Why Journalism Keeps Losing (And Can’t See Why)

Late one night, as I started reading Ed Yong’s An Immense World, I encountered a concept that stopped me cold. Yong—the science journalist who won the 2020 Pulitzer Prize for explaining Covid with exquisite clarity—was describing something I’d never heard of: Umwelt.

It’s a German word coined by biologist Jakob von Uexküll to describe an animal’s perceptual world—the sliver of reality that creatures can actually sense and experience. And suddenly, I understood something about journalism I’d been circling for years but couldn’t quite name.

A Different Kind of Bubble

The book explores how animals perceive reality through radically different sensory experiences—how a tick’s world of heat and scent bears no resemblance to a robin’s world of magnetic fields and song. Yong makes complex science accessible.

The book’s introduction opens with a 2½-page passage so vivid you can feel yourself standing in that classroom—every sensory detail placing you there. I won’t do it justice—you should read the book—but imagine an elephant in a school gym. Now add a mouse, a robin, an owl, a bat, a rattlesnake, a spider, a mosquito, a bumblebee. Add a human named Rebecca. Yong walks you through this hypothetical menagerie, showing how each creature perceives the same physical space in wildly different ways. The elephant raises its trunk, the rattlesnake flicks its tongue, the mosquito cuts through the air. All three are smelling, but experiencing completely different sensory information. The robin’s chest looks red to Rebecca but not to the elephant, whose eyes can’t perceive that color. The flower is only yellow to Rebecca; she can’t see the ultraviolet bullseye at its center that guides the bee. The mouse squeaks in alarm at a pitch the bat can hear but the elephant cannot.

“These seven creatures share the same physical space but experience it in wildly and wondrously different ways,” Yong writes. ”Earth teems with sights and textures, sounds and vibrations, smells and tastes, electric and magnetic fields. But every animal can only tap into a small fraction of reality’s fullness. Each is enclosed within its own unique sensory bubble, perceiving but a tiny sliver of an immense world.”

That sensory bubble—that perceptual world unique to each creature—is what Uexküll called the Umwelt. The tick’s Umwelt is built from body heat, hair texture, and butyric acid. Trees, roses, clouds—they simply don’t exist in the tick’s universe because it cannot sense them. The robin lives in a world of magnetic fields and complex song. The elephant inhabits a universe of infrasonic rumbles and limited color.

Same gym. Completely different worlds. Each creature trapped within the house of its senses, unable to perceive what lies beyond its windows.

Reading this, I couldn’t help but think: humans don’t just experience the world through our senses. We experience it through media, through information, through the written word. And that world—the world of text-based information—is as varied among humans as the sensory worlds Yong describes are varied among species.

We’re not different animals, but we might as well be inhabiting different Umwelten when it comes to reading. Some of us move through the world of written information the way an eagle soars—taking in vast landscapes, spotting connections from great distances, synthesizing information across multiple sources with ease. Others navigate more like the tick, able to detect only the most immediate, concrete information, struggling with anything that requires piecing together multiple threads or holding complex ideas in suspension.

And here’s the uncomfortable irony embedded in my late-night reading: Ed Yong’s writing, brilliant as it is at making complexity accessible, still requires significant reading skill. That opening passage I just described demands sustained attention, tracking multiple perspectives simultaneously, holding a complex scenario in your head while absorbing specific details about sensory biology. Most Americans will never read Ed Yong. Not because they’re not curious. Not because they’re not intelligent. But because the world of words he inhabits is largely inaccessible to their reading world.

This isn’t a metaphor. The data is stark and getting worse.

The Numbers We Don’t Want to See

The Programme for the International Assessment of Adult Competencies (PIAAC) is an international study that measures how well adults can use reading, math, and problem-solving skills in everyday life. It doesn’t test academic knowledge—it tests whether you can actually do things with text: follow instructions, compare information, evaluate arguments, synthesize multiple sources.

The assessment divides reading ability into six levels—from ”Below Level 1” (struggling with simple sentences) up to Level 5 (integrating information across multiple dense texts, evaluating evidence, constructing arguments). In practice, researchers often collapse these into broader categories to report results. And that’s where things get interesting.

New data released in December 2024 shows American reading ability declining faster than expected. Between 2017 and 2023, average literacy scores dropped 12 points. Here’s where Americans now fall:

28% of adults read at Level 1 or below. These readers can handle simple sentences and short paragraphs on familiar topics, but struggle with anything requiring multi-step instructions or inference. They can read a medicine bottle or a basic email. A news article with competing claims or conditional reasoning? Largely inaccessible.

Another 28% read at Level 2. They can handle straightforward tasks—finding information on a website, reading dosage instructions, following a clear narrative. But ask them to compare information across multiple paragraphs, make inferences, or evaluate which source deserves more credibility, and the cognitive load becomes work. Hard work. When reading feels this difficult, we avoid it. We all do this in domains where we feel inadequate.

That leaves 44% reading at ”Level 3 or above.”

This is where federal reporting becomes misleading. By lumping Level 3 with Levels 4 and 5, they make it sound like nearly half of Americans are strong readers. But Level 3 is proficient, not advanced. Level 3 readers can navigate lengthy texts, handle basic inference, follow a well-reported story from start to finish. What they struggle with is precisely what defines in-depth journalism.

