Reliable Partners: How Democracies Have Made a Separate Peace [Course Book ed.] 9781400850723

Citation preview

RELIABLE PARTNERS

RELIABLE PARTNERS H O W D E M O C R A C I E S H AV E M A D E A S E PA R AT E P E A C E

Charles Lipson

PRINCETON UNIVERSITY PRESS

PRINCETON AND OXFORD

Copyright  2003 by Princeton University Press Published by Princeton University Press, 41 William Street, Princeton, New Jersey 08540 In the United Kingdom: Princeton University Press, 3 Market Place, Woodstock, Oxfordshire OX20 1SY All Rights Reserved Library of Congress Cataloging-in-Publication Data Lipson, Charles. Reliable partners : how democracies have made a separate peace / Charles Lipson. p. cm. Includes bibliographical references and index. ISBN 0-691-11390-4 (alk. paper) 1. Democracy. 2. Peace. I. Title. JC423 .L583 2003 327.1'7—dc21 2002031740 British Library Cataloging-in-Publication Data is available This book has been composed in Times Roman Printed on acid-free paper.∞ www.pupress.princeton.edu Printed in the United States of America 10 9 8 7 6 5 4 3 2 1

For Susan

Contents

List of Tables and Figures

ix

Acknowledgments

xi

1. The Argument in a Nutshell

1

2. Is There Really Peace among Democracies?

17

3. A Contracting Theory of the Democratic Peace and Its Alternatives

47

4. Why Democratic Bargains Are Reliable: Constitutions, Open Politics, and the Electorate

77

5. Leadership Succession as a Cause of War: The Structural Advantage of Democracies

112

6. Extending the Argument: Implications of Secure Contracting among Constitutional Democracies

139

7. Conclusion: Reliable Partners and Reliable Peace

169

Notes

191

Index

249

Tables and Figures

Tables 3.1 6.1 7.1 7.2

Alternative Theories of the Democratic Peace: Summary Table Table of Conjectures: Implications of Reliable Contracting between Democracies Evidence from the Conjectures Causal Theories and Policy Recommendations

74 140 172 187

Figures 1.1 and 7.1

Contracting Basis of the Democratic Peace

15 and 170

Acknowledgments

MY THANKS go to friends and colleagues who have aided me throughout this project. Some commented on portions of the manuscript, others answered specific questions, and still others debated their views with me. My largest debt is to scholars at the University of Chicago, past and present. The University is a cauldron of scholarship and animated discussion about the core issues of world politics. James Fearon, John Mearsheimer, and Duncan Snidal all made extensive contributions. I received valuable assistance from Daniel Drezner, Charles Glaser, Jack Goldsmith, Lloyd Gruber, Robert Pape, Anne-Marie Slaughter, Daniel Verdier, Stephen Walt, and Alexander Wendt. No aspect of teaching and learning at Chicago has been more rewarding than working with graduate students, especially those at PIPES, the Program on International Politics, Economics, and Security, which I codirect. Many students who discussed this book with me are now colleagues at fellow institutions. I wish to thank Michaela Dabringhausen, Xinyuan Dai, David Edelstein, Hein Goemans, Seth Jones, Barbara Koremenos, Andrew Kydd, Keir Lieber, James Marquardt, Brian Portnoy, Alex Thompson, and Yael Wolinsky. Special thanks go to Sabastian Rosato, who read the entire manuscript and offered sharp, thoughtful criticism. This book builds on two decades of scholarship about the democratic peace, beginning with Michael Doyle’s pathbreaking work. I have benefited greatly from the work of Bruce Bueno de Mesquita, Christopher Gelpi, John Oneal, John Owen, James Lee Ray, Dan Reiter, R. J. Rummel, Bruce Russett, Kenneth Schultz, and Allen Stam, as well as some of their most incisive critics, especially Joanne Gowa and Christopher Layne. I also wish to thank those who commented on individual chapters or specific issues: Kenneth Abbott, Emanuel Adler, Lisa Bernstein, Charles Gochman, Judith Goldstein, Joseph Grieco, Russell Hardin, Benedict Kingsbury, Deborah Larson, Christopher Layne, Walter Mattli, Andrew Moravcsik, James Morrow, Henry Nau, John Odell, Kenneth Oye, Jon Pevehouse, Michael Schudson, Martin Sherman, Kathryn Sikkink, and Beth Simmons. I had long, useful conversations with many of them, particularly Debbie Larson. My deepest intellectual debt, aside from that owed to my colleagues at Chicago, is to Joseph Nye and Robert Keohane. I have often benefited from their guidance and, most of all, from their examples as serious scholars concerned with fundamental issues. And then there are more personal debts, to my children, Michael and Jonathan, my extended family, and my lifelong friends. I dedicate this book to my wife, Susan Bloom Lipson, with love.

