Book Search:  


Google full text of our books:


Deterrence by Diplomacy
Anne E. Sartori

Book Description | Reviews | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2005, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format

Chapter 1


This book is about deterrence and diplomacy. An attempt at deterrence is a counter-threat. It is a threat by one state to use military force if another state does not refrain from some undesired, threatened action. For example, a state may threaten to invade another state or to subject it to an economic blockade. The threatened state may attempt deterrence by counter-threatening to start a war if the first state goes through with the invasion or blockade.

At a casual glance, it is not clear that deterrent threats ever should succeed, because states that do not intend to fight have strong incentives to bluff. In the early 1990s, prior to NATO military intervention in the Yugoslav War, the Bush and Clinton administrations, the European Community, and the United Nations threatened to impose sanctions and then to use force to influence the behavior of Serb president Slobodan Milosevic. If the threats had worked as intended, they would have prevented the slaughter of many civilians and perhaps made NATO participation in the war unnecessary (Keesing's Contemporary Archives 1992, 1993; Silber and Little 1997). The United States and others had an incentive to claim a willingness to take actions that they were, in fact, proven unwilling to take, at least in the short run. Similarly, when Turkey threatened to invade Cyprus in June 1964, President Johnson sent a "brutal note" to deter this attack.1 This time, deterrence worked. But Johnson, too, had an incentive to bluff if he did not intend to take action; a successful bluff would have changed Turkey's behavior in exactly the same way as a genuine threat. When states' leaders hear a threat, they must ask themselves, "Is this a bluff, or is my adversary truly willing to do what its leaders say?"

As the saying goes, talk is cheap. So why do leaders pay so much attention to each other's threats? Why does deterrence ever succeed? When is it more likely to work, and when is it more likely to fail?

In this book, I study deterrence as a form of diplomacy. Diplomacy is the use of language and other signals by one state in an attempt to convey information to another. It is a kind of communication--the use of language by representatives of one state, aimed at influencing the actions of one or more others. Deterrence is the use of a particular subset of language--deterrent threats--to attempt to convey the information that a state is willing to fight over a disputed issue or issues.2 Thus, deterrent threats are a form of diplomacy.

I study deterrence in order to shed light on the question of why diplomacy so often is effective. Studying deterrence leads me to a theory of communication between states. Thinking of deterrence as a form of talk also leads to insights into why deterrence itself is effective and when and why it is likely to succeed.3

Many of the existing books about diplomacy are written by professional diplomats, whose studies are concerned primarily with the activities of the diplomatic corps (e.g., Nicolson 1939; de Callières 1919; Stearns 1996). Like many of those authors, I see a distinction between the making of foreign policy on the one hand, and diplomacy on the other; diplomacy concerns the implementation of foreign policy.4 However, I would argue that diplomacy includes the work of top foreign-policy leaders as well as of members of the diplomatic corps. A nation's top foreign-policy leadership formulates and implements policy. For example, when a foreign minister announces to the press that her country is likely to join a war, she is implementing a policy (which may be broader than the decision to join the war). Such a statement never is purely policy formulation.

Deterrence is an important phenomenon in its own right. States sometimes attain their foreign-policy goals, while at other times they achieve little of what they desire. States' allies sometimes remain safe, while at others they are conquered. States sometimes themselves lose territory, at others gain. When states gain territory, this process may be peaceful, or it may involve a long and bloody war.

In order to understand the foundations of foreign-policy success and the causes of war, we must understand what leads states to try and to succeed in deterrence. To change the status quo, a state usually makes a demand of another state. If one state makes a demand accompanied by a threat and the recipient of the demand does not try deterrence, then the first state gains something and the second loses something. The most famous example of "deterrence not tried" is Britain's and France's acquiescence to Hitler's demands at Munich. If the threatened state tries to deter an attack but fails, then either the states go to war, or the aggressor gains at the expense of its foe. However, if the threatened state tries to deter an attack and succeeds, then the state that is trying to change the status quo fails to attain its goal. To understand which states get what, and when, one must study deterrence. Similarly, one must study deterrence to know what leads to its failure and to war.

