SELLEST OSAST
Liz Pelly is our foremost journalist/critic on the Spotify beat. Her byline has appeared at the Baffler, Guardian, NPR, and many other outlets. She is also an adjunct instructor at NYU Tisch School of the Arts. Liz is also been making the media rounds lately, talking about her new book Mood Machine: The Rise of Spotify and the Costs of the Perfect Playlist (One Signal Publishers).
The book is both a history of Spotify and an argument that Spotify is not, in fact, a music company, but rather an advertising company focused on manipulating user behavior to maximize time on platform. As a consequence, Spotify not only pushes musical aesthetics towards banal, “lean-back listening,” it also makes musicians themselves expendable: replaceable by ghost musicians, AI slop, and behavioral algorithms that keep people just barely engaged at the lowest cost. In this show, Liz details how platforms shape listening and music making alike. We also discuss the tension between frictionless music consumption and meaningful cultural engagement.
And remember, there’s an extended version of this interview which features a bunch of bonus material including a listener question, a deep dive into Liz’s reporting methods, and the backstory of how she got into journalism and got a major book deal, plus her book and music recommendations. It’s available to our Patrons for a mere $3 a month. Sign up at Patreon.com/phantompower.
Liz Pelly: [00:00:00] When I hear something like the founder of an AI company saying “Making music is too hard.
People don’t want to learn how to play instruments,” or even this idea that a streaming platform should help people reduce cognitive work. It’s like, that essentially means we should help people not have to think. And I think that, you know,
Mack Hagood: Yeah.
Liz Pelly: As critics, what we do is encourage people to think, you know, thinking and making decisions is an important part of processing life in the world and information and culture and figuring out how you actually feel about someone’s art.
Introduction:
This is Phantom Power.
Mack Hagood: Welcome to another episode of Phantom [00:01:00] Power, a podcast about sound. I’m Mack Hagood. My guest today is journalist Liz Pelly, someone I’ve been reading avidly and having my students read for almost a decade now. Pelly is our foremost journalist and critic on the Spotify beat. Her byline has appeared in the Baffler, the Guardian, NPR, and many other outlets.
She’s also an adjunct instructor at NYU’s Tisch School for the Arts. Liz has been making the media rounds lately, talking about her new book Mood Machine, the Rise of Spotify and the Cost of the Perfect Playlist, out on One Signal Publishers, an imprint of Simon & Schuster. The book is both a history of Spotify and an argument that Spotify is not in fact a music company, but rather an advertising company focused on manipulating user behavior to maximize time on platform. As a consequence, Spotify not only pushes musical aesthetics towards banal, lean back [00:02:00] listening, it also makes musicians themselves expendable, replaceable by ghost musicians, AI slop, and behavioral algorithms that keep people just barely engaged at the lowest cost.
I am super excited to have Liz on the show and get into the weeds of how platforms shape listening and music making alike.
And remember, there’s an extended version of this interview that features a bunch of bonus material, including a listener question, a deep dive into Liz’s reporting methods, and the backstory of how she got into journalism and got a major book deal. We’ll also have her book and music recommendations.
It’s available to our patrons for a mere $3 a month. Sign up at patreon. com /phantom power. All right, let’s get to it.
All right. Liz, welcome.
Liz Pelly: Hey, thank you so much for having me.
Mack Hagood: So I [00:03:00] thought we could start off by talking about the title of your book. For those of you folks out there who aren’t familiar with your years of research on Spotify and your journalistic pieces on it, why name a book about Spotify, Mood Machine?
Liz Pelly: That’s a great question. I think when I first started thinking about the book, I was thinking about it in two sections. Actually, the first book proposal that I wrote was a proposal for two books. One was going to be about the impact of the streaming economy on listening. And one was going to be about the impact on artists. I quickly realized that it made much more sense to just write one book, but I shifted to this idea of writing a book in two parts where the first part was going to be about how streaming had reshaped listening and the second part was going to be about the material impact on musicians. That structure didn’t quite hold, by the time I got to the final table of contents, things shifted a little bit, but [00:04:00] when I was thinking about, originally, when I was thinking about mood, to me, that word sort of evoked the way that streaming has impacted listening and the shift from albums to playlists, the championing of playlists that are mood playlists or connected to emotions in some way.
And then, when I thought about the word machine, I really thought about the relationship between, like labor and the music industry or labor and capital even. You know, I was thinking about the way
in which the music industry squeezes musicians more and more under this model and, just the trajectory of the music business.
