Popular music

My friend Jon Bash is a very talented musician/composer, so it's safe to assume that he thinks very seriously about music in general. Recently he posted this article ("How Popular Music's Lyrics Perpetuate American Idiocy") from a site called The Anti-Media, about a study examining the "reading level" of Billboard Top 40 songs. Stories about the study were passed around other places too, including The Guardian ("Hip hop ain't top: new study reveals reading levels behind music lyrics") and The Onion's A.V. Club ("Popular songs are dumb and only getting dumber, says dumb study"), with varying levels of skepticism and/or snark. Before getting in to any substantive discussion, however, I will first make fun of The Anti-Media for writing a really dumb article.

Claire Bernish (Twitter handle: @subversive_pen — Twitter bio: a Julian Assange quote) opens her turd-de-force with the following incredible paragraph-and-a-half:
A recent study served to confirm the patently obvious: song lyrics for the most popular genres of music are ridiculously obtuse — and getting worse over time. Though this might not be a revelation, the figures are distressing indicators of both an intellectually vapid societal and cultural future as well as its apparent inevitability.

If you’ve already moved away from Billboard music, congratulations, you refuse to be insulted. But if you haven’t, or if you’re concerned about pop culture trends acting as portents of systemic dysfunction, you should probably pay attention.
Her fingers must have been trembling as she forced them away from typing "decadence" or "degeneration." Even still, she managed to cram a heroic amount of smug elitism with general "these days" present-and-future-anxiety in those four sentences.
So how did this happen and why is it getting even worse? For the sake of brevity, this is a systemic issue being reinforced across the board by pandemic anti-intellectualism. Some have argued there is no harm in a bit of mindless distraction, but this is incontrovertibly false. When just six corporations control 90% of the media, and 80% of radio stations have identical playlists, mindless content isn’t a choice — it’s a virtual mandate. In this self-propelled cycle of banality, the conglomerates dictate content to be promoted by radio, which in turn pushes it endlessly, creating a false perception that what is being played is due to listener demand. But this insidious marketing ploy is more akin to kidnapping and is every bit as dangerous.
After muddling through some cynical platitudes about the American public education system (shorter version: Another Brick In the Wall, Pt. 2), Bernish arrives at her goal, the political angle:
From a political standpoint, all this ‘dumbing down’ makes sense: indoctrination creates obedience. If music and culture focus on mindless diversion, and education lacks, well, education, then people lack the acuity necessary to question the absurdity of the system. Those who manage to liberate themselves from this mold and have the gumption to question official authority will find a cozy spot on the government’s watch list. So while we bemoan our country’s lack of intellectual prowess, it isn’t by a failure of design.
Again, one can easily conjure an image of Bernish, sweating over her PGP-encrypted laptop, fingers tensed up millimeters above the keys as she manages to avoid typing "infowar" only by sheer force of subversive will. Wake up sheeple! she whispers to an uncaring and oppressive night, as she clicks "Publish"...


The only thing more offensive than the Anti-Media piece is the fact that the mainstream media apparently took this seriously. From The Guardian, an actual serious newspaper that reports on important issues and also this for some reason:
Is lyrical intelligence getting lower? A new study suggests the complexity of lyrics is declining, comparing words used in songs to reading levels in US schoolchildren.
By the Guardian's standards, I have done "studies." In a just world, the entire blog post (and it is just a blog post... says a guy in a blog post) would be dismissed out of hand, because averaging readability scores? Tests designed for prose and especially technical writing? Readability scores are justifiably important for, say, museum displays and legal briefs, hospital pamphlets and government missives. It's about conveying the most accurate information with the greatest ease. It's worth noting that a high reading-level score corresponds to low readability.

Last time I checked, pop music (or country, or rock, or metal, or...) wasn't created with education in mind, except for They Might Be Giants, maybe. Pop music is for entertainment. So the whole readability-analysis endeavor is sort of like if someone dismissed Van Gogh because his work wasn't as "detailed" as, say, Rembrandt. (Oh wait, that was a pretty common,  and dumb, criticism of art, especially in the Modernist and Post-modernist era...)

