On May 6th, 2024, a research team published a number that should bother anyone who thinks they make up their own mind about right and wrong. Across more than three thousand people, the psychologists Matthew Stanley and Christopher Neck ran a simple experiment. They described the same act — a bribe, a fraud — to two groups. One group got the blunt word. The other got a softer label for the exact same behavior. Same crime. Same facts. The only thing that changed was a word.
And the people who heard the soft word judged the act as less wrong. They wanted the wrongdoer punished less. Same deed, lighter sentence, just because somebody picked a gentler noun.
That's the thing to sit with. Not "words are powerful" in some vague poster-on-the-wall sense, but this specific, measurable effect: a euphemism is a dial, and when someone turns it, your moral judgment moves with it — often without you noticing the hand on the dial. That's what this chapter is built around. Euphemism isn't politeness. It's a quiet lever on the most important judgment you make: whether something is okay.
Start with what a euphemism actually is, because most people file it under "manners." A euphemism is a mild or indirect word swapped in for a blunt one. "Passed away" instead of "died." "Let go" instead of "fired." The word itself comes from Greek — "eu," meaning good, and "pheme," meaning speech. Good speech. Speaking with good words. For most of human history that's exactly what it was for. The ancient Greeks used soft language to avoid naming the gods directly, for fear of summoning their attention. Here's a detail that captures the whole instinct: the English word "bear" comes from a Germanic root that just means "the brown one." Hunters were so afraid that saying the animal's real name would make it appear that they refused to say it at all. So they renamed the most dangerous thing in the forest after its color.
That's the deep origin of the move — a belief that words have literal power over reality. Most of the time, the everyday version is harmless and even kind. When you tell a grieving friend their mother "passed," you're not manipulating anyone. You're cushioning a blow. The evolutionary biologist Robert Trivers, who's spent a career studying deception in nature, makes the point that softening language is a form of deception, but a mild one — and deception, in his view, runs through all of life, from the gene up to the group. The trouble starts when that same cushioning instinct gets pointed at your moral judgment on purpose.
So how does swapping one word for another actually move what you believe? This is the part that trips people up, because it feels like it shouldn't work on a thinking adult. The mechanism Stanley and Neck point to is something psychologists call construal — basically, how you mentally picture a thing. You never react to the raw event. You react to the version of it your mind builds, and the words you're handed are the blueprint for that version. A blunt word makes you picture the act concretely. You see the hands, the harm, the person on the receiving end. A soft word pulls the camera back, makes it abstract, fuzzy, bloodless. And an abstract harm is easier to wave away than a concrete one.
Think of it like the difference between hearing "your neighbor hit a dog with his car" and "there was an incident involving a vehicle and an animal." Same event. One puts you at the scene. The other files it under paperwork. Euphemism is the second sentence, deployed on purpose, to keep you out of the room where the harm happened.
Here's the detail from that study that should genuinely unsettle you. The researchers tried to break the spell. In some versions, they told participants outright — look, this soft label and this blunt label mean the same thing, here's the plain translation. And it helped. The effect shrank. But it didn't disappear. Even when people were explicitly shown the trick, they didn't adjust their judgments all the way back. As the researchers put it, participants "did not sufficiently adjust." Read that twice. Knowing you're being framed isn't enough to fully un-frame you. The soft word leaves a residue even after you've caught it.
That's worth pausing on, because it changes the whole stakes of this course. You might assume that once you can name a manipulation, it stops working — and for a lot of the moves you'll learn, naming really does most of the job. But euphemism is stubborn. Spotting it weakens it; it doesn't erase it. Which means the only reliable defense is to slow down at the word itself, before the picture forms in your head.
Now take this out of the lab, because this is where it stops being a curiosity and starts being a strategy. Stay with this for one step, because the examples build on each other. During the Vietnam War, a phrase entered the language: "collateral damage." It replaced a blunter description — the accidental killing of civilians. Notice what the swap does. "Accidental killing of civilians" puts dead people in your mind. "Collateral damage" sounds like an accounting line. Same outcome, completely different moral weight, and not one false fact in either phrase. That's the eerie part. The euphemism doesn't lie. It just files the truth somewhere you won't look.
Then there's the phrase that reached what one writer for Made Up Mind called a clinical peak: "enhanced interrogation." In the 21st century, that term was used to rebrand techniques the international community recognized as torture. Sit with the engineering of it. "Torture" is a word with a thousand years of moral horror packed into it. "Enhanced interrogation" sounds like an upgrade — like premium economy. The relabeling did real work. It created, in that writer's words, a legal and moral loophole by simply changing the label. People who would never sign off on torture could sign off on enhanced interrogation, because their own minds had been handed a gentler picture to react to.
