on Kubrick films and lots of other well-known films. Those guys had been around so long so that they might have been jaded if they hadn't had us around to energise them. The younger people were always asking questions and bothering them.

We obviously derived a huge benefit from being able to ask them questions and find out how things had been done in the past, and I wish there was that kind of chemistry on more films. The inclination, at least in the US, is to think you always have to hire the most qualified people you can find, and I think that that's not always true. Often the best hire you can make, whether it's in movies or anywhere else, is somebody who's on the edge of knocking at the door of being established, who has a huge amount of enthusiasm and who is smart and has lots of ideas, but hasn't quite gotten there yet, because they're going to be so much more enthusiastic about doing the work than somebody who has a lot of history doing it. At least it's fairly likely that they will.

So, I couldn't help but notice after the first couple of films for which I had been the sound designer that none of them had been Apocalypse Now. I began to think: is Walter Murch really that much smarter than I am? How was it that he was able to devise these wonderful sounds to put into Apocalypse Now to make it the groundbreaking classic soundtrack that it is, and I haven't been able to do that same thing with these other films? I always knew Walter was not smarter than me, so I started thinking about what it was about not only Apocalypse Now but about lots of films that have broken new ground in sound and have used sound in powerful and subtle and innovative ways.

I discovered that that those films have lots of things in common and that virtually none of the ones I had subsequently worked on had: nobody on the crew was called a sound designer. Essentially, what I figured out, I think, is that if you want to create a film that uses sound extremely well, it's more important that you design the film for sound than design sound for the film. That's exactly what happened on Apocalypse Now.

Apocalypse Now was designed for sound and I think in one way or another you can say the same thing about every film that has used sound well. Ridiculous idea number 2: "Sound is something you add to the film". Sound is not something you add to the film. That is a completely wrong headed notion. There has never been a great soundtrack that was pasted onto the outside of a finished product. In order for the sound in a film to contribute in the most powerful way to the film - not only loud sound, but also subtle sound - ideas about sound have to inform creative decisions in every other craft. So, if you have a film like Apocalypse Now where you have Walter Murch as one of the picture editors (Walter has been a picture editor on all of the films that he is famous, justifiably, for doing the sound for) then you have a person who is in a position to actually shape the film to use sound effectively and efficiently; powerfully and subtly.

If you have Orson Welles directing the film, you have somebody who came to film out of radio drama and who had a wonderful imagination for using sound in storytelling before he showed up on the set of Citizen Kane. So, you don't need a sound designer, but what you do need is either one person or a team of people who are open to sound having an effect on designing the film. That was the most over-arching conclusion I came to after looking at lots and lots of films that I thought used sound well.

Point-of-View in Apocalypse Now & Forrest Gump

But there are many more specific things, and as a jumping-off point to start talking about those, I want to discuss the opening of Apocalypse Now. Having seen it, can you remember what guided the editing of those visual images? That is, what was the organising principle of that sequence? The helicopter and the fan are certainly important elements of it, but they are not what guided the editing. Simply, the opening sequence is edited to the music. The Doors didn't come in and make a new recording especially for this. We did get the master recording of "The End", but one of the few things that Francis knew about Apocalypse Now was that he wanted to use that piece of music somewhere.

He wasn't sure that it was going to be at the beginning, it turned out that it was, and though the music itself has some edits in it, essentially, that whole opening sequence is structured to work with The Doors' "The End". I wish all movies would use pop music half that well. Most film-makers' idea of how to use pop music is just to buy the CD and spin it in over a montage. The first sound that we hear in the sequence is this odd synthesized helicopter. Why not just have a real helicopter? Because the sequence somehow moves between objective reception and Captain Willard's (Martin Sheen's) experience - even though I'm not sure that any of it is actually objective. But as far as I can tell, certainly the easiest way to use sound effectively - that is, to design a film sequence to use sound effectively - is to make it a POV (point of view) sequence.

Everything that happens in the opening sequence, at least until the two soldiers arrive there, is Martin Sheen / Captain Willard's point of view. He is remembering, dreaming, hallucinating all of those images that we see, some of them literally, some not so literally. When the audience gets the idea - and God knows they shouldn't get it consciously - but when they get the feeling that what they are seeing and hearing is being filtered through somebody else's consciousness (ideally, one of the characters in the scene) suddenly we as film-makers, and especially we as sound people, have enormous latitude to be creative with what the audience sees and hears. If the audience gets the message that Captain Willard is dreaming and hallucinating and remembering all of this stuff - which I think most audiences do, although they might not be able to articulate it - then they're willing to believe almost anything that's presented to them.

We can distort the images and they'll just attribute it to the idea that that's the way they're seeing the images. We can turn one sound into another sound and they'll buy it because they assume that that manufacturing process is going on inside the head of that character. The hardest shot or scene to do sound design for is one that is shot objectively: that is a locked-down steady camera, brightly-lit scene, nothing ambiguous about it. But, basically, ambiguity in one form or another is what makes a shot or a sequence point of view. There are lots of ways to get at that kind of ambiguity, and most of them are represented in the opening sequence to Apocalypse Now.