It's worth noting at this point, though, that I totally feel where Jon is coming from: I think that Top 40 songs are mostly dumb garbage. But I long ago stopped trusting broadcast radio, because for various reasons even stations that play music I enjoy, still only play the same half-dozen singles at best. If I like an artist I usually like lots of that artist, and usually their singles (if they even release singles) aren't my absolute favorite tracks. So I have to apply that same logic to pop music and hip-hop, predicting that the other tracks on their albums are probably different from the singles.

This is often true to a bizarre degree. I'll get to that later.

The other thing to note is that the Billboard charts are a really strange metric for whether a song is good or bad, and this has been true since the 40s. In the 1950s, when Christmas songs managed to top the charts, and even after Elvis Presley came on the scene, the year-end number one songs were often not the ones people remember: Elvis songs were only year-end #1 twice, and those were the not-very-lyrically-sophisticated "Heartbreak Hotel" and "All Shook Up." Does anyone remember this 1955 ten-week-chart-topper, "Cherry Pink (and Apple Blossom White)"? Or this thirteen-week smash hit from 1950, "Goodnight Irene"? And that was a year when freakin' "Rudolph the Red-Nosed Reindeer" was the top single for a week!

For what it's worth, the band that recorded that rendition of "Goodnight Irene" also did "On Top of Ol' Smoky," which you might have first encountered in kindergarten, or possibly pre-school. Maybe in it's "Spaghetti" version, which would probably bump the Flesh-Kincaid readability up by a few points what with the "spaghettis" and "meatballs" flying around.

Moreover, it's not as if the main stylistic issues of Top 40 songs are so incredibly different from today. Mediocre repetitive beats? Listen to that mambo tune "Cherry Pink", it's really kinda dumb (especially with that try-hard warbling horn sustain). Auto-tune killing music? Music has been dead for half a century by that reckoning, the goddamn Chipmunks Song was top of the charts for four whole weeks in 1958. It won three Grammy Awards. Calibrate your expectations accordingly. (And gain a new appreciation for why the 60s were an insane revolution in popular culture.)


Jon made a point about something a bit different, though:
The problem is, though, that none of the songs are saying anything. It's not explaining complicated subjects in an easily understandable way, a la Mark Twain or something; it's saying nothing with nothing. It's mindless ear candy garbage, ooh-ing and wow-ing for maximum profit. That's its undeniable purpose. There's the rare bit of self-expression and complexity, but even *it* tends to often be fed through the profit machine and the question "That's a cool song, now how can we make it a HIT," which usually drains it of its uniqueness in favor of something only slightly different from a cookie-cutter.

So yes, you can say profound things without coming across as a Thesaurus-monger. But that's not what's happening. And it's concerning that the vast majority of the music industry and its consumers don't seem to care.
While the historical evidence suggests that this is nothing new in pop music, Jon's criticism is independent of the reading-level analysis. Even if it's no big deal to have a song with third-grade-level lyrics, it might still matter if the most popular songs on the radio are literally content-free.

I'm sympathetic to this concern. Though it's also worth noting that popular "message" songs are a decidedly mixed bag: You have CSNY's "Ohio" getting to #14 on the one hand, and on the other you have Toby Keith's "Courtesy of the Red, White and Blue (The Angry American)" getting to #25:
The reaction was so strong that the Commandant of the Marine Corps James L. Jones told Keith it was his duty as an American citizen to record the song. "It's your job as an entertainer to lift the morale of the troops," Jones said to Keith. "If you want to serve, that is what you can do."[1] As the lead single from the album Unleashed (2002), "Courtesy of the Red, White, & Blue" peaked at number 1 on the country charts over the weekend of July 4.
That was a weird time to be alive.

Anyway, yeah, songs are usually about relatively straightforward emotions rather than high-complexity concepts—love, loss, joy, depression, nostalgia, optimism, faith, all are popular themes. That said, of course emotions aren't simple, but it's the difference between describing nuanced instances of a broader emotion, and simplifying a complex and intricate topic. We describe emotions and other phenomena of that conceptual level by adding specifics, and hoping that it touches on the greater truth.

You can definitely have narrative songs, though even then they have to be pretty broad if you want to be super popular. Unless the genre is metal and your band is called Sabaton, because they can sing about a specific battle in the 1938 invasion of Poland to a festival audience and are just that awesome. (Take that Toby Keith, some Swedes made me feel patriotic for Poland!)

Okay, okay. I didn't write this post to talk about my favorite genres.


