This is a study guide to the literature on influence: how people are persuaded, pressured, deceived and coerced, and what the evidence says about resisting each of those. It runs to 360 short chapters in twelve parts, and it is meant to be used rather than read straight through.
What each chapter contains. A chapter opens with several hundred words of prose, then shows its working: the specific study or documented case it is built on, the mechanism as the researchers described it, the beats the chapter covers, and the sources. The apparatus sits below the prose deliberately. It is not decoration, and it is the part you should check when something here sounds too tidy.
Where the sources come from, and how much they are worth. The 1,077 citations are not of uniform quality, and it would be dishonest to present them as though they were. Roughly, they fall into three bands. Peer-reviewed papers and government reports — Milgram's 1963 paper, the 2014 Senate report on torture, the National Academies review of the polygraph — carry the most weight and are the backbone of the mechanics and detection chapters. Investigative journalism and court records carry the case studies; they are usually reliable on what happened and weaker on why. Encyclopedia entries and secondary summaries appear where a chapter needs an orienting fact rather than an evidential claim, and they should be treated as a starting point for your own checking rather than as an authority.
The replication problem is part of the subject. Several findings that this field was built on have not held up: power posing's hormonal effects, the NLP eye-accessing model, voice stress analysis, behavioral deception detection at airport scale, and most of the popular reading of Mehrabian's 7-38-55 figures. Those chapters are in the guide precisely because they failed, and they are marked as failures rather than quietly dropped. A guide that only listed the findings that survived would give you a false sense of how much is actually known. Where the evidence is contested, the chapter says so.
What this is not. It is not clinical advice, legal advice, or a diagnostic instrument. The Situation tool matches a description to chapters; it does not tell you what is happening to you or what to do about it. If you are in danger, the chapters on separation and safety planning explain why the period around leaving is the highest-risk one, but they are a reason to contact a professional service, not a substitute for one.
The dual-use problem. Part 10 states thirty techniques of applied influence as operating laws, which is an uncomfortable thing to publish. The argument for doing so is the one the last chapter of the guide makes: these techniques are already documented, already taught commercially, and already in use, and the asymmetry between people who know them and people who do not is the thing that makes them work. Every law is paired with its counter, and each carries the boundary at which its use stops being persuasion.
Your data. Progress, saved chapters and highlights are stored in this browser using `localStorage`. Nothing is sent to a server, there is no account, no analytics and no cookies. Clearing your site data clears all of it, and it does not sync to your other devices. The guide is a set of static files; the only network requests it makes are for its own stylesheet, its script, and the search index.
Provenance. The 360-chapter structure, the case anchors, the mechanism summaries and the source lists come from a book outline. The prose in each chapter, the routes, the attack-and-defense pairings, the situation tools and the seven counters written for laws with no defense chapter were produced for this guide. Where something was written here rather than taken from the outline, the page says so.