first and second Persian Gulf Wars. Nations may also place such different values on the matter in dispute that there is no real overlap in the range of outcomes they will accept. If Beijing, for example, values Taiwan far more highly than Washington because of the symbolic and historical meaning of uniting China, the expected values of the two countries might not make a settlement possible.
Lack of a supranational government makes agreement even more difficult because nations cannot trust the information that they receive. If nations do not know important variables, such as the probabilities of winning a conflict, the value that their rivals place on a contested resource, or expected war costs, they will be unable to decide accurately whether to go to war or to settle. The most important factor in this calculus is a nation’s probability of winning a conflict, which depends on military capabilities and political determination. Information in the public domain, such as military size, defense budgets, and economic growth, can provide some clues about a nation’s military strength. But even these relevant public facts may prove difficult to collect and analyze. During the 1970s and 1980s, for example, the CIA badly mistook the size of the Soviet defense budget and underestimated Moscow’s large amount of spending necessary to keep up with the U.S. The Soviet Union’s quick collapse in 1989-90, therefore, came as a surprise to most of the American national security establishment.61 Even if accurate figures are publicly available, economic growth may not directly translate into military effectiveness because of weakness in military equipment, training, or culture.
Other relevant information will fall primarily within the control of the opponent. The United States, for example, will have private information on the quality of its armed forces and the superiority of its strategies and tactics. Indeed, nations will go to great lengths to conceal military abilities in order to preserve tactical advantages or strategic surprise. The United States keeps performance data on many of its weapons systems classified, which makes it more difficult for the enemy to develop effective countermeasures. Imperial Japan concealed its advances in aircraft carrier operations, which allowed it to project force as far as Hawaii, well beyond American estimates at the time. Nations will also have private information on the political willingness of their leadership, elites, and people to fight. One nation may be willing to suffer vastly higher casualties than the other, which affects their probability of winning a conflict. While the United States suffered about 58,000 deaths in the Vietnam War, North Vietnam and the Viet Cong bore losses estimated at least ten times that number.62
Lack of knowledge of an opponent’s military capabilities and political resolve creates an information asymmetry. Information asymmetries inhibit the reaching of agreements, whether they are domestic contracts or the settlement of international disputes. First, imperfect information will lead to mistakes in bargaining. If nations overestimate their probability of winning a conflict, and correspondingly underestimate their opponent’s odds, they will not realize there is a broader range for agreement. This lack of information will result in less settlement and more war. Second, nations will also have an incentive to bluff. A nation might seek to hide its military abilities in order to gain a tactical or strategic advantage. Or a nation will exaggerate its resources in order to bluff its way to a better deal. Great Britain and France mistook Germany’s capabilities in 1938 and 1939, which allowed Berlin to seize Czechoslovakia and invade Poland without response. Faced with possible bluffs, nations will have few means of gaining credible information about their opponents’ true capabilities. Such uncertainty will undermine the ability to reach a deal.
Third, nations will have few ways to credibly reveal private information. In order to avoid the costs of war, a nation may wish to communicate information on its true capabilities to its opponent. This picture will allow parties to reach a more accurate prediction of a dispute’s outcome, which should smooth the way to a settlement. In domestic litigation, for example, the parties to a lawsuit can reveal information through discovery in federal court that provides credibility. But under conditions of anarchy, nations will have difficulty revealing private information in a credible manner.
More precise, less destructive uses of force can help overcome the obstacles of imperfect information. Coercive measures can signal political will, the value placed on the resources at stake, or military capabilities that could influence the outcome of a broader armed conflict. The more costly the signal, the more credible the information becomes. A nation’s leader can make a threat of war and send military forces near disputed territory or a potential conflict zone. Deployment eats up resources that would go to waste if the nation is bluffing. It also incurs “audience costs” domestically, because a leader will suffer politically if he aggressively deploys force but then backs down.63 Escalating steps of force will provide the opportunity to send more precise signals that gradually consume more resources, reveal more military capability, and edge closer to war. More signaling should reduce the chances of bluffing and reveal more reliable private information.
A good example is the Cuban Missile Crisis. The U.S.S.R. sought a rapid change to the balance of power in its favor by stationing intermediate-range nuclear missiles in Cuba. At this time, the United States had an ample deterrent capable of striking Soviet territory, while Russian forces could not yet match U.S. levels—this would change after the crisis when Moscow embarked on a program to build a large arsenal of ICBMs. President Kennedy decided to prevent the deployment. But before launching an all-out attack on the missiles, the United States effectively sought to force Soviet premier Nikita Khrushchev to agree to cancel the move. Kennedy ordered a limited use of force, a naval blockade of Cuba, which sent a signal of his willingness to use force to remove the missiles, and the U.S.’s military capabilities. He reinforced these messages by placing U.S. nuclear forces on high alert. But using limited coercive means allowed Washington to reach a negotiated settlement with Moscow that avoided the costs of a direct conventional conflict.
As Schelling suggested, the Korean War may provide another example. The United States directly fought against North Korea, which was supported initially by the Soviet Union. After Chinese intervention in December 1950, the United States conducted hostilities directly against another great power. The conflict, however, remained limited both in geography and means. Hostilities never left the Korean peninsula, despite the proximity of U.S. forces in Japan and Chinese bases in Manchuria. Both sides only resorted to conventional weapons, despite a large advantage in U.S. nuclear weapons and delivery systems. After Chinese intervention, both nations engaged in grueling ground combat for two years, even though the front settled early around the original dividing line between North and South Korea. Acceptance of high casualties by both the U.S. and China signaled their unwillingness to accept a peace that deviated from the original 38th parallel. It also demonstrated their credibility in respecting an armistice or cease-fire, because the alternative would be a renewal of costly fighting for little benefit.
If nations engage in such signaling as part of the bargaining over a settlement of their disputes, means of exerting limited force will prove valuable as ways to demonstrate resolve without choosing between complete acquiescence to enemy demands or all-out war. Instead of a naval embargo, or costly ground tactics, the United States could bargain with Russia or China with new types of weapons, such as drones or cyber. New technologies might not prevent a conflict from breaking out, but they will provide more opportunities to reach a negotiated settlement to avoid full great power hostilities. Conversely, limiting the ability to use lower levels of force might have the unintended consequence of rendering war more harmful. A ban on new weapons, for example, could narrow the range of targets and the means of coercion to produce more destructive signaling and ultimately more lethal conflicts. One nation may want to send a signal during a crisis that inflicts a precise cost on its opponent. With a broader set of targets and more levels of harm, the nations can send more discrete signals. But if nations limit their signals to kinetic attacks on military targets, they will have to employ more destructive levels of force. They might develop even more devastating kinetic weapons to produce the same effects as the precision offered by cyber or robotic weapons. Limits on new weapons technology might even destabilize crises by encouraging nations to use offensive weapons early in a crisis which might themselves be vulnerable to attack.64
Take, for example, disabling an opponent’s financial markets or transportation and communications networks. During the Kosovo War, the United States Air Force dropped graphite on Belgrade’s electrical grid, which temporarily