The Filter

The Filter Your critical thinking gym
Join the Club
Home
Propaganda Techniques
Cognitive Biases
The Commandments Join the Club

● Propaganda technique

Labeling

← 1938: a specialized reading of the IPA's original name-calling device, splitting off to target whole categories of people rather than a single opponent.

Put a whole group of people in one box, and the box does the arguing for you.

Also called: category-labeling · identity-labeling · group branding · German: Etikettierung, a distinct technique in German-language scholarship, not just a translation (see Section 06)

Identity & social pressure IPA classical / multi-tradition Offline + online

Labeling constructs a category and attaches a shared name to it, such as "the elites," "welfare queens," or "coastal liberals," so that anyone sorted into that category inherits everything the label implies, without anyone checking whether it's actually true of them.

TL;DR

Instead of arguing about one idea or one person, you get handed a box with a name already written on it. Once you're sorted into the box, people already think they know what to expect from you.

"...a device to make us form a judgment without examining the evidence." (Institute for Propaganda Analysis, 1938, describing the parent name-calling device; via Propaganda Critic)

Aliases

Category-labeling · identity-labeling · group branding. German-language propaganda scholarship treats Etikettierung as its own separately-named technique, distinct from Name-Calling; see Cross-cultural notes in Section 06.

Scope

Rhetorical fallacy Psychological weapon Media tactic

Gets confused with

Name-Calling, Stereotyping, Scapegoating, Dehumanization, Guilt by Association: precise distinctions in Section 09.

Detectability 3 / 5

The label itself is said out loud, so it's more visible than most techniques on this site. But "the elites" or "coastal liberals" sound like ordinary political vocabulary, not a manipulation move, and once you dislike the people wearing the label, that identity-level charge makes it harder to step back and check whether the category holds up.

Can you see it?
4 / 5
The label is stated outright, in the open. There's no hidden mechanism to uncover, just a word.
Innocent cover?
3 / 5
A category name sounds like ordinary descriptive language, not an attack, especially once it's been in circulation for years.
Blocks your thinking?
3 / 5
If you already dislike the people the label points at, or share the identity it's attacking, that charge makes it harder to check the category calmly.

Primary bias

Social categorization: the brain's automatic habit of sorting people into groups. Normally this is a fast, useful shortcut for a complicated social world. Labeling hijacks it: once a category has a name, membership in it gets treated as if it tells you something real about a person, even before you've met them.

Secondary biases

Outgroup homogeneity effect: groups you're not part of look more "all the same" than groups you belong to, which makes a flattening label feel more accurate than it is.

Illusory correlation: a few vivid, memorable examples get treated as proof the whole category behaves that way.

Confirmation bias: once you've accepted the label, you notice the cases that fit it and skip past the ones that don't.

How it activates

1. A speaker names a category, often anchored to one real (or exaggerated) case.
2. They attach a label that's easy to say and carries a clear charge, positive or negative.
3. The audience starts sorting new people and stories into that category on sight.
4. The label replaces the need for an actual argument: "they're an elite" does the work a claim with evidence would normally have to do.
5. Repeated use across speakers and years hardens the category into something that feels like common sense rather than a rhetorical move.

Susceptibility factors

Strong: low personal contact with real members of the labeled group. This is the single biggest factor.

Moderate: existing economic stress or status anxiety that the category can absorb and explain.

Moderate: a media environment that repeats the label often, across many different messengers.

Moderate: a political identity already organized around opposing whatever the label represents.

What does not predict it

Intelligence and education don't fully protect against this one. Social categorization is a fast, automatic habit that even careful thinkers rely on for groups they don't know well. Education mostly changes how quickly someone can be talked out of a category once it's pointed out, not whether they form one in the first place.

Primary purpose

Discredit and mobilize at the same time. A good label does both: it discredits everyone sorted into it, and it mobilizes everyone sorted outside it against them.

Immediate effect

The audience stops needing to evaluate a specific claim or person and starts sorting instead. "Which box does this belong in?" replaces "is this true?"

Long-term effect

The category outlives any single use of it. Terms like "the elites" or "welfare queens" become part of everyday political vocabulary, reused for years on new targets without anyone re-justifying whether the category is accurate.