Obviously, a lot of this book also covers the shift from the playlist era into the era of streaming curation, being more driven by machine learning and algorithms and personalization. So there’s surely like a point of this all, or there’s a way of interpreting the title that also evokes that, but yeah, it’s really interesting.
Like, you know, when I thought of it, I was thinking a lot about the relationship between musicians and this model.
Mack Hagood: Yeah. Well, I love the title and I definitely want to dig into a lot of what you just mentioned there, but I was thinking maybe we could go back to 2007. Is that when Spotify launched?
Liz Pelly: Technically, the company was created in 2006 and it launched in its first markets in 2008.
Mack Hagood: Okay. So back then, what was the state of the music industry? What was Spotify, the alleged solution to? What was the problem?
Liz Pelly: So this is still the era where the music industry was trying to figure out how to recover from the impact of file sharing. Around 1999 to 2001 was the time when Napster was, , according to the people in the music business, wreaking havoc on the global recorded music industry. By the time Spotify came along, the music business had already spent years and years trying to figure out solutions that would work in the digital music era, you know, also taking individual fans to court over file sharing.
The music business had tried to launch some of its own streaming services as an alternative to piracy. There’s a whole sort of, fumbled strategy, on the behalf of the major record labels and the mainstream music business trying to figure out how to solve this problem. It’s really interesting because the impacts of file sharing, I think , were and continue to be felt differently by different musicians and different corners of music.
But something else that was also going on was in the United States by the mid 2000s, the iTunes library model had taken off or taken hold a little bit more here than in other parts of [00:07:00] the world. Spotify was founded in Stockholm, Sweden by two men with backgrounds in the advertising industry, in Sweden, even by 2005, 2006, while the industry was somewhat successfully starting to figure out ways of selling digital music to consumers, piracy was stronger there.
Around that time, you still had the Pirate Bay, which is a really big cultural force in Sweden. Sweden had a pirate party. There’s a politicized element of music piracy in Sweden. And yeah, I think the founders of Spotify, Daniel Ek and Martin Lorentzon just saw an opportunity to build on their backgrounds and add tech to try to build a product that would appeal to both consumers that loved music piracy and the music industry that was like, you know, floundering.
Mack Hagood: You know, it’s really fascinating to think about the Sweden piece, why it pops up there specifically, there were so many American companies that [00:08:00] were trying to take a crack at this, I mean, back in the day, if you had forced me to guess, like this concept of just like streaming music, , who would, , be able to do it.
Well, I probably would have guessed something like MySpace because there were, that’s where all the indie bands were. And it just seemed like something like band camp could have happened there. But Sweden winds up being the hotbed of it. And as you say, something to do with piracy, the role piracy plays there.
And it seemed from what I gleaned from your book that partly the music labels were willing to let Sweden experiment, because they already saw it as a lost cause in terms of everyone was just pirating music. Is that right?
Liz Pelly: Yeah, according to people who were in Sweden at the time, the music industry had started to see Sweden as a lost market, to use the quote, the phrase that came up in some of my interviews. , specifically the [00:09:00] free tier, you know, that was a period of time when the music business, , and the major labels who, , control the rights to so much of what we think of when we think of the history of recorded music, which is its own issue to unpack the major labels were really allergic to anything that had the word free in it that involved, the concept of giving music away for free or providing free access to music. In Sweden, though, which the music business saw as a sort of lost market, they were more willing to take a chance on something with a free tier.
Of course, the original Spotify model, like the original model that Daniel Ek and Martin Lorentzon came up with, was a product that would, It was exclusively a way of delivering content for free, supported by ads. So that was the original business model. And originally they weren’t even sure if it would just be music.
The original idea was some sort of media delivery platform that would provide access to a library of content for free, funded by [00:10:00] advertising. And Even if the, , the music business was more likely to, or more willing to allow for experimenting with this sort of free tier in Sweden.
Mack Hagood: Yeah. And I was really struck in your book by what you said a minute ago, which is that these guys weren’t interested in music per se. They were interested in some kind of streaming media content to run ads against. And so, they were an ad selling business, not a music selling business from the get go.
Liz Pelly: That was their background. Daniel Ek’s background was that he worked at an SEO firm after he graduated from high school. You know, in high school, he had applied for a job at Google and they said, try again when you have a college degree. So he built his own open source search engine.
And then he got a job at an SEO firm. And then by the time he was 22, he was the CTO of this cartoon doll gaming website for teen girls. And then it was Martin Lorentzon [00:11:00] background was in ad tech. In selling automated advertisements.