And it's not only states. Corporations run the same play constantly, just with lower stakes and more syllables. "Layoffs" became "downsizing," then "right-sizing," and in the most absurd version, "career transition opportunities." Your technology does it too. When an app breaks because somebody wrote bad code, it doesn't say that — it says it "experienced an unexpected error." Algorithms don't censor or delete what you post; they "de-prioritize" it, or "moderate" it. Each one of those is the same move as "collateral damage," scaled down: take a thing somebody would object to, and hand them a word that doesn't summon the objection.
So if someone stopped you right here and asked what these phrases all have in common — what would you say? … They all describe harm in language that keeps the harm out of the picture. That's the whole trick, from the bear in the forest to the press release.
Which connects to a darker idea, and this is the load-bearing one for the chapter. Psychologists talk about moral disengagement — the mental moves people use to do harmful things while still feeling like good people. Euphemism is one of the cleanest tools for it. If you're the one carrying out the harm, the soft word isn't just for your audience. It's for you. It lets you stand at a distance from your own actions. The clerk who "processes a transaction" sleeps better than the clerk who "takes a bribe," even when it's the identical act. The language doesn't just shape what observers think is right and wrong. It shapes what the actor lets himself do in the first place. That's why institutions reach for it so reliably — it greases the whole machine, top to bottom, the people giving orders and the people receiving them.
So here's where it gets stranger, and where a real fight breaks out among smart people. If euphemisms are this powerful, can't you use them for good? Swap an ugly old word for a kinder one and improve how society treats people? This is where Steven Pinker, the Harvard psychologist, makes a famous and contested argument. He coined the term "euphemism treadmill" to describe what actually happens. You replace a harsh word with a polite one. For a while it's better. Then the polite word soaks up all the negative feeling attached to the thing it describes — and now it's just as loaded as the word it replaced. So you need a new one. And another. Pinker's examples in the medical and disability world trace the path: "mad" to "insane" to "mentally ill" to "neurodivergent," each one fresh for a while, each one eventually tainted.
Pinker's deeper claim is the one that bites. He argues the treadmill is powered by a false theory — the idea that you can change how people think by changing the words they use. His view: it's backwards. People have concepts, and they attach words to concepts. They don't invent new concepts by inventing new words. So when you swap the label, the feeling people had about the thing just slides right over onto the new word, because the concept underneath never moved. The treadmill never lets you step off.
Now, plenty of people push back on Pinker here, and the disagreement is real. The writer at Intellectual Takeout, walking through the immigration debate — "illegal alien" to "illegal immigrant" to "undocumented immigrant" — lands somewhere more useful than either extreme. The honest concession is that Pinker's probably right about concepts: you can't rewire what people fundamentally feel just by sanding down a word. But the same writer points out that's not actually why most political operators reach for euphemism. They're not trying to change the concept. They're doing framing — making certain beliefs jump to the front of your mind the moment you hear the phrase. "Undocumented immigrant" pulls up a bureaucratic, paperwork picture, an ordinary person missing a form. "Illegal alien" pulls up law-and-order, a foreign threat. Same people, same situation, two different sets of beliefs activated. And framing, unlike concept-change, demonstrably works — at least in the moment, on the ambivalent listener who hasn't made up their mind. So the sharper read is this: the euphemism treadmill is a fool's errand if your goal is to permanently change hearts, but a live weapon if your goal is to tilt the undecided right now. The mechanism the next chapter digs into is exactly that — framing — so for now just hold the verdict: changing concepts, no; steering the moment, yes.
There's one more thing the treadmill quietly reveals, and it's almost reassuring. The fact that soft words keep decaying tells you the concept underneath is the real thing, and the label is just paint. People aren't fooled forever. The disgust finds the new word eventually. Which means the residue from that study — the part of the framing that survives even after you spot it — fades over time as the culture catches up to each new euphemism. The treadmill is exhausting for the manipulator, too. They have to keep running.
So strip all of it down to what's doing the work. A euphemism doesn't change a single fact — it changes the picture your mind builds from those facts, pulling the camera back from concrete harm to bloodless abstraction. That shift measurably lowers how wrong you think something is and how much you want it punished. And catching the trick helps, but it doesn't fully undo it, which is why the defense has to happen at the word, not after.
Here's the one line to carry out of this chapter: when someone hands you a soft word for a hard thing, they're not protecting your feelings — they're adjusting your verdict. The next time a phrase sounds suspiciously clinical, do the swap yourself. Translate "enhanced interrogation" back into "torture," "collateral damage" back into "dead civilians," "right-sizing" back into "fired," and watch your own reaction move. That small act of translation is the whole skill.
And notice what just happened in that last move. The manipulation didn't depend on hiding a fact from you. Every word was technically true. Which raises the question that runs straight into what comes next: if someone can bend your judgment without ever lying — just by choosing which true thing to put in the spotlight — how do you defend against a lie that isn't a lie?