Lament for the Declining Art of Editing

Thoughts inspired by Taylor Swift’s 31-song notebook dump

Taylor Swift performs during "Taylor Swift | The Eras Tour" at the National Stadium on March 2, 2024 in Singapore. (Photo by Ashok Kumar/TAS24/Getty Images for TAS Rights Management)

Well, those drifter’s days are past me now
I’ve got so much more to think about
Deadlines and commitments
What to leave in, what to leave out
—Bob Seger, “Against the Wind”

I don’t want to join in the critical pile-on against Taylor Swift’s new 31-song double album, The Tortured Poets Department: The Anthology. I’ve listened to the album and find a lot to like on it, with a heavy emphasis on a lot and like (as opposed to love).

Years ago, Rolling Stone assigned a star rating to every new album it reviewed: Five stars was something approaching perfection, while one star was reserved for foulest garbage. The new Swift album sounds to me like it’s jam packed with three and three-and-a-half star songs, one after another after another for just over two hours. That means: Hardly any junk. But also: Hardly anything great. But then again, I could be wrong. I’ll need at least a few weeks to form firm judgments of all this new material. It’s going to take a good bit of work.

And that’s part of the problem: The burden of judgment—separating out the great from the good from the mediocre—falls to me and Swift’s countless millions of fans because she has opted against doing that work herself. She’s hardly the only artist in recent years to do the same.

Writing Under Constraint

You’ll hear aging rock stars talk about it in interviews—how back in the first decades of the rock-album era, the constraints imposed by the medium helped to ensure quality control. Vinyl LPs could fit only a bit more than 20 minutes of music per side. That meant a new single-album release would add up to, at most, 42 or 43 minutes of music. Any longer and sound quality would quickly begin to degrade from having to compress the grooves carved into the surface of the vinyl.

Want to squeeze a lot of songs onto your album? Then they need to be short. (In 1980, Elvis Costello released an album with twenty two-minute songs.) Writing long? Then you’ll have to limit yourself to a small handful of songs. (I’ve just described almost every prog-rock record ever released.) The only other option was to put out a double (or triple) album, if the record company would go along with it—which it might not, given the added expense of pressing two records instead of one, designing and printing a gatefold album jacket, and shipping a product twice as heavy as a single album. The biggest acts were permitted the occasional double-album indulgence, but less successful artists weren’t.

That’s where editing came in.

“Editing” can mean many things: streamlining individual songs (cut this guitar solo; get rid of that uninspired verse) or adding embellishments or adjustments (let’s try a string arrangement on this; I wonder how this one would sound in 6/8 time; this song would be better if we repeated the chorus once more at the end after a key change). This is the kind of editing a record producer typically brings to a project. It’s a blend of what, on an essay or book, is called developmental editing and line-editing: the first looks at the overall structure of a piece of writing; the second streamlines and cleans it up at the level of the word, sentence, and paragraph.

But there’s another kind of editing that’s more like curation—deciding what to include and exclude in putting an album together. If a band records twenty songs for a project, but only has room for 12 on the final album, decisions need to be made. Which are the 12 best songs—not just in absolute terms, but in terms of the whole (the album) being constructed out of the parts (the songs)? What kind of statement are the artists trying to make? What kind of sound and mood do they want to become immortalized at this moment in their career? Which of these songs do they want to play live dozens of times on their upcoming concert tour? Which song would make for the best opening track, and which the best closer? And how about pacing? How many upbeat tunes, how many ballads, and in which order?

Bob Dylan is one of the greatest curatorial editors in rock history. He frequently recorded multiple, radically different versions of each song on every album, along with many songs that ended up on the cutting-room floor and were only compiled and released decades later in outtakes collections for the most obsessive fans.

Bruce Springsteen gave Dylan a run for his money as a curatorial editor, especially in the years when he was most prolific. The Boss reportedly recorded as many as 70 songs for his fourth record, Darkness on the Edge of Town (1978). The final album included 10 of them. Springsteen gave away several songs to other artists, held onto others for his next album, and left the overwhelming majority of them in the vault. When 20 or so of them were released in a few batches decades later, fans were shocked by how many of the abandoned songs were gems. But Springsteen considered them either too derivative of other artists or too overtly commercial to fit his stark, uncompromising vision for the Darkness album.  

The same thing happened with his next album, The River (1981). Springsteen and the E Street Band recorded something on the order to 50 songs in several sessions. One version of the final album included ten songs. Then Springsteen changed his mind and expanded the project into his first and only double album of new material. The final version included 20 songs, which meant another 30 were left behind. Once again, the extraordinarily high quality of the abandoned songs thrilled his fans when many of them were released a number of years later.

Now imagine Springsteen’s early career took place in the streaming era, without the constraints imposed by vinyl pressings and the need to produce and ship a physical product. In this alternative timeline, the Boss puts out almost everything. In addition to the 30 songs he actually released in those years, he releases 40 more. That could have meant four more single albums of new songs from the Boss in these crucial years of his career. As I’ve noted, there’s an abundance of great material there. Many fans would have been ecstatic. But what would have been the artistic consequence of flooding the market in this way?

Most likely, Springsteen and his fantastic band would have come to be known as prolific craftsmen of highly enjoyable pop songs and expert musical ventriloquists capable of mimicking the sound of a 1950s ballad on one track and the amped-up punk aggression of The Clash on the next. They might have sold a lot more records, but they also might have sold fewer, as Springsteen lost much of his distinctiveness as an artist, and his universe of fans was kept fat and happy with a steady diet of new material and never left hungry for more.