Mack Hagood: I mean, I think this is really interesting because I think on the one hand, because They weren’t music people. They didn’t have certain attachments to music that probably allowed them to innovate in the way that they did. Right. But on the other hand, a business model that is downstream from running ads rather than downstream from how we work with music as music, You can imagine, a lot of cultural and technological effects of that.
Liz Pelly: Totally. One of the things I write in the introduction is, you know, there’s so many different ways to understand them. The story of Spotify, the story of the streaming era. But one of them, I think, is the story of these advertising men bringing the logic of their industry to music in new ways.
And I feel like that comes up time and again. I mean, in some ways I think it could be [00:12:00] boiled down to purely just the reality that now what we have is a model that views music so much as just , a cheap content source that fuels, , this, a broader, , advertising or subscription model.
and I think one of the things that I tried to trace in the book, there’s multiple different threads running through the book, but one is sort of investigating, Different efforts on the part of Spotify to lower the cost of content, , or to, , lower the amount of money that they have to spend on the audio that circulates on their platform and allows them to sell ads and sell subscriptions.
Mack Hagood: Yeah. Yeah. One interesting sort of leap that I think the company took is you talk about how, in 2012, the company had, , I think 5 million paid subscribers. And in 2017, , they had more than 60 million [00:13:00] and they were , basically in 2012, this company that had a lot of major investors, but wasn’t making any money. And, , I’m really interested in how you treat sort of that five year span, , and think about what shifts happened in the company and then what the cultural effects of those shifts have been. And I’m. Particularly interested in this because, , you were asking me online or offline what my research was about.
And I do this research into how people use sound to control their effect on things like noise canceling headphones, white noise machines, nature, sound apps. And what I was noticing around 2013 or 2014, was that it seemed to me that the streaming platforms were starting to treat music. In a very similar way, and that I was noticing that instead of organizing around genre, like they were in the music, in the record stores that I grew up going to, [00:14:00] they were really, pitching music playlists in terms of like the mood or the function that it has.
So this is music to work out by, this is music to do this, to have a dinner party with or whatever. Right. And I remember mentioning this to my friends, like in popular music studies. And I was like, and saying, this is the, they’re changing the way music is purchased basically and organized and they’re like, ah, that seems like a bit of a stretch to me. They’re like, you know, genre and cultural capital and taste making is still like the coin of the realm in popular music. , you got to remember all the editorial playlists, like rap caviar and pitchforks playlist. And they were kind of like talking me down from this. And I was like, Oh, okay.
you guys. Maybe you know better than I do. But then when I saw your articles come out in the Baffler, like the first one that came out, I don’t know if that one was called like the problem with Muzak or something, I was so psyched because [00:15:00] you were taking this premise seriously, and examining it really closely.
And so I’ve just been a super big fan of your work ever since then. And what I really like to focus on in this interview is, , we know, At least I think we know that Spotify in a lot of ways has been bad for musicians, but I would really like to talk a lot about this user experience and the ways that music’s capacities for affecting mood or shifting attention or managing attention or feelings seems to have gained precedence over music’s capacities as art or as the object of our attention that the, you know, active attention. So going back to 2012, what. Started to happen at Spotify that’s changed, that [00:16:00] gave them the insight to move in this direction.
Liz Pelly: So something that I spent a while doing while trying to trace the specific moments of when strategy changed. I mean, I did a lot of interviews for this book. It’s interesting hearing you talk about it. Because I started covering Spotify in 2016. So it was, you know, kind of a little bit after. This moment, and a lot of my initial research was interviewing musicians and interviewing people who’ve worked at independent record labels.
And by the time I started writing about streaming and Spotify musicians and people who worked at independent labels, it had become clear to them that these types of mood playlists were having such a more dominant role in how people were presented music. Especially things like chill playlists or workout playlists.
Things like functional music. but trying to trace, you know, earlier than the 2016, 2017 era, [00:17:00] I went and interviewed people who were close to the company at the time. And I also spent a lot of time on the internet archive, looking at the front page of Spotify and trying to trace when things changed and when the company’s presentation of its products specifically changed.
Mack Hagood: That is, I mean, can I just pause there? Like for, we have a lot of people who do research in this who listened to this podcast. I mean, that is such a great mode of doing research is just going through the internet archive and looking at those different front pages. I love that
Liz Pelly: Thank you. And you know, it was super interesting because there was a specific moment. It was around December 2012, where before that moment, the front page of Spotify had been really simple. It was like a white background and green and black and gray text. And it was all about explaining how easy and simple and frictionless it was to subscribe to Spotify, or to access [00:18:00] Spotify for free and get access to a world of music.
