1
GROUPTHINK
by Irving Janis (early draft)
"How could we have been so stupid?" President John F. Kennedy asked after he and a close
group of advisors had blundered into the Bay of Pigs invasion. For the last two years I have been
studying that question, as it applies not only to the Bay of Pigs decision-makers but also to those
who led the United States as such other major fiascos as the failure to be prepared for the attack on
Pearl Harbor, the Korean War, and the escalation of the Vietnam War.
Stupidity is certainly not the answer. The men who participated in making the Bay of Pigs
decision, for example, comprised one of the greatest arrays of intellectual talent in the history of the
American Government--Dean Rusk, Robert McNamara, Robert Kennedy, Douglas Dillon, McGeorge
Bundy, Arthur Schlesinger, Jr., Allen Dulles, and others.
It also seemed to be that explanations were incomplete if they concentrated only on
disturbances in the behavior of each individual within a decision-making body: temporary emotional
states of elation, fear, or anger that reduce a man's mental efficiency, for example, or chronic blind
spots arising from a man's prejudices or idiosyncratic biases.
I preferred to broaden the picture by looking at the fiascos from the standpoint of group
dynamics as it has been explored over the past three decades, first by the great social psychologist,
Kurt Lewin, and later in many experimental situations by myself and other behavioral scientists. My
conclusion after poring over hundreds of relevant documents--historical reports about formal group
meetings and informal conversations among the members--is that the groups that committed the
fiascos were the victims of what I call "groupthink."
"Groupy." In each case study, I was surprised to discover the extent to which each group
displayed the typical phenomena of social conformity that are regularly encountered in studies of
group dynamics among ordinary citizens...For example, some of the phenomena appear to be
completely in line with findings from social-psychological experiments showing that powerful social
pressures are brought to bear by the members of a cohesive group whenever a dissident begins to
voice his objections to a group consensus. Other phenomena are reminiscent of the shared
illusions observed in encounter groups and friendship cliques when the members simultaneously
reach a peak of "groupy" feelings.
Above all, there are numerous indications pointing to the development of group norms that
bolster morale at the expense of critical thinking. One of the most common norms appears to be
that of remaining loyal to the group by sticking with the policies to which the group has already itself,
even when those policies are obviously working out badly and have unintended consequences
which disturb the conscience of each member. This is one of the key characteristics of groupthink.
1984. I use the term groupthink as a quick and easy way to refer to the mode of thinking that
persons engage in when concurrence-seeking becomes so dominant in a cohesive in-group that it
tends to override realistic appraisal of alternative courses in action. Groupthink is a term of the
same order as the word in the newspeak vocabulary of George Orwell used in his dismaying world
of 1984. In that context, groupthink takes on invidious connotation. Exactly such a connotation is
2
intended, since the term refers to the deterioration in mental efficiency, reality testing, and moral
judgments as a result of group pressures.
The symptoms of groupthink arise when the members of decision-making groups become
motivated to avoid being too harsh in their judgments of their leaders' or their colleagues' ideas.
They adopt a soft line of criticism, even in their own thinking. At their meeting all the members are
amiable and seek a complete concurrence on every important issue, with no bickering or conflict to
split the cozy, "we-feeling" atmosphere.
Kill. Paradoxically, soft headed groups are often hard-hearted when it comes to dealing with outgroup
or enemies. They find it relatively easy to resort to dehumanizing solutions - they will readily
authorize bombing attacks that kill large numbers of civilians in the name of the noble cause of
persuading an unfriendly government to negotiate at the peace table. They are unlikely to pursue
the more difficult and controversial issues that arise when alternatives to harsh military solution
come up for discussion. Nor are they inclined to raise ethical issues that carry the implication that
this fine group of ours, with its humanitarianism and its high minded principles, might be capable of
adopting a course of action that is inhumane and immoral.
Norms. There is evidence from a number of social-psychological studies that as the members of a
group feel more accepted by the others, which is a central feature of increased group cohesiveness,
they display less overt conformity to group norms. Thus we could expect that the more cohesive a
group becomes, the less the members will feel constrained to censor what they say out of fear of
being socially punished for antagonizing the leader or any of their fellow members.
