Influence - The Psychology of Persuasion, New and Expanded - Cialdini, Robert B. (2021) (EN)

Résumé de lecture, chapitre par chapitre — rapporte ce que le livre dit.

Reading summary

🔬 Built from _dossiers/Influence - The Psychology of Persuasion, New and Expanded - Cialdini, Robert B. (2021) (EN) — DOSSIER.md
(150 candidates of 2 207) + — SONDES.md. ⚠️ 2 057 candidates cut by the cap.

⚠️ 10 modules of kb/17_Influence_manipulation cite this edition. Nothing here re-verifies them.

⛔ Edition matters here, and the corpus holds two

library/17_Influence_manipulation/ contains two editions: 2009 and 2021 (New and Expanded). Measured: 10 modules cite the 2021, one cites the 2009. This summary is of the 2021.

⚠️ I first built the dossier on the wrong one — a name match collapsed the two editions. Caught before writing. **A module citing Cialdini, Influence without its year is ambiguous in this corpus.**

What kind of book it is

A trade book by the researcher, reporting a large experimental literature — much of it his own — in narrative form. Same status as Make It Stick: the author is a principal investigator, the book is not the study. ⚠️ Several of the classic effects it popularised have been the subject of replication controversy; this corpus has assessed none of that.

Where the book bounds itself

⛔ The ethical distinction the book itself draws — and the one this corpus most needs, since the domain is called Influence / manipulation:

Verified quotations

"It only becomes so when the lever is not a natural feature of the situation but is fabricated by
the practitioner." (probed, 1 occurrence)

⚠️ Cialdini's own line between influence and manipulation is fabrication, not persuasion as such. A module that uses this book to describe manipulation techniques without carrying that distinction misrepresents its source — and the source is explicit.

Verified quotations

"Although such a review of events might be informative, it is not a necessary step in protecting
ourselves from the liking rule." (probed, 1 occurrence)

Verified quotations

"There is no evidence that it did anything of the sort" (probed, 1 occurrence)

⚠️ Said of a real institutional decision taken on a unity/identity rationale. The author applies his own evidential standard to a case that would have flattered his theory, and finds nothing.

⛔ One quotation to handle with care

Verified quotations

"Since 95 percent of the people are imitators and only 5 percent initiators" (probed, 1 occurrence)

⛔ This is Cialdini quoting a sales consultant (Cavett Robert), not a finding. The figure has no study behind it in the file. A module reporting 95 % of people are imitators as social-proof evidence would be citing a sales trainer's slogan as science.

Good for / must not be used for

Good for: the principles as an organising vocabulary; the author's own influence/manipulation distinction; a large set of named studies to go and read.

• ⛔ As primary evidence for any effect — go to the studies it names.
• ⛔ The 95 % figure, and any other quoted slogan, as data.
• ⛔ Citing Cialdini, Influence without the edition year — two are in the corpus.
• ⚠️ Replication status of the classic effects: not assessed by this corpus.

Provenance