Like this was sort of the way in which it was pitched. It was instant, simple, and free. You know, this product gives you easy access to all the music you could ever imagine. And then there was this really specific moment where it really shifted and it was less about, looking like kind of a tech company website.
And all of a sudden it was this montage of hazy images, like moody, sort of washed over like, you know, someone, driving with a sun flare and there was actually a song that auto played in the background. It was like this really kind of moody generic, like folk song playing in the background. So I would auto play when you went to the website
Mack Hagood: Really?
Liz Pelly: Yeah, it’s really interesting.
And there’s like, you know, sun flares, people driving, people kind of like hanging out with their friends, with their loved ones, like in a hammock together. And it was really all about showing you these distinct moments in your life where you might [00:19:00] be listening to music or might be listening to Spotify.
In my book, I described the photos as like stock photos in a picture frame, like waiting to be replaced by real life or something like that.
Mack Hagood: Yeah. Yeah. It’s like the early Instagram aesthetic.
Liz Pelly: Exactly, definitely. Like the early Instagram filter aesthetic, or even like, it sort of looked like images you would see on Tumblr or Pinterest at the time or something like that. So to me, it became clear that something strategically had happened at that moment, where they had decided that, in order to reach more people, They had to think of themselves more as sort of pitching a lifestyle or pitching different ways of users relating to what they were offering.
And in one of the interviews that I did, it came up that around this time, Spotify had commissioned a research agency to do a study on its product and its own users, to figure out how people were actually using [00:20:00] Spotify. This was also about , just over a year after they launched in the United States, where there was not only more competition than in, , other markets in terms of, selling the specific idea of how to relate to digital music, because, there were other streaming services in the United States and there also was,, more people used the iTunes store.
More people listen to mp3s and buy digital music. So they were kind of competing, also with Pandora, Rhapsody. And according to this person who was close to the company at the time, who told me about this research study that they commissioned, apparently they Found out through this research that a vast majority of their listeners were coming to the platform for more of a lean back experience, , and that they started to think of what they were offering less as, access to this world of music, access to this fully stocked, , iTunes library, but streams through the Internet, maybe similar to the experience , users [00:21:00] had come to expect in the post file sharing era, , there’s this interview that I referenced early on in the book, as a conference appearance by Andres Ehn who is the first, Spotify’s first employee.
And he talks about how in the early days, they didn’t see their competition as other music services, but they saw their competition as the pirate Bay and how, you know, it seems like what they, talk about how, what they were trying to offer initially was a product that was better than piracy, not competing with piracy, but better than piracy.
So they were really thinking of the experience of the file sharing music listener, as the person they were trying to win over. And it seems like after there was this specific moment where they realized that actually , their new target listener was becoming more of like the Pandora listener or the person who is coming to the service and looking for, you know, a button that they can press and listen to a feed of music organized around, a specific, theme, or who maybe needed [00:22:00] more guidance in terms of what to listen to.
So I think it was distinctly related to having to grow in the United States. Even in the press at the time, they talked about how they were embracing this moods and moments strategy as part of needing to grow beyond early adopters and reach a more mainstream audience, which I think is pretty interesting.
Mack Hagood: Yeah, and there was definitely something in the air around then because I remember Beats Music, which eventually became Apple Music, had something called The Sentence. I don’t know if you ever came across this, but it was this really bizarre thing where there was like a sent-, a literal sentence, but it had pull down menus and it was like, I’m, Blank and you could pull down like hanging with my homies or like chilling with my pets or whatever. And I want to, you know, feel like, and it ended up like lifted or like, they’re just like all of these [00:23:00] weird, you know, adjectives for the emotional feeling that you wanted to have at that moment. And then you would just fill this sentence with things from the pull down menu and then it would just generate, you know, an on the fly radio station for you at that moment. I don’t really know how that happened, how it worked in the background, but there was, I’m not sure precisely what year that was. That might’ve been, it was probably after what Spotify did, but I’m not, I’m really not a hundred percent sure, there was definitely this kind of strategy in the air.
Liz Pelly: Yeah, there’s also Songza, which was another company that existed at the same time. That was doing sort of like concierge type mood playlist Recommendations and they were a company that got mentioned in the press at the time, you know So following this shift, Spotify bought this company Tunigo, which was at first a third party app on its platform for what they called ready made playlists that [00:24:00] were themed to different moods and moments and .
They were compared a lot to Songza in the music business press at the time.