Though the details of this book are about deterrence, statesmen and members of the diplomatic corps face the same credibility problem when they engage in other forms of diplomacy. Diplomacy is particularly necessary when states' interests are not perfectly aligned, when leaders have something to "work out" and often something to hide. For example, in trade, environmental, or arms control negotiations, diplomats usually have an incentive to dissemble about precisely how much their state is willing to give up; pretending to be unmovable can result in a more favorable bargain. Negotiating partners must ask, "Is this a bluff, or is my negotiating partner truly willing to walk away from the table if s/he does not get this bargain? Should I believe this promise, or does my negotiating partner have some ulterior motive for making it? Will s/he renege somewhere down the line?"

For this reason, it is not clear that diplomacy ever should be an effective tool of state. Yet leaders spend hours in conversation, and most countries employ extensive diplomatic corps. Why does diplomacy ever succeed? When is it more and less likely to work?

Talk is cheap, but it has repercussions. This book argues that states are able to use deterrence effectively precisely because it is so valuable. States often are tempted to bluff, or dissemble, but a state that is caught bluffing acquires a reputation for doing so, and opponents are less likely to believe its future communications. With less credibility, a state is less likely to get its way in future disputes. Thus, the prospect of acquiring a reputation for bluffing--and reducing the credibility of its future deterrent threats--keeps a state from bluffing except when doing so is most tempting. In other words, to maintain their ability to use deterrent threats in future disputes, state leaders often use them honestly today. They sometimes even acquiesce to opponents' demands; they do so in order to stress that they are serious on those occasions when they insist they will not acquiesce.

As with deterrence, states are able to use diplomacy effectively because it is so valuable. The prospect of acquiring a reputation for lying--and lessening the credibility of the state's future diplomacy--keeps statesmen and diplomats honest except when fibs are the most tempting. That is, to maintain their ability to use diplomacy in the future, representatives of states usually use it honestly.

The explanation of deterrence and diplomacy that I propose in this book has implications about who gets what, and when. Contrary to much of the literature on international relations, my theory implies that diplomacy often is effective. It also implies that deterrence is a tool for the weak, as well as the strong. Thus, weak states, too, often can use threats to attain their goals. While some theories of deterrence suggest that states must fight to gain credibility for their threats, this book suggests that the honest use of diplomacy is an alternative road to credibility.

Three Misconceptions About Diplomacy

Many scholars of international relations write that diplomacy works. Nevertheless, I see the implications of the international-relations literature as largely pessimistic about the prospects for effective diplomacy. This pessimism comes in part from three misconceptions: (1) that the effectiveness of a threat is equivalent to the communicator's military strength and/or its resolve, or willingness to fight; (2) that "cheap talk" is ineffective; and (3) that disbelief requires a psychological explanation. While I am not the first to argue against these ideas, they nevertheless remain prevalent. This section discusses them in turn.

The Effectiveness of the Communication Is Equivalent to a Communicator's Capabilities and/or Resolve

The first misconception--that the effectiveness of a threat is equivalent to the communicator's military strength and/or its resolve--arises from a school of thought called deterrence theory and from the broader, related school referred to as realism. Both of these schools of thought emphasize the importance of military force and/or power. Each consists of a large number of loosely related works.5 Here, I merely discuss a few fairly common themes that are relevant to my arguments; I do not attempt to summarize the work of all scholars in these literatures.

According to what is often called "rational" deterrence theory, state leaders are rational actors who, in a crisis, try to decide whether or not war is in their interest.6 They do so, according to the theory, by calculating the expected costs and benefits of going to war and of remaining at peace. A crisis begins when one state (a "challenger") threatens to use force to obtain something of value to another (a "defender"). The challenger may wish to obtain something concrete, such as a piece of territory, or something abstract, such as a change in the defender's policy. The goal of deterrence as a strategy is to convince a challenging state to back down from its threats by counter-threatening the use of force (issuing a deterrent threat). If it works, the counter-threat indicates to the challenger that the costs of using force will be high and/or the benefits will be low. Deterrent threats may be threats to impose pain on the challenging state if it proceeds to attack, or they may be threats to deny the challenger its goals by fighting.7 A central question among scholars of deterrence is that of when deterrent threats will be credible. "Credible," of course, means "believable"; deterrence theorists are concerned with when an adversary will believe a state's threats to fight.8 This question is crucial because a deterrent threat only affects the challenging state's assessment of the costs and benefits of an attack insofar as the challenger believes the threats.