Good-Enough Songs

The past five years have been a period of mind-boggling productivity for Taylor Swift. By my rough count, she has released 138 “new” songs since August 2019. (This includes the track listings of the new studio albums Lover, Folklore, Evermore, Midnights, and The Tortured Poets Department: The Anthology, plus the vault tracks/outtakes/b-sides included on the re-recordings of her albums Fearless, Red, Speak Now, and 1989.)

In terms of quantity, that’s an extraordinary songwriting accomplishment. But I wonder if the volume of output—especially with her latest release—speaks to a failure or disregard of curatorial editing. It’s one thing to release 17 songs on Folklore and then another stylistically similar 17 on Evermore five months later. It’s a lot, but at least listeners had some time to digest the first batch before the second was dropped into their laps. But now imagine she combined the two albums into a single record with only the very best songs from each included, holding the rest for release years or decades from now. Wouldn’t that have been better, elevating this imagined single album above the extremely accomplished records she actually did release?

What happened on Friday of last week is as far away from such an approach as one could imagine. That’s when the previously announced 16-track new album was released but then became a 31-track magnum opus two hours later. This is a bigger version of what happened when Midnights was released in October 2022 with 13 tracks that became 20 later that night, when then the “3am Edition” dropped.

At this point, I can’t help but wonder if Swift is holding anything back. If she writes a song with a serviceable tune, does she just automatically put it out, even if it’s the 13th song in a row to reflect on some aspect of her latest failed romance and sounds remarkably similar to the last couple dozen songs she’s co-written with collaborators Jack Antonoff or Aaron Dessner?

Speaking of Dessner, his indie rock band The National put out a decent album with 11 songs on it in April 2023. Then the band released another decent album with another 12 songs five months later. The albums are sonically interchangeable. If you put on any song from either record at random, I’ll probably enjoy it. But put on either album from the beginning and let it play through and I’ll quickly become a little bored, as the songs start to blend into each other, with one solid but undistinguished atmospheric mid-tempo track following another, and another, and another.

Wouldn’t it have been better to do the work of curation and put out a single album with the very best tracks instead of putting all of them out and thereby diluting the band’s art? Can the members of The National still hear and judge, or reach agreement on, the difference between a good-enough song and a great song? Do they still have a vision of the kind of band they want to be, and the kind of music they want to put out into the world—a vision that can be used to hone their public presentation? Or have they now reached such a high level of professionalism that nothing they produce is bad, and all of it is good enough that they see no reason not to release it?

I wondered the same thing in listening to the two albums pop songwriter Ed Sheeran released in May and September of 2023. (Once again Dessner was deeply involved in both records as co-writer and co-producer.) With bonus tracks, the two albums together add up to 32 songs. Many of them are good. Few, if any, are great.

Avoiding Notebook Dumps

Journalists have a name for an article written and published without the necessary curatorial oversight. It’s called a notebook dump. The reporter does research, conducts interviews, jots down thoughts, talks to some more people, reads some more on background—and then just pours it all into a long, sprawling mess of a draft that includes everything. A master prose stylist like Tom Wolfe at his youthful best could make a notebook dump work, just as a master songwriter like Swift can sometimes pull it off (as she largely did with Folklore and Evermore). But most cannot. In most cases a skilled editor will have an enormous amount of work to do in cutting, organizing, and shaping that draft into a whole vastly greater than the unwieldy sum of its possible parts.

What to leave in? What to leave out?—Those are the curatorial editor’s guiding questions. They imply that leaving everything in—putting everything out for the public to consume—isn’t always, or even usually, better. Less can be more. The clay needs to be sculpted. The fat trimmed. Distinctions made. Decisions rendered.

In writing this post, for example, I decided to limit myself to a few contemporary examples (Swift, The National, and Sheeran) and a few contrasting examples from the past (Dylan and Springsteen). I also decided to focus exclusively on music when I might have brought in many more examples from other genres—like the proliferation of overly long, undisciplined 3-hour Hollywood films; flabby, overwritten novels and biographies; and self-indulgent TV series filled with red herrings, extraneous characters, and unresolved plot points. But I thought readers would be better off making those kinds of connections on their own rather than piling up too many examples from too many forms of popular art that might lead the post to become repetitive, boring, or self-indulgent. I know music best, so that would be the genre I’d write about to make a broader point about the importance of editing and its waning place in our culture.

That was my curatorial decision, and I think it was the right call.

Note, though, that unlike Dylan and Springsteen putting out vinyl LPs in the 1970s—or a long-form journalist writing for a print magazine or newspaper during that same era—I am not constrained by physical limitations in writing this Substack. Instead of this being the roughly 2,300-word essay it is, it could easily have grown to 4,000 or 6,000 or 10,000 words if I allowed it to. Nothing’s preventing me from putting out a journalist’s equivalent of 31 new songs all at once for my subscribers to read, adore, revile, argue about, or nod-off over. Except that I’ve opted to impose some discipline on myself—and in the process to make the essay better than it likely would have been at longer length.

It's an old modern story: As external, received constraints on our choices are removed (through political reform, moral and theological liberalization, or technological advances), we are left with the burden of imposing constraints of our own choosing on ourselves—or else opting to give up on limitations altogether. I fear some of our greatest popular artists are showing signs of taking the latter path, and with less-than-entirely-positive results.

Editing isn’t just about fiddling around the margins with something already great. It’s the process of finding what’s potentially great amidst the merely good or even pedestrian and refining it until it achieves its highest potential. Without it, we run the risk of drowning in a sea of the merely “mid.”