RELIABLE PARTNERS

1 The Argument in a Nutshell

DEMOCRACIES almost never fight wars against each other. This simple observation is one of the most powerful findings in international politics and one of the most throughly tested. But what explains it? The answer, I think, is that democracies have unique “contracting advantages,” which allow them to build stable, peaceful relations, based on multiple self-enforcing bargains. The basic finding—that democracies rarely, if ever, fight each other—is one of the most compelling in modern international politics. Discovered by accident and initially overlooked, the “democratic peace” has been vigorously debated and exhaustively tested. Most statistical studies and case histories have found the same robust relationship. Democracies often go to war but very seldom against each other.1 Thanks to all this testing, the democratic peace is now one of the bestestablished regularities in international politics, perhaps the best-established.2 The absence of war between democracies, Jack Levy concludes, “comes as close as anything we have to an empirical law in international relations.”3 Another recent article notes that when all interacting states are democracies, that is “a near-perfect sufficient condition for peace.”4 Not surprisingly, these findings have prompted a flurry of research. Some extend the initial conclusions, showing that democracies are distinctive in other ways. They generally win wars, for example, probably because they take special care in choosing opponents.5 Others show the limits of the democratic peace. New and unstable democracies may well be more war-prone than other states, although they, too, are reluctant to fight among themselves.6 This does not mean that democracies are pacifists, however, even when dealing with each other. They sometimes threaten to use force against fellow democracies, and they have come close to war several times. Even with these qualifications, the democratic peace is a powerful finding with far-reaching implications for both policy and theory. It means that international interactions are profoundly shaped by the way states are governed. Naturally, this finding inspires democracy’s many advocates. But it deeply puzzles theorists, who want to know why it occurs. Their puzzlement echoes an old academic joke: “We know it works in practice. Now we have to see if it works in theory!’ ” That is exactly the question about peace among democratic states. It works well in practice, but there is considerable confusion about how it works in

2

CHAPTER 1

theory. The lack of an answer is no joke, however. Despite extensive research, all we have is a remarkable correlation. We still lack a convincing explanation about why democracies do not fight each other. Knowing why they do not fight is important for both practitioners and theorists. It bears directly on two central issues of international politics: the reasons for war and peace and the problems of sustaining cooperation. It also has serious implications for any general theory of international relations. It poses a specially pointed challenge to those who deliberately ignore the character of domestic governments, a crucial simplifying assumption of neo-Realism.7 The answer to this puzzle may also say something important about how countries can build relationships of greater trust and reassurance, laying the foundations for enduring peace. In short, it is a prominent puzzle in every sense. How do we explain this apparent relationship between governmental forms and international outcomes? So far, three basic explanations have been advanced: 1. citizens’ reluctance to bear the costs of war; 2. shared values among democracies; and 3. unique domestic institutions, which restrain elected leaders. The cost explanation, initially developed by Immanuel Kant, argues that citizens of a republic are less war-prone because they must bear the burdens themselves and can vote to avoid them. Monarchs and dictators, by contrast, can shift the costs of war onto their subjects, who have no voice in the decision.8 The normative explanation, also suggested by Kant, is that liberal democracies share certain basic values, grounded in their domestic political life. They settle disputes through neutral courts rather than through blunt force or status differences.9 They place a high value on individual life and liberty. As they look abroad, they recognize that other democracies also have governments based on popular consent and hold similar values. They then apply these liberal values in dealings with other democracies, where they can expect reciprocity.10 Relying on these common values, they can adjudicate disputes and compromise voluntarily rather than resorting to military force.11 Finally, the institutional explanation underscores the constraints facing democratic policymakers.12 They face constitutional limits, must share power with other elected leaders, and can remain in office only by winning periodic elections, openly contested. Although there are important differences between parliamentary and presidential systems, both include legitimate opposition, genuine public debate, and procedural limitations on how public policy is made.13 These institutional arrangements inevitably slow down decisionmaking and make it difficult to launch military operations. If other states are similarly constrained, then neither will fear surprise attacks, and both will have the luxury of time to resolve international crises.14