Now, at the same time that this is happening, it’s not like genre based playlists have gone away. There are the Rap Caviars and the Pitchfork curated playlists and things of that ilk. One thing I was really interested to learn about was that there was almost like a farm league of playlists with the smaller playlists that they would, maybe A B test songs against one another in these smaller playlists and see things like skip rates, that a song could sort of… through these different metrics work its way up from a smaller playlist to a more popular playlist, , and I guess, between the mood based and the human tested curated playlists, I guess you could say, like, “What’s wrong with that?” I [00:25:00] think Daniel Ek would say this is democratic form of programming, right? Like, like we’re letting people’s stated choices and, or, at least the choices that their behavior tells us through examining the data, choose what’s popular. So, what’s wrong with that?
I think that narrative around the democratic, meritocratic data driven playlist ladder climb, where they would say that, the music that ends up on these really popular playlists is informed by user data. We’re looking at putting songs on these smaller feeder playlists and testing them and seeing which ones react.
And if it reacts, we’ll move it up. In some ways, , I also think that was sort of like a narrative that they sold to the independent music world to convince independent labels and independent artists that this was a democratic system when in reality, there were so many other factors that would influence whether or not [00:26:00] a song ended up on a popular playlist and, to this day, talking to people in the music world, it seems that there’s no better way to get onto a Spotify playlist than to just simply know one of the people who works there.
Or to have some sort of direct relationship due to a marketing team relationship, the major labels.
Mack Hagood: So there’s still some old school gatekeeping going?
Liz Pelly: Absolutely. You know, like the very first article that I wrote about streaming was about the privileged relationship that the major labels had with playlist curators. And at the time I interviewed a major label, , someone who worked in digital marketing for a major label who talked about how it was just a different level of access, how major labels had dedicated reps at these companies and dedicated employees at their labels who would take meetings, send over Excel spreadsheets of priorities, those phone calls. Maybe not like guaranteed placement because of some marketing relationship, but a different level of access.
And I think even now, you know, [00:27:00] musicians are told if you want to get onto Spotify playlist, there’s this form that you fill out and you have to pitch your songs two to three weeks in advance and put the proper tags and a description and how you’re trying to target this. It’s like, if you are an artist on a major label especially if you are an in-demand artist on a major label, like you’re not filling out the Google form, you’re not filling out the form on Spotify for artists website. You know, there’s a different type of relationship there. So that is, even to this day, I think it continues to be interesting to remember.
Mack Hagood: So in fact, it’s not as democratic or demotic equity lead us to believe. Let’s just say it was, would that solve the problem? would that, make it
Liz Pelly: Well, I guess it depends, the context that you’re discussing, I think, because something also that I think in the streaming era, there’s, this idea that [00:28:00] these playlists are helping people listen to more music than they would be listening to otherwise that, you know, maybe you don’t, know what you want to listen to, but putting on like a morning commute playlist could introduce you to 50 new artists that you’d never heard of before.
And sometimes you’ll hear people say things like, “I listened to 6,000 new musicians last year. How could you say I’m not discovering new music?” And I think something to remember is the way in which coming to a platform and listening to these sort of mood and moment curated playlists encourages this kind of more passive relationship with what you’re listening to. More functional relationship with what you’re listening to, and it’s like how, you know, much of a connection is actually being made between the listener and these 6,000 artists that you might have listened to through listening to Spotify curated playlists or niche mixes, for example. [00:29:00] And you know, in the trace the history of music being used as a tool of mood stabilization. And I think that the idea that this kind of way of engaging with music would be a means to discovery is almost funny when you put it in the context of other ways that music has been used as a means of mood stabilization throughout history.
Mack Hagood: Yeah. Yeah. And you do talk about things like Muzak the old BBC radio show, the music to work by and that sort of, affective control that music can offer. you talk a little bit about. The calculation that Spotify made, like there was a, there seems to be a moment when they realized that if we can make the experience of, listening to music, like utterly frictionless through the interface and perhaps make [00:30:00] it a little less more frictionless aesthetically, where it’s not quite as demanding of your attention, we can maximize the number of hours that people are using our product, which is what we really want, because either, either they’re going to integrate our product throughout their day and they’re going to paying for their subscription, or they’re going to hear as many ads as possible.
And that’s going to help us that way. Right. So, I think the term that you use for this, you mentioned earlier is “lean back listening”. Can you maybe really delve into that a little bit more?
