Two things you can use

Reading about how influence works does very little. Being able to test a specific thing in front of you does a lot more, and it is a different kind of knowledge: a skill rather than a conclusion.

Both tools below are already used elsewhere on this site. They are collected here because a tool described in passing is a fact, and a tool set out on its own is something you can pick up.

Neither one tells you what to think about anything. They tell you what to check.

The four questions

Put these to any source: a study, a report, a documentary, a think tank paper, a news article, a course syllabus. They are the questions this whole site is built on, and they are on the method page for that reason.

1. Who paid, and what did they want? Not who wrote it. Who funded the work, and what would count as a useful result for them. Funding shapes findings through selection far more often than through instruction. Nobody has to be told what to conclude.

2. What was the institution's actual job? Not its mission statement. What it exists to do, day to day, and what it is measured on. A bureau measuring radio audiences for advertisers produces a different theory of the public than a seminar writing against fascism, even when the two share a building.

3. What crisis was in the room? What was going badly at the time this was made. Ideas about control get written when control is visibly failing or visibly being built, and the crisis usually explains the shape of the argument better than the author's biography does.

4. What could not be said there? What would have ended a career inside that institution. The boundary of the sayable is often the most informative thing about a place, and it never appears in the document itself.

A worked answer looks like this. The Congress for Cultural Freedom funded good magazines that published honest writers, and most of those writers did not know the money came from the CIA. Nobody lied. The four questions still return something: the money set which topics had a well paid venue, which journals could afford translation, and which careers became possible. That is what patronage does instead of censorship.

Run the questions on this site too. The answers are on the reflexive page, including the ones that are awkward.

Eight signs of coercive persuasion

Robert Jay Lifton set these out in 1961, after interviewing people who had been through thought reform programmes in China. They were written about an extreme case and they work on ordinary ones, which is why they have outlasted their original subject.

The point is not that a group showing one of these is sinister. Most institutions show one or two. The point is the count, and the direction of travel.

1. Milieu control. How much of what you hear comes through one channel. Who decides what reaches you, and whether you have contact with people outside it.

2. Mystical manipulation. Events arranged to look spontaneous, presented as proof of the group's rightness. Things that were planned, experienced as things that happened.

3. The demand for purity. The world split into pure and impure with nothing between. Since nobody can reach the standard, guilt becomes permanent and useful.

4. Confession. Personal failings disclosed and recorded. The disclosure is framed as relief and functions as leverage.

5. Sacred science. A doctrine held to be both scientifically true and morally absolute, so that questioning it is both an error and a sin. This one is easy to miss because it can wear the language of evidence.

6. Loading the language. Thought-terminating clichés: short phrases that end an argument rather than answer it. The vocabulary shrinks, and questions that cannot be asked in the available words stop being asked. This is the single most useful item on the list.

7. Doctrine over person. When your own experience contradicts the doctrine, your experience is what gets corrected.

8. The dispensing of existence. The group decides who counts as a real person. Outsiders are a lower category, and leaving means ceasing to exist in a way that matters.

Lifton's own warning applies to the list itself: it describes a direction, not a verdict. Used as a checklist for labelling people, it becomes the thing it describes.

How to use either one

Run them on something you agree with first. Both tools are easy to point at opponents and that is the use with the least value, because it returns the answer you already had.

Pointed at your own sources they return something you did not know, which is the only reason to have a tool at all.