We should probably consider the structure of music production and consumption in America, too, because I think that's where Jon's point is most on-target. True to this generation of viral videos and ultra-brief superstardom (lucky 60s and 70s and 80s where artists might have enjoyed a whole week at the top!) it does seem like most top songs are lazy vehicles for one catchy hook or one particular chorus. Certainly that's what I remember of most of the top songs, if I remember anything at all: just a few lines, maybe the main beat. That's it.

This isn't new, but you can see how music commodification has become ever more infinitesimal over time. Two-track singles were once the smallest unit of music one could buy; now it's individual tracks—and pretty much no expectation that you'll buy a whole album. Just chuck the artist into Spotify.

Moreover, production has become more individualized. Random (or "random") people on YouTube can make it big, regardless of actual merit. There's something like startup culture around pop music now; whereas the old evil was record executives getting their hooks in a band and bleeding them dry over the course of years, now it's everyone (publishers and artists both) putting more effort into "virality" than enduring quality. If so many people can just produce music, your only hope (so the market wisdom goes) is to go out in a blaze of venture-capital glory. Get a billion views on YouTube and ride the ironic appreciation all the way to the bank.

You can get a sense of the divide between album content and the mass-market singles, by actually listening to an entire album by a Top 40 artist. I have it on good authority (i.e. my friend Adam, a huge fan of Lady Gaga) that Lady Gaga's albums are a notable example of this.

This cements my personal aversion to pop music: the radio singles are far too "sugary" for my taste, and the stylistic dissonance between them and the B-sides (why didn't we invent new terms when we moved away from vinyl?) is too jarring. A huge part of my music collection is albums with coherent musical themes, often to the point of repetition—in service of a story or character, say. In other words, I think that rock operas and concept albums are cool.


More quantitatively, I think that the conclusions of this pseudo-study might have fallen prey to summary statistics. Scott Alexander at Slate Star Codex (one of my favorite idea-blogs) coincidentally posted about this just the other day: ]
Here [in this example] all crime is committed by low IQ individuals, but the correlation between IQ and crime is still very low, r = 0.16. The reason is simple: very few people, including very few low-IQ people, commit crimes. r is kind of a mishmash of p(low IQ|criminal) and p(criminal|low IQ), and the latter may be very low even when all criminals are from the lower end of the spectrum.

The advice some people on Tumblr gave was to beware summary statistics. “IQ only predicts 3.6% of variance in crime” makes it sound like IQ is nearly irrelevant to criminality, but in fact it’s perfectly consistant with IQ being a very strong predictive factor. [...] Based on the way you explain things, you can use different summary statistics to make things look very important or not important at all.
The potential problem in this case is that there's an assumption that all people who listen to Top 40 music only listen to Top 40 music. We don't know this at all, even though we know that such-and-such pop song is the #1 Billboard hit for so many weeks. We don't know this, even though we know that such-and-such track is played so many millions of times on Spotify.

It could easily be the case that "everyone" likes Top 40, but each person likes other genres more than Top 40, but not enough people like any particular non-Top-40 genre.

This is reinforced by something like pluralistic ignorance: if you're in a group whose aggregate musical preference you don't know, you'll probably suggest something you believe to be popular, because that's the most likely to be correct—not to mention it's at the forefront of your mind. And so you'll end up listening to the latest viral hit, even if it turns out that you're all closet post-rock fans.

This, in turn, leads to one of my all-time worst pet peeves: when people claim they listen to "pretty much any kind" / "all kinds" of music, and then go on to list artists/bands that are very solidly in the "indie" or "alternative" category, for example. The "outside-the-box box"is pretty small in that situation. It's a really bad pet peeve of mind because it hits right at the intersection of my non-central musical tastes (see YouTube links above) and my mathematical-logical anxiety around "for all" type statements. So fragile! So dangerous! It only takes one counterexample!


So here, at the end, we conclude that the study was barely one at all. Raw totals of Billboard rankings tell you nothing about even the aggregate individual's listening habits. Readability scores are Not Even Wrong when applied to song lyrics, and even if modern pop music uniquely represents a degeneration compared to the pop music of earlier eras, it's still a bad argument of the "video games make people violent" caliber.

Dumb songs aren't making people dumb. Posts about how This Song Perfectly Describes Something are making people dumb, but everyone knew that already.