That's a word that sounds like a Roman emperor pronouncing a death sentence and then forgetting about you.
It lands with this heavy, public, almost theatrical scorn — and yet almost nobody uses it. Or its adjective, opprobrious. Daniel sent us this one. He's asking: opprobrium is one of those great words you rarely see used, so what are some others? And I think the deeper question here is — why do words like this rot in the lexical attic while flimsier ones thrive?
Because this isn't just a vocabulary list episode. We're diagnosing the disease, not just admiring the patients.
Today we're going to look at what makes a word become rare, how fossil words survive in plain sight inside idioms you already use, and — this is the part I like — how you can actually start deploying these words tomorrow without sounding like you're auditioning for a Dickens adaptation.
The key is strategic rarity. One opprobrium per paragraph, not per sentence. But before we get to the how, we need to understand the why. Why do perfectly good words just... fall out of the language?
Some of them don't even fall out entirely. They get trapped. Dictionary dot com has this great piece on fossil words — terms that only survive inside fixed phrases. You say "kith and kin" but you'd never say "my kith came over for dinner.
Or "hem and haw" — when was the last time you hemmed without hawing? "Eke" survives only in "eke out." "Dint" survives in "by dint of." These words are technically dead as standalone terms, but they're still walking around inside idioms like linguistic zombies.
Opprobrium isn't one of those, though. It's not a fossil. It's fully functional — you can say "the CEO faced widespread opprobrium" and it works perfectly. It's just culturally abandoned. Nobody picks it up.
That distinction matters, because it tells us there are different categories of neglect. A fossil word like "kith" died because its meaning — friends, acquaintances, familiar people — got absorbed by other words. There was no gap left. But opprobrium names something specific that "shame" and "disgrace" don't quite capture. It's public scorn. It's the crowd's judgment, not your conscience.
We're losing precision. Every time you say something is "shameful" when you mean it drew public condemnation, you've blurred a distinction English worked hard to create.
This connects to a broader phenomenon. English has roughly a hundred and seventy thousand words in current use, but the average educated speaker uses about three thousand in daily conversation. We're leaving a hundred and sixty-seven thousand words on the table.
That's a lot of lexical attic space.
And what we want to do today is go up there with a flashlight, pull down the boxes worth opening, and figure out which ones deserve a spot in your active vocabulary.
We're going to walk through a catalog of these rare-but-great words — defenestration, petrichor, sesquipedalian, perspicacious, and several more — but not as a trivia list. Each one tells us something about how language evolves and why it abandons its own creations.
Some of these words are phonetically awkward. Some are too specialized. Some carry a social register that makes people uncomfortable. And some just never got the cultural moment they needed.
Then, in the second half, we're going to look at the one that did get its moment. Petrichor — the smell of rain on dry earth. Coined in nineteen sixty-four by two Australian scientists, obscure for decades, and then suddenly everywhere. That's our case study in how a word comes back from the dead.
Because some of them can. The question is which ones, and how.
Let's start with the diagnosis. What actually kills a word?
If we're diagnosing what kills a word, the fossil words are the easiest place to start because the body is still warm. " It comes from Old English "cythth," meaning knowledge or acquaintance — your kith were the people you knew. But over centuries, "friends" and "acquaintances" did the same job with less friction. "Kith" narrowed into that one phrase and then stopped existing outside it.
It's like a hermit crab that found its shell and never left.
That's actually a perfect image. The idiom is the shell, and the word can't survive without it. "Eke" is another one — it meant "to increase" or "to add." You eke out a living by stretching it, by adding just enough to survive. But the standalone verb is gone. Nobody says "I eked my salary with a side gig.
"dint" — "by dint of hard work." "Dint" originally meant a blow or a strike, so you achieved something by force of effort. Now it's just that one prepositional phrase, preserved in amber.
Which brings us to the first category of word death: semantic narrowing into a fixed phrase. The word didn't die because it was bad. It died because the language found more efficient ways to say the same thing, and the only reason we remember it at all is the idiom.
Opprobrium isn't in that category. It didn't fossilize. It's just sitting there, fully operational, and nobody launches it.
And that's the second category — cultural abandonment. The word works fine. It's phonetically decent. It names something specific. But the culture just... stopped reaching for it. And that's the category Daniel's question really points to, because these are the words we could actually revive.
The diagnostic question is: what makes a word culturally abandoned versus truly obsolete? A word like "wain" for wagon is obsolete because the thing it named is obsolete. But opprobrium names something we still experience constantly — public condemnation. We just reach for weaker words instead.
That's where the mechanisms get interesting. Some words die because they're phonetically ugly — they trip off the tongue like a stumble. Others die because their meaning drifts into awkward territory. And some die because they carry a social register that makes people uncomfortable. We'll see all three of those in the catalog coming up.
