Tuning Up: How do you write Music For Podcasts?
- 16.02.2024
What’s it like to compose for a hit podcast? How does music fit alongside voices? Do you need a catchy theme? And what can small children add to the process? To find out more, Andrew Dickson spoke to Novel composers Luisa Gerstein and Julian Lynch about how their world works.
Of the numerous elements of that make up a podcast, for many listeners music is maybe the most mysterious. Sure, we get that producers record interviews with subjects and atmosphere out in the real world, and that a host sits behind a microphone in a studio to record voiceovers and scripted links. But for many audio documentaries – especially the richly atmospheric kind that Novel specialise in – music is such an integral part of the soundscape that it’s almost another character.
The Girlfriends, for instance, is all but narrated by a choir of female voices, whose breathy vocals almost seem to comment on events. And in Deliver Us From Ervil a sinister score of guitars, percussion, saxophones add a sense of dissonance and tension, like a 1940s noir movie that’s drifted from Southern California to the wilds of Mexico, amidst the sage brush and sand. And think how powerful a theme tune can be: the upbeat, whomp-heavy number that signals the long-running always-on Freakonomics, say, or the cool, immediately recognisable title music for the New York Times The Daily, which turns TV news cliches on its head.
But how do you actually make that come together? Is composing for audio like composing for film or TV, or are there crucial differences? And how do musicians, producers and editors work together to shape the finished product? Who better to ask than the brilliant musicians behind Girlfriends and Ervil? A few weeks ago, Luisa Gerstein, Julian Lynch and I got together to discuss all this and more. Slip on your headphones and listen in ...
Andrew Dickson: Luisa, how did you first get involved with The Girlfriends? Had you worked in podcasting before?
Luisa Gerstein: Not really! I haven't got a huge amount of experience in making music for podcast or film or other forms. Primarily I'm a musician and arranger. I set up [the female vocal group] Deep Throat, and I play in bands and stuff. [Novel producer] Anna Sinfield came to me because she wanted something that centered around female vocals. It chimed with what she wanted to do with the podcast, which of course focuses on female stories. I think we started talking about six months before the show was finally released.
AD: How about you, Julian? How did you get involved in Deliver Me From Ervil? Was podcasting also a fairly new space for you?
Julian Lynch: Fairly similar, actually, to a lot of what Louisa said. My background also is in performing: I play in a band and make solo music. I have done sporadic score work. But I was in touch with [lead producer] David Waters from a long time ago because he’d wanted to use some of my prerecorded music, and then in 2020 we collaborated on a podcast called Witnessed: Borderlands. And then more recently we started talking about Ervil and he was looking for someone to score it. I read about the story, this weird tale of Mormon fundamentalists in the desert and the crime wave they unleashed, and it was immediately fascinating to me. He sent me some edited audio material, and we went from there.
AD: Luisa, how do you work with the audio, once you start composing? Are you getting raw cuts of interviews and things like that, or do you work on your own? How do you start to build up the various elements you need?
LG: Before anything else, Anna and I had a conversation about instrumentation. Anna has a background in music, so was very direct in what she thought would work well. We settled on keeping it narrow: a bit of piano, some drums, voices, strings. At first I didn't have that much audio to work with, but what I did have was a cue sheet of very specific prompts and themes they wanted to work towards. I think there were about 10 or so, a mixture of characters in the story and particular dramatic moments. I just started to work through them. I tend to sit at a piano and fiddle around. I can't really think my way into anything: I have to just start playing something and hopefully it feels kind of good. I’m thinking, OK, what's the emotion? What's the function? What’s the right pace and feel?
JL: It was pretty similar for me. Nick Alexander, who’s Novel’s senior sound designer, was on the emails with David and I. They shared some reference points or tracks, things that had the right feel or vibe. So I went from there, composed a little, offered initial ideas, got some feedback, and it went on like that. We also figured out a sonic palette that was quite limited, as it happens; that’s an aesthetic that really appeals to me, and it also seemed to suit Ervil too. What made it slightly crazy is that I have a 2-year-old son, and at the time when I was doing this project he was even younger than that. So I was spending a lot of time with him with a guitar. working stuff out in front of him.