Here are some of them. Darkness. When you starve the eye of information, then it's natural for the brain to go to the ears to help figure out what's going on. Dark scenes are much easier to do sound for. Dark scenes are much easier for the sound people to make a significant contribution to. Smoke, haze, fog. A similar thing: you're starving the eye of information and so you're opening the way for the ears to participate. Moving camera. Moving camera shots are almost always easier to do sound for. Moving camera usually suggests POV, but it's not that easy because almost any of these things that I'm mentioning can suggest the POV of the person you're looking at, or they can suggest the POV of some character you haven't even seen yet who may be viewing these images and hearing these sounds. The way that you figure out whose POV you are seeing and hearing is a very complicated process and involves always a lot of experimenting. Slow motion. Slow motion is always a field day for sound, because it's ambiguous. You don't quite know why it's happening and so the audience's brain opens up to almost any possibility that's thrown at it in terms of explaining what's going on. Black and white photography. Very good for sound. There are several others I'll talk about later. But I want to stick to this idea of POV for a minute.

Forrest Gump has a battle sequence in Vietnam and what happens, essentially, is that the Viet Cong soldiers ambush American soldiers and beat the crap out of them. So, in doing the sound for the sequence, it was up to me to figure out how best to deal with that. One of the things I needed to do was to make sure that the weapons being fired at the Americans seemed more powerful than the weapons being fired by the Americans.

One of the main things to notice about that sequence is its camera angles. Almost all of them are very low, and what that does is reinforce this idea of POV. It puts us with the American soldiers, while you never even see the Viet Cong soldiers. The sound of the bullets whizzing by is probably the most powerful sound in the sequence because it's the one that gives you the strongest impression that you're really in the middle of a battle and the bullets are going right over your head.

Movies are lies

I recently did a presentation on film sound in London at the B.F.I.'s School of Sound. There I saw a presentation by an experimental film-maker. The films that he makes are very odd: pieces of film stock with scratches on them and that sort of thing. It's nothing that I have any philosophical problem with. But he explains the reason that he made films like that rather than narrative films or films that depict any form of reality as we normally perceive it with our eyes and ears is that it's impossible, he said, to match reality.

He gave us an example of having been at one point on some savanna in Africa at sunset and he said it was one of the most extraordinary moments in his life, there were elephants in the distance and there were people around who were all aware of how magical this moment was and it was transcendentally beautiful and he said no film could ever duplicate that, so why even bother trying to depict reality in film? We might as well do something totally different from what we normally perceive with our eyes and ears. I couldn't help but think isn't it strange that this guy seems to have no idea of what movies are about? Movies are never, I think, about depicting reality. Usually when they try to be about that, they're very bad.

What makes a movie powerful is not that any one moment in the movie is more beautiful than anything you've ever seen before or sounds better than anything that you've ever heard before. What makes a movie wonderful is that it's a continuum of experiences. That is, any individual shot in a film, if you think of it in strictly visual terms, is usually barely mediocre photography. If you take literally a frame from a film and compare it even with the best of still photography, it really won't compare because a moving picture camera makes certain compromises and the lighting of a scene in a moving picture sequence makes certain compromises to make it work as a whole. It's not based on getting a perfect single image, and you could certainly say the same thing for sound, but in a different way.

Movies are about making connections between things that couldn't possibly be connected in a single real life moment or, at least, in a way that you could be aware of in any sense. Sound is one of the best ways to make those connections. It's about making connections between characters and places and ideas and experiences. We all know that a musical theme for a certain character can link scenes together and periods of time together, and sound effects and dialogue can certainly do the same thing. In my view, movies are lies. There's almost nothing real about a movie. In reality, there are no cuts, there are no picture cuts.

In reality, when you see a car 100 feet away, you do not hear the people inside the car talking, as you often do in film. But one of the wonderful things about certain kinds of lies or misrepresentations of reality is that if you put them together in an imaginative way, they can tell you truth.

Writers tend to make the worst sound films

Maybe I should say as a way of explaining my snide remark about writers, another thing that great sound sequences in films have in common is very sparse, efficient dialogue. There are no great sound sequences in films that have people talking constantly. There are lots of films these days with people flapping their lips from the beginning of the scene to the end of the scene and you can certainly make a wonderful film that way, but there's no way that it's going to be a great sound sequence. So, one of my frustrations is that directors come to me with a film that's already been shot and edited and have people talking non-stop from the beginning of a sequence to the end of a sequence and they say, "can you make this sequence more sinister with sound?"

You can usually make it one or two per cent more sinister but it could be made much more sinister if people weren't talking all the time. So writers, or at least certain kinds of writers, tend to think that having lots of clever dialogue in films is what makes films good. I'm not saying for a second that I don't love the Tracey/Hepburn comedies and there are lots and lots of films that are dialogue-driven films that I like, but they're only powerful sound films in the most limited sense, in the sense that there's cleverly written dialogue and dialogue is sound. Dialogue, or at least, human vocalisations, are a vast, unexplored palette of sound design that I hope I and various other people will explore in the near future.