THE ARGUMENT IN A NUTSHELL

3

All three explanations highlight important features of democracy, but each is incomplete. If costs alone were the answer, then citizens would be equally reluctant to fight both democracies and nondemocracies. That does not explain why they choose to fight some types of states and not others. Large, powerful democracies have waged wars against small dictatorships, but not against small democracies. Moreover, their sensitivity to costs is not so straightforward. Public opinion in democracies sometimes favors war, as it did in 1898, when the United States fought Spain, which was blamed for sinking the USS Maine in Havana harbor. Monarchs and dictators, on the other hand, are not always eager to fight, since they, too, must weigh the financial costs, political risks, and chances of success. Defeat will not hurt them at the polls, but it could lead to their overthrow, exile, or death.15 If norms alone were the answer, democracies would also be more peaceful toward all other states, not just other democracies. There is now some evidence to support that assertion, at least in mild form. Democracies are probably a bit less war-prone than other states. Even so, it is clear that democracies go to war often. In the twentieth century, for instance, Britain went to war more frequently than Germany.16 The United States has obviously been willing to use force repeatedly, both to defend its material interests and to extend its values. Moreover, democracies do not merely respond to provocations by others. They often seize the initiative and sometimes attack first. We need to understand, then, how norms and beliefs could lead democracies to make war against some states but not against others. If their norms encourage them to negotiate, compromise, and use judicial mechanisms, why don’t they do exactly the same thing with nondemocracies, achieving peace with them, too? Equally important, if democracies are really bound by such strong, pacific norms of conduct toward each other, why do they sometimes threaten fellow democracies with force? Contrary to the normative explanation, democracies sometimes issue military threats against each other and display troops and weapons to make them more credible. They probably do so less often than nondemocracies, but they definitely do issue military threats against each other. They have occasionally launched covert military operations against other democracies or supported coups from within. A purely normative explanation cannot easily grasp this kind of belligerence or fold it into an account of the democratic peace. What we need is a much better understanding of how normative concerns operate, how they dovetail with material interests to define national preferences, and, ultimately, how they shape the choices that yield war and peace. In fact, common norms, shared values, and even shared interests may not be sufficient to produce peace. They may foster cooperation, but they do not ensure it. That is the fundamental lesson of the Prisoners’ Dilemma. Both players value mutual cooperation over mutual defection, but they still cannot achieve it in a finite game. To achieve stable cooperation, the game itself must be