AD: Amazing. You were taking notes on this podcast about a Mormon death cult from your infant son?
JL: Yeah. It does sound a bit crazy when you put it like that. But in fact I was sleep-deprived for so long that I was kind of psychotic anyway, so maybe it helped! But seriously, I found a lot of inspiration in late-70s, early-80s ECM jazz records. That music had the right vibe: slightly spectral, minimalist, eerie. It’s not too on-the-nose, it’s just quite unsettling. That was what we wanted.
LG: I had no idea about your son before we began this conversation, Julian, but I was going to say: strong relate! I've got a one-and-a-half-year-old and I'll play her all sorts of things. I’m not sure I was playing her stuff for The Girlfriends directly, but she’d hear it while I was working, for sure.
AD: Luisa, one of the things that’s really striking about The Girlfriends score is that the soundscape is so much about female voices, partly because it’s the Deep Throat choir we hear. They almost sound like characters at some points, like a Greek chorus, observing the action.
LG: It's the risk of bringing the human voice into music that you’re using for scoring. You’re immediately asking, “Who is that? How do they relate to the story?” That’s partly why we were quite sparing with words and lyrics. But there’s an intentional aspect to using a chorus of voices: you do feel they are people. And actually recording with the singers was great. By the time we got into the studio, we had a lot of the music composed, but we also made time to play around and make things together, explore different sounds and textures. And a lot of that ended up coming into the podcast.
AD: So some of it was improvised?
LG: Yeah. I If we'd have had infinite time, I would've loved to have done more of that. But we certainly did explore it as much as we could. There’s a composer I really admire, Joan La Barbara, who wrote exercises for listening and meditation. One is a game called Cathedral, where you improvise together by following different notes from within the group. We ended up using a lot of that material.
AD: Once you’ve got that kind of music, whether it’s live or made electronically, how does it work with the final podcast mix? Are you simply sending in tracks to the producers, or weaving them in yourself?
JL: Often you’re providing what we call stems, which is to say tracks that can be broken down into their constituent parts for editing or remixing. That gives the producers something to work with, because they need to edit and time things, layering them into other elements such as interviews and voiceover or effects. A lot of the late-stage stuff is down to the producers, working with your material. It’s a really active collaboration.
AD: So it’s not even finished tracks? It’s more like giving them a palette of musical options?
JL: Well, in the mixes I sent I’m generally giving some indication of how I think something should sound. But ultimately it’s theirs to use how they want.
AD: What was the process like overall, Luisa? Was there anything that surprised you?
LG: Actually it’s really liberating to create music that has a specific purpose rather than being simply a means of self-expression, if that makes sense. There were things that I played with that I wouldn't necessarily ordinarily work with. And exactly as Julian said, it’s fun to deliver a load of stems and say to someone, “Do whatever you want.” They heard things in the music that I hadn’t heard, explored new options. It’s great to be able to work in a different way.
JL: That resonates with me too. It's probably commonplace for composers whose primary kind of work is this sort of thing. But for me it was fascinating. You’re contributing to something much bigger. And for me sometimes having restrictions imposed on the process can be a source of inspiration in itself. I really enjoyed it.
AD: Last question for now – there sometimes seems to be a debate in podcastland about whether you need a catchy theme tune, something that’s instantly recognisable and which sends people away humming. What’s your take on that? Julian first.
JL: I think it depends on the podcast, really. Um, I think some can make it work really nicely without one. But Ervil kind of did: it wasn’t catchy in a traditional sense, but there was a theme. And it felt like we needed a theme, because it would give listeners a sense of the tone and what kind of experience they were going to have. But it doesn’t need to be corny.
AD: Luisa? Where do you stand on that question? Are you theme? Anti-theme?
LG: I'm probably quite biased because I respond to music quite strongly. But I think it partly depends on listeners: what can the music add to the storytelling? What will resonate? That’s ultimately the question you’ve got to ask, no matter what the music is.
The Girlfriends is available to listen now, and the music from the series is available as a digital album