The precepts in this group are advice rather than findings, and the interesting thing about them is that they do not agree. Each one quietly assumes a goal. Once the goal is named, most of the apparent contradictions between them dissolve into a choice about what the speaker is trying to do.
Four goals, four kinds of rule
The largest cluster comes from Dale Carnegie's How to Win Friends and Influence People (1936): avoid direct criticism, give specific appreciation, let people talk about themselves, and leave an opponent a way to change position without losing face. The goal throughout is to change one person's behaviour while keeping the relationship. Later research on reactance and face has supported the core of several of these, which is more than their critics expected of a self-help book.
The craft rules have a different goal: holding an audience's attention to a work. Chekhov's Gun says every element introduced must be used. Hemingway's Iceberg Theory says the opposite in one sense, that deliberate omission strengthens a story when the author knows what is left out. Show, Don't Tell and the Rule of Three belong here too. They are rules of economy, and they conflict only at the edges.
The rules for disagreement aim at something else again: getting at the truth, or at least keeping a dispute about the right thing. Grice's Maxims describe the cooperative assumptions any conversation relies on. Graham's Hierarchy of Disagreement ranks responses from name-calling up to refuting the central point. The Rationalist Taboo bans a disputed word so that both sides must say what they mean, and Understand-Before-Persuading asks each side to state the other's view first.
Two precepts aim at winning in public, and they sit uneasily beside the rest. Alinsky's Ridicule Rule recommends exactly what Carnegie forbids, and for the same reason: ridicule provokes the defensive reaction Carnegie warns about, and a campaigner may want that reaction on display. Orwell's Political Language as Concealment comes from the other side of the fight, as a rule for noticing when abstraction and the passive voice are doing the arguing.
Where they apply
The applications follow the goals. A manager giving feedback needs Carnegie. A novelist or speechwriter needs the craft rules. A moderator running an online forum, a teacher running a seminar or a committee trying to settle a dispute needs the rules for disagreement, and several of those rules were written for exactly those settings.
The mistakes also follow the goals. Applying the rules for truth-seeking to a relationship can make someone correct and friendless, which Carnegie called the Argument Paradox. Applying Carnegie to a question of fact can leave an error in place because nobody wanted to embarrass the person who made it. Applying Alinsky to a colleague is rarely forgiven.
Where they are argued over
Most precepts in this group have never been tested as stated, and a few invite particular criticism. Carnegie is regularly charged with teaching manipulation, a charge his defenders answer by pointing out that the advice works only if the attention is sincere. The Bechdel Test is a deliberately minimal proxy for representation, and it is criticised both for passing weak films and for failing good ones, which is what a minimal proxy will always do. The 90-9-1 Rule is quoted as though its numbers had been measured for the community at hand when they rarely have.
Why it is worth knowing
Precepts are the form in which communities store what they have learned about talking to each other. Knowing a range of them is less useful than knowing which goal each one serves, because most failures of communication are failures of that choice rather than of technique. The person who tells a grieving friend the facts, or who flatters a colleague out of a correction they needed, has usually followed a good rule for the wrong purpose.
The entries behind this essay: Precept in Communication & Media. Every subject has its essays on the essays page.