The skills that emerge at Level 4 and above—and only at Level 4 and above—are:

• Synthesizing information across multiple sources
• Evaluating competing claims and assessing which deserves more weight
• Recognizing subtle rhetorical moves and argumentation techniques
• Distinguishing between what’s stated and what’s implied
• Assessing source credibility and reliability
• Following complex conditional reasoning across extended text
• Integrating evidence to construct or evaluate arguments

These aren’t nice-to-have skills for reading journalism. These ARE reading journalism at the level quality journalism demands.

And only about 10–15% of American adults read at Level 4 or 5. Federal agencies had already collapsed Levels 4 and 5 into a single category because Level 5 readers were so rare. Now they’ve absorbed Level 3 into that grouping too, obscuring the reality: roughly 1 in 7 Americans can fully engage with the kind of text-based journalism we produce.

The remaining 85–90% aren’t living in our reading Umwelt. They’re inhabiting a fundamentally different information world. And we’re designing for ours, not theirs.

The Blind Spot

Level 3 readers—those proficient-but-not-advanced readers—aren’t unintelligent or uneducated. They hold college degrees. They run businesses, manage teams, raise families, contribute to their communities in meaningful ways. They’re informed, engaged citizens. They’re just not comfortable with the kind of reading that journalists—who are, by definition, expert readers—take for granted.

I’ve watched this play out in consumer research over the years. Young people telling interviewers, with genuine conviction, that they’re ”not smart enough for The New York Times.” It doesn’t matter whether that’s objectively true. It’s what they believe. And belief shapes behavior. When something feels difficult or unpleasant, humans avoid it. All of us do this, constantly, in domains where we feel inadequate.

Consider how many people stop reading for pleasure after finishing school. Not because they suddenly became incurious, but because the reading they were forced to do felt like work. The skill atrophies without use. The gap between their reading ability and the complexity of quality journalism widens. The avoidance deepens.

Journalists design for our own Umwelt without realizing it. We write for people who can effortlessly integrate information from multiple sources, who notice when competing claims need reconciliation, who can distinguish news from opinion even when the boundary is subtle, who track complex arguments across thousands of words and find the experience pleasurable rather than taxing. We write for ourselves and people like us.

We do exactly what Ed Yong warned against: we frame the information world in terms of our own perception rather than others’. We commit a kind of anthropomorphism, mistaking our sensory bubble for universal reality.

From inside our reading Umwelt, quality journalism feels accessible. Clear, even. We’ve worked hard on that clarity—shorter sentences, active voice, concrete details. But we’re making text clearer for people who already read at advanced levels. We’re rearranging furniture in a house that most of the population can’t fully enter.

Where They Go Instead

The stakes aren’t abstract. When fact-based journalism can’t reach people, they don’t simply go uninformed. They place their trust elsewhere—in individuals rather than institutions, in sources that feel more accessible even when those sources are systematically misleading them.

Research shows partisan media outlets use simpler language than mainstream news organizations. This isn’t an accident; it’s a competitive advantage. While we craft carefully sourced stories that require Level 4 reading skills to fully comprehend, they’re meeting audiences where they are. We dismiss this as ”dumbing down.” They call it reaching people.

The manipulation thrives precisely where fact-based journalism fails to reach. Not because these Americans are gullible, but because we’ve made verified information functionally inaccessible to them. When The New York Times feels like translating a foreign language but a charismatic podcaster or cable news personality speaks in terms you can follow, where does your trust go? When investigative reporting requires holding six competing claims in your head simultaneously but a Facebook post gives you a clear villain and a simple story, which one feels true?

This is how democracies fracture. Not through dramatic censorship or government control of information, but through a quiet sorting where accurate information becomes accessible only to an elite minority while the majority navigates by different stars entirely.

We wonder why people trust unreliable sources over established journalism. But what if they’re not choosing misinformation over truth? What if they’re choosing the information they can actually access?

The Question We’re Not Asking

Journalism keeps losing—readers, trust, revenue, influence—and we generate endless theories about why. Platforms stole our business model. Algorithms destroyed our reach. Political polarization made people retreat to tribal bunkers. Media literacy collapsed. Technology moved faster than we could adapt.

We’re not wrong about any of these factors. They’re all real. They all matter. But we’re missing something more fundamental, something we can’t see because we’re trapped inside our own perceptual bubble.

Until journalism recognizes that most Americans inhabit a different reading Umwelt than we do, we’ll keep writing for a shrinking audience of people like us and wondering why our work doesn’t land. We’ll keep losing readers to simpler, more manipulative sources and frame it as their failure rather than ours. We’ll keep investing in visual storytelling and multimedia and podcasts—all good ideas—while still designing our core written journalism for the 10–15% who can fully access it.

Ed Yong writes that our human Umwelt is limited—“it just doesn’t feel that way. To us, it feels all-encompassing. It is all that we know.” Journalists live in a reading world so completely that we forget most people don’t. We forget there are other Umwelten. The tick doesn’t know about roses. The elephant can’t perceive the robin’s red chest. And journalists—expert readers all—don’t realize that the carefully reported, thoroughly sourced, beautifully written story we just published is, for most Americans, as invisible as ultraviolet light is to human eyes.

The question isn’t ”Why won’t they read?”

The question is: Why do we keep designing news for a perceptual world that 85–90% of Americans don’t fully inhabit? And what happens to democracy when truth becomes accessible only to an elite few?