But all of the sequences, that I admire in terms of their use of sound, have very sparse dialogue. Among other things, when you're listening to dialogue I think it probably shuts off the part of your brain that is open to certain other kinds of experience. But the more important thing is that it gives you the message that because the characters in the film are talking, they're not listening and it's crucial if you want to design a film for sound to find a way that your characters can actually listen, because it's when they're listening and when they're affected by the sound in the film that the audience is going to be affected most powerfully.

You put sounds in the film in a way more for the characters to hear than for the audience to hear, because if you can find a way for the characters to hear them and be affected by them, though not necessarily in any kind of obvious, overt way, then the audience is automatically going to be affected by them, and more powerfully than if you just say here, audience, isn't this a cool sound.

The Black Stallion & Allan Splet

David Lynch couldn't make an uninteresting sound film if he tried. I find that visual artists, and that's the background David Lynch comes from, often have a very powerfully intuitive feeling for sound in their films. David worked on most of his films with a guy named Alan Splet who died of cancer three or four years ago. Alan was another of my mentors.

Alan got into the movie business mostly through David Lynch. He was an accountant in Philadelphia, but also a musician, and Lynch was hanging out in Philadelphia and they somehow connected, I've forgotten exactly how, but they went on to move to Los Angeles and to work on Eraserhead - a film totally different from those I am here discussing, like Apocalypse Now, Forrest Gump and The Black Stallion.

Alan and David worked for five years on Eraserhead, mostly because David didn't have enough money to make the film in one shot. He was making a living delivering newspapers in those days, and so they'd shoot a little bit and do the sound for a little bit and then they'd have to go back to work for a while and get some more money to try to finish the film.

So, Alan then went on to work on Carroll Ballard's films, most notably Never Cry Wolf (for which I did the location sound) and The Black Stallion, for which he was the supervising sound editor, the creative sound force on the movie. The Black Stallion was being worked on at American Zoetrope at the same time that Apocalypse Now was.

In a sequence near the beginning of The Black Stallion, a little boy named Alec is on a trip across the Mediterranean with his father. This is right after World War II and there's a storm and the ship sinks. That sequence certainly has most of the elements I've talked about already, lots of darkness, lots of ambiguity, and next to no dialogue. Later, both Alec and the horse end up stranded on an island, where Alec finds the horse struggling to get loose from some boulders that he's tangled in with the ropes that are around him. Incidentally, they didn't record any sound on the island for this film. So, when I said earlier that sound is not something that you add to a film, I didn't mean it literally, of course. For certain kinds of film sequences, it can actually be an advantage if no sound was recorded. I say that with some trepidation because I think it's almost always a good idea to record the sound when you're on location and then you can always decide not to use it later. But the advantage can be that you start with a clean palette and you don't have to worry about problematic sound recorded on location, then it can sometimes be easier to construct a sound sequence. In the initial scene on the island where Alec discovers the horse, once again, very low camera angles were used. Alec's point of view is apparent all the time, which makes the horse seem all that much bigger and more mythical a character.

Once Upon A Time In The West & the many roles of sound

Sergio Leone's Once Upon A Time In The West finds an entirely different way to use sound creatively. The opening scene where the three dusters are waiting for Charles Bronson to arrive on the train demonstrate this. In some ways I think David Lynch might have borrowed from the film, because like a Lynch film, you're not sure whether it's funny or sinister, but it's probably both at the same time. When you ask, "what does sound add to a film?" most people will say well, it makes you feel a certain way, and that's certainly true. Sound can do lots of other things. So here's the beginning of a list - containing many things covered in that opening sequence from Once Upon a Time in the West.

Sound can also modify the tempo of a sequence. If a director thinks that a sequence is running slow, he or she will often ask the composer to compose something upbeat to make it seem as though it's going faster than it actually is. You can direct attention to or away from something. Directors often ask me to create some sound on the right side of the screen to distract the audience from something on the left they don't want them to see. Sound can smooth transitions between shots or scenes. Sound can also clarify things or make them more ambiguous when that's what you want to do. Sound can establish geography. A common problem in battle sequences - especially in the jungle, and more especially at night - is that it's very difficult for the audience to know exactly where they are at any given time, or where each character is relative to everybody else. Sound is a natural way to indicate something about that geography by having certain voices further away and certain sound effect further away than others at any given moment. Sound can indicate a time period. By that I mean, say, 18th century. It can indicate transitions in time where they might not otherwise be obvious. It can startle or soothe and finally, connect characters, places and ideas within the film or between the film and the culture at large.

Sound people as second-class citizens

You may wonder why I'm talking more about how to make the movie than I am about how to do sound for a movie. Well, it's because it's tragic that sound people for so long have gotten used to thinking of themselves as second class citizens in film and video - because we shouldn't be. It's really only stupidity, ignorance - both theirs and ours - that perpetuates that situation. So, it's crucially important for all of us to continue figuring out what it takes to make a powerful sound sequence in a movie, and what it takes to design a sequence for sound. After we've figured out a little bit about it - others and I have only begun to scratch the surface - we'll be able to raise the consciousness of those around us: directors, producers, editors, etc. who really have never taken sound seriously, even though most of them are fond of saying that they do.