Cui bono

Whoever needs a durable enemy category or a durable in-group identity. This shows up across the political spectrum, not on one side of it.

Offensive or defensive

Mainly offensive, since it attacks a category of people. But it plays defense too: once a group is labeled, anything that group says can be pre-dismissed as "that's just what they'd say."

How it works

Setup: a real or exaggerated group is identified, often anchored to one memorable case. Delivery: a label gets attached and repeated across many speakers and contexts. Effect: the audience files new information under the label automatically instead of evaluating it fresh.

Preconditions

The label needs some real anchor to feel credible: a genuine incident or behavior pattern, even if it then gets generalized far beyond what the evidence supports. A label with zero anchor in reality tends to get laughed off rather than adopted.

When it fails

It fails when the label collides directly with someone's own lived experience of real individuals in that group. It can also backfire through reappropriation: the labeled group adopts the term as a badge of identity instead of an insult, which drains its power (see Section 07 for a documented case of exactly this).

Amplifiers

Repetition across many independent-seeming speakers, media coverage that treats the label as a normal descriptive term rather than a loaded one, crisis conditions that increase the appeal of a simple enemy category, and algorithmic platforms that reward short, charged phrases with more engagement.

Natural habitat

Political speeches Talk radio & cable news Op-eds Social media hashtags

Resources needed

Low. A single memorable coinage from one speaker can spread on its own if it's catchy and taps an existing resentment. No large campaign infrastructure is required to originate one, though sustained repetition by many messengers is what makes it stick long-term.

Scale dependency

A label can go viral from one instance, but it only hardens into "common sense" vocabulary through repetition across many speakers and years. That repetition is what separates a one-off insult from a durable category.

Digital mutation

Hashtags now function as instant, ready-made labels that spread far faster than a phrase in a speech ever could. Because engagement-driven platforms reward short, emotionally charged text, a new category label can go from one post to widespread use in days rather than years.

First documented use

Labeling as a distinct category grows out of the Institute for Propaganda Analysis's 1938 "name-calling" device, one of the seven original propaganda devices identified from analysis of 1930s political and Nazi propaganda. Later computational taxonomies, including the widely-used NLP4IF-2019 propaganda-detection schema, still list "name calling or labeling" as one combined technique.

Named and theorized by

Institute for Propaganda Analysis (1937/38), as part of name-calling. The computational-linguistics research community has continued to use the combined name in later shared tasks built on that same schema.

Tradition

Western rhetorical / IPA classical, as a specialization of name-calling. Also active in the modern-digital tradition through hashtag-based labels.

Key evolution

Pre-mass-media: durable category labels already existed as tools of persecution long before modern propaganda theory named them (see the kulak example in Section 07). Mass media: labels like "welfare queen" spread through television and print and stayed in circulation for decades. Social media: hashtag labels can now be coined and spread within a single news cycle.

Cross-cultural notes

German-language propaganda scholarship treats this as its own separately-named technique. Sources documenting the German Wikipedia article on propaganda and manipulation techniques distinguish "Labeling/Etikettierung" from "Name Calling" using almost exactly the distinction this project draws: labeling sorts people or groups into a category using a stereotyped catchword, with its own examples being "Elite" and "Querdenker" (a German label for COVID-lockdown skeptics), while Name Calling degrades a specific target directly, with "Volksverräter" ("traitor to the people") as its example. That's a useful confirmation this project's split isn't an invented distinction: an existing academic tradition draws the same line independently.