Liz Pelly: I think that one of the things in that part of the book is trying to sort of look at what it means for a streaming service to sort of, champion this way of relating to music and what some of the consequences of it might be for listeners and for musicians.
You know, it was really interesting talking to some former employees.
And I had [00:31:00] this one conversation where someone said, you know, some of these bigger sweeping changes that have happened in the history of Spotify or on streaming platforms, these things that in retrospect might seem sort of like nefarious or like someone is trying to control listeners.
A lot of times it’s people being tasked with moving a metric and their job is just to kind of look and say, well, when we put this playlist on the front page, , , people clicked on it and the listening sessions were healthy. So let’s just keep putting this on the front page, that kind of thing.
Someone else explained one of the early goals of the playlist ecosystem as trying to reduce the cognitive work that the user had to do when they opened the platform, which obviously would also, for a user who may not know what they want to listen to would potentially lead to helping someone know what to click on, what they want to listen to, extending the [00:32:00] amount of time that they spend on the platform or extending a session length.
It’s like these things that may be in the moment to the people who work there don’t seem like a big deal. are planting the seeds for these bigger cultural shifts where when you have an environment where all you’re doing is following the data and trying to optimize for this , frictionless user experience, to grow time on the platform, to grow session length, to in some ways, help shape user behavior around your own product. These things that might seem not that consequential. In the moment, add up and do you end up, I think, having a cultural impact.
Mack Hagood: Absolutely. And I’m so glad you said that because we tend to always find villains around every corner. In my own research into these kinds of sound technologies that help us have affective control, I’ve really liked every single person I’ve met working in this space who are developing new forms of noise [00:33:00] cancellation or who developed, Practice of listening to artificial nature sounds, you know, it’s like I think cumulatively we have a social problem in which we’re we’ve become accustomed to the idea that are listening is something that’s supposed to be completely managed and we’re not supposed to experience anything we don’t want to hear. And I think that has major, bad effects for public discourse and interpersonal discourse and a lot of things. But the individuals who are working on these technologies are just trying to solve a particular problem generally. Like how do we eliminate noises that bother people?
And you know, that it’s not… there’s no evil intent there.
Liz Pelly: Right.
Mack Hagood: Another thing that struck me about what you just said. It was the real emphasis on things being frictionless, which again, if you’re designing a platform that totally makes sense, friction is [00:34:00] good, right? Friction makes us stronger.
Like, you know, if people, whatever, lift weights, like it’s the resistance that allows you to grow stronger. And the frictions that are involved in pursuing something are a big part of its aesthetic. Enjoyment and its meaning, right? the friction of going all the way to the record store and investing that time and flipping through stacks of stuff you’re not interested in and finding that one gem, that’s part of the enjoyment and it, calls to mind. this guy who designed one of the AI platforms that just spit out music for you. And he recently said in an interview that making music is too hard and musicians don’t enjoy it. And you have to do all these hours of practicing and you might not ever be good. And it’s like, as a guitar player, who [00:35:00] I’ve been playing since I was like 15 years old, like I’m 55 now. and I’m still not great. Right. Like, and, but it’s that struggle to get better. That is like part of the pleasure of playing. I’m like, this guy is totally missing the point of what musical experience is all about.
That was a bit of a ramble, but, I mean, this dichotomy between making information flow as frictionlessly as possible seems to be at odds with giving culture meaning and having a rich experience with it.
Liz Pelly: For sure. And I also think about it from another perspective, which is thinking about the role music critics have historically played in recommending and contextualizing music. And, I think about this way of recommending music in the most frictionless way possible and this idea, when you, when I hear [00:36:00] something like, “The founder of an AI company saying making music is too hard”
People don’t want to learn how to play instruments or even, this idea that a streaming platform should help people reduce cognitive work. It’s like, That essentially means we should help people not have to think. And I think that, you know, like critics, what we do is encourage people to think, you know, thinking and making decisions is an important part of processing life in the world and information and culture and figuring out how you actually feel about someone’s art.
As critics, we want to encourage people to think. And that’s why I describe this pursuit of a frictionless experience, this idea that people don’t want to make decisions, this idea of following the data and just being pushing this really data driven experience of receiving music.
It’s like it’s not only not fulfilling [00:37:00] the role that critics, I mean, it’s obviously not fulfilling the role that critics have played historically in presenting into contextualizing music, but I also think it’s actively anti critical because it’s telling people, it’s encouraging people not to think and encouraging people to not think is, Extremely anti critical, anti intellectual.
