Home

All Ears

Streaming music services have made record reviewing superfluous. But are we really better off without it?

Like many creative types, writers believe their work is resistant to automation. Their language is too rich and complex to be reduced to data and code. But what if their readers are less interested in complexity than straightforward, direct information?

Automation and artificial intelligence don’t necessarily have to emulate human behavior perfectly in order to profitably approximate the functions humans perform. They don’t have fool users into thinking they are human; they just have to adequately meet a set of practical expectations. Perhaps it is a uniquely human skill to rhapsodize in prose over the achievements of a particular musical composition, but when a set of algorithms can let you choose among songs that are consistently in the general ballpark of your tastes, record reviews seem to lose some of their purpose.

On-demand music streaming has obviated the need for purchasing advice. Why read even a few paragraphs about a song when you can just listen to it?

The modern record review was conceived in the late 1960s, with Rolling Stone, Melody Maker, Creem, and other publications emerging to help listeners sort through the explosion of recorded music at the dawn of the album era, and they have stayed largely the same since. Reviewers functioned as de facto ideologists for the music industry, helping to stimulate demand by rationalizing a boredom with the familiar and championing novelty as an expression of innovation and creativity.

The streaming model would seem to require a different sort of ideology to sustain demand. With streaming services, it costs consumers no more to listen to 100 albums than it does to listen to one album 100 times. Whereas the music press has traditionally hyped the quality of new music to preserve its role as tastemaker, streaming services hype quantity and access to its pseudo-Babelian music libraries.

At their best, reviews provided cultural context that could help imbue certain kinds of music with the power to foster greater social connection. People who read music reviews made up a small percentage of listeners even before digital streaming, but those readers often evangelized for the records they read about. As a result, music reviews have shaped the discourse around music for half a century. But even as they articulate an ecology of cultural influence, such reviews are still ultimately conceived as a buyers’ guide (hence the numerical and star ratings), and the way we buy music has changed. On-demand music streaming has obviated the need for purchasing advice, and as a result, publications like Pitchfork and Spin have ceded influence to the likes of Spotify and YouTube. Why read even a few paragraphs about a song when you can just listen to it?

In its mid-aughts heyday, Pitchfork had kingmaking power in the indie world. Even a bad review could increase album sales or attendance at shows, and a positive one could push a band to the upper echelons overnight. In a 2006 Wired piece by Dave Itzkoff, Broken Social Scene’s Kevin Drew talked about the effect a positive Pitchfork review had on his then-unknown band. “The next tour we went on, we suddenly found ourselves selling out venues. Everyone was coming up to us, saying, ‘We heard about you from Pitchfork.’ It basically opened the door for us. It gave us an audience.” Now, placement on a prime Spotify playlist can do for a band today what a favorable score from Pitchfork did for Arcade Fire and Interpol a decade and a half ago.

With more access to more music than ever — Spotify boasts more than 30 million songs in its library — listeners are still in need of filters. But with streaming services, influential editors at taste-making publications have been replaced by a more diverse set of algorithms that draw on different kinds of data, making the filters at once more personalized and more standardized. Where reviews once served as a proxy for the sentiments and attitudes of a certain milieu — explaining why music might matter as well as what it sounds like — Spotify draws directly on cohort analysis to guide its listeners, as several articles have detailed, but it serves those recommendations without much supplementary context. It overlays social connectivity on music consumption, but deskills it at the same time. It assumes no learning curve; it demands no intricate subcultural navigations to find your way in.

If expanding one’s horizons is simply a matter of clicks, the broader purposes that musical affiliation has often served — not merely the generation of cultural capital but also the sense of oppositional solidarity and tolerance for friction, difficulty, and difference necessary for social resistance — may be in danger of being lost.


Both streaming sites and reviewing sites are in the business of guiding listeners toward new music, but they differ in how they view those listeners. Outlets like Pitchfork conceive of their readers as aspirationally cool, reading to learn which bands their friends will be talking about, so they can talk about them first. It assumes people want to discover new music in part to impress and influence others, and demonstrate how they are ahead of a trend curve. Accordingly, its editors favor music that they perceive to be innovative or indicative of emerging trends. Writers for these publications are telling readers not what to listen to so much as how to seem relevant.

If sites like Pitchfork offer coolness as aspirational, Spotify suggests users are already cool enough. Spotify doesn’t have to prove it is a tastemaker you should follow. Instead, it flatters: Your taste is already great, and here’s some tracks to show how tasteful your listening data has already proved you are. In Spotify, taste requires no effort, no cultural capital, no imaginative synthesis or rationale. It becomes simply empirical, latent in the data and waiting to be extracted by Spotify’s algorithmic analysis.

Spotify doesn’t have to prove it is a tastemaker you should follow. Instead, it flatters: Your taste is already great

Streaming services’ incentives are not to create a sense of aura and scarcity around music and musical taste. They are premised on the principles of abundance and access. Spotify makes its money not by making music appreciation seem demanding, requiring arcane knowledge and background reading, but by making it seem automatic, to keep you listening.

It’s unclear to what extent Spotify can be profitable, but bands certainly aren’t making much money off of it directly. The music press, by contrast, had mutual incentives with the bands they covered, to establish each other’s credibility. When positive coverage in the music press translates to record sales and show attendance, it translates back into a demonstrable sense of a publication’s value to the industry: The co-ascendance of indie rock with independent music blogs in the early 2000s exemplifies this. Streaming services, however, seek mainly to increase plays on their platforms. Hyping specific bands is incidental to their business model (and without a particular niche, it has no allegiance to any genre or scene). Streaming services have no reason to push anyone out of their comfort zones. It hopes instead to prove to us how well it knows those comfort zones and convince us that comfort matters more than taste.

