Euphemism
A Socratic walk-through of euphemism — reasoned out one step at a time, not lectured.
The question we started with
THE QUESTION #Why does renaming a thing change how people feel about the thing itself?
The room with the toilet in it has been called, in turn, the privy, the water closet, the lavatory, the bathroom, the restroom, the facilities. Each name was adopted because the previous one had become too blunt to say at dinner. Nothing about the room changed. Something about the words did — and it kept happening, in the same direction, to name after name. That regularity is the interesting part. A single polite substitution is a matter of manners; a cycle suggests a mechanism.
Reasoning it through
REASONING #Ask first why a new name feels cleaner at all. A word carries more than its dictionary sense: it carries the company it has kept. Where you have heard it, who said it, what they were doing when they said it. A freshly minted or freshly borrowed term has none of that history, so it does the same referring work with none of the accumulated residue. That is a real effect, not an illusion, and it is why the substitution works at first.
Now follow it forward. The new word becomes the standard, expected name for the thing. What does it start keeping company with? The thing. Every occasion on which the referent is unpleasant is now an occasion on which this word is used, and after enough of them the word has acquired precisely the associations its predecessor had. Not because the word decayed, but because the associations were never really about the word. Connotation attaches to the referent and flows back into whatever term currently points at it.
If that is right, we can predict the cycle without knowing anything about the particular words — and this is what Pinker named the euphemism treadmill. Its clearest evidence is the words that started out as neutral technical vocabulary. "Idiot", "imbecile" and "moron" were once clinical classifications, chosen deliberately as dispassionate replacements for cruder speech. They are now schoolyard insults. Their replacement followed them down the same slope, and so did its replacement. The sequence for physical disability — crippled, handicapped, disabled, and the several proposals after it — ran the same course.
Then ask the question that tests the model: does the treadmill ever stop? It does, and the exception is exactly as instructive as the rule. Where the underlying attitude changes, the current term stops turning over and simply settles. Where the attitude persists, no naming will outrun it; each replacement is granted a few decades of freshness and then inherits the same load. So the treadmill is not a fact about language's inherent instability. It is a readout of an attitude that has not moved.
One complication worth admitting: the traffic is not one-way. A group can take a term used against it and wear it deliberately until its charge shifts — the direction runs the other way, and it too works by changing what the word has kept company with. And plenty of renamings are not euphemism at all but a group's own request about how it is described, which is a question of who gets to choose the label rather than of masking.
The analogy
THE ANALOGY #Think of repainting a rusting iron railing. The fresh coat is genuinely better — it looks clean, it is clean, and for a season nobody can tell the difference between this railing and a sound one. But the rust underneath is still working, and in time it blisters through the new paint exactly as it did the old, so the choice is between repainting every few years and treating the metal.
Paint is inert, whereas a word is not merely a coat over a fixed thing — a new term can genuinely alter what speakers notice and how they treat people, so renaming is not purely cosmetic; and rust is an objective fact of the iron, while the stigma here exists only in the speakers and can therefore change in a way corrosion cannot.
Clarifying the model
THE MODEL #The interesting misconception is the one buried in the original question, which said renaming changes how people feel about the thing itself. The mechanism argues almost the opposite. Feeling flows from the thing to the word, and the short-lived improvement a euphemism buys comes from its lack of history, not from any power it has over attitudes. Where a euphemism does change behaviour durably, it is generally by changing what is foregrounded — "collateral damage" makes it harder to picture the people, "person with a disability" puts the person first grammatically as well as rhetorically — rather than by sweetening the referent.
Two further refinements. First, the treadmill's speed varies with how much stigma is being carried: taboo words for death and bodily functions turn over slowly and comfortably, while terms for stigmatised groups turn over fast, which is the pattern you would expect if the load, not the language, sets the pace. Second, none of this makes euphemism pointless. A word that has not yet acquired its predecessor's associations really does spare people something in the meantime, and courtesy is not nothing. It is just worth being clear about what has been bought: time, not a change of heart.
A picture of it
THE PICTURE #How to readStart at the top and follow a single term through its life. The loop on the left is the treadmill — fresh, standard, tainted, avoided, replaced — and it turns as long as the referent keeps its stigma, which is why the arrow back to Fresh is a new word rather than a recovery of the old one. The only exit is the branch out of Tainted: when the attitude toward the thing changes, the current word stops taking on charge and settles, and the cycle stops.
What became clearer
WHAT CLEARED #A euphemism works, briefly, for the same reason it fails, eventually. It works because a new word has no history; it fails because it immediately begins acquiring one, and what it acquires is whatever people already feel about the thing it names. Connotation lives with the referent and lodges in whichever term is currently pointing at it. So the rate at which a society churns through names for something is a fairly good measure of how much unresolved feeling it has about it — and the only thing that reliably stops the churn is the feeling changing, at which point the naming looks after itself.
Where to go next
ONWARD #- Why some taboos produce stable euphemisms that last centuries while others turn over in a decade.
- How reclaimed slurs run the mechanism in reverse, and what conditions make reclamation stick.
Key terms
TERMS #| Term | What it means |
|---|---|
| Euphemism | a milder term substituted for one felt to be blunt, offensive or distressing. |
| Euphemism treadmill | Steven Pinker's term for the cycle in which each replacement acquires the connotations of its predecessor and is replaced in turn. |
| Connotation | the associations a word carries beyond its literal reference, accumulated from the contexts it has been used in. |
| Dysphemism | a deliberately harsh term chosen in place of a neutral one, the treadmill's mirror image. |
Every term the collection defines is gathered in the glossary.