The taxonomy we're building is: fossil words trapped in idioms, culturally abandoned words that still work, and truly obsolete words that named things we no longer need to name. Opprobrium is in the second bucket. The question is what else is in there with it.
Which ones are worth pulling back out.
Let's pull some of these down from the attic. Merriam-Webster has a list called Words Worth Using More Often, and it's a good starting point. But I want to go beyond just reading definitions — each of these illustrates a different failure mode.
Start with the one that's almost too specific to live.
The act of throwing someone or something out of a window. It exists because of one historical event — the Defenestration of Prague in sixteen eighteen, when Protestant nobles threw two Catholic regents out of a castle window. That triggered the Thirty Years' War.
Which is a lot of weight for one word to carry.
That's exactly why it's rare. The semantic space it occupies is vanishingly narrow. How often do you need to specify that something was thrown out a window, as opposed to just thrown? But here's the thing — when you do need it, nothing else works. "He was thrown out of the meeting" is metaphorical. "He was defenestrated from the meeting" is...
Technically inaccurate, unless the conference room is on the ground floor and someone opened the window.
So defenestration illustrates the first failure pattern: hyper-specificity. A word so surgically precise that occasions for its use almost never arise. It's not dead — it's just waiting. Like a fire extinguisher behind glass.
What's the opposite failure pattern? A word that's too ugly to live?
It means physically beautiful. And it sounds like a respiratory condition.
That's the one where the sound actively undermines the meaning. You can't call someone pulchritudinous and have it land as a compliment.
It comes from Latin "pulcher," meaning beautiful, so the etymology is sound. But the consonant cluster — pulch — is alien to English phonetics. It feels like you're clearing your throat mid-word. Compare it to "beautiful," which flows. Or "lovely," which is practically a sigh. Pulchritudinous is a word that trips on its own feet.
Phonetic ugliness is the second failure pattern. Even if the meaning is useful, if saying the word makes you sound like you're choking, people will avoid it.
That's a real barrier to revival. A word has to survive the mouth before it can survive the culture. Which brings us to callipygian.
I was hoping you'd get to that one.
Callipygian means having shapely buttocks. From Greek "kallos" — beauty — and "pyge" — buttocks. It's a perfectly legitimate word. It appears in classical sculpture criticism. And you can almost never use it in conversation.
Because the social register is all wrong. It's simultaneously clinical and leering. You sound like an art historian who's also being inappropriate at a dinner party.
That's the third failure pattern: register mismatch. The word is too formal for casual use but too specific to feel academic. It falls into the gap between "nice" and "I'm calling HR." Compare it to something like "curvaceous," which is informal enough to feel natural. Callipygian has no natural habitat.
We've got hyper-specificity, phonetic ugliness, and register mismatch. What about words that just never naturalized in English?
It's a nineteenth-century borrowing from French. Literally "fly-swallower" — gobe-mouche. It means a credulous person, someone who'll believe anything, who walks around with their mouth open swallowing flies.
That's a fantastic image.
But it never took root in English because it requires French literacy to even parse. If you don't know that "mouche" means fly, the word is just noise. And even if you do, it sounds vaguely like an insult you'd get detention for saying. So it's culturally obsolete — the fourth failure pattern. The cultural bridge it needed never got built.
It's also competing with words like "gullible" and "credulous" and "sucker," which are all perfectly serviceable and don't require a French dictionary.
And that's where the concept of lexical opportunity cost comes in. Every time a word like gobemouche fails, we don't lose much, because we've got alternatives. But when opprobrium fades, we lose something real. We lose the ability to name public scorn as distinct from private shame.
That's the core of it. "Shameful" points inward — it's about what you feel or should feel. "Opprobrious" points outward — it's about what the crowd is doing to you. Different vectors entirely.
When you default to "shameful" for both, you collapse a distinction. The language gets blurrier. That's the opportunity cost — not just a missing word, but a missing precision.
Among the words we've cataloged, which ones are actually worth reviving? Defenestration is too narrow. Pulchritudinous is too ugly. Callipygian is too awkward. Gobemouche is too French.
Opprobrium clears all the hurdles. It's phonetically solid — that rolling "opp" into "probe" feels weighty. It names something we frequently need to name. And its register is formal without being clinical. You can say it in a meeting without people looking at you strangely. Well, some people.
What about sesquipedalian?
That's the self-aware one. It means "given to using long words." From Latin "sesquipedalis" — a foot and a half long. Using it is inherently a joke, because the word itself is sesquipedalian.
It's a word that performs its own meaning. That's rare.
