Consensus fabrication presents a false picture of widespread agreement — making a contested, minority, or purchased position look settled and universal. It works because agreement is a shortcut: if everyone (or every expert) already concurs, the individual reasoning feels unnecessary. Manufacture the appearance of that agreement and you can borrow its persuasive force without earning it. Walter Lippmann’s account of how public opinion is shaped, later sharpened in Herman and Chomsky’s “manufacturing consent,” describes the large-scale version; the same move runs in a boardroom or a comment thread.
Two mechanisms give it teeth. Repetition drives the illusory-truth effect — Hasher, Goldstein, and Toppino (1977) showed that statements simply feel more true the more often we encounter them, independent of their accuracy — so a claim echoed across many outlets acquires unearned credibility. And coordinated or fake “independent” voices, often supplied by astroturfing (→ T3.14), simulate the many disinterested agreers that real consensus would produce. Dissent is then hidden or framed as fringe, so the manufactured agreement looks total.
Because the whole effect depends on you not checking, sampling independent sources is the core defense. Ask who actually agrees, how many, and how you can verify it — a genuine consensus can answer those questions, while a fabricated one deflects. Treat near-identical wording across supposedly independent sources as a coordination tell rather than corroboration, and look for the missing dissent: ask what credible disagreement exists and why it isn’t in front of you. Finally, discount your own sense of familiarity — repeated exposure is not evidence, and recognizing that is what disarms the illusory-truth lever.