Skip to content
SSOCIOSOPHIA

Pillar II — Group Dynamics

Wisdom of Crowds.

Under the right conditions, a crowd's aggregated guesses beat its cleverest member — but the conditions are strict, and when they fail the same crowd becomes a herding machine.

Aggregation of Independent Judgements125+ found this interestingExamine in the index →

01Overview

The wisdom of crowds is the finding that independent, error-prone judgements can be aggregated into something remarkably accurate. Galton's famous 1907 observation that a country fair crowd's median estimate of an ox's weight was within a pound of the truth is the founding anecdote; the modern treatment is Surowiecki's, which sets out the conditions. A crowd is wise when the judgements are diverse, independent, decentralised, and aggregated by a mechanism capable of turning opinion into a single answer. Each condition guards against a specific failure: diversity prevents shared blind spots, independence prevents early errors from cascading, decentralisation draws on private knowledge, and aggregation ensures the signal is not lost. The striking contrast with the herd entries in this archive is deliberate — the same psychology that produces information cascades can produce collective accuracy, and the difference lies almost entirely in whether judgements are made independently before they are combined.

02Key theorists

01Francis Galton (1907)

02James Surowiecki (2004)

03Scott Page (2007)

Galton's letter to Nature reported that the crowd's pooled guess at a livestock competition was more accurate than the estimates of individual experts present. The finding sat largely unused until the 1990s and 2000s, when prediction markets, polling aggregation and online forecasting revived it. Surowiecki's 2004 book gave the conditions their currency. Tetlock's forecasting tournaments later added important qualifications, showing that aggregation works best with people who reason from evidence and update in proportion to it.

03How it works

  1. 01Error cancellation: independent errors point in different directions and cancel when averaged
  2. 02Diversity bonus: varied backgrounds bring information a homogeneous group lacks
  3. 03Independence: uncorrelated judgements prevent one mistake from propagating to all
  4. 04Decentralisation: private, local knowledge is drawn on rather than suppressed
  5. 05Aggregation mechanism: a rule converts many estimates into one, whether a mean, median or market price
  6. 06Condition failure: when independence or diversity collapses, aggregation amplifies error instead of correcting it

04Where it shows up

Prediction markets that outperform pundits; polling averages more stable than any single poll; open-source projects whose quality exceeds any individual contributor's reach; crowdsourced estimates of quantities no expert can measure directly; the flip side — social media trends that are pure cascade and contain no aggregated information at all.

05The propaganda link

The Party would find the wisdom of crowds intolerable, because it depends on independence of judgement — the very quality the regime spends the novel eradicating. Ingsoc replaces aggregation with direction: the crowd is wise only if it has not been told what to think.

06How to resist

  • Preserve independence: collect judgements before people can see or hear each other's
  • Protect diversity and avoid selecting for the same kind of mind
  • Prefer simple aggregation rules, since they are hard to game and resistant to outliers
  • Distinguish crowds that aggregate from crowds that cascade, and check which one you are watching
  • Weight or track accuracy over time when the aggregation is repeated