Chapter 4

Updated 2026-07-02

"The most common name is at the level of usual utility." — Roger Brown, "How Shall a Thing Be Called?" (1958)

Nobody points out the window and says "look at the mammal." A parent crouches next to a stroller, finds a golden retriever trotting past, and says "Look at the dog!" — never "Look at the animal," never "Look at the golden retriever." Not because the parent doesn't know the word retriever. Because dog is the size of word that fits the moment.

Every object you've ever owned lives at several levels of generality at once, stacked like a layer cake. The chair in your kitchen is a chair, but it's also furniture, and it's also, more narrowly, a kitchen chair, and if you wanted to get specific you could call it a 1962 Eames-style shell chair with a walnut dowel base. All five of those are true. You will use exactly one of them by default, almost every time, without thinking about the other four: chair. Not furniture — too vague, doesn't tell you whether to sit on it. Not Eames shell chair — too specific, nobody needs that much precision to ask someone to pull one up to the table.

This default size has a name — psychologists call it the basic level — and once you start looking for it, it explains an enormous amount of otherwise mysterious behavior in how people talk. It's the first word a child learns for a thing ("dog" arrives well before "animal" and well before "poodle"). It's the fastest word an adult can produce, the easiest one to picture clearly with your eyes closed, the one two strangers will draw almost identically if you hand them each a crayon and ask for "a bird." A psychologist in 1958 named Roger Brown noticed this years before anyone had a theory to explain it, just from listening to how parents talk to children: a dog on a lawn is also a quadruped, also an animate being, also, if it happens to be a boxer, also a boxer — and parents almost never use the technical or the generic word, they reach for the one in the middle every time. "The most common name," he wrote, "is at the level of usual utility." He was listening to families talk and accidentally discovered a law.¹

It's not a law about laziness. The basic level isn't the lazy choice — it's the choice that does the most work for the least effort. You have one motor program for "sit in a chair" and none at all for "sit in a furniture." You can picture "chair" with your eyes shut and get something useful; picture "furniture" and you get a blur, picture "Eames shell chair, walnut base" and you get a specific image so narrow it leaves out half the chairs that count. The basic level is the largest size at which a category still has a single, recognizable shape and a single thing you can do with it. Zoom out past it and the shape dissolves into a blur. Zoom in past it and you've stopped saving anyone any effort at all.

---

The part that shouldn't be true: this default size isn't a quirk of English, or of one culture's habit of talking to children. It shows up, independently, on the other side of the world.

The anthropologist Brent Berlin spent decades cataloguing how unrelated cultures — societies with no shared history, no contact, separated by oceans and millennia — sort the plants and animals around them. He found the same architecture again and again: a handful of very broad categories at the top (plant, animal), a layer of basic, everyday names in the middle (oak, trout, robin), and increasingly fine technical distinctions underneath for anyone who needs them (white oak, post oak, bur oak). The middle layer — what Berlin called the generic-species rank — is the one that gets a short, simple, one-word name in every language he studied. It's the one children learn first. It's the one ordinary speakers reach for by default and reserve the fine distinctions for people who need them. Researchers who later reviewed this body of work across dozens of unrelated languages summed it up bluntly: folk taxonomy everywhere is "composed of a small number of absolutely distinct hierarchical levels," and the generic-species rank is, everywhere, "most easily recognized, most commonly named and most readily learned."²

This is the same shelf Rosch found with a stopwatch in a psychology lab, discovered independently by a different field studying a different question: which plants do societies have words for, and at what size. Two unrelated investigations — one timing reaction speeds in a lab, one cataloguing plant names across continents — landed on the same answer about where a category sits most naturally. That kind of convergence is hard to explain as coincidence.

---

The basic level isn't fixed, though — it slides, and it slides toward whoever's looking.

To someone who doesn't watch birds, a robin and a chickadee and a heron are all, basically, "a bird"; that's the size at which the differences stop mattering to them. To a birder, "bird" is the vague, useless word — too blurry to be worth saying — and "warbler," or more precisely "yellow warbler," is the word that arrives first and fastest, the one that pops out before the brain even finishes processing the shape on the branch. Researchers studying expert bird-watchers and dog-show judges found just this: experts verify a fine-grained name — warbler, not bird — as fast as novices verify the broad one.³ Practice doesn't just add more facts to the same basic level. It drags the basic level itself down a notch, narrows the size of "the obvious word," for that one corner of the world the expert has spent years inside.

Which means the basic level was never really a property of the dog, or the chair, or the warbler. It's a property of the relationship between a thing and somebody's accumulated attention to it — the size at which this viewer's experience has carved the deepest groove. Everyone has a few corners like this. Most people have one for cars, or wine, or their own children's friends, where the "obvious" word for a stranger is the word they'd never bother using.

---

So the basic level is real, and it's shared across cultures — and it also bends, person to person, language to language, the moment you look closely enough at the edges. Both of these things are true, and the second one runs deeper than expertise.