In contrast, the groupthink type of conformity tends to increase as group cohesiveness
increases. Groupthink involves non-deliberate suppression of critical thought as a result of
internalization of the group's norms, which is quite different from deliberate suppression on the basis
of external threats of social punishment. The more cohesive the group, the greater the inner
compulsion on the part of each member to avoid creating disunity, which inclines him to believe in
the soundness of whatever proposals are promoted by the leader or by a majority of the group's
members.
3
Stress. I do not mean to imply that all cohesive groups necessarily suffer from groupthink. All ingroups
may have a mild tendency toward groupthink, displaying one or another of the symptoms
from time to time, but it need not be so dominant as to influence the quality of the group's final
decision. Neither do I mean to imply that there is anything necessarily inefficient or harmful about
group decisions in general. On the contrary, a group whose members have properly defined roles,
with traditions concerning the procedures to follow in pursuing a critical inquiry, probably is capable
of making better decisions than any individual member working alone.
The problem is that the advantages of having decisions made by groups are often lost
because of powerful psychological pressures that arise when all members work closely together,
share the same set of values, and, above all, face a crisis situation that puts everyone under intense
stress.
The main principle of groupthink, which I offer in the spirit of Parkinson's law is this: The
more amiability and esprit de corps there is among the members of a policy-making in-group, the
greater the danger that the independent critical thinking will be replaced by groupthink, which is
likely to result in irrational and dehumanizing actions directed against out-group.
Symptoms. In my studies of high-level governmental decision-makers, both civilian and military,
I have found eight main symptoms of groupthink.
1. INVULNERABILITY. Most or all of the members of the in-group share an illusion of
invulnerability that provides for them some degree of reassurance about obvious dangers and leads
them to become over-optimistic and willing to take extraordinary risks. It also causes them to fail to
respond to clear warnings of danger.
The Kennedy in-group, which uncritically accepted the Central Intelligence Agency's
disastrous Bay of Pigs plan, operated on the false assumption that they could keep secret the fact
that the United States was responsible for the invasion of Cuba. Even after the news of the plan
began to leak out, their belief remained unshaken. They failed to consider the danger that awaited
them: a worldwide revulsion against the United States.
A similar attitude appeared among members of President Lyndon B. Johnson's in-group, the
"Tuesday Cabinet," which kept escalating the Vietnam War despite repeated setbacks and failures.
"There was a belief," Bill Moyers commented after he resigned, "that if we indicated a willingness to
use our power, they (the North Vietnamese) would get the message and back away from an all-out
confrontation...There was a confidence - it was never bragged about, it was just there - that when
the chips were really down, the other people would fold."
4
A most poignant example of an illusion of invulnerability involves the in-group around Admiral
H. E. Kimmel, which failed to prepare for the possibility of a Japanese attack on Pearl Harbor
despite repeated warnings. Informed by his intelligence chief that radio contact with Japanese
aircraft carriers had been lost, Kimmel joked about it: "What, you don't know where the carriers are?
Do you mean to say that they could be rounding Diamond Head (at Honolulu) and you wouldn't
know it?" The carriers were in fact moving full-steam toward Kimmel's command post at the time.
Laughing together about a danger signal, which labels it as a purely laughing matter, is a
characteristic manifestation of groupthink.
2. RATIONALE. As we see, victims of groupthink ignore warnings: they also collectively
construct rationalizations in order to discount warnings and other forms of negative feedback that,
taken seriously, might lead the group members to reconsider their assumptions each time they
recommit themselves to past decisions. Why did the Johnson in-group avoid reconsidering its
escalation policy when time and again the expectation on which they based their decisions turned
out to be wrong? James C. Thompson, Jr., a Harvard historian who spent five years as an
observing participant in both the State Department and the White House, tells us that the policymakers
avoided critical discussion of their prior decisions and continually invented new
rationalizations so that they could sincerely recommit themselves to defeating the North Vietnamese.