A common theme in the literature is that crisis behavior and credibility are heavily influenced by two factors: the military capabilities and resolve of the states involved in the crisis.9 Resolve is an amorphous concept, used in different ways by different scholars; Fearon (1992, 69) defines the composite well as the "willingness to use force." Scholars of deterrence differ about what creates a willingness to use force; some argue that it stems from having greater interests at stake in the dispute (Snyder and Diesing 1977; George and Smoke 1974; Maxwell 1968), while others maintain that it is a dispositional quality, perhaps a willingness to take risks (Kahn 1965; Schelling 1966).10 Though "capabilities" and "resolve" are separate terms in the deterrence literature, few deterrence theorists would deny that capabilities also affect crises by influencing states' willingness to use force.

Why do deterrence theorists believe that capabilities and resolve increase a state's credibility? If deterrence fails, a state with relatively stronger or more technically advanced military forces is better able to deny its adversary its goals and/or to impose greater costs on the adversary (Huth 1988, 35-9). At the same time, a militarily stronger state is more willing to defend, since it is more likely to prevail in war. Since militarily stronger states are more willing and more able to follow through on their threats, their adversaries are more likely to believe their deterrent threats and to believe that the threatened action actually will harm them and/or deny them their goals. Similarly, more resolved states are more likely to fight if necessary, and to fight hard enough to win. Their adversaries also are more likely to believe their threats to fight and to believe that the threatened action will harm them and/or deny them their goals. Thus, states are more likely to attain their goals after issuing deterrent threats if they are stronger and/or more resolute. Of course, they also are more likely to attain their goals without deterrent threats if they are stronger and/or more resolute, since others know that strong, more resolute states are more likely to win wars if push comes to shove.

Realist scholars as a group are less concerned with crisis behavior than are deterrence theorists, and more concerned with macro events, such as the prevalence of war in the international system. However, as in deterrence theory, one of the unifying ideas in realism is that power and/or military force is extremely important in deciding the course of international events (Doyle 1997, 43).11 Many realists argue that a more powerful state's diplomacy will be more effective than that of a less powerful one; for example, Hans Morgenthau argues, "[D]iplomacy must determine its objectives in light of the power actually and potentially available for the pursuit of these objectives" (Morgenthau 1967, 539-40).12

In sum, most deterrence theorists and many realists argue that deterrence and/or diplomacy are effective tools of state. However, while these scholars pay lip service to the importance of diplomacy, many of their arguments suggest that diplomacy is essentially irrelevant. According to their arguments, credibility comes from might or resolve. Diplomacy is superfluous for a powerful or resolute state, which will attain its goals anyway; it is ineffective for a weak or irresolute state, which will not attain its goals with or without diplomacy.

The arguments about credibility and diplomacy that I have just reviewed contain a core of truth but confuse three ideas. In a dispute, the challenger begins with some assessment of the defender's willingness to use force to prevent the challenger's desired change in the status quo (figure 1.1). As the dispute progresses, the challenger's assessment may change or stay the same. The defender's deterrent threat, if effective, provides new information that changes the challenger's beliefs about the defender's willingness to use force, or to take some other threatened action. After hearing the defender's threat, the challenger then has new beliefs about the defender's willingness to use force over the disputed issue.

An adversary's initial assessment of a state's willingness to fight depends upon the military balance and on what is known about a state's resolve, for reasons that the deterrence theorists suggest. However, the concept of so much concern to deterrence theorists--the credibility, or believability, of the threat--is the listener's assessment of the threatener's willingness to fight after hearing the threat, box number 3 in figure 1.1. Deterrence theory does not adequately explain how a state can change its adversary's beliefs about its capabilities and resolve during the course of a dispute.13 Thus, in arguing that diplomacy is primarily the instrument of power or resolve, realists and deterrence theorists effectively suggest that the adversary's initial assessment of the situation determines the effectiveness of the threat.