That's what makes it worth keeping. It's not just a synonym for "verbose." It specifically names the tendency to use long words as a stylistic choice. You can be sesquipedalian without being verbose — it's about word length, not word count.
From Latin "perspicax" — sharp-sighted. It's rare because "perceptive" does most of the same work with fewer syllables. But perspicacious carries a connotation of penetrating insight that "perceptive" lacks. A perceptive person notices details. A perspicacious person sees through to the structure.
That's a useful distinction. And it's phonetically pleasant — the "spic" in the middle gives it a crispness.
Tintinnabulation is the opposite case. It means the ringing of bells. From Latin "tintinnabulum" — a bell. And it's almost entirely associated with Edgar Allan Poe's poem "The Bells," where he uses it as a refrain.
It's a word that exists inside a single literary reference. Like a fossil word, but the amber is a poem instead of an idiom.
And that's a fifth failure pattern we haven't named yet: literary entrapment. The word is so strongly associated with one work that using it anywhere else feels like quotation, not speech. You can't say "the tintinnabulation of the church bells" without someone thinking you're doing Poe.
Even if you've never read Poe, the word just sounds like it belongs in a Gothic poem. It's got that quality.
Which brings us to petrichor. And this one is different. It's the success story.
The smell of rain on dry earth.
Coined in nineteen sixty-four by two Australian scientists, Isabel Joy Bear and Roderick Thomas, in a paper for the journal Nature. They were studying the oil exuded by certain plants during dry periods, which gets absorbed into soil and rock and then released into the air when rain falls. They named it petrichor — from Greek "petros," stone, and "ichor," the fluid that flows in the veins of the gods.
Which is a wildly poetic etymology for a scientific paper.
It sat in obscurity for decades. It was a technical term in geochemistry. But then, starting around twenty ten, it exploded. Reddit threads, Tumblr posts, Twitter. People discovered this word that named a universal experience — that distinct smell when rain hits dry ground — and nothing else named it.
It's phonetically beautiful. Pe-tri-kor. It sounds like what it describes.
That's the trifecta. It names a universal experience. It's phonetically pleasant. And it filled a genuine lexical gap — there was no other word for that smell. Compare that to callipygian, which fails on social register, or pulchritudinous, which fails on phonetics.
Petrichor succeeds on all three. And that's what makes it our case study. It went from an obscure Nature paper in nineteen sixty-four to Merriam-Webster's online dictionary in roughly fifty years, with the real acceleration happening in about a decade.
What lit the fuse? Because plenty of technical terms are phonetically pleasant and name universal experiences and they still never break out.
Digital word-of-mouth. Specifically, a Reddit thread around twenty ten asked something like "what's the most beautiful word you know?" and petrichor surfaced. Then Tumblr picked it up. It became one of those words that makes people feel smart for knowing it, which is a powerful distribution mechanism.
It's the linguistic equivalent of a song going viral because someone put it in the right playlist.
And the conditions had to be right. Condition one: it names something everyone has experienced but almost no one had a word for. That's the lexical gap. Condition two: it sounds good. The soft "pet" opening, the liquid "ri," the resonant "kor" — it's almost onomatopoeic. Condition three: it had scientific credibility. It wasn't made up by a poet. It came from Nature.
It had both aesthetic appeal and intellectual permission. You could use it at a party and sound precise, not pretentious.
That's the formula. Compare it to callipygian, which fails on condition one — how often do you need to describe shapely buttocks in a register that isn't available through other words? And it fails on condition three — it sounds like you're trying too hard. Petrichor sounds effortless.
What about the German comparison you mentioned? Schadenfreude versus epicaricacy.
This is the cleanest illustration of phonetic aesthetics determining survival. Schadenfreude — pleasure at others' misfortune — was borrowed into English wholesale. We didn't translate it. We just took it. And it thrived because it sounds weighty and clinical, which gives people permission to name a slightly shameful emotion without feeling petty.
"Schaden" lands like a judgment. "Freude" lands like a diagnosis. The whole word feels like it belongs in a psychology textbook.
And English had its own word for the same thing — epicaricacy. From Greek "epi," upon, "chara," joy, and "kakon," evil. Joy upon evil. It's etymologically sound. It's been in English dictionaries since the seventeen hundreds. And it's dead.
Because it sounds like a sneeze.
Ep-i-car-i-ca-cy. It trips over its own syllables. There's no gravitas. If you say "I felt a twinge of epicaricacy," the listener is too busy processing the word to register the meaning. If you say "I felt schadenfreude," they know exactly what you mean and they don't judge you for the word choice.
The German word won because it sounded like it deserved to win. That's almost unfair.
Language isn't fair. It's a marketplace where aesthetics and social signaling matter as much as utility. Epicaricacy was the better English word on paper — native roots, precise meaning — but it failed the mouth test and the cachet test. Schadenfreude had the cultural weight of German academic tradition behind it.