In the fall of 1964, before the bombing of North Vietnam began, some of the policy-makers
predicted that six weeks of air strikes would induce the North Vietnamese to seek peace talks.
When someone asked, "What if they don't?" the answer was that another four weeks certainly would
do the trick.
Later, after each setback, the in-group agreed that by investing just a bit more effort (by
stepping up the bomb tonnage a bit, for instance), their course of action would prove to be right.
The Pentagon Papers bear out these observations.
In The Limits of Intervention, Townsend Hoopes, who was acting secretary of the Air Force
under Johnson, says that Walt W. Rostow in particular showed a remarkable capacity for what has
been called "instant rationalization." According to Hoopes, Rostow buttressed the group's optimism
about being on the road to victory by culling selected scraps of evidence from new reports or, if
necessary, by inventing "plausible" forecasts that had no basis in evidence at all.
Admiral Kimmel's group rationalized away their warnings, too. Right up to December 7,
1941, they convinced themselves that the Japanese would never dare attempt a full-scale surprise
assault against Hawaii because Japan's leaders would realize that it would precipitate an all-out war
which the United States would surely win. They made no attempt to look at the situation through the
eyes of the Japanese leaders - another manifestation of groupthink.
5
3. MORALITY. Victims of groupthink believe unquestionably in the inherent morality of their
in-group. This belief inclines the members to ignore the ethical and moral consequences of their
decisions.
Evidence that this symptom is at work usually is of a negative kind - the things that are left
unsaid in group meetings. At least two influential persons had doubts about the morality of the Bay
of Pigs adventure. One of them, Arthur Schlesinger, Jr., presented his strong objections in a
memorandum to President Kennedy and Secretary of State Rusk but suppressed them when he
attended meetings of the Kennedy team. The other, Senator J. William Fulbright, was not a
member of the group, but the President invited him to express his misgivings in a speech to the
policy-makers. However, when Fulbright finished speaking, the President moved to another agenda
item without asking for reactions of the group.
David Kraslow and Stuart H. Loory, in The Secret Search for Peace in Vietnam reported that
during 1966, President Johnson's in-group was concerned primarily with selecting bomb targets in
North Vietnam. They based their selection on four factors - the military advantage, the risk to
American aircraft and pilots, the danger of forcing other countries into fighting, and the danger of
heavy civilian casualties. At their regular Tuesday luncheons, they weighed these factors the way
school teachers grade examination papers, averaging them out. Though evidence on this point is
scant, I suspect that the group's ritualistic adherence to a standardized procedure induced members
to feel morally justified in their destructive way of dealing with the Vietnamese people - after all, the
danger of heavy civilian casualties from U.S. air strikes was taken into account on their checklist.
4. STEREOTYPES. Victims of groupthink hold stereotypes views of the leaders of enemy
groups: they are so evil that genuine attempts at negotiating differences with them are unwarranted,
or they are too stupid or too weak to deal effectively with whatever attempts the in-group makes to
defeat their purposes, no matter how risky the attempts are.
Kennedy's group thinkers believed that Premier Fidel Castro's air force was so ineffectual
that obsolete B-26s could knock it out completely in a surprise attack before the invasion began.
They also believed that Castro's army was so weak that a small Cuban-exile brigade could establish
a well protected beachhead at the Bay of Pigs. In addition, they believed that Castro was not smart
enough to put down possible internal uprisings in support of the exiles. They were wrong on all
three assumptions. Though much of the blame was attributable to faulty intelligence, the point is
that none of Kennedy's advisors even questioned the CIA planners about these assumptions.
The Johnson advisors' sloganistic thinking about "the Communist apparatus" that was
working all, around the world (as Dean Rusk put it), led them to overlook the powerful nationalistic
strivings of the North Vietnamese government and its efforts to ward off Chinese domination. The
crudest of all stereotypes used by Johnson's inner circle to justify their policies was the domino
theory ("If we don't stop the Reds in South Vietnam, tomorrow they will be in Hawaii and next week
they will be in San Francisco," Johnson once said). The group so firmly accepted this stereotype
that it became almost impossible for any advisor to introduce a more sophisticated viewpoint.