In a dispute, both a state initially believed to be irresolute (less willing to fight) and one initially believed to be resolute (more willing) wish to convince an adversary that the adversary's initial assessment of the situation is mistaken; the threatener is more resolved than previously thought. For example, in the Yugoslav crisis discussed earlier, the United States had an incentive to appear more resolute--whether it was, in fact, quite resolute, or whether it was bluffing in an attempt to provide safety for Milosevic's victims at little cost. There is no a priori reason to believe that the state initially believed to be resolute will be more effective in convincing an adversary that the adversary has underestimated its resolve, that it is even more resolute than the adversary at first believed.

Political realists and deterrence theorists are correct when they argue that power and resolve affect the course of disputes. They are wrong to infer that a more powerful state's diplomacy therefore must be more effective in changing an adversary's mind or that it must be more credible. Understanding the circumstances under which threats will be credible and deterrence will be effective requires a theory that explains the value-added of diplomacy. This book aims to provide such a theory.

Cheap Talk Is Ineffective

It is the current (though perhaps passing) trend among scholars of international politics to claim that talk is a waste of time. More technically, some scholars believe that signals must be costly to convey information and that "cheap talk" is ineffective.14

The international-relations arguments about the ineffectiveness of cheap talk draw on a literature on costly signals in economics. In the economics literature, a "costly signal" is one that has a direct (and negative) effect on the sender's well-being. A cheap-talk signal may have eventual negative consequences, but the message itself is costless to send. For example, if one must pay $1,000 to make a speech, that speech is a costly signal. If one simply holds a press conference or sends a diplomatic note, then the speech is cheap talk.15

If signals must be costly, then diplomacy is a monumental waste of time. Diplomacy is the epitome of cheap talk; it includes speeches, communiqués, and diplomatic notes.

The widespread dismissal of cheap talk as ineffective in international relations is the consequence, probably unintended, of part of a literature in international relations on "audience costs." The point of the audience-costs literature is to emphasize the impact of domestic politics on international behavior; scholars in this school argue that the existence of domestic audiences allows states to signal their intentions. Some works in this vein even argue that cheap talk can work (Smith 1998a; Ramsay 2003). However, as I discuss in more depth in chapter 3, part of the literature gives the impression that signals must be costly to convey information, so that cheap talk does not work.16

In actuality, the dismissal of cheap talk has neither a theoretical nor an empirical basis. While the economics literature does show the effectiveness of costly signals (see, e.g., Spence 1974), a related literature in economics shows that signals that carry no cost often convey information (Crawford and Sobel 1982; Farrell and Gibbons 1989). In common usage, the phrase cheap talk refers to talk that probably is unreliable. Yet every person in human society knows that talk can be a very useful means of communication. Speakers choose their words carefully when it matters, because even cheap talk can have repercussions.

In international relations, talk that technically is "cheap"--diplomacy--can be very powerful. When states do try to deter actions by means of diplomatic threats, they often succeed in persuading their challengers to back down. For example, the Anglo-Russian Treaty of 1942 specified that both Britain and Russia were to withdraw their troops from Iran by March 1946. When the Soviets threatened to remain in parts of Iran after that date, the U.S. chargé d'affaires in Moscow delivered a diplomatic note to the Soviet government calling for the withdrawal of troops from Iran, and the Soviet Union withdrew its troops.17

This book reconceptualizes diplomacy and deterrence as cheap talk. While deterrence theory focuses on the possession of military capabilities and the resolve to use them, I focus on threats as a form of communication. In the work that follows, I show that states can use even cheap diplomacy to change their adversaries' minds in international crises.

Disbelief Requires a Psychological Explanation

One important critique of realist views of international relations comes from scholars applying ideas from the field of psychology (e.g., Jervis 1976; Lebow 1981). These scholars argue that states are led by human beings who misperceive incoming information to such an extent that communication and deterrence are difficult or impossible. For this reason, these critics are at least as pessimistic as the realists about the prospects for diplomatic solutions to international conflict.

