Kelefa Sanneh, the New Yorker:

[…] In 2018, the social-science blog “Data Colada” looked at Metacritic, a review aggregator, and found that more than four out of five albums released that year had received an average rating of at least seventy points out of a hundred — on the site, albums that score sixty-one or above are colored green, for “good.” Even today, music reviews on Metacritic are almost always green, unlike reviews of films, which are more likely to be yellow, for “mixed/average,” or red, for “bad.” The music site Pitchfork, which was once known for its scabrous reviews, hasn’t handed down a perfectly contemptuous score — 0.0 out of 10 — since 2007 (for “This Is Next,” an inoffensive indie-rock compilation). And, in 2022, decades too late for poor Andrew Ridgeley, Rolling Stone abolished its famous five-star system and installed a milder replacement: a pair of merit badges, “Instant Classic” and “Hear This.”

I have quibbles with this article, which I will get to, but I will front-load this with the twist instead of making you wait — this article is, in effect, Sanneh’s response to himself twenty-one years after popularizing the very concept of poptimism in the New York Times. Sanneh in 2004:

In the end, the problem with rockism isn’t that it’s wrong: all critics are wrong sometimes, and some critics (now doesn’t seem like the right time to name names) are wrong almost all the time. The problem with rockism is that it seems increasingly far removed from the way most people actually listen to music.

Are you really pondering the phony distinction between “great art” and a “guilty pleasure” when you’re humming along to the radio? In an era when listeners routinely — and fearlessly — pick music by putting a 40-gig iPod on shuffle, surely we have more interesting things to worry about than that someone might be lip-synching on “Saturday Night Live” or that some rappers gild their phooey. Good critics are good listeners, and the problem with rockism is that it gets in the way of listening. If you’re waiting for some song that conjures up soul or honesty or grit or rebellion, you might miss out on Ciara’s ecstatic electro-pop, or Alan Jackson’s sly country ballads, or Lloyd Banks’s felonious purr.

Here we are in 2025 and a bunch of the best-reviewed records in recent memory are also some of the most popular. They are well-regarded because critics began to review pop records on the genre’s own terms.

Here is one more bonus twist: the New Yorker article is also preoccupied with criticism of Pitchfork, a fellow Condé Nast publication. This is gestured toward twice in the article. Neither one serves to deflate the discomfort, especially since the second mention is in the context of reduced investment in the site by Condé.

Speaking of Pitchfork, though, the numerical scores of its reviews have led to considerable analysis by the statistics obsessed. For example, a 2020 analysis of reviews published between 1999 and early 2017 found the median score was 7.03. This is not bad at all, and it suggests the site is most interested in what it considers decent-to-good music, and cannot be bothered to review bad stuff. The researchers also found a decreasing frequency of very negative reviews beginning in about 2010, which fits Sanneh’s thesis. However, it also found fewer extremely high scores. The difference is more subtle — and you should ignore the dot in the “10.0” column because the source data set appears to also contain Pitchfork’s modern reviews of classic records — but notice how many dots are rated above 8.75 from 2004–2009 compared to later years. A similar analysis of reviews from 1999–2021 found a similar convergence toward mediocre.

As for Metacritic, I had to go and look up the Data Colada article referenced, since the New Yorker does not bother with links. I do not think this piece reinforces Sanneh’s argument very well. What Joe Simmons, its author, attempts to illustrate is that Metacritic skews positive for bands with few aggregated reviews because most music publications are not going to waste time dunking on a nascent band’s early work. I also think Simmons is particularly cruel to a Modern Studies record.

Anecdotally, I do not know that music critics have truly lost their edge. I read and watch a fair amount of music criticism, and I still see a generous number of withering takes. I think music critics, as they become established and busier, recognize they have little time for bad music. Maroon 5 have been a best-selling act for a couple of decades, but Metacritic has aggregated just four reviews of its latest album, because you can just assume it sucks. Your time might be better spent with the great new Water From Your Eyes record.

Even though I am unsure I agree with Sanneh’s conclusion, I think critics should make time and column space for albums they think are bad. Negative reviews are not cruel — or, at least, they should not be — but it is the presence of bad that helps us understand what is good.

⌥ Permalink

Greg Storey begins this piece with a well-known quote from Plato’s “Phaedrus”, in which the invention of writing is decried as “an elixir not of memory, but of reminding”. Storey compares this to a criticism of large language models, and writes:

Even though Plato thought writing might kill memory, he still wrote it down.

But this was not Plato’s thought — it was the opinion of Socrates expressed through Thamus. Socrates was too dismissive of the written word for a reason he believed worthwhile — that memory alone is a sufficient marker of intelligence and wisdom.