In the documents of Pearl Harbor, it is clear to see that the Navy commanders stationed in
Hawaii had a naive image of Japan as a midget that would not dare to strike a blow against a
powerful giant.
6
5. PRESSURE. Victims of groupthink apply direct pressure to any individual who
momentarily expresses doubt about any of the group's shared illusions or who questions the validity
of the arguments supporting a policy alternative favored by the majority. This gambit reinforces the
concurrence-seeking norm that loyal members are expected to maintain.
President Kennedy was probably more active than anyone else in raising skeptical questions
during the Bay of Pigs meeting, and yet he seems to have encouraged the group's docile, uncritical
acceptance of defective arguments in favor of the CIA's plan. At every meeting, he allowed the
CIA's representatives to dominate the discussion. He permitted them to give their immediate
refutations in response to each tentative doubt that one of the others expressed, instead of asking
whether anyone shared the doubt or wanted to pursue the implications of the new worrisome issue
that had just been raised. And at the most crucial meeting, when he was calling on each member to
give his vote for or against the plan, he did not call on Arthur Schlesinger, the one man there who
was known by the President to have serious misgivings.
Historian Thomson informs us that whenever a member of Johnson's in-group began to
express doubts, the group used subtle social pressure to "domesticate" him. To start with, the
dissenter was made to feel at home provided that he lived up to two restrictions: 1) that he did not
voice his doubts to outsiders which would play into the hands of the opposition; and 2) that he kept
his criticisms within the bounds of acceptable deviation, which meant not challenging any of the
fundamental assumptions that went into the group's prior commitments. One such domesticated
dissenter was Bill Moyers. When Moyers arrived at a meeting, Thomson tells us, the President
greeted him with, "Well, here comes Mr. Stop-the-Bombing."
6. SELF-CENSORSHIP. Victims of groupthink avoid deviating from what appears to be
group consensus; they keep silent about their misgivings and even minimize the importance of their
doubts.
As we have seen, Schlesinger was not at all hesitant about presenting his strong objections
to the Bay of Pigs plan in a memorandum to the President and his Secretary of State. But he
became keenly aware of his tendency to suppress objections at the White House meetings. "In the
months after the Bay of Pigs, I bitterly reproached myself for having kept so silent during those
crucial discussions in the cabinet room," Schlesinger writes in A Thousand Days. "I can only
explain my failure to do more than raise a few timid questions by reporting that one's impulse to blow
the whistle on this nonsense was simply undone by the circumstances of the discussion."
7. UNANIMITY. Victims of groupthink share an illusion of unanimity within the group
concerning almost all judgments expressed by members who speak in favor of the majority view.
This symptom results partly from the preceding one, whose affects are augmented by the false
assumption that any individual who remains silent during any part of the discussion is in full accord
with what the others are saying.
When a group of persons who respect each others' opinions arrives at a unanimous view,
each member is likely to feel that the belief must be true. He arrives at a unanimous view; each
member is likely to believe that the belief must be true. This reliance on consensual validation within
the group tends to replace individual critical thinking and reality testing, unless there are clear-cut
disagreements among the members. In contemplating a course of action, such as the invasion of
Cuba, it is painful for the members to confront disagreements within their group, particularly if it
becomes apparent that there widely divergent views about whether the preferred course of action is
7
too risky to undertake at all. Such disagreements are likely to arouse anxieties about making a
serious error. Once the sense of unanimity is shattered, the members can no longer feel
complacently confident about the decision they are inclined to make. Each man must then face the
annoying realization that there are troublesome uncertainties, and he must diligently seek out the
best information he can get in order to decide for himself exactly how serious the risks might be.
This is one of the unpleasant consequences of being in a group of hardheaded critical thinkers.