It’s like encouraging people to just trust the data. Just, you know, yeah, the music that we’re showing you, we’re showing you because the data has told us that you will like it. So you should just believe us and you should just. listen to it and like it, ,
Mack Hagood: Yeah, no, that’s right. And it’s encouraging us not to think and also sort of encouraging us not to feel, , you know, because I remember, gosh, it’s been probably 10 years. Now, I remember asking my friend, music journalist, Eric Harvey about things like, “Eric, what’s going on with hip hop?”
Like , it all sounds so sad and [00:38:00] blasé and kind of mellow. Like it’s kind of aesthetically beautiful now, but it also is so dispassionate and I was like, what’s happening with that? And I think your journalism has really helped me connect the dots between the incentives of the platform and the kinds of music that succeeded.
And I don’t know if you’ve noticed that about hip hop in particular, but I mean, You’ve certainly talked about the aesthetic of chill. So, can we maybe talk about what kinds of aesthetics seem to get privileged in this era of lean back listening?
Liz Pelly: Well, I could say I also do think that a similar thing has happened, in the musical realm of like indie rock too, you know, like, they’re even talking to people who run indie rock labels for my book. There is this idea that music that was softer or that had softer vocal deliveries or more , emotionally, not flattened, [00:39:00] like less adventurous in terms of dynamic range, or something is the kind of music that does well in the streaming era.
And, you know, it’s not to say that I think that’s the only type of music that people are making, but I think for people who aren’t making this sort of music like whispery, more chill, more soft music. It’s harder to break through. And I think it does encourage a general softening around the edges.
There’s a chapter of the book called Streambait Pop, which is an extended take on this essay that I wrote in 2018 for the Baffler called Streambait Pop. They’re sort of looking more at this, in terms of the way in which the rise of mood playlists and playlists determined by emotions and affect were influencing the sound of music.
Liz Pelly: For that piece, I talked to A pop songwriter who was talking about how it had been become common at the time to go into record, writing sessions and hear people say that they wanted to make, a Spotify song or a song specifically targeting specific [00:40:00] Spotify playlists. how there was this, you know, specific sounds that he felt like had emerged that time.
And that was also around the time that a musician like Billie Eilish became really popular. It was this kind of whispery sort of pop music that could do well on these sort of like chill vibe or sad vibe playlists and it was interesting, going back to that piece in 2022/2023 when I was working on the book.
I think by that point it became, really clear to me that it’s not just that streaming services, are shaping the sound of pop music, but there is a sort of like broader history that needs to be there to contextualize it, that, mediums for recording and releasing music and the technology that musicians and artists use to create their work have always impacted the sounds of, their music.
Music and yeah, I know this is more of like an academics oriented podcast, but before I started writing my book, I read “Capturing Sound: How Technology Has Canged Music” by Mark [00:41:00] Katz. And that was really instructive and sort of like understanding this like broader history there.
So that is referenced in my new updated version of Streambait pop in the book. but also, you know, it was interesting. Looking at these, effects of the playlist environment on not just pop music, but other sorts of genres as well. And yeah, speaking with people in other corners of music or operating on smaller scales about the impact of this as well.
And I also took that chapter for the new expanded version of it. Talk to some songwriter, a songwriter who had participated in a Spotify songwriting camp, which was, really interesting to, hearing how some of these, aesthetic trends discussed within like the realm of, Spotify core were sort of like true to the kind of music that was like, at least in the case of this one musician, like true to the music that was coming out of those sessions as well.
Mack Hagood: That’s fascinating. And then this songwriting camp, was it actually sponsored by [00:42:00] Spotify?
Liz Pelly: Yeah. Yeah. It was hosted by Spotify.
Mack Hagood: That’s wild. I remember some quotes from Greg Saunier from Deerhoof. Like the sort of, I can’t remember specifically, but he always has so many fiery takes on things, but particularly like this world, I mean, it’s funny in our last episode, we wound up mentioning Deerhoof for some reason. You can imagine Deerhoof in the way it leaps from one genre to another and demands your attention in a lot of interesting ways because it does so much unexpected musical work. Not really being a band that would work as well in the Spotify era.
Liz Pelly: Yeah, I’ve interviewed Greg for the first piece that I wrote for the Baffler back in 2017, and then did another interview with him for my book. And I feel like, well, one, Greg just always has like so [00:43:00] many interesting things to say about the state of the world and even not just music.
But yeah, I think they’re a really interesting band to kind of comment on this trajectory of the way that independent music has been impacted by the streaming era because they’re a band that has been making music really like on their own terms since the nineties and worked with so many of the labels that we think of when we think about the idea of independent music in the popular imagination and they also are an example of a band that like, you know, had a really dedicated fan base before the impacts of the streaming era, emerged.
