The Quiet Return of the Long Read

Sometime around the third week of February, my old editor sent me a screenshot of her phone's screen-time report. She had read for nineteen hours that week — long pieces, mostly, on a Kobo she'd bought used for forty-two euros. She framed it as a confession. "I think I'm reading more than I did in college," she wrote, "and I'm forty-six." It was the tone of someone admitting to a vice she had quietly stopped feeling guilty about.

I started asking around. A literary agent in Brooklyn told me her clients' Substack essays — the kind that run six thousand words and end on a footnote — are, for the first time in her career, outearning their print contracts. A radio producer in Lisbon mentioned that her partner had begun reading aloud from The New York Review of Books at dinner, the way some couples watch a series. A friend who runs a small magazine in Edinburgh said his subscriber numbers had grown four quarters in a row, and that almost no one cancels.

None of this is supposed to be happening. We are, all of the available data tells us, in the late innings of a long disaster of attention. The TikTokification of everything. The brainrot. The slow erosion of any reading practice that requires more than a thumb. And yet here, on the ground, in the rooms I keep walking into, something else is happening too — something subtler, harder to chart, and easier to miss because it does not announce itself.

The long read, pronounced dead at least a half-dozen times in my working life, is quietly coming back.

I — The ConditionsThe exhaustion underneath the scroll

You have to start with the exhaustion. Not the kind we already talk about — burnout, doomscrolling, the fatigue of bad news — but a finer-grained one. The exhaustion of having every spare minute of your life converted into a small, jittering interaction. The exhaustion of arriving at the end of an evening and being unable to remember what you read, or watched, or thought.

For about fifteen years the social internet ran on the premise that this was actually what people wanted. The metrics said so. Engagement was up. Time on app was up. The dashboards were green. What the dashboards never asked, of course, was whether people liked being engaged — or whether engagement, as a feeling, is closer to compulsion than to pleasure.

The phone is the only object I own that I sometimes hide from myself, and the only one I have ever had to apologize to other people for using.
Maya Holm, Notes Toward a Quieter Year, 2024

The first sign, I think, was the rise of what the marketers later called "considered media" — Substack, of course, but also Are.na, Read.cv, the small renaissance of independent magazines you can hold in your hand. Then came the strange resurgence of the e-reader, a device pronounced obsolete in roughly 2014 and now back in such force that Kobo's parent company posted record quarterly results in March.

Something was being asked for, and the market — slowly, awkwardly, and with the usual amount of overshoot — was beginning to supply it. The thing being asked for was not, strictly, more. It was longer.

II — The EvidenceNumbers that don't behave

Here is where the conventional wisdom starts to misbehave. Pew's 2025 reading survey found that the share of American adults who finished a book of more than 300 pages in the previous year ticked up for the first time since the survey began in 2003. Modest — a percentage point and change. But the line had been falling for two decades. A flat line, after a long fall, is itself a piece of news.

The more interesting data is in the metadata. Pocket, the read-it-later app, told me on background that the average length of articles saved by their users grew by 38% between Q1 2022 and Q4 2025 — and, more importantly, the completion rate on those long saves grew too. People are not just hoarding longer pieces. They are, increasingly, finishing them.

Audiobook listening has nearly doubled since 2019. Book clubs, in the bottom-up, no-marketing sense, are at numbers Penguin Random House's research arm last saw in the late nineties. The most-read piece on The Atlantic last year was a 14,000-word essay on the history of the American interstate, of all things, by a writer almost no one had heard of two years ago.

None of this is a tidal wave. None of this overturns the broader story. But the broader story — that we are simply, irreversibly, becoming worse readers — is not really holding up.

What we mistook for the death of attention was something more specific: the death of cheap attention. The expensive kind, the kind a person pays in deliberately, has been quietly returning all along. — Henry Aldred

III — The ArgumentWhy now, and why this

Why is this happening, and why now? The easy answer is reaction — that any cultural movement large enough to define an era will eventually trigger its own opposite. The honest answer is more interesting. I think long-form reading is coming back because it is, increasingly, the only place left to do a particular kind of thinking.

The short post, the thread, the reel — these are media that excel at delivery. They are extraordinary at moving an idea, fully formed, from one head to thousands. What they are bad at is the in-between work: the drift, the digression, the reader-side construction of an argument out of fragments. Reading a long essay, even a mediocre one, is one of the few activities left in modern life in which your own mind has to do the assembly.

You can feel this in the room when people talk about a piece they actually read. There is a different quality to the conversation. Not better, necessarily — sometimes wrong, often incomplete — but theirs. They have built a small private model of the writer's argument and they are now defending or revising it. This is the experience that platforms cannot fake, and have stopped trying to.

IV — The ThreatThe risks of the recovery

I do not want to oversell this. The economics of long-form writing remain catastrophic. Most of the writers I know who can sustain a long piece can do so only because of some combination of inherited money, a working spouse, a teaching job, or the brutally luck-based attention of an algorithm. The infrastructure of patient attention — the patient editors, the magazines that paid by the word, the publishers that gave a book three printings to find its audience — is mostly gone.

What is returning is a culture of reading without a corresponding culture of being read. The demand side has moved faster than the supply side, and the supply side has been decimated. We are watching a generation of readers grow more sophisticated than the institutions left to feed them.

And there is the obvious, looming risk: that AI-generated longform — slop in formal dress — drowns the recovery in its cradle. There is a real possibility that the thing being asked for is now too easy to fake, and that the fakes will glut the channel before readers have time to develop the discrimination to sort them.

V — The StakeWhat gets lost if we miss it

And yet. I keep coming back to my old editor, and her forty-two-euro Kobo, and her quiet nineteen-hour week. I keep thinking about the people I know who are reading more than they have in years, almost in secret, almost guiltily, as if they have stumbled on something they aren't supposed to have time for anymore.

The last decade asked us to mistake the speed of information for its meaning. Long reading is one of the few practices that lets you tell the difference. Not because long is automatically better — most long things are not — but because the act of reading at length is itself a way of asking your own attention what it values. It is a slow no, repeated for an hour, to everything else.

If the long read really is coming back, even in this stutter-step, half-hidden way, the question is not whether to celebrate it. The question is whether we are paying enough attention to notice — and then, at the necessary risk of ruining the thing by talking about it, whether we are willing to defend it.

Because the next thing that comes for our attention will be louder, faster, and better-funded than anything that has come before. And the only thing that has ever stood up to that, reliably, in the whole history of literate cultures, is the long, patient sentence — and the reader willing to follow it to the end.