Propaganda is not a relic of wartime posters and state broadcasting. The devices that shaped public opinion in the twentieth century are running right now in social feeds, comment sections, and video thumbnails. What changes is the delivery speed. What stays the same is the structure. The main propaganda types were named nearly a century ago, and learning them is less about spotting liars than about recognising the shapes persuasion takes when it stops being honest.

Why Propaganda Types Follow a Playbook

These devices are not random. They have been refined across different regimes, industries, and political movements because they work on the same basic hardware — a human brain that takes shortcuts under pressure.

The pattern matters more than the source. The same technique works whether it comes from a government ministry, a pharmaceutical company, or a neighbourhood Facebook group. Identifying the device tells you something useful regardless of who is using it.

This article is not about who is lying to you. It is about what the lying looks like when it happens.

The Seven Classic Propaganda Types

In 1937, a group of American journalists and academics formed the Institute for Propaganda Analysis. They identified seven devices — a framework that has outlasted the organisation itself, which closed in 1942. These seven remain the clearest starting point.

Name-calling

A loaded label gets attached to a person or idea — extremist, traitor, elitist — so the audience reacts emotionally before thinking. The label does the work the argument cannot.

Glittering generalities

A claim gets wrapped in vague but appealing words: freedom, security, our values. Nobody argues against these words because they mean nothing specific. That is the point.

Transfer

Authority or prestige gets borrowed from a trusted symbol — a flag, a religious institution, a respected organisation — and attached to a message that would not stand on its own.

Testimonial

A respected or famous person endorses something outside their actual area of knowledge. A celebrated athlete recommending a financial product is a clean example. The credibility is real; the relevance is not.

Plain folks

A leader or brand presents themselves as ordinary people — eating at diners, wearing work clothes, using everyday language — to build false common ground with the audience.

Card stacking

Only the facts, statistics, or examples that support one side appear. Anything that complicates the picture gets quietly left out. The individual facts may all be true. The picture they create is not complete.

Bandwagon

The impression is created that everyone is already on board. The audience feels social pressure to follow rather than think. Millions of people can’t be wrong is the sentence this device wants you to finish yourself.

Two More Worth Adding

The Institute’s list stopped at seven, but two further moves are common enough today that they belong alongside them. These are not part of the original framework — they are later additions that the 1930s list did not anticipate.

Fear appeal

A threat — real, exaggerated, or invented — is presented, and the speaker’s preferred solution is offered as the only way out. The fear does not need to be false to be exploited.

False dilemma

A complicated situation gets reduced to two options, usually one reasonable and one alarming. A third or fourth option may exist. You are not shown it.

If you want the modern, digital-native versions of these moves — the firehose, manufactured consensus, preemptive discrediting — we covered those separately in the nine moves to recognise. This piece is the older, more durable framework the newer ones are built on.

How These Devices Are Packaged for Digital Audiences

Short-form video and social feeds compress messages to the point where context cannot survive. Name-calling and bandwagon land in under three seconds. There is no room for a counterargument to follow.

Algorithmic amplification rewards emotional reaction. Fear appeals and outrage travel further than calm analysis without any human editor deciding to push them — the system does it automatically because engagement is the metric.

AI-generated content can now produce persuasive text, realistic images, and synthetic voices at scale. The same devices can be deployed across thousands of posts simultaneously, with minor variations to avoid detection. The techniques are not new. The production capacity is.

The Role of Repetition and Narrative Control

A claim does not need to be true to feel true. Repeated exposure across different sources creates a familiarity that the brain misreads as credibility. Psychologists call this the illusory truth effect — hearing something multiple times makes it feel more plausible even when the content has not changed.

Narrative control works by setting which questions get asked, not just what answers are given. If a frame is accepted — say, that a problem is caused by one particular group — then the argument is already half won before any evidence appears.

Coordinated inauthentic behaviour, where networks of fake or manipulated accounts push the same message, uses repetition to manufacture the appearance of consensus. What looks like public opinion forming organically may be a small number of actors creating the impression of many.

Practical Tests You Can Run on Any Piece of Content

These steps are specific and achievable. They work on content you distrust and, more importantly, on content you already agree with.

• Pause before sharing. Ask which device is present. This takes about ten seconds.
• Check what is missing. Card stacking only works if you do not notice the absent evidence. Actively search for the counterargument before accepting the picture.
• Trace the source of authority. If a testimonial figure is being used, ask what that person actually knows about this specific subject.
• Name the emotion being triggered. If you feel urgency, fear, or tribal belonging, identify it before acting on it. Naming an emotion reduces its grip.
• Look for the missing options. If a piece of content presents two choices, ask whether a third or fourth exists that it never mentions.

Why Smart People Still Get Caught

Intelligence is not a reliable defence. Motivated reasoning means people apply critical thinking selectively — questioning sources they already distrust, accepting sources they already like. The device does not need to defeat your intelligence. It just needs to reach you when your intelligence is pointed elsewhere.

Emotional state matters too. Tiredness, anxiety, and anger all reduce the mental effort available for checking claims. These techniques tend to work better late at night and during periods of collective stress — not because the content changes, but because the audience does.

It was long believed that correcting a false belief could entrench it — the so-called backfire effect. Larger studies since have mostly failed to reproduce that result: corrections generally do work, even on politically charged claims. The real problem is not that correction backfires. It is that a correction almost never reaches as many people as the original claim did, and arrives long after the impression has settled.

Building a Habit Rather Than Playing Catch-Up

Media literacy is not a one-time checklist. It is a disposition applied consistently, including — especially — to content you agree with. Agreement is exactly when the guard drops.

The single most effective change is slowing down. Almost all of these devices depend on fast, unreflective reaction. A three-second pause before sharing is not a small thing. It is the gap in which recognition becomes possible.

Frequently asked questions

What are the seven types of propaganda?

Name-calling, glittering generalities, transfer, testimonial, plain folks, card stacking, and bandwagon. They were identified by the Institute for Propaganda Analysis in 1937 and remain the standard framework taught in media literacy courses. Fear appeal and false dilemma are common modern additions but were not part of the original seven.

Are propaganda types only used by governments?

No. The same devices appear in commercial advertising, political campaigns, activist movements, and personal arguments. Governments are often the most visible users, but the techniques are available to anyone trying to persuade an audience without engaging their full reasoning.

Can I spot propaganda in content I already agree with?

Yes, and that is where the skill matters most. Content that confirms your existing views is the content you are least likely to examine. Applying the framework to material you find convincing is a more useful exercise than applying it to obvious opponents.

Does recognising these devices mean every persuasive message is dishonest?

No. Persuasion is a normal part of communication. The difference is whether a message engages your reasoning or tries to bypass it. An argument that presents evidence, acknowledges counterarguments, and lets you reach your own conclusion is persuasive without being manipulative. The devices described here are the ones designed to skip that process.

You cannot see who is behind the message — no name, no accountability, no cost to whoever wrote it. Manufactured consensus is only possible when identity is free and infinite. Get Wes Kussmaul’s Escape the Plantation free.