When I'm introduced to somebody, it makes no difference what kind of job they do in the movie business. I'm introduced as a sound person, and I can't tell you how often the person feels called upon to say "well, you know, sound is really very important to film." I wish just once somebody would be introduced as a director of photography and somebody would say "well, you know, images are very important to film." But we sound people get used to being patronised and not taken seriously. It's a tragedy not only for us but for film. I want to have a cloud with a silver lining here: the great thing is that the very fact that we know so little about how sound functions in film is wonderful in a sense because it means that there are these vast territories out there waiting to be discovered.

Questions

Who decided on the songs for Forrest Gump?

I have a pet theory about whose choice it was: I think it was Forrest's choice. I mean that in a literal sense. I've never asked Bob Zemeckis about this, but even though it was up to Bob to make the final choice of what pieces of music would be used in the film, I have a suspicion that Bob chose pieces that he thought that character would have appreciated. They certainly are not the pieces you would choose if you trying to be 'cool'.

I find myself often in the position of defending Forrest Gump against my friends on the Left who think the film is part of a corporate conspiracy to convince people that it's good to be stupid and just buy lots of things. I think Forrest Gump is much more subtle than most people realize it is. One of the things that helped me in Forrest Gump is the fact that Forrest sits on the bus stop bench and describes his life to a succession of people who sit next to him. You may have noticed that he almost never looks at the person he's talking to; he's always looking straight ahead. That fact performs two central functions. Firstly, it makes him seem stupid. Just as a child will often not look at the person they're talking to in an appropriately adult way, you might assume that a stupid adult would do the same thing. The other function is that it puts him into the place he is describing. As the sound designer, I could start playing the sound of the place that he's describing before you actually see the place. You can justify that because he's already there, staring into space thinking about it. It's all being filtered through his brain. We do the same thing on the transitions out as we cut back from what he's describing to seeing him sitting on he bench. You can continue some of the sounds associated with the things he was describing even though obviously they wouldn't be happening there.

There's another very subtle visual thing about the movie that I think is cool. In terms of visual composition, one of these is something you are 'not' supposed to do. Let's say you have a person on the left side of the frame. You're never supposed to have that person looking into the short side of the frame unless there is some very good reason to have them looking that way. There are several places in the film where Forrest does exactly that, and it makes him seem 'stupid'. That's the message you get: not that the film maker is stupid for composing the shots that way, but that the character is stupid - which is exactly the message that we wanted the audience to get. There are many more things like that in Forrest Gump. Bob Zemeckis is a very sharp guy.

The opening scene you mentioned from Apocalypse Now with Captain Willard hallucinating, etc. - what was the decision-making process for sound employed by the director? Was that scene fully covered, taken into the mixing theatre and then decisions made there?

Certain crucial decisions were made in the writing of the film. John Milius wrote the first script, and he certainly designed quite a bit for sound. Once of the things I advocate is that somebody who thinks about sound get involved at some point in the writing or the rewriting stage of a film. Like everything with a film, the way it is written tends to lock you into certain approaches - or lock certain other apporoaches out. It is crucial for sound in one way or another to inform the writing of the script. I said earlier that Coppola wanted to use The Doors' "The End". He also certainly knew he wanted images and sounds of warfare.

Other than that I think there were very few preconceptions about what sound would go with the film. But the fact that it took such a long time to shoot the film and to edit the film meant that there were lots of opportunitites to explore things. There is this thing in Hollywood that I sometimes call 'The Tyranny of Competence': the idea that if you work in the film business you'd better know damn well what you're doing at all times, and that anybody who would ever go into any situation without knowing exactly how to deal with it is unprofessional. Well of course that attitude is anathema to the creative process. No creative person - a painter, writer, film maker or film sound person - knows exactly what they're going to do. One of the dumb myths about the creative process is that certain people just get fully-realized wonderfully perfect ideas zapped into their heads by God, and that's where art comes from. Anybody who's tried to do anything creative knows that you almost always discover more interesting things in the process of doing something than whatever that kernel of an idea was that you had when you started the process. When I talk at film schools in the US, one of the most impotant things I say is: don't be afraid to make mistakes. You are not 'unprofessional' if you make mistakes. In fact, you have to make mistakes. You won't be able to do anything interesting if you don't make mistakes. The trick is to make the mistakes as early and as inconspicuously as possible.

Just as a good film maker will try out photographic ideas by using video cameras very early on when nobody else is watching, you should do the same thing with sound. And you do that because you don't know exactly how to accomplish what you want to accomplish, and you're going to have to find a way to do it. And in the process of finding that way to do it, if you're lucky at all, you come up with these wonderful things that would never have happened otherwise.

