WikiSummary, the Social Science Summary Database

Jervis: War and misperception

From WikiSummary, the Free Social Science Summary Database


Jervis. 1988. War and misperception. In The Origin and Prevention of Major Wars, eds. Robert Rotberg and Theodore Rabb, pp. 101-126..

In Brief

  • Y: War (and related behaviors, like bluffing)
  • X: Misperception (especially of intentions, but also of capabilities).

War is most likely if you overestimate others' hostility but underestimate their capabilities. War can occur without misperception, but rarely. (See Fearon 1995 on these same variables.)

Misperception is inaccurate inferences, miscalculations of consequences, and misjudgments about how others will react to one's policies, and may include military optimism, pessimism about long-term diplomatic and military prospects, incorrectly anticipated consequences. Scholars (Jervis included) generally focus on misperceptions of intentions, not situations.

On average, states are more likely to overestimate others' hostility, inferring threatening motives from actions that a disinterested observer would consider at least partially cooperative, and underestimate the extent to which their own actions can be seen as threats. When others do feel threatened and react, the first state views these moves as confirmation of aggression.

Methodological Issues

How do you measure misperception and test these arguments? The obvious method is comparative case studies, but there are measurement problems. Should perceptions be compared to what was later shown to be reality, or to what information was available at the time? How do we determine which perceptions are accurate? If states' true military balance can only be determined by war, it may be impossible to determine states' intentions. Also, statesmen's assessments are often probabilistic, not definitive

Since we prefer to study wars rather than peace, we know little about the degree to which peace is characterized by accurate perceptions

Two Models of Misperception and War

  1. WW I ("spiral") Model: states exaggerate one another's hostility when their differences are, in fact, bridgeable. States will rely on threats to prevent war, not conciliation
    • All participants thought the war would be short, and were optimistic about its outcome--had either side known the truth, they would have bluffed or sought a more limited war
  2. WW II Model: aggressors underestimate the willingness of status quo powers to fight. This model underlies deterrence theory
    • States misjudged capabilities (effectiveness of blitzkrieg, German economic vulnerability, etc.), intentions, and the behavior of neutral countries
    • Without these misperceptions, war would still have occurred. UK/FR would have gone to war earlier; Hitler would have gone about war in a more winnable manner


Jervis then makes it virtually impossible to apply this, by arguing that the existence of a spiral may be a reflection of the underlying conflict, not a cause of it. He also argues that if the initial conflict of interest does not justify a war and it is the process of conflict itself which generates the impulse to fight, misperception may not be the crucial factor

A Nuclear World War III?

Jervis applies his theory to predict whether the US and USSR are likely to engage in a nuclear war. Misperceptions could increase the probability of conflict (US vs. USSR), but it is a game of chicken: nuclear war will not happen if both sides are minimally rational and control their behavior, but will be more likely if both sides conclude war is inevitable

Commitment can inhibit the flexibility to control its behavior, and may be a psychological as well as tactical phenomenon as states become convinced that their policies are morally justified and politically necessary. This in turn can lead to misperceptions: belief that a policy is necessary may lead statesmen to believe it will work.

Perceptions of inevitability are not objective, but are based on each side's perceptions of the other. To keep the peace, a state would have to convince the other that it won't start a war and that it doesn't believe the other will, either.

Comment and Criticism

Similarities to Prospect Theory

Jervis's ideas seem to have much in common with prospect theory.

  • Risk and what states make of it
  • Status quo forms reference point--states willing to take big risks to recoup recent losses
  • Preemptive war (to avoid a loss, state accepts the risk of a greater loss) may occur if players view war as likely but not inevitable. The danger in IR is even greater if both sides feel they are losing, especially if there is a slight first-strike advantage.
  • Exaggerating the danger of crisis instability makes a severe confrontation more dangerous, but if statesmen believe crises are uncontrollable, they will be reluctant to create them, thus keeping states from the brink of war
  • States should design policies that will not fail disastrously even if they are based on incorrect assumptions