If anything, I think Storey’s error in attribution actually reinforces the lesson we can draw from it. If we relied on the pessimism of Socrates, we might not know what he said today; after all, human memory is faulty. Because Plato bothered to write it down, we can learn from it. But the ability to interpret it remains ours.

What struck me most about this article, though, is this part:

The real threat to creativity isn’t a language model. It’s a workplace that rewards speed over depth, scale over care, automation over meaning. If we’re going to talk about what robs people of agency, let’s start there. […]

Thanks to new technologies — from writing to large language models, from bicycles to jets — we are able to dramatically increase the volume of work done in our waking hours and that, in turn, increases the pressure to produce even more. The economic term for this is “productivity”, which I have always disliked. It distills everything down to the ratio of input effort compared to output value. In its most raw terms, it rewards the simplistic view of what a workplace ought to be, as Storey expresses well.

⌥ Permalink

Speaking of podcasts, Michael Hobbes dove into Jonathan Haidt’s “The Anxious Generation” — previously mentioned — for his “If Books Could Kill” podcast. At two hours, it is the longest single episode they have done, but it is worth it for Hobbes’ careful exploration. There is some profanity.

⌥ Permalink

Apple finished naming what it — well, its “team of experts alongside a select group of artists […] songwriters, producers, and industry professionals” — believes are the hundred best albums of all time. Like pretty much every list of the type, it is overwhelmingly Anglocentric, there are obvious picks, surprise appearances good and bad, and snubs.

I am surprised the publication of this list has generated as much attention as it has. There is a whole Wall Street Journal article with more information about how it was put together, a Slate thinkpiece arguing this ranking “proves [Apple has] lost its way”, and a Variety article claiming it is more-or-less “rage bait”.

Frankly, none of this feels sincere. Not Apple’s list, and not the coverage treating it as meaningful art criticism. I am sure there are people who worked hard on it — Apple told the Journal “about 250” — and truly believe their rating carries weight. But it is fluff.

Make no mistake: this is a promotional exercise for Apple Music more than it is criticism. Sure, most lists of this type are also marketing for publications like Rolling Stone and Pitchfork and NME. Yet, for how tepid the opinions of each outlet often are, they have each given out bad reviews. We can therefore infer they have specific tastes and ideas about what separates great art from terrible art.

Apple has never said a record is bad. It has never made you question whether the artist is trying their best. It has never presented criticism so thorough it makes you wince on behalf of the people who created the album.

Perhaps the latter is a poor metric. After Steve Jobs’ death came a river of articles questioning the internal culture he fostered, with several calling him an “asshole”. But that is mixing up a mean streak and a critical eye — Jobs, apparently, had both. A fair critic can use their words to dismantle an entire project and explain why it works or, just as important, why it does not. The latter can hurt; ask any creative person who has been on the receiving end. Yet exploring why something is not good enough is an important skill to develop as both a critic and a listener.

Dan Brooks, Defector:

There has been a lot of discussion about what music criticism is for since streaming reduced the cost of listening to new songs to basically zero. The conceit is that before everything was free, the function of criticism was to tell listeners which albums to buy, but I don’t think that was ever it. The function of criticism is and has always been to complicate our sense of beauty. Good criticism of music we love — or, occasionally, really hate — increases the dimensions and therefore the volume of feeling. It exercises that part of ourselves which responds to art, making it stronger.

There are huge problems with the way music has historically been critiqued, most often along racial and cultural lines. There are still problems. We will always disagree about the fairness of music reviews and reviewers.

Apple’s list has nothing to do with any of that. It does not interrogate which albums are boring, expressionless, uncreative, derivative, inconsequential, inept, or artistically bankrupt. So why should we trust it to explain what is good? Apple’s ranking of albums lacks substance because it cannot say any of these things. Doing so would be a terrible idea for the company and for artists.

It is beyond my understanding why anyone seems to be under the impression this list is anything more than a business reminding you it operates a music streaming platform to which you can subscribe for eleven dollars per month.

Speaking of the app — some time after I complained there was no way in Apple Music to view the list, Apple added a full section, which I found via foursliced on Threads. It is actually not bad. There are stories about each album, all the reveal episodes from the radio show, and interviews.

You will note something missing, however: a way to play a given album. That is, one cannot visit this page in Apple Music, see an album on the list they are interested in, and simply tap to hear it. There are play buttons on the website and, if you are signed in with your Apple Music account, you can add them to your library. But I cannot find a way to do any of this from within the app.

Benjamin Mayo found a list, but I cannot through search or simply by browsing. Why is this not a more obvious feature? It makes me feel like a dummy.