I think there are good reasons why we are most intelligent when we're in the shower. Have you ever noticed that people's best ideas come to them there - when they're doing something which has nothing to do with the obvious part of the creative process? Trying to be creative is a very mysterious thing. But one thing that I know from my own experience is this kind of process. Initially I'll play with an idea for a while, and try to physically acomplish something because I'm really eager to not just imagine the sound but to have it right. So I first have to try things that don't work and go down dead ends. Then I stop playing with it for a while and go to do something else: either work on some other film or some other aspect of sound, or not work on sound at all - go take a shower for week. And then the most interesting ideas come to me almost always when I've gone away from it.

I think the source of this serendipity is the fact that we're not smart enough to invent a world out of whole cloth. Every artist is a thief in a way. We're borrowing ideas, or pieces of ideas from everything we have experienced and tried to do, and if you're lucky you're able to synthesize all those stolen pieces of things into something you can legitimately call your own. Interacting with the real world and actually trying to do whatever it is you're about instead of just thinking about it is a crucial part of that. It proves the cliche about writers' block: you just have to start writing something. What kills you with writers' block is thinking that the next thing you have to write has to be true and wonderful and exactly to the point. If you're dominated by that you're never going to do anything interesting because it's going to paralyze you. It's the same with working with film sound. Question: I'm working on the music for a film which has been delivered to me whole and complete, and I have 4 weeks to finish the job. How do I work with the sound as a composer without turning it all into a sonic 'dog's breakfast'? Randy Thom: I'm sort of in your situation almost all the time. Composers are often in no better position than I am as the sound effects guy. The composer is often in a better position that I am. There is a group of white men who make huge amounts of money by composing music for probably 60-70% of the major films made in the United States and Europe. Yet even they don't have it so good because they're constantly being asked to do the impossible. They as much as any of us in sound are hindered by this idea that they should know exactly what it is they have to do because they're great composers and they don't need time to experiment and make mistakes. So the typical composer on a huge feature film (being paid betwen $500,000 to $1m) only has 4 weeks to compose the music - the same amount of time as a low budget movie.

So it's a miracle the music is as good as it is - and lots of film music isn't very good, just as lots of film sound effects aren't very good. How could you possibly expect to consistently compose 90 minutes of great music in 4 weeks? So you're pushed into that thing that all of us have to do on one level or another, which is borrowing things or musical ideas from other films and composers. Many big composers' scores are similar to other scores they have done. Sometimes that's not a bad thing. I'm not saying it's impossible to do a reasonably good score for a film even if you're only working 4 weeks on it, and even if that 4 weeks is at the very tail-end of the process and you've had no opportunity to have any effect on the creative process up to that point.

We can all think of scores that work just fine and which were created under exactly that kind of scenario, but it's a shame that it has to be that way. The expensive composers have painted themselves into a corner: they make so much money that the producers can justifiably say that they can't afford to have them on for any more than 4 weeks. Some of the composers who hate temp scores (music made up from the music from pre-existng movies) in fact thank God for those temp scores, because if they didn't exist they would have no idea of what to do - partly because the communication between a director, composer and sound designer is a very difficult thing. Essentially no commmon vocabulary exists. I find that usually the most instructive thing a director can tell me is what he or she wants the audience to feel at any given moment if it isn't totally obvious from what I know already about the film. It's usually a waste of time to draw up technical terms, and that would be even more silly with music.

Working on music or sound in such a short time is very tough: you're essentially being asked to do the impossible, and they will tend to blame you on one level or another. You'll just have to live with that idea and let it roll off your back. One of the good things about the fact that it's so hard to get into the movie business is that it filters out people who aren't tough and peristent enough to survive in the movie business once they're there. Each film is a test. If you can live through it, and even if it turns out badly and you're resilient and not angry and go after the next job and do as well as you can - you pass the test. But if you say: well, they didn't take me seriously, they didn't give me enough time, so I quit - then that's fine. But if you have that attitude you're not going to survive in the movie business, because no matter how much success you have in the movie business - whether you be a director, composer or sound designer - people are always going to judge you by the most recent thing you did. It's hell being somebody like Francis Ford Coppola who had an enormous amount of success early in life and is expected by people to do more films like The Godfather and Apocalypse Now.

Could you talk a bit about the differences between location dialogue and ADR, and how you blend them in a mix?

Location dialogue - what we call in the US 'production sound' - is the sound that's recorded on set when the cameras are rolling. It's mostly the actors talking, but it also includes incidental sounds that happen to be present there - grips and gaffers cursing in the background, dollies creeping, noisy generator trucks. I used to do production sound for movies, but got out of it after 4 or 5 films because it tends to be a pretty demeaning job. People who do location sound for films tend to be thought of as being in the way. It's not true on all film sets. There are few angelic directors and producers who know how important it is to get good sound on a set, and they go out of their way to make that possible. But on most film sets, sound people are very near the bottom of the totem pole in terms of people's priorities. The prevailing attitude especially on action films is: pffff! we'll loop it. "Looping it" is a term that means the same as post-syncing dialogue or ADR (automated dialogue recording in the US). This is when we bring the actors in in post-production and have them watch themselves in a scene and re-do their dialogue into a microphone.