4

CHAPTER 1

changed in fundamental ways. Assuming both players value the future highly enough, they can sustain cooperation only if the game is played repeatedly, without a clear end point, and if defection by either player is deterred by the threat of punishment (that is, by the threat of noncooperation later in the game). Finally, if domestic institutions were a sufficient explanation, why aren’t democracies constrained from using force against all types of regimes? Not only do they use force against nondemocracies, they sometimes strike the first blow. The larger problem here is that institutional explanations do not spell out the causal connection between domestic limitations and international outcomes. How exactly are constraining institutions supposed to prevent democracies from going to war? And why do they operate to prevent wars only with other democracies? All three explanations—costs, norms, and institutions—highlight unique elements of modern democracy and provide insights into the democratic peace. They suffer, however, from the same basic problem. Each is focused on the properties of an individual democracy. The democratic peace is fundamentally an interactive phenomenon. It is not about why one democracy or another is peaceful. It is about why two democracies so seldom fight each other. The existing explanations merely allude to this central, interactive element of the democratic peace without developing and pursuing its implications. Essentially, they say, “if state A behaves this way, or holds these values, or has these institutions, and if state B does too, then we will have peace.” Sometimes these explanations go one step further and say that states A and B must recognize their shared values or constraints. That is correct, as far as it goes, but it does not go far enough. It glides over the most distinctive feature of the democratic peace: its emergence from the durable bargains and mutually profitable relationships democracies have forged with each other. That is the heart of my own explanation. My explanation is that constitutional democracies have a special capacity to make and sustain promises with each other, including those about war and peace. They are better equipped to find and capture gains from mutual interests, to sustain them, and to forge durable, mutually profitable relationships. Democratic norms and institutions facilitate this by making bargains easier to identify, less risky, and more reliable in practice. Because democracies have unique “contracting advantages,” they can usually avert or settle conflicts with each other by reliable, forward-looking agreements that minimize the dead-weight costs of direct military engagement. To do that, states must be confident their partners will live up to their promises or, if they do not, that they can protect themselves from the risks. After all, in matters of war and peace, there are no effective global courts to punish violators or protect innocent parties. It is a self-help system. States must look first to themselves for protection in a dangerous world. That usually

THE ARGUMENT IN A NUTSHELL

5

means arms and fortifications. But states can do more than deter, defend, and attack. They can also help themselves by making mutually profitable deals with reliable partners, both as military allies and as trading partners. This contracting explanation recognizes that democracies have mixed interests, sometimes overlapping, sometimes diverging. They have good reasons to push hard for their own favored outcomes and may bluff and threaten to achieve them. On the other hand, they also have strong incentives to capture joint interests and prevent the breakdown of cooperation. One of the most important of these joint interests is avoiding the barren costs of war. Their ability to achieve such cooperative outcomes by peaceful negotiations should increase over time as partners learn more about each other and develop profitable working relationships. As they learn, their contracting relationships evolve. This explanation does not discard the extensive research on democratic norms and institutions. Quite the contrary; it systematically incorporates that work as part of a more comprehensive contracting explanation. My explanation begins by noting that states must negotiate and secure their own bargains, ranging from formal treaties to informal working relationships. To do that, they must locate some common interests, if any are to be found, and divide the potential gains. They must also take care to ensure that their deals (and, indeed, their wider relationships) are self-enforcing. Self-enforcement is vital because there are seldom neutral third parties to interpret and enforce international bargains. They must seal their bargains in the harsh world of international anarchy, where states must help themselves, interpret their own bargains, and enforce whatever deals they make. In this setting, perceptions and beliefs matter because states are not completely informed about each others’ preferences and behavior, much less about the future contingencies they will face. They must also be concerned about their partners’ institutional capacity to make commitments today and fulfill them tomorrow. The danger is that partners are unable or unwilling to live up to their promises, perhaps because the leader who made the agreement changes his mind, perhaps because circumstances have changed, or perhaps because there is a new ruler or a fundamental change of government. These concerns arise because in the real world bargains are necessarily “incomplete contracts.” No matter how elaborate treaties are, they cannot anticipate all future possibilities, all conceivable changes or external shocks. Negotiators either fail to foresee them or deliberately choose not to spend scarce time and money dealing with remote eventualities. If these contingencies arise, or if the parties later interpret their bargain differently, then they have to sort out their disputes, adjust the terms, settle on a common interpretation, or, failing that, abandon the agreement. When disputes arise, then, they must look after themselves. If they wish to capture the gains of cooperation and facilitate their ongoing relationship, they have to create some kind of governance mechanisms—formal or informal—to cope with such problems.17