And I think they can sort of, but I think it’s always interesting. I think that lots of critics have different perspectives on this, on how we should consider the perspective of musicians whose careers span back to before the dawn of the streaming era something I’ll hear a lot is, critics saying, “Well, these artists are just complaining about the streaming era because they were, one of the lucky few who [00:44:00] happened to have careers in the pre streaming era” And it’s always been hard for everyone. So I don’t know. I think it’s actually kind of the opposite.
I think that musicians whose careers span back to the pre streaming era are really well positioned to show. how things have changed and to speak to it. That’s something that is interesting about their perspectives.
Mack Hagood: I agree with that. And I think these digital trends hit differently, generationally. So ultimately if Spotify could get people to tap on moods, it wouldn’t really matter that much what specific songs were found in those playlists. And then that paves the way for two phenomena that you have also written about, which is. Ghost artists and AI generated music. Can you talk about what’s been happening in that regard?
Liz Pelly: Yeah. And I think that one thing that I tried to emphasize in the book is [00:45:00] how, there is this era where Spotify curated editorial playlists, often categorized by mood, had a really big cultural impact. And in some ways, There’s one former employee I interviewed who refers to 2016 to 2019 as the peak playlist era, but right off the bat, correctly points out that in the music business.
Now the top of the discovery funnel has shifted away from these curated playlists made by Spotify editors. And now there’s a lot more influence coming from personalized feeds and algorithmic recommendations and platforms like TikTok and short form video as sort of, illustrating how discovery has really shifted in this next era of the digital music economy.
But something that I try to point out or show is how there have been all of these effects of all of these sort of other impacts of listening being reshaped [00:46:00] around this concept of, you know, click here for happy, click here for sad, click here for chill, click here for upbeat party vibes, and how it’s not just, you know, that these platforms had a role reshaping listening around their own editorial playlists, but also that it paved the way and like helped prime this idea of listening to music, according to these types of buttons that you can press.
And I think that, one of the, long term impacts of that does have to do with like the way in which in some ways it sort of encourages people to have more of a relationship with the playlist they’re listening to or the mood category that they’re they’ve organized they’re listening around than necessarily the artists and it again is like one of those things that could seem subtle but actually you know reorganizing music in a way where it is more about the , the vibe or the playlist category, then necessarily, , having any sort of [00:47:00] connection with the artists that are, , being played does in some ways, I think, set people up for, , being more desensitized to the way in which music made by generative AI could start either filling these playlists or, you know, like you think of something like earlier, you were talking about Suno or these kinds of generative AI music platforms, even something like Endel, which I talk about in the book, or, like these products that are basically just like click here for a vibe and we’ll give you an AI generated feed of music that fits that vibe.
And you have no relationship at all with the people who are creating the music. In some ways I think the general playlist economy, the devaluing of the artists making the music you’re listening to, pointing people towards these, anonymous ghost artists, you know, essentially like low royalty stock music that has started to be over the years, has filled up some of these more leanback playlists. [00:48:00]
Like I think it does have these bigger cultural impacts of just generally devaluing the role of the artist, in and prioritizing the role of, , the tech product or the, playlist name or the shaping a relationship around the product that’s delivering the music and not the music itself.
Mack Hagood: Okay, folks, that’s it for this version of my interview with Liz Pelly. Remember, you can get the full interview and bonus content at patreon.com/phantompower. The Patreon has been growing slowly, but surely. And it’s a great motivator to keep these episodes coming. We also have free membership. So if you just want to join our email list and keep up with what’s happening around here, that’s a great way to do it.
Again, it’s all at patreon.com/phantom power. And that’s it for this episode of Phantom Power. Huge thanks to Liz Pelly for being on the show, see show notes, links, and [00:49:00] more for this episode at phantompod.org. Please remember to subscribe to us and rate and review us on the platform of your choice platforms.
And I want to thank my Miami University assistants, Katelyn Phan, Nisso Sascha, and Lauren Kelly for their help with the show.
Josiah Wolf of the band WHY helped me make my intro soundscape and the one and only Alex Blue, aka Blue the Fifth, did our outro music. Got feedback? Send it to mac at mactrasound.com or hit me up on socials at Mactra. See you next time.
The post How Spotify Dulls the Musical Mind (Liz Pelly) appeared first on Phantom Power.
Learn more about your ad choices. Visit megaphone.fm/adchoices