As psychological theorists emphasize, the success of language is by no means a given in the context of international disputes. However, most difficulties that arise in communication require no psychological explanation (Fearon 1992, chapter 2). Communication is difficult because states have incentives to lie about their willingness to use force. A United States that was not willing to protect "safe areas" in Yugoslavia had an incentive to threaten to do so, since bluffs sometimes succeed. As long as states sometimes bluff, it is perfectly rational for threatened states sometimes to disbelieve the threats they hear.

This book explains why states often do believe each other's diplomacy. Nevertheless, while my explanation focuses on states' incentives to use threats honestly, it also explains why states sometimes do give in to temptations to bluff. Since diplomacy is not 100 percent honest, it also is not 100 percent credible, nor is it always effective. As I explain later in the book, diplomacy works, but not all the time or under all circumstances.

How Can a State Communicate That an Adversary Has Misjudged Its Resolve?

In 1968, the United States acquiesced to the Soviet invasion of Czechoslovakia. This decision was not a high point of moral policy making, but any threats over Czechoslovakia would have been bluffs. U.S. leaders seemed to realize the benefits of honesty; when Russian ambassador Anatoly Dobrynin told President Johnson that U.S. interests were not affected by the Soviet action in Czechoslovakia, "[i]n response he was told that U.S. interests are involved in Berlin where we are committed to prevent the city being overrun by the Russians."18 Johnson's words reveal that he saw a difference between Czechoslovakia, where he was honestly admitting that there were no strong U.S. interests, and Berlin, where he was threatening and prepared to go to war.

In contrast, Russian leaders bluffed in the early Balkan conflicts, and these bluffs may have harmed the effectiveness of their subsequent diplomacy. In the conflict of 1908-9, Russia backed down from its initial support for Serbia, leading Serbia to recognize Austria-Hungary's annexation of Bosnia-Herzegovina.19 When the Austro-Serbian conflict flared up again in 1912, Russia bluffed again, accepting Albanian independence when it became clear that Germany would support Austria-Hungary. 20 Russia's threats in 1914 then were ineffective. Based on the issues, one might have expected Russia's threats to be more credible in 1914. The issue in 1914 was the invasion of a Slavic state by Austria; based on estimates of Russia's interests, other states should have believed that Russia was likely to fight. Yet this was not so.21

These examples illustrate two facts. First, states sometimes honestly acquiesce to each other's demands, despite incentives to bluff. Second, when a state is caught bluffing, it acquires a reputation for doing so, and such a reputation hinders its ability to use diplomacy in the near future.

As I argued earlier, one purpose of diplomacy is to communicate to an adversary that a state is more resolved to fight over disputed issues than the adversary believes to be the case. Earlier, I noted that the effectiveness of diplomacy is puzzling because states' representatives have incentives to bluff, or dissemble. This book explains the effectiveness of diplomacy as, in part, the result of another strong incentive--the incentive to use diplomacy honestly in order to avoid a reputation for bluffing. Because much of diplomacy is honest, states are able to learn from each other's words.

Leaders often believe each other's deterrent threats because these threats are often--though not always--honest. States tend to be honest when they are willing to fight; they have no incentive to hide that fact. But, like the United States in the Czechoslovakian crisis of 1968, states sometimes also honestly acknowledge that they are not prepared to fight, even when that means acquiescing to unwanted change. States do bluff, but they bluff only rarely. Like Russia at the beginning of the last century, they acquire reputations for bluffing when they bluff and are caught. When a state is caught bluffing, others are unlikely to believe its diplomacy soon thereafter. The state then cannot use diplomacy honestly to convince another to back down; nor can it bluff successfully. It is less likely to attain its goals. The desire to avoid a reputation for bluffing, or to maintain a reputation for honesty, induces states to use their diplomacy honestly much of the time.