6

CHAPTER 1

Before relying on such incomplete bargains, prudent states must make judgments about their partners’ integrity and reliability. They want to reap the promised benefits, but they also want to be sure others are dealing in good faith and will continue to do so after the bargain is struck. Is their potential partner stable and dependable? Can its performance be monitored effectively at a reasonable cost? Is it able to fulfill its commitments, and will it try? Is it likely to discard its obligations if they become inconvenient, or cheat if its actions are hidden or too costly to punish? When new leaders come to power, will they be bound by earlier promises? Can the participants forge governance arrangements to sort out disputes and fill in gaps when they arise? These are hard questions with serious consequences. For all these reasons and more, states are concerned about the types of partners they are dealing with. That is particularly true in sensitive national security issues, where cheating and betrayal can be a matter of life or death.18 That is why states look to their partners’ reputations for compliance or opportunism, their transparency to outside observers, their continuity of government, and their institutional capacity to make solid commitments about future behavior. There are always risks, of course. But they are lower and more manageable when partners are well-established constitutional democracies. Such states are simply better equipped to make long-term commitments, to sustain them across changes of political leadership, and to show others their true preferences and their compliance with existing bargains. What drives the democratic peace, then, is not so much democracies’ distinctive policy goals—although they may well be less greedy, aggressive, and conflictual—but rather their special ability to locate and capture joint gains and to avoid deadly escalation in dealing with each other. The key to this explanation is the reciprocal exchange of promises. To work, these promises must be credible, reliable, and durable. Otherwise they would not reassure anxious neighbors. Stable, constitutional democracies have a greater institutional capacity to make such reassuring, self-binding promises, both formally and informally. Their constitutions are designed to make some policies very difficult to reverse, sometimes by requiring supermajorities to change key decisions. They also make major strategic choices visible to all, including other countries, largely as a by-product of public discussion among elected officials and their communication with voters. Because political opponents can openly challenge government policies, other states can judge the depth of support for these commitments and the likelihood, if any, that they will be overturned. That gives partners some protection against devastating surprises. The mass electorate supports these arrangements in several ways. First, elected leaders generally pursue voters in the center of the political spectrum. This concern with the “median voter” serves as an anchor against dramatic policy changes despite turnover in elected leadership. Second, to win votes, politicians must articulate major policies and commitments to the public,

THE ARGUMENT IN A NUTSHELL

7

which spreads information to friend and foe alike. Third, promises made to the electorate serve as a kind of surety bond. Politicians know it is costly to break major promises, including treaty obligations. That allows them to make some foreign commitments more credible and trustworthy by deliberately creating high “audience costs” among the electorate and often among legislators as well. These audience costs mean that democratic leaders have political incentives to fulfill their commitments, and their partners know it. These incentives, plus the transparency of democratic government, lessens the chances they are bluffing or deceiving. Even so, democracies do change course and occasionally break promises. When they do, however, the process is typically slow and open to outside scrutiny, which diminishes the risks to partners. The process is slow because democracies have more elaborate, codified procedures for policymaking and more locations where policy can be vetoed or amended. The process is more open because democracies are organized to communicate policy alternatives to their own citizens and their multiple representatives. Democracy is a vast, information-generating machine, much like the market as Friedrich von Hayek describes it. Engaging the public in major policy debates is a fundamental part of the electoral process, which is why press freedom is essential in constitutional democracies (and a danger to autocracies). Legislatures also require extensive information to perform their duties. Their debates and oversight of the executive continuously disclose information to the public and, as an inevitable by-product, to foreign friends and foes. Opposition parties promote this flow of information in several ways. They prod the government to release data, forecasts, and evaluations. They debate the wisdom of specific policies and broad strategies. They support or oppose government policies and, in doing so, signal the depth of national support for them. Again, all this information is freely available to both allies and adversaries.19 The resulting transparency means foreign partners can gauge the depth of a democracy’s commitment to specific policies, not only today but over time, as policies are implemented. This combination of open debate and slow policymaking gives partners ample notice of potential changes. That lowers the risks of devastating surprise, which makes it more feasible to make bargains in the first place. As the comedian Henny Youngman once said, “It’s good to see you. It means you’re not behind my back.” That is exactly the point about transparency, and that is why it is so valuable in making secure international bargains. These institutional arrangements make it hard for democracies to act nimbly, but they also make it easier for democracies to contain risks and make durable commitments. When democracies work with partners that can reciprocate, they can profitably exchange promises of peace—promises they fully expect to be mutually rewarding and self-enforcing for the long haul.