Which brings us to the practical question. If you're someone listening who wants to revive a word like opprobrium, what do you actually do? What's the petrichor playbook?
You don't need a Reddit thread to go viral. You need to deploy the word in conditions where it can succeed. And that means understanding the signaling function of rare words.
Every word choice sends a social signal. "Hello" signals one thing. "Greetings" signals another. Rare words signal membership in a literate in-group. When you say "opprobrium" instead of "backlash," you're telling the listener "I assume you know this word, and I assume you value precision." It's a compliment disguised as vocabulary.
Overdo it and the signal flips. You're not saying "we're in the same club." You're saying "I'm in a club you can't enter.
That's the pretension trap. And the way to avoid it is strategic rarity. One precise rare word per paragraph, surrounded by plain language. The rare word is the spice, not the meal. If every sentence has a word the listener has to process, you've lost them. If one sentence in six has a word that makes them think "that's exactly right," you've earned their attention.
The rule is: make the rare word the payload, not the wallpaper.
Choose words that clear the three-condition test. Does it name something you frequently need to say? Is it phonetically pleasant? Does it have a natural social register? If the answer to two or more is no, leave it in the dictionary.
Perspicacious passes — it's crisp, it's useful, it doesn't sound like you're performing.
Tintinnabulation fails the frequency test. Gobemouche fails the register test. Callipygian fails all three unless you're an art historian with a specific brief.
Which means the revival list is actually quite short. Out of the hundred and sixty-seven thousand words we're not using, maybe a few hundred are genuinely revivable. The rest are in the attic for good reason.
Let's make this usable. If I'm walking away from this episode wanting to actually do something, what's the playbook?
First, the one-per-paragraph rule. Use exactly one rare word per block of text or chunk of conversation, and surround it with plain language. The CEO faced public opprobrium for the layoffs, and rightly so. Everything around "opprobrium" is ordinary. The word lands because it's isolated.
The spice, not the meal, like you said. And if you drop two in the same paragraph?
Then you sound like you swallowed a thesaurus. Density is what triggers the pretension alarm, not the words themselves. One per paragraph signals precision. Three per sentence signals insecurity.
The second thing — you mentioned a revival list.
Build a personal list of five words that fill genuine gaps in your vocabulary. From today's catalog, I'd recommend opprobrious for public shame, perspicacious for penetrating insight, tintinnabulation for bell sounds, gobemouche for a gullible person, and petrichor for the smell of rain on dry earth. Practice using one per day. In conversation, in an email, in a text.
The third thing is the filter. The three-condition test before you adopt any rare word.
Does it name something you frequently need to say? Is it phonetically pleasant? Does it have a natural social register? If yes to all three, adopt it. If no to two or more, let it stay in the dictionary. That's the difference between reviving a word and just showing off.
Most of the attic fails that test. That's the comfort. You're not ignoring a hundred and sixty-seven thousand words. You're ignoring a hundred and sixty-six thousand nine hundred that don't clear the bar, and keeping a hundred that do.
There's a question underneath all of this that I can't shake. What happens when the language that surrounds us is generated by models that optimize for the most probable word?
That's the AI angle Daniel would want us to sit with. If you're reading text produced by a language model, and that model is trained to predict the likeliest next token, it's never going to reach for opprobrium. It's going to reach for "backlash" or "criticism" or "outrage.
Because those are more frequent in the training data. That's the whole mechanism. And the more we read AI-generated text, the more those frequent words get reinforced, and the rarer the rare words become. It's a feedback loop that shrinks the active lexicon.
The models aren't neutral. They're an accelerant for lexical narrowing. They don't kill words on purpose — they just starve them of oxygen.
Unless you prompt them otherwise. You can say "use opprobrium" and they will. But the default is the death of precision by probability.
Which means the counterforce is us. Language is a commons. If we don't use these words, they die. The models won't save them. We have to.
Next time you feel public scorn, say opprobrium. Next time you smell rain on dry earth, say petrichor. You're not being pretentious. You're being precise.
Precision is a form of generosity. You're giving your listener a distinction they might not have had words for.
Now: Hilbert's daily fun fact.
Hilbert: In the nineteen hundreds, researchers analyzing Aztec patolli game pieces from the Chatham Islands found their scoring disks were coated in a manganese-rich slip that hardened into a dark, glassy surface when fired — chemically distinct from the terra-cotta bodies of the pieces themselves.
...right.
This has been My Weird Prompts. Thanks to our producer Hilbert Flumingtop. If you enjoyed this, email the show at show at my weird prompts dot com.
Use a rare word in the subject line. We'll know.