ADR isn't taken any more seriously than any other aspect of film sound. You march the actor onto a cold sterile room - usually give them not time to get into the character or rehearse. They're expected to just start performing typically a few minutes after they walk into the room. The emphasis is almost always on getting the dialogue in sync. It's not in the right performance. Of course the majority of the actor's brain is then occupied with whether their lips are flapping at exactly the same rate as they were on the day that the camera was rolling. It's no wonder that most ADR is not very good. If you work very hard at it, you can get it to sometimes barely acceptable level, but when you're mixing a film it almost never happens that the director pleads to use the ADR. The director almost always begs for any way possible for the production sound to be used. The reason is that the performance is almost always better in the production sound than the post-sync sound.

One of the things you do with ADR to make it sound more like production sound is to pitch it up. ADR is almost always delivered at a lower pitch because there's adrenalin running through an actor on set. There's lights and cameras all around, they're nervous, and they tend to talk louder and higher. In an ADR session, the director really has to push the actor to get them anywhere near the level of vocal performance they gave on set. In that sense, for the sake of the movie, I would also nearly always rather use production sound than ADR. Sometimes ADR can be better. I did the production sound for Coppola's Rumblefish and virtually none of the production sound made it into the movie - mostly because of ambient noise problems on set. It was a very noisy set; every exterior was immediately adjacent to a freeway even though you never saw it in any of the shots. It was imposible to record it so that you didn't hear those cars. But that film was so stylized that it was helped by ADR. ADR gave it a kind of disembodied feel which was perfect. I could say the same thing about certain biblical epics. It's hard to imagine gritty production sound being used in any of those Charlton Heston movies.

The way that ADR is treated and approached is symptomatic of how little respect sound gets in movies. It's treated as a technical process, and sound in general gets pigeon-holed as a technical process. Sound people are thought of as knob-turners for the most part, and composers are the one occasional exception to that. Occasionally on some films, some of us sound designers try to get what's called a main title credit, which is your name at the beginning of the film with the other so-called important people who work on the film. In the States it is very difficult - ironically because of the Directors' Guild of America (DGA). Because of its contract with the producers, the DGA has control over who gets a credit in the main titles, and the DGA tends to be adamantly against anybody out of the ordinary getting a credit in the main titles. (Most of the people in the DGA, by the way, are not directors: they're assistant directors and production managers.) So if I propose to the producer and director of a particular film that I made an extraordinary contribution in terms of sound effects work, the producers propose it to the DGA, and the DGA says no. The negotiation always come down to: well, OK, but only if the assistant director can get a main title credit too. Of course the producer/director of the film are never willing to do that, so that's where the negotiation usually ends. I got a main title credit on David Lynch's "Wild At Heart" mainly because it was a mistake. At one point Lynch and I were going to share the credit, but at some point he decided he didn't want to do that, but my name stayed there even though his was removed. Nobody noticed it until all the prints were made, yet still various people threatened with law suits and so on.

Do you still do much location sound effects recording, and what kind of equipment do you use for recording sound effects?

We still record location sound effects as much as we can - largely because of those factors of serendipity I discussed earlier. Inevitably when I go somewhere to record a sound, the most interesting sound I find is never one I thought was going to be there. I once went to a nuclear reactor to record sound, and the most interesting sound there was the malfunctioning water cooler in a hallway outside the manager's office. In fact I recorded that sound, slowed it down, and told the director that that was the sound of the nuclear reactor. And he believed it and loved it. He didn't care anything about authenticity so I knew that the lie I was telling him was not important. But I broke down at one point and I told him what it was and he had me taken out of the film. So never tell them what it is.

One of the great things about sound is that it is such a malleable thing. And one of the first things we learn as a sound designer is to forget about the literal and treat sound as music, and appreciate its musical qualities. If you're going to design the sound for a sequence the main thing to go after first is the right emotional and dramatic notes to push the feelings you're interested in pushing. After you find those notes - and they're usually going to come from places that have nothing to do with anything youn see on the screen. An obvious example is say a plane is screaming down the air abut to crash: you bring in wild animals screaming. The reason is that the sound you need there is going to evoke a feeling of out-of-control, wild, frightening. So you think: elephants trumpeting; horse vocalizing in some bizarre way. Only after you find that stuff do you then go and find the more realistic or literal sounds to put in, and then a lot of the art involved is in blending the two, so you have the appropriate emotional impact out of the non-realistic sound, but without anyone in the audience asking where's the elephant in the aireplane. It's amazing how far you can push it: people will tend to accept how far you can push them, and unless it's abolutely blatantly wrong or weird, it won't occur to the audience.