8

CHAPTER 1

Extensions: Implications of a Contracting Theory of the Democratic Peace If this explanation is correct, it should not only clarify why democracies so rarely fight each other, it should tell us much more about democracies in world politics. We should expect a series of other major empirical findings, closely related to the contracting explanation. These extensions are important for both substantive and methodological reasons. First, they indicate that the democratic peace has wider ramifications for international politics. Second, they allow us to evaluate the contracting explanation beyond simply retrofitting it to familiar evidence. These extensions are crucial to what Imre Lakatos calls a progressive research program.20 Instead of simply tailoring a theory to fit well-known facts, he urges us to see if the theory also yields new and unexpected discoveries. That is a hard test, but meeting it gives us confidence that the theory has real explanatory power. If secure contracting really does explain the democratic peace, then we should also expect to find the following: • Bargaining among democracies should show less concern with sudden betrayal or opportunistic abandonment than similar bargaining involving one or more nondemocratic states. Negotiations among democracies should pay less attention to mechanisms to guard against cheating, betrayal, and renunciation.21 • Bargaining among democracies should be less concerned with problems of compliance, verification, and safeguards (although they are unlikely to ignore these matters entirely). These issues should be much more prominent when nondemocracies are involved. When security bargains are reached with nondemocratic states, which are more opaque and less trustworthy partners, they should include much more extensive mechanisms for self-protection. Agreements with them must be amply buttressed against potential breach, if they are to be relied on. • Although the risks of opportunism and nonperformance are lower among democracies, they are not entirely absent. Since democracies lack perfect information about each other and are not perfectly trustworthy, bargains among them should include some safeguards. Nevertheless, these measures should be less extensive, less costly, and less vital to sealing the bargain than those needed when nondemocracies are involved. • Alliances among democracies—a crucial type of security agreement—should be more stable and reliable than alliances involving nondemocratic states. • Changes of political leadership should have a greater impact on bargains involving nondemocracies. Leadership changes should have less effect on bargains and relationships among democracies since new policymakers are more constrained by prior commitments and constitutional rules (including voting rules that heighten the influence of median voters and therefore stabilize policy choices).

