What Is Groupthink? Why Capable Groups Make Bad Decisions
By the BrainSnail editorial team. How these articles are written and checked, and how to tell us when one is wrong.
In April 1961 the most talented group of advisers ever assembled in the White House approved a plan to land 1,400 Cuban exiles at the Bay of Pigs and overthrow Fidel Castro, and it failed within three days. Afterwards several of them said they had had doubts and not voiced them. The psychologist Irving Janis read the accounts and recognised a pattern he had seen in other disasters, in which a close-knit group of intelligent people, wanting to agree, stops examining what it is about to do. He called it groupthink, and the word has since entered the language.
How it happens
Janis's account, published in 1972, began from the observation that the groups involved were not stupid, lazy or corrupt. They were cohesive, they respected their leader, they were under pressure to decide, and they were insulated from outside opinion. In those conditions the wish to preserve the group's good feeling becomes stronger than the wish to get the answer right. Members who doubt keep quiet, taking others' silence as agreement; the leader's early preference sets the direction; and the group comes to believe that its unanimity is evidence of its correctness, when it is evidence only that nobody spoke.
The symptoms
Janis listed eight signs, and the list has been used since as a checklist in boardrooms and cockpits:
- •An illusion of invulnerability, encouraging excessive risk
- •Collective rationalisation of warnings that should prompt reconsideration
- •An unquestioned belief in the group's own morality
- •Stereotyped views of opponents as too weak or stupid to matter
- •Direct pressure on any member who argues against the consensus
- •Self-censorship of doubts
- •An illusion of unanimity, silence being read as consent
- •Self-appointed mindguards who shield the group from troubling information
The cases
Janis built the theory on American foreign policy: the attack on the Pacific Fleet in Hawaii in 1941, where commanders dismissed warnings; the Bay of Pigs; the escalation of the Vietnam War. Later writers added the decision to launch the space shuttle Challenger in January 1986, when engineers' warnings that the seals would fail in cold weather were overridden by managers who had convinced each other the risk was acceptable, and the Columbia loss seventeen years later, where the same organisation repeated the pattern. Corporate collapses such as Enron and Swissair and the intelligence assessments before the 2003 invasion of Iraq, which a Senate committee explicitly described as groupthink, have been analysed in the same terms.
The counter-example Janis used was the Cuban missile crisis of 1962, eighteen months after the Bay of Pigs, in which the same president deliberately changed how the same advisers worked: he absented himself from meetings so they would speak freely, split them into subgroups to develop rival plans, brought in outsiders, and assigned his brother to argue against every proposal. The result was a decision most historians regard as sound, reached by a group that had recently made a disastrous one.
What the research says
The theory is more popular than it is proven. Laboratory studies that try to produce groupthink by making groups cohesive and stressed get mixed results, and critics note that Janis chose his cases after the fact, from decisions already known to have failed, so that the symptoms could be found in them. What has held up is the underlying mechanism: groups do tend to polarise, moving towards a more extreme version of their initial lean, and members do suppress private doubts when they believe the rest agree. The related work on conformity by Solomon Asch and on obedience by Stanley Milgram shows how powerful the pressure to go along is even among strangers, and a cohesive group under a respected leader supplies more of it.
The cures
Janis's remedies were the ones Kennedy stumbled on, and they have become standard practice in organisations that take decisions seriously. The leader withholds their view until others have spoken. Someone is assigned the role of devil's advocate and required to make the strongest case against the plan. Outside experts are brought in, and members are encouraged to test the plan on trusted colleagues outside the group. Rival subgroups develop alternatives independently. And after a decision is reached, a second-chance meeting is held at which every doubt is aired before it is final. Aviation added the crew resource management training that followed the 1977 Tenerife disaster, in which a co-pilot's unvoiced doubt cost 583 lives, and which teaches junior crew to challenge a captain and captains to listen. The common principle is that dissent has to be manufactured, because in a group that likes itself it will not arise on its own.
The takeaway
Groupthink is the pattern, described by Irving Janis in 1972, in which a cohesive group under pressure suppresses disagreement to preserve harmony, mistakes silence for consent, and makes a worse decision than its members would individually; its signs include an illusion of invulnerability, self-censorship and pressure on dissenters. It is countered by leaders who speak last, assigned devil's advocates, outside opinion, rival subgroups and a second look before a decision is final.