A reviewer can urge you to stick with music even if it’s not immediately accessible or within your typical purview. A playlist offers no such rationale, just the opportunity to skip ahead. A reviewer can explain why a piece of music might matter. An algorithm’s logic will often remain formulaic or opaque. The net effect is that Spotify crystallizes musical taste as it is, for the individual and for the culture at large.

Internally, Spotify measures success for its Discover Weekly playlists in terms of how many users they reach, how much time they spend on the playlist, and how many of those users come back each week. Any tweaks to the algorithm or the balance between code and human curation will be made toward optimizing for these metrics. Whether it furthers musical innovation or pushes boundaries or cultivates listeners’ ability to appreciate “difficult” music — the music that critics might secure capital by championing — is incidental. Instead, streaming services tend to imply there is nothing difficult in discovery and that exposure to music is tantamount to the competency for appreciating it.

The algorithmic modus operandi, in which our revealed preferences dictate what is offered to us, in some ways corresponds to an early internet fantasy that the tyranny of tastemakers could be upended through our broader access to the means of media distribution. Mainstream reviews and marketing could be countered with a more diverse flow of information, which some assumed would lead to a flourishing of niches and niche reviewers. Pop culture would no longer necessarily congeal around its lowest common denominators.

Pitchfork’s prominence reflected that idea, but it was an outlier. As cultural products became easier to access, the spectrum of consumer tastes for the most part narrowed rather than broadened. In The People’s Platform, Astra Taylor notes that “while peer-to-peer networks can be used to share amateur and independent creativity, in practice they are more often used to trade Lady Gaga or Game of Thrones.” In part, this is because consuming pop culture, even when alone, is a social activity, an attempt to belong to the moment. Social media (and the social components of streaming services) up the stakes for this: They let us showcase our consumption for immediate recognition. This makes the idea of being uninfluenced in one’s tastes seem undesirable and implausible as well as impractical. Having “refined” tastes might only make more clear and palpable one’s sense of cultural isolation. Streaming services, meanwhile, deploy the idea of collective participation as a form of personalization, drawing recommendations from what others with similar taste are listening to, offering a sociality without requiring any socializing.

Spotify insists that, despite pressure from bands and labels, it never intentionally puts anyone’s music on anyone’s Discover Weekly playlist. Like other social media platforms, it asserts neutrality, following only the information its users provide, but those playlists are sourced heavily from playlists that are prone to all sorts of outside influence. Zach Fischel, vice president of label relations and business development for independent music distributor MVD Entertainment, says that Spotify has become the primary marketing focus for modern indie labels. Unlike Pitchfork, which is “going to write about what they feel,” he says, Spotify is, in his view, more easily influenced. “With Spotify, you can build a case,” he says. “They are looking at streaming numbers, or SoundCloud, or Bandcamp. They’re going to look toward information you can present, whether it be a support slot on a big tour or a bunch of great press. You can influence them if you have that direct communication.”

Streaming services’ incentives are not to create a sense of aura and scarcity around music and musical taste. They are premised on the principles of abundance and access

If you can generate some momentum, you can get your band on prominent playlists, which will in turn lead to more exposure via Discover Weekly. That in turn leads to greater exposure beyond Spotify. Sirius XM, for example, makes decisions about its rotation based on Spotify numbers. “The Spotify curated playlist is where we see the biggest, fastest spike,” Fischel says. So he instructs his artists to do everything they can to direct listeners to their Spotify pages and drum up follows and listens, hoping to reach the critical mass that will push them to the next level. Bands must serve and promote Spotify if they want its influence. Their music, regardless of what it sounds like, becomes an ad for Spotify subscriptions.


The consumption of complex art has always required privilege. Understanding “difficult” music, for example, often requires not just a specialized education, disposable income and plenty of leisure time, but also a surplus of intellectual capital. Difficult music takes work to consume. For everyone else, there is “escapist” media that lets a consumer “turn their brain off.” A hard day’s work necessitates shallower art consumption, or at least the escapist sort that one can dissociate into for a little while.

In “The Culture Industry: Enlightenment as Mass Entertainment,” Max Horkheimer and Theodor Adorno argue that capitalism produces hollow art to condition people for hollow lives. “Amusement under late capitalism is the prolongation of work,” they write. “It is sought after as an escape from the mechanized work process, and to recruit strength in order to be able to cope with it again.” The choices offered by the marketplace serve not to liberate listeners with freedom of consumption but rather to sort them into more easily manipulated subgroups. “Something is provided for all so that none may escape; the distinctions are emphasized and extended,” they write. “The public is catered for with a hierarchical range of mass-produced products of varying quality, thus advancing the rule of complete quantification.”

Streaming services modernize this rule. The mesmerizing freedom of choice traps us in a cycle of deskilled consumption that greases the wheels for deskilled production. The more choices offered, the more we depend on algorithmic sleight-of-hand to make those choices for us. Transgressive music will have a more difficult time finding an audience, while readily accessible music may be superficially diverse yet carries the same message that cultivating “taste” is a worthless, unnecessary pursuit. Spotify offers not just escapism after work, but often, a lubricant to more easily get through the workday, background music specially fitted for any desk job. By dispensing with the contextual ornamentation, streaming services align themselves with other Silicon Valley productivity apps. The music may entertain but it also numbs.

Breaking that cycle may possibly require “difficult” music, music that exists not in reaction or response to the predictable and familiar but wholly separate from it. It must not simply syncopate the rhythm of modern life but seek to dismantle it entirely. If we are to step out of the marching line and into new social possibilities, we need to question and challenge the comfort zones of streaming.

Adam Clair is a writer currently based in Philadelphia. He tweets infrequently at @awaytobuildit.