In the highlands of New Guinea, the Dani have two words for color. Not eleven, not a dozen — two, one that covers the dark, cool end of everything and one that covers the light, warm end. By the bare logic of language, a Dani speaker shouldn't be able to tell a focal, top-of-the-line red from an off-shade of orange-red any more easily than an off-shade of green — they have no words to mark the difference. When Eleanor Rosch tested this directly, showing Dani speakers colors and later testing what they remembered, the result wasn't what either side of the language-and-thought debate expected.⁴ Dani speakers remembered focal colors — the reds, the blues, the most "red" red — better than off-shades, in essentially the same pattern English speakers show, despite having no words at all for most of those colors. Whatever was doing the sorting, it wasn't the vocabulary.

And yet the vocabulary isn't nothing, either. Look across the world's languages and the color words don't arrive in any order at all — they arrive in something close to a fixed sequence. A language with only two color terms divides dark-and-cool from light-and-warm, the way Dani does. The third term to arrive, almost without exception, is red. After that, green and yellow, in either order; then blue; then brown; then, last, the fine distinctions — purple, pink, orange, grey. Berlin and Kay catalogued this pattern across nearly a hundred languages in the late 1960s, and while the methodology has since taken serious hits — critics argue the data were cleaned up, that real fieldwork finds messier boundaries, that some languages sort by brightness rather than hue at all — the broad sequence has held up reasonably well under decades of further testing.⁵ Something underneath is constraining the order languages discover color in, even while two-color Dani speakers are quietly tracking distinctions their language never bothered to name.

The same tension — universal scaffolding, culturally specific furniture on top of it — shows up in places with much lower stakes than color perception. An American asked to sort leisure activities will file bowling with bocce and darts, games you play for fun, no particular skill class implied. A German cataloguing the same activity for a library or a sports federation is liable to file Bowling under organized, competitive sport, closer to bobsledding than to backyard horseshoes — a different shelf entirely, for the same act of rolling a ball at pins. Soup runs into the same wall in the other direction: English distinguishes "soup" from "chowder" by a thickness and chunkiness that French doesn't mark at all, so a recipe for hearty New England chowder, translated word for word into French, just becomes soupe — correct, and missing the entire point of what made it chowder.⁶ Nothing about a thick, chunky, potato-and-clam stew demands the word "chowder." A culture decided that distinction was worth a separate name, and another culture, just as reasonably, decided it wasn't.

Put the color sequence and the bowling shelf and the soup pot next to each other and the picture sharpens: there is structure underneath — a basic level that different minds keep rediscovering on their own, a color sequence that keeps arriving in roughly the same order — and there is choice layered on top of it, different in every culture, about where to draw the next line down. Different minds carve differently. The carving still has joints.

---

What's left, after three chapters of this, is the shape of the instinct in full: fast, automatic, graded rather than binary, organized around centers rather than fences, partly universal and partly local, and — every example so far, from the fire list to the warbler — capable of being revised in an afternoon without anyone grieving the loss. A four-year-old re-sorts dinosaurs by lunchtime. A birder's "bird" shrinks to "warbler" over years of practice and nobody experiences that as a small tragedy.

There is one way to see what all of this is actually protecting against, and it requires a thought experiment rather than a study, because no real mind works this way. In 1942, the writer Jorge Luis Borges invented one that does — a short story, not a case — but a useful diagram. After a fall from a horse, a young man named Ireneo Funes wakes up with a memory that forgets nothing and a perception that misses nothing. He can reconstruct, on command, the exact shape of every cloud he has ever seen. And it costs him the one thing every category does for free: he cannot generalize, and so, in the story's own terms, he cannot think. "To think is to ignore (or forget) differences, to generalize, to abstract," Borges writes. It irritates Funes that "the dog of three-fourteen in the afternoon, seen in profile" gets called by the same word as "the dog of three-fifteen, seen frontally," as if a single name could possibly cover two such different dogs.⁷ He tries, for a while, to invent a private number system that gives every integer its own unrelated name — not "one, two, three" but something closer to a name per number, the way you might name a child — and even that, eventually, strikes him as too general, too forgetful of the differences between any one count of a thing and the next. He starts a project to catalogue every memory of his entire life and abandons it, because it would take a lifetime to recall a lifetime, and the project, like every fully complete category, would finish only when there was no time left to use it.

Funes is fiction, deliberately so — nobody perceives the world at that resolution, and the story's point depends on it being impossible. But it draws, in negative space, the exact shape of everything chapters one through four have been describing as automatic and nearly invisible. The philosopher Friedrich Nietzsche, writing seventy years before Borges, described the ordinary cost of an ordinary concept in almost the same terms, without the fiction: "Every concept arises from the equation of unequal things," he wrote. "Just as it is certain that one leaf is never totally the same as another, so it is certain that the concept 'leaf' is formed by arbitrarily discarding these individual differences and by forgetting the distinguishing aspects."⁸ Funes refuses to discard anything and loses the ability to think. Everyone else discards constantly, instantly, mostly without noticing, and that discarding is not a flaw in the instinct. It is the instinct. Every leaf really is different from every other leaf, and the word "leaf" is a small, useful, perpetual act of forgetting — sorted with conviction, held lightly, exactly the way a four-year-old holds a pile of dinosaurs, and the thing Funes, perceiving everything and forgetting nothing, could never bring himself to do.

That is what the mind does, left to itself: sort fast, sort in degrees, sort differently depending on who's looking, and let it all go the moment the purpose passes. It is, for an instinct, a remarkably forgiving way to run a filing system. Then someone handed that impulse a filing cabinet that wasn't allowed to forget.

---