Milgram experiment

The Milgram experiment refers to two distinct but related programs of research conducted by social psychologist Stanley Milgram in the 1960s: the famous obedience studies (1961-1963), in which participants were instructed to administer what they believed were increasingly painful electric shocks to a learner, and the small-world studies (1967), in which participants attempted to forward letters to distant targets through acquaintance chains. Though the obedience studies dominate popular memory, the small-world studies are arguably more consequential for the development of network science.

The small-world experiments began in 1967 with a study in which 296 randomly selected residents of Omaha, Nebraska were asked to send a letter to a stockbroker in Boston, Massachusetts, passing it only through acquaintances on a first-name basis. Of the letters that reached their destination, the average path length was approximately six intermediaries. This result — the empirical origin of the "six degrees of separation" concept — demonstrated that the social network of the United States was a small-world network long before the term existed.

Milgram's methodology was ingenious but flawed. The completion rate was low: only 29% of letters reached their target. The sample was small and geographically constrained. The "six degrees" figure was an average of completed chains, not of all possible paths. Later replications using email (Dodds, Muhamad, and Watts, 2003) and digital social networks (Facebook, 2011) confirmed the qualitative finding — short paths in large networks — while revealing that the exact path length depends on network density, communication technology, and measurement methodology.

The obedience studies and the small-world studies share a deeper theme: both reveal how social structure constrains individual behavior in ways that individuals do not perceive. In the obedience studies, the authority structure of the laboratory overrode personal moral judgment. In the small-world studies, the topological structure of the social network enabled coordination that no individual could plan. Milgram's work is a sustained demonstration that social outcomes are not aggregations of individual choices but emergent properties of relational architecture.

The popular reduction of Milgram's small-world finding to the catchphrase "six degrees of separation" has done more harm than good to network science. It transforms a profound empirical discovery about navigable topology into a party trick about celebrity connections. The important finding was not that six is a magic number — it isn't, and the exact value varies with context. The important finding was that large social networks are navigable by local greedy search, a property that requires specific topological structure and has profound implications for decentralized coordination. The party trick has survived; the topology has been forgotten.