“Everything in Between” is about the systems, institutions, and practices that people build, “things” of a sort that sit in between us, between groups of us, between “us” and “them,” and between us and other systems and institutions that seem terribly far away: “the market,” “the state,” the universe, and so on.
Last week’s essay, which introduced my interest in how institutions end, intersected this week with a recent book that I picked up at the suggestion of new acquaintances who live full-time in various niches of the music production and distribution “ecosystem,” mostly in and around independent music: “Mood Machine: The Rise of Spotify and the Costs of the Perfect Playlist,” by Liz Pelly (Simon & Schuster 2025).
The question prompted by the book, borrowed in a way from Don McLean’s “American Pie” (remember: the plane crash that killed Buddy Holly was “the day the music died”), is this: did the music industry “end” in any interesting way, and did we not notice? Either way (yes? no? maybe so? again?), how does exploring the question advance a story about institutions and governance?
Liz Pelly is a music journalist, and the book is the product of a lot (and I mean a lot) of great reporting. Thoughtfully, here and there the book incorporates insights from scholarship. Journalists often struggle to get that part of their work right; Pelly has done a nice, nuanced job there. If you’re interested in the nitty-gritty of how Spotify came to be “Spotify,” as a business, as a platform, and as a force in the lives and careers of musicians and listeners, and what that all adds up to in art and culture, then you should take a look.
My brief summary of the thesis is something along the lines of: by virtue of its heavy reliance on creating and distributing playlists, Spotify encourages, encodes (literally), leverages, and amplifies a shift in the experience of music and sound from “music as Art” (musician and listener in active engagement with each other) to “music as background” (producer and consumer transacting passively through a sociotechnical music “market”). That is Bad, above all for musicians themselves, and for the time, effort, and expertise that they cultivate, deploy, and risk when they make music.
My interest here isn’t in the music as such. I’m a long-time music listener, mostly of the passive variety; I understand how musicians often focus on music as Art, how they often want to engage with audiences (whether live or via recordings) who are really, really into the music, and how they – like a lot of people – want the best of all possible worlds: stable careers, full of opportunity, recognition, and reward, along with the opportunity to do what they really, really love.
But I am, among other things, a copyright lawyer, and in the land of copyright, music is mostly about money. Music is one category of “expression,” among many, and the core principle animating legal “protection” for “expression” is to give the producers of “expression” an opportunity to recoup the costs of producing it while making it accessible and available to audiences, including next generations of producers. Copyright is part of a knowledge- and culture-producing flywheel.
And as music copyright developed over the course of the 20th century, it trafficked in “the song” above all things. Just as “the work of authorship” is the canonical object and artifact of copyright generally, “the song” (the composition, for certain legal, artistic, and cultural purposes; the recording of “the song,” for others) is the canonical object and artifact of music copyright.
Back to Spotify, and to “Mood Machine”:
Has Spotify, and the playlist, and datafication, and all of the other features that make Spotify “Spotify”, ended “the song” as the key unit of cultural production in music?
Are there “songs” on Spotify? Absolutely. Does Spotify succeed by manufacturing and/or rewarding audience demand for “songs”? Do artists succeed on Spotify (or otherwise, or not) by manufacturing or rewarding audience demand for “songs”? Given Pelly’s book, I am not confident that the answers are clearly “yes.” And if we take “songs” out of the commercial and cultural equation, as Spotify may have done, then what roles do composers and performers play in that equation – where for the last 100 years, those human beings have directed their attention significantly (if not exclusively) to “songs”? What remains of the craft of music if that craft does not produce artifacts to be woven into broader cultural experience?
All of that brings to mind a distinction that scholars and practitioners of systems design draw between “stocks” (resources measured at one particular time) and “flows” (resources measured over an interval of time), and a distinction that scholars of common pool resources (CPRs) draw between “resource pools” (such as a forest) and “resource units” (the trees in the forest).
Is it possible that when we observe what seems to be the end of an institution (such as system), we engage – perhaps subconsciously – in a shift of measurement perspective, so that what we once saw as “stock” might turn out to be “flow,” or what we understood as a collection of distinct “resource units” is, instead, a “resource pool”?
If my suggestion about changes in perspective seems right – a big “if” – then Spotify does not signify the “end” of “the song,” let alone of “music.” The song remains the same, in a way, but with a different meaning [“since you’ve been gone”]. I just finished a semester of teaching a law school seminar titled “Copyright and Music,” and I’m wondering whether we ended the syllabus – still focused, as copyright is, on “the song” – just as the legal, economic, and cultural conversation was getting truly new, interesting, and problematic. Was I teaching from a demised point of view?
Is it right or fair to associate such a change in perspective with Spotify’s business models and tactics?
Because music is far from the only institution where changes in perspective - stocks and flows, pools and units - have been at work, through history and in modern life. You know what else is a “resource pool”? The data underlying a GenAI model. Which is to say that resource pools are institutions of a sort, sometimes built well and fairly, sometimes deployed for good, sometimes governed effectively and wisely, and sometimes … not.
We do not always “see” pools and flows for what they are. Seeing them, and trying to understand them, has been a big part of my work over the last 20 years. Despite the interesting and culturally and commercially important questions surrounding music, I will return later to wonder about how the changes in perspective that I see in Pelly’s Spotify case study might be at work as we frame new and different questions about the ends of other institutions, such as colleges and universities.