"Kulak" (Soviet Union, 1929-1933)
Soviet authorities defined "kulak," a loosely-defined category for supposedly wealthy or exploitative peasants, then used the label to justify what Stalin announced on December 27, 1929 as the "liquidation of the kulaks as a class." In practice the label was applied arbitrarily to any peasant who resisted collectivization, not just the genuinely well-off. Between 2.1 and 2.3 million people were deported to remote settlements between 1930 and 1933, with death toll estimates running from roughly half a million to several million once executions, deportations, and the resulting famine are counted (Sciences Po Mass Violence and Resistance Research Network). This is the starkest available example of how much weight a constructed category can carry once institutions start acting on it directly.
Historical USSR 1929–1933
"Welfare queen" (United States, 1976 onward)
The term surfaced in 1974 press coverage of Linda Taylor, a Chicago woman later convicted of a comparatively modest welfare fraud. Ronald Reagan repeated an exaggerated version of her story throughout his 1976 presidential campaign without naming her, claiming she collected far more in benefits than investigators ever substantiated (Washington Monthly). The "welfare queen" label, widely understood as coded for poor, Black, single mothers, became a durable category used to justify welfare-policy cuts for decades, even though Taylor's own conduct was never representative of welfare recipients generally (NPR Code Switch).
Historical USA 1976–1980s
"Basket of deplorables" (United States, 2016)
At a September 9, 2016 fundraiser, Hillary Clinton said half of Donald Trump's supporters could be put "into what I call the basket of deplorables," describing them as racist, sexist, homophobic, xenophobic, or Islamophobic (TIME transcript). She said the next day she regretted saying "half." The label immediately backfired through reappropriation: within days, Etsy sellers and rallygoers were producing "Deplorables" T-shirts, mugs, and memes, and Trump supporters wore the label as a badge of identity rather than an insult (CBC News). A clean, well-documented example of a category label, and of a labeled group draining the label's power by claiming it themselves.
Contemporary USA 2016
Individual
When you hear a group-label, ask: does this describe an actual shared trait, or is it doing the work an argument should be doing? Could you name three real, different people the label supposedly covers?
Communicator
Don't repeat the label uncritically, even to criticize it. Describe the actual group and behavior in plain terms instead, since repetition strengthens a label whether you agree with it or not.
Systemic
Newsroom style guides and platform moderation policies that flag loaded group-labels for added context, rather than letting them run as neutral descriptors.

Pre-bunking note

Broader inoculation-theory research on stereotyping and outgroup labeling generally finds that pre-exposing audiences to a weakened version of a label, alongside the reasoning trick behind it, makes them meaningfully more resistant when they encounter the real thing later. There isn't a large body of research specific to political category-labels by name; most of the inoculation literature covers the broader stereotyping mechanism this technique draws on.

Backfire risk

Yes, and it cuts an unusual direction here: repeating a label in order to debunk it can still spread and normalize the label itself. It can also backfire the other way, as "deplorables" shows: a target group can adopt a label as a positive identity, which strengthens their cohesion rather than discrediting them.

Commonly paired with

Often precedes

Scapegoating and Dehumanization. Once a category exists and has a name, it's much easier to blame that category for a problem, and further along, to strip it of individual humanity.

Often follows

A single vivid incident or an Appeal to Prejudice that gives the eventual category something concrete to anchor to, even before the label itself gets coined.

Gets confused with

Name-Calling attaches a derogatory word to one specific target, a person, a proposal, an opponent in a debate, for use in that moment. Labeling constructs and maintains a durable category applied broadly, across many different individuals, over years. "You're a traitor" is name-calling. "The elites" is labeling. The German-language tradition draws this exact line independently (see Section 06).

Stereotyping is the underlying belief that a category has certain traits. Labeling is the rhetorical act of assigning people to that category with a memorable word, so the stereotype gets applied without having to be spelled out each time.

Scapegoating blames a group for a specific bad outcome. Labeling doesn't need a specific outcome to blame; it just needs a category people can be sorted into.

Primary sources

Institute for Propaganda Analysis (1938), name-calling device. Excerpt via Propaganda Critic.
Da San Martino et al., "Findings of the NLP4IF-2019 Shared Task on Fine-Grained Propaganda Detection" (2019). Lists "name calling or labeling" as a combined propaganda-detection category.
Sciences Po Mass Violence and Resistance Research Network, "Dekulakisation as mass violence". Scholarly account of the kulak category and its use in Soviet repression.
CBC News, "How Trump turned 'deplorables' into a campaign rallying cry" (September 2016). Documents the reappropriation of the label.

Accessible reading

Washington Monthly, "The Tyranny of the Welfare Queen" Beginner. Plain-language account of Linda Taylor and the label's real origin.
NPR Code Switch, "The Truth Behind The Lies Of The Original 'Welfare Queen'" Beginner. Short audio segment and companion article.
Christian Science Monitor, "'Les Deplorables': why some Trump supporters have embraced the label" Intermediate. Covers the reappropriation mechanism with expert commentary.