As to the technical question about what gear I use (though I really hate answering technical questions). Fairly soon I'm going to be recording on a portable hard disc recorder. A new one coming out is the Diva. It records 4 channels simultaneously; it can take mic lines from a mixer; it's battery-powered and very small like a DAT machine. One of the neat things is that it can record into a RAM buffer without putting anything onto the hard drive, so that you can be sitting somewhere waiting for the sound to happen, then as soon as you hear it happen you push record and it automatically incorporates the previous 10 seconds onto a hard drive so you don't miss any of the good sounds. I also use DAT machines - usually with mics in an MS set-up. MS recording makes for a small, high quality stereo package with the 2 MS mics in the one wind shield. If you're out in the field, a small package is important because you find yourself hanging off freight trains and that sort of thing, so you don't want a lot of heavy leads around you.

Could you talk about your feelings on the pros and cons of using temp scores?

I usually don't have much to do with putting together temp scores. Alan Splet did on several films he worked on. He certainly was a much more knowledgable person about music than I am. By temp scores we're talking about selecting pieces of pre-existing music and artfully cutting them together to compose a pseudo-score for a new film just to give yourself an idea about how it might work in sequences and how those sequences might play better. Very often a hired composer's score will not be present in the first few test screenings, and a temp score can do an amazing job of making it sound like the score has been done for the film, even though it hasn't. I think there are pros and cons both ways. I think it can be a valid way for directors to give composers an idea of what they want. The terrible thing, though, is that the director and editor tend to fall in love with the temp score, so that then the composer is in the position of having to duplicate something that somebody else has already done, which is not creatively interesting when the director and/or editor insist that that is what they should do.

My proposal to get around temp scores is to bring the composer on earlier, and I think that's something more films are tending to do. A couple of films I'm working on now have brought on the composer very early - sometimes even before they've finished shooting - and have him do rough sketches with synthesizers, then use that as the temp score. Obviously the composer learns a lot about what does or doesn't work in that process, and what the director likes or doesn't like. A few weeks or months later the composer can come back and elaborate on or fix what was lacking in the first pass. For a lot of films that would be the ideal scenario, rather than using pieced-together elements from other film scores.

I'm interested in how you place elements in a surround sound mix.

There's a very slick magazine with expensive advertising in it called "Surround Professional" which is up to its 3rd issue now. They asked me to write an article for them, because they thought I would present the curmudgeonly not-totally-in-love-with-surround view, whereas all the other people who were writing for them were advocating putting everything in the surrounds at all times. I think the question about how to place sounds spatially is the same kind of question that we're faced with in just about every other aesthetic problem that we address in movies: it's about figuring out what you want to do and where you want to direct attention. Movies are a lot about directing the attention of the audience. Probably the biggest myth about mixing for surround sound promulgated by these dumb covers of so-called professional sound magazines that always try to have a photo of the largest possible mixing desk - this one is a mile long, they have to have a telephone to communicate from one end to the other - is that mixing is about literally mixing together as many sounds as possible. I think mixing is not about that at all.

There are two periods of mixing in film: pre-mixing (or pre-dubbing) and final mixing. In pre-mixing, essentially what you're trying to do is boil down huge numbers of tracks into a manageable number. In the Forrest Gump battle sequence, we started with 300 or so separate sound recordings that had to be played in a fairly short period of time. If you were to try to mix all those sounds together without premixing them into categories, you literally wouldn't have enough tracks on a mixing console to plug all those sound into. That's one reason to pre-mix. The final mixing stage I think is mostly about getting rid of stuff. It's absolutely not about seeing how many sounds you can mix together: it's about deciding from moment to moment what you want to hear and what you want to focus on. It's about seeing what is the minimum you can use to get the desired effect. When inexperienced people cut and mix the sound for a complex action sequence, the resulting sound is often very akin to a giant waterfall or standing in the middle of a huge surf. It's a wall of pink noise: that's what happens when you try to mix too many sounds together simultaneously. The first few moments of a final mix of a big action film are very frightening for that reason, because that's what's been brought into the mix. You have lots of tracks of music; way more tracks of sound effects; and people talking and screaming all the time. You put all the faders up, play it all, and everyone wants to commit suicide! You have no idea how you're going to make sense of this mess. This happens in most of the films that get made that have any complexity to them in terms of sound. Everyone in the room looks at everyone else, asking what will we do now. I can't hear the dialogue; I can't hear the sound effects because the music is playing all the time; and I can't hear the music because people are screaming all the time. The way to work yourself out of that conundrum is to make decisions as to what you can get rid of, and those decisions have the purpose of directing the audience's attention from moment to moment. Mixing a complex scene like that is really another movie lie that in the end 'tells the truth' and gives the audience the impression that they're hearing everything at the same time.

The dangerous thing about putting certain kinds of things in the surrounds - transient sounds which get your attention, sounds that come and go quickly as opposed to continuous homogeneous sounds - is in having the audience turn their heads to see where that sound is coming from when they hear it. It takes the audience out of the film. It reminds them that they're being fiddled with, and you usually don't want to do that. That doesn't mean that you never put transient sounds into the surrounds, but it does mean that you experiment with it and think about what you're doing. Home Theatre magazines are notorious for rating movies based on how aggressive their surround presence is. It's like an orthodontist having a web page where he criticises movies based on the quality of people's dental work! A narrow way to judge movies.