To avoid such an unpleasant state, the members often become inclined, without quite
realizing it, to prevent latent disagreements from surfacing when they are about to initiate a risky
course of action. The group leaders and the members support each other in playing up the areas of
convergence in their thinking, at the expense of fully exploring divergences that might reveal
unsettled issues.
"Our meeting took place in a curious atmosphere of assumed consensus,” Schlesinger
writes. His additional comments clearly show that, curiously, the consensus was an illusion--an
illusion that could be maintained only because the major participants did not reveal their own
reasoning or discuss their idiosyncratic assumptions and vague reservations. Evidence from
several sources makes it clear that even the three principals--President Kennedy, Rusk, and
McNamara--had widely differing assumptions about the invasion plan.
8. MINDGUARDS. Victims of groupthink sometimes appoint themselves as mindguards to
protect the leader and fellow member from adverse information that might break the complacency
they shared about the effectiveness and morality of past decisions. At a large birthday party for his
wife, Attorney General Robert F. Kennedy, who had been constantly informed about the Cuban
invasion plan, took Schlesinger aside and asked him why he was opposed. Kennedy listened coldly
and said, "You may be right or you may be wrong, but the President has made up his mind. Don't
push it any further. Now is the time for everyone to help him all they can."
Rusk also functioned as a highly effective mindguard by failing to transmit to the group the
strong objections of three "outsiders" who had learned of the invasion plan--Undersecretary of State
Chester Bowles, USIA Director Edward R. Murrow, and Rusk's intelligence chief, Roger Hilsman.
Had Rusk done so, their warnings might have reinforced Schlesinger's memorandum and jolted
some of Kennedy's in-group, if not the President himself, in not reconsidering the decision.
Products. When a group of executives frequently have most or all of these symptoms, a detailed
study of their deliberations is likely to reveal a number of immediate consequences. These
consequences are, in effect, products of poor decision making practices because they lead to
inadequate solutions to the problems under discussion.
First, the group limits its discussions to a few alternative courses of action (often only two)
without an initial survey of all the alternatives that might be worthy of consideration.
Second, the group fails to reexamine the course of action initially preferred by the majority
after they learn the risks and drawbacks they had not considered originally.
Third, the members spend little or no time discussing whether there are non-obvious gains
they may have overlooked or ways of reducing the seemingly prohibitive costs that made rejected
alternatives appear undesirable to them.
8
Fourth, members make little or no attempt to obtain information from experts within their own
organizations who might be able to supply more precise estimates of potential losses and gains.
Fifth, members show positive interest in facts and opinions that support their preferred policy;
they tend to ignore facts and opinions that do not.
Sixth, members spend little time deliberating about how the chosen policy might be hindered
by bureaucratic inertia, sabotaged by political opponents, or temporarily derailed by common
accidents. Consequently, they fail to work out contingency plans to cope with foreseeable setbacks
that could endanger the overall success of their chosen course.
Support. The search for explanation of why groupthink occurs has led me through a quagmire of
complicated theoretical issues in the murky area of human motivation. My belief, based on recent
social psychological research, is that we can best understand the various symptoms of groupthink
as a mutual effort among the group members to maintain self-esteem and emotional equanimity by
providing social support to each other, especially at times when they share responsibility for making
vital decisions.
Even when no important decision is pending, the typical administrator will begin to doubt the
wisdom and morality of his past decisions every time he receives information about setbacks,
particularly if the information is accompanied by negative feedback from prominent men who
originally had been his supporters. It should not be surprising, therefore, to find that individual
members strive to develop unanimity and esprit de corps that will help bolster each other's morale,
to create and optimistic outlook about the success of pending decisions and to reaffirm the positive
value of past policies to which all of them are committed.
Pride. Shared illusions of invulnerability, for example, can reduce anxiety about taking risks.
Rationalizations help members believe that the risks are not really so bad after all. The assumption
of inherent morality helps the members to avoid feelings of shame or guilt. Negative stereotypes
function as stress reducing devices to enhance a sense of moral righteousness as well as pride in a
lofty mission.