THE ARGUMENT IN A NUTSHELL

9

• Democratic bargains should be most reliable when they involve ❍ Two states dealing directly with each other about ❍ Bilateral issues that the two parties themselves can control. They should least reliable when multiple partners are involved and when issues lie beyond their immediate control. Agreements involving multiple partners pose problems because all states, including democracies, have incentives to free ride, to let others shoulder the burdens of collective action. Democracies, like other states, have incentives to shirk individual expenses, pass the buck to others, and avoid the costly task of punishing those who violate their commitments. • The democratic peace should have the weakest hold over (a) new and unstable democracies and (b) marginal democracies. Both are less reliable partners. Even if they share the normative commitments, transparency, and formal institutions of older democracies, their preferences and future behavior are still highly uncertain. Reliance on them is risky. Potential partners cannot be sure if the domestic regime will survive, if it will remain democratic, or if it will be replaced by a radically different alternative. As Vice President Dan Quayle once remarked, “We are on an irreversible trend toward more freedom and democracy, but that could change.” It certainly can change for individual states. Aside from the risk of rebellion and coups d’e´ tat, new democracies are less reliable because they have no track record or long-term relationships to build on. This same shroud of uncertainty surrounds states that are only marginally democratic, especially if partners are unsure about what type of government they will have in the future. They, too, should be largely outside the democratic peace because of their instability and unreliability as partners. • As a closely related proposition: the democratic peace should be weaker when democracies are internally threatened and unstable, when they must confront enormous economic and political shocks. During such periods (or in unstable regions), greater uncertainty about the permanence of democratic regimes weakens their reliability as long-term partners. • Among well-established democracies, peace should become increasingly stable over time as they develop confidence in their relationships as contracting partners. This learning effect is an integral feature of contracting. It develops as states gain experience in their relationships and gain knowledge about their partners. They learn from their successes (or failures) in resolving conflicts short of war. They gradually acquire more information about their partners’ preferences and capabilities. As they construct a nexus of successful bargains and sustain them over time, they also build confidence in each other’s trustworthiness. This evolution resembles bargaining and cooperation among firms. “The literature is agreed that interpersonal trust in business relations is rarely offered spontaneously but requires an extended period of experience,” according to industrial sociologist Christel Lane. “During this time, knowledge about the exchange partner is accumulated through direct contact or is acquired indi-

10

CHAPTER 1 22

rectly through reliable third parties.” We expect the same kind of learning among states. Partners that resolve a series of major problems successfully, without fighting, should gradually develop deeper, more stable forms of cooperation. Thus, lengthy periods of joint democracy should produce increasing peacefulness, rationally grounded in experience. Even though states may still have significant policy differences—differences that could grow as well as diminish—military threats should decrease markedly within such long-standing democratic partnerships. This learning effect predicts, for example, that the unbroken relationship between the United States and United Kingdom should produce a stronger, deeper peace in the twentieth century than in the ninteenth. • Bargains and long-term relations among democracies should not be free of diplomatic rivalry or strife. After all, these are independent states, and they have a mixture of common and conflicting interests. These mixed interests have important consequences in the contracting explanation of the democratic peace. States want to capture joint gains through cooperation, but they also want to win the lion’s share of any gains. To prevail in these distributional struggles, democracies may signal their resolve by issuing threats, even against fellow democracies. Likewise, they may use threats to ensure compliance with existing bargains. Our contracting explanation does not preclude military threats or even the occasional use of force in relations among democracies, but it does expect them to be very rare. Democracies should be especially reluctant to threaten military action against each other for at least two reasons. First, democracies are more transparent, so there is simply less private information to reveal. Second, when democracies need to show their private level of commitment, they have more policy instruments to do so. Because their societies, polities, and economies are more open, they have more extensive transnational ties with each other.23 That allows them to use a wide range of sanctions to show their resolve.24 It gives them valuable options short of military action, which carries high costs and risks. • Because established constitutional democracies are better able to govern their relationships and settle conflicts by reliable agreements, they should perform better toward each other than do nondemocracies at every level of interstate conflict. Relationships among democracies should involve ❍ ❍ ❍ ❍

fewer military threats, fewer actual displays of force, fewer skirmishes or low-level conflicts, and fewer outright wars,

compared to relationships involving nondemocracies. • The contracting explanation is fully consistent with defensive anxiety about national security, even in relationships among democracies. Yet these Realpolitik fears should be dampened among democracies for two main reasons.

THE ARGUMENT IN A NUTSHELL

11

❍ First, the inherent openness of democracies allows outsiders to observe their foreign-policy choices and military posture. Their openness should reduce “worst-case” fears and lessen any exaggerated concerns about their aggressive motivations and expansive (offensive) policies. That is, it should facilitate more realistic and informed assessments. ❍ Second, as positive contracting experience builds over time, partners’ fears should recede, assuming the experience is pertinent.