The same thing for left and right mixing of sounds. It can be great to have stereo movement with sounds, but it can also be dangerous. Let's say you have somebody walk into the room from the left of the screen. They open a door as they walk in and they are talking continually. Fairly often the shot immediately after that will be from a reverse angle. If you have put the sound over on the left, for some odd reason it seems very weird for the sound to then jump over to the right side of the screen. You're able to rationalize that kind of radical picture cut - seeing the person change spatially that much - but hearing the sound jump across the screen is far weirder. It doesn't even have to be a person on the screen: it could be the sound of a dog barking. In fact it tends to be more of a problem when dealing with off-screen sounds, where you don't even see the sound on the reverse of the cut. So it's just another one of those things that if you're not careful, take you outside the movie. Another thing about multi-channel mixing: you don't need to make multi-channel recordings to mix multi-channel sound. You can do wonderful 7.1 channel sound composed entirely of monaural recordings. You just manipulate them in certain ways to fill the space so that it doesn't sound like every sound was recorded by itself. There's this myth that it's a wonderful thing to go out and record 4 to 8-channel sounds in the field, but those recordings almost never get used in the movie. As I said earlier, a movie scene or shot is never about trying to duplicate reality. No matter how wonderful a discrete multi-channel recording you make in the Amazon jungle, you're always going to want to fiddle with it once it's in the movie - spatially, how loud each channel is in terms of the others, and so on. I wouldn't say it's a waste of time to make multi-channel remote location recordings, but it's one of those things that's better on paper than it is in real life.

How exactly does the sound design you produce get blended with the work a composer does? How are they mixed together?

I hate the distinction between music and sound, and I look forward to the day when composers feel free to use sound effects as part of their musical palette and sound effects people will play with things more musical. It's a silly line drawn between the two: sound is sound in so many essential ways on the film soundtrack.

On the next Bob Zemeckis film, I will be working with the composer Alan Silvestri. We are going to be trying to basically compose for it together, which will be a really exciting thing for me and I hope for him too. The object is to come up with something that isn't clearly either music or sound effects but has the qualities of both.

How I usually work with composers varies quite a bit. On most films I've worked on I've hardly had a conversation with the composer. It's partly because the composer is so crazed trying to do what he or she is attempting in those 4 weeks that the last thing they want is the sound effects guy bothering them. In terms of who mixes the sound effects and the music, that happens in the final mix. In the traditional Hollywood film, there were originally three mixers sitting in the booth - one doing dialogue, one music and one sound effects. In recent years the trend has been toward two people doing it instead of three, and the usual thing is that the dialogue and the music will be mixed by one person, and the other person will mix the sound effects - partly because there are usually more sound effects tracks to deal with than music and dialogue.

Typically the person who mixes the dialogue is the most experienced mixer, because dialogue mixing in film is technically the most demanding, at least in terms of a final mix: you're usually being asked to do the impossible. The hardest thing anybody is ever asked to do in mixing movies is to combine ADR and production sound. Very often in a scene not only do you have certain lines that are post-sync and certain lines that are production, it's very common to have one word be post-sync, the next word be production, the next word be post-sync, and to try to make that seem believable and smooth and real in any sense is just about impossible. Sometimes you can come close enough so that it at least doesn't jump out and grab you and say I'm fake. But, that's the reason traditionally that dialogue mixers were the most senior mixers, because they had the most experience and they had some idea of how to begin to do that. Question: Could you talk about some of the work you have done for scenes in films which you would particularly proud of? Randy Thom: Well, the battle scene in Forrest Gump which I mentioned earlier is one of them. A couple of scenes in Contact directed by Bob Zemeckis worked really well - once again mostly because they were designed for sound; it wasn't necessarily that I was a 'sound genius' in those scenes. I really do think that if a film or a sequence is designed to use sound, whether on purpose or inadvertently, like a lot of them have been in the past, that many people can do a wonderful job with the sound. You don't have to be Walter Murch. If they're not designed for sound, it doesn't matter if you are Walter Murch, they're not going to be that good. Walter and Alan Splet and many other heroes of mine have done sound for movies that did not have groundbreaking or interesting soundtracks. The movie really has to be designed to use it.

Working with David Lynch was quite an experience. After I worked on Wild at Heart I said that I was never going to work on a film again because I was so drained. He demands a lot of you and it's really exhilarating and frightening. He didn't want to do any pre-mixing on Wild at Heart and so we improvised the mix of the film as we went and at least as much music was composed by Lynch and me just noodling on samplers as by Angelo Badalamenti - the film's composer. It wasn't because Angelo had done a bad job, it's just that David wanted to improvise. He's a big believer in serendipity and experimenting, but it put a huge amount of pressure on me, being the sound designer and the supervising mixer. There were many days during the post production when I was pretty close to a nervous breakdown. But, as you can tell, I went back to working on movies.


Text © Randy Thom & Philip Brophy 2000.