The mutual enhancement of self-esteem and morale may have functional value in enabling
members to maintain their capacity to take action, but it has maladaptive consequences insofar as
concurrence-seeking tendencies interfere with critical, rational capacities and lead to serious errors
in judgment.
While I have been limited by study to decision-making bodies in government, groupthink
symptoms may appear in business, industry and any other field where small, cohesive groups make
the decisions. It is vital, then, for all sorts of people--and especially group leaders--to know what
steps they can take to prevent groupthink.
9
Remedies. To counterpoint my case of the major fiascos, I have also investigated two highly
successful group enterprises, the formulation of the Marshall Plan in the Truman Administration and
the handle of the Cuban missile crisis by President Kennedy and his advisors. I have found it
instructive to examine the steps Kennedy took to change his group's decision-making processes.
These changes ensured that the mistakes made by his Bay of Pigs in-group were not repeated by
the missile-crisis in-group, even though the membership of both groups was essentially the same.
The following recommendations for preventing groupthink incorporate many of the good
practices I discovered to be characteristic of the Marshall Plan and missile-crisis groups:
1. The leader of the policy-forming group should assign the role of critical evaluator to each
member, encouraging the group to give high priority to open airing of objections and doubts. This
practice needs to be reinforced by the leader's acceptance of criticism of his own judgments in order
to discourage members from soft-pedaling their disagreements and from allowing their striving for
concurrence to inhibit critical thinking.
2. When the key members of the hierarchy assign a policy-planning mission to any group
within their organization, they adopt an impartial stance instead of stating preferences and
expectations at the beginning. This will encourage open inquiry and impartial probing of a wide
range of policy alternatives.
3. The organization should routinely set up several outside policy-planning and evaluating
groups to work on the same policy question, each deliberating under a different leader. This can
prevent the insulation of an in-group.
4. At intervals before the group reaches a final consensus, the leader should require each
member to discuss the group's deliberations with associates in his own unit of the organization--
assuming that those associates can be trusted to adhere to the same security regulations that
govern the policymakers--and then to report back their reactions to the group.
5. The group should invite one or more outside experts to each meeting on a staggered
basis and encourage the experts to challenge the views of the core members.
6. At every general meeting of the group, whenever the agenda calls for an evaluation of
policy alternatives, at least one member should play devil's advocate, functioning as a good lawyer
in challenging the testimony of those who advocate the majority position.
7. Whenever the policy issue involves relations with a rival nation or organization, the group
should devote a sizeable block of time, perhaps an entire session, to a survey of all warning signals
from the rivals and should write alternative scenarios on the rivals' intentions.
8. When the group is surveying policy alternatives for feasibility and effectiveness, it should
from time to time divide into two or more subgroups to meet separately, under different chairmen,
and then come back together to hammer out differences.
9. After reaching a preliminary consensus about what seems to be the best policy, the group
should hold a "second-chance" meeting at which every member expresses as vividly as he can all
his residual doubts, and rethinks the entire issue before making a definite choice.
10
How. These recommendations have their disadvantages. To encourage the open airing of
objections, for instance, might lead to prolonged and costly debates when a rapidly crisis requires an
immediate solution. It could also cause rejection, depression and anger. A leader's failure to set up
a norm might create cleavage between leader and members that could develop into a disruptive
power struggle if the leader looks on the emerging consensus as anathema. Setting up outside
evaluation groups might increase the risk of security leakage. Still, inventive executives whew know
their way around the organizational maze probably can figure out how to apply one of another of the
prescriptions successfully, without harmful side effects.
They also could benefit from the advice of outside experts in the administrative and
behavioral sciences. Though these experts have much to offer, they have had few chances to work
on policy-making machinery within large organizations. As matters now stand, executives innovate
only when they need new procedures to avoid repeating serious errors that have deflated their self-
images.
In this era of atomic warheads, urban disorganization and eco-catastrophes, it seems to me
that policymakers, should collaborate with behavioral scientists and give top priority to preventing
groupthink and its attendant fiascos.