Both effects should reduce the security dilemma for any partner of a democratic state. The effects should be greater, however, in relations with other democracies, where mutual transparency and diplomatic experience should diminish suspicions and encourage more accurate inferences. Among democratic partners, these effects short-circuit the cycle of fear and response that so often characterizes international politics. • If democracies do go to war against each other, they are much more likely to fight about fundamental, existential issues, which are extremely difficult to resolve, rather than about peripheral matters. They are unlikely to fight each other because of confused perceptions or inflated fears. ❍ Democracies should be better at locating profitable compromises, where they exist. Such compromises should be more accessible for peripheral matters. As a result, costly conflicts over such matters should be rare. ❍ Because democracies have better information about each other and less private information about their own preferences and intentions, they should be less likely to fight each other because of confusion and misinformation.

Taken together, these mean that if democracies do fight each other, their disputes should deal with basic issues of physical and territorial security and core political issues, central to the organization of state and society, not secondary issues like colonial expansion. These inferences cover a wide range of international relationships. If they are borne out, they would give us more confidence in the contracting explanation. They would meet Lakatos’s test of uncovering new and unexpected facts. They would also indicate our theory has real breadth, a useful test proposed by Mancur Olson. Theories are much more persuasive, he says, when they can explain a diverse array of facts.25

Four Sources of Democracies’ Contracting Advantage All these inferences—which we will test later—assume that democracies have a systematic advantage in locating and sustaining international bargains with each other. This contracting advantage is no fluke or historical accident. It is deeply embedded in the character of democratic states and their setting

12

CHAPTER 1

within civil society. Four basic traits are most important for securing international agreements: 1. 2. 3. 4.

democracies’ openness to outside scrutiny; the continuity of democratic regimes; visible electoral incentives for leaders to keep important promises; and the constitutional capacity to make enduring commitments, which cannot be easily overturned by current elected leaders or their successors.

These are deep-seated features of established constitutional democracy, features that facilitate international agreements and lessen fears. They allow democracies to reassure each other in ways that nondemocracies simply cannot, creating an informal governance structure for their bilateral relationships. In other words, the institutional capacities of constitutional democracies allow them to strike mutually profitable bargains, erect mechanisms for durable cooperation, and resolve conflicts peacefully. While no international promises are ironclad, those of democracies are more reliable because they come from institutions that transcend individual leaders, are likely to persist, and are relatively open to outside inspection. The result is that democratic promises are generally more convincing and enduring. They are less subject to surprise breach. This does not eliminate the pulling and hauling of international politics, the divergence of interests, or even the occasional threat of force among democracies. Diplomacy among democracies is not a Quaker Meeting. But it is not a brutal Hobbesian anarchy either, where lethal danger lurks behind every tree, promises are worthless, and treaties are mere scraps of paper. Rather, democracies have a unique capacity to exchange promises based on strong foundations of rational confidence and lower risk, relying on credible commitments, not mere hopes and dreams.26 It is this foundation that allows democracies to move beyond the immediate task of safeguarding individual promises and short-term relationships. As democracies become more confident in their mutual dealings, they reap a major benefit: they can promote wider, more enduring ties of interdependence at lower risk. Their dealings give them considerable information about each others’ political and economic interests and calculations. By hard experience, they gain a better sense of whether their partners will adhere to bargains, even when those bargains are inconvenient, or drop them at an opportune moment. Moreover, the density of their mutual ties gives them powerful levers to ensure compliance. Close, multiplex ties may even promote shared expectations and beliefs, which, in turn, strengthen the bargains themselves in a virtuous circle. This cooperative framework, together with the protection of property and contractual rights in liberal democracies, profoundly affects the global organization of production and trade. States and corporations can seek out the larger gains that come from specialization and an international division of labor.

THE ARGUMENT IN A NUTSHELL

13

These gains are secured within a web of voluntary, self-enforcing cooperation, at least among established democracies. These are not just gains from economic exchange. They are gains from exchanging all kinds of commitments, from trade to environmental issues, from national security to criminal justice. In