Reputations for honesty play a key role in explaining the success of diplomacy. When a state has a reputation for honesty, it is better able to use diplomacy successfully to communicate that it is more resolved than an adversary previously had believed to be the case. Since other work on deterrence discusses reputations for having resolve, it is worth emphasizing here that reputations for honesty differ from those reputations.22 A state acquires a reputation for honesty when others observe it acting honestly. This happens when a state actually acts honestly, as well as when it bluffs and is not caught. A state can act honestly in two ways. Under some circumstances, it honestly can claim to be resolute about a particular issue (i.e., threaten, and then fight if deterrence fails). Alternatively, it honestly can acknowledge that it is irresolute; it can acquiesce to an adversary's demands.23

Unlike many theories of international relations, this book leads to optimistic conclusions about diplomacy. In demonstrating why states often are able to use deterrence, a type of diplomacy, effectively, I suggest an explanation for why diplomacy works more generally. Like deterrent threats,diplomacy is likely to work because it is so valuable; states that misrepresent their positions may be caught, and may suffer decreased ability to communicate in the near future.

This explanation of diplomacy--elaborated later in the book--also provides clues as to when diplomacy and deterrence are more or less likely to be successful. A state's diplomacy is less likely to be effective when the state recently has been caught bluffing, more likely to change an adversary's mind when the state recently has used diplomacy honestly or engaged in a successful bluff.

While traditional deterrence theory suggests that a state should fight or strengthen its military to increase its credibility, my work shows that there is another path: honest acquiescence, too, can increase a state's ability to use diplomacy in the future, by letting an opponent know that the state will threaten only when it really does intend to fight. Moreover, both powerful states and weak ones can increase the effectiveness of their future diplomacy by using their diplomacy honestly today. Thus, diplomacy is a tool for the weak as well as the strong.

Overview of the Book

In the remainder of this book, I first present my argument as to why and when diplomacy works, then offer evidence in support of that argument.

Chapter 2 uses a case of failed deterrence to illustrate a central feature of my explanation: reputations for bluffing hurt a state's credibility and lessen the effectiveness of its diplomacy. This chapter examines China's attempt to deter UN forces from crossing the 38th parallel during the Korean War. The case is a fascinating one because the Chinese made numerous and varied attempts at deterrence and the United States probably did not want to fight China, but China's attempt at deterrence failed anyway. This study details how U.S. officials dismissed China's threats as bluffs, and argues that China's difficulties in using diplomacy stemmed, in part, from recent, unfulfilled threats vis-à-vis Taiwan. Because they came in the context of these threats, China's threats to enter the Korean War were less credible. Because the United States did not believe China's threats, China had to go to war in order to attain its goal--keeping U.S. forces away from its border.

Chapter 3 formalizes my explanation of successful deterrence and diplomacy in a game-theoretic model of repeated international disputes. Using the model, I prove that "cheap-talk" threats often are effective in deterring an attack because states wish to maintain their reputations for using diplomacy honestly.

Chapter 4 presents statistical evidence that this theory helps to explain actual disputes and crises. I analyze data from more than 1,300 international crises, and many additional cases in which states did not threaten the use of force. I find that these disputes, on average, progress in ways implied by my explanation of diplomacy. In particular, reputations for honesty help to make diplomacy effective. A state with a reputation for honesty is more likely to succeed in deterring an attack. As the theory suggests, this is because its diplomacy is, in fact, more likely to be honest. The fact that these patterns arise in the data, as the theory predicts, helps to confirm the usefulness of the theory.

The theory uncovers reputations for honesty or for bluffing as an important variable for helping to explain states' behavior. Specifically, it explains why these reputations are critical for understanding the escalation of existing international disputes. In chapter 5, I speculate that this variable also affects states' decisions to become involved in disputes in the first place and I investigate this speculation empirically. I also explore the impact of the military balance on dispute escalation.

Chapter 6 summarizes the findings of the book. It also returns to a discussion of how this theory expands our understanding of diplomacy more generally. Finally, I discuss ideas for future research.


Current scholarship in international relations encourages the dangerous view that diplomacy is superfluous, does not work, or can be made to work only through the use of force or other costly signals. This book shows that diplomacy does work--though, like any other tool of state, it is not a sure thing.

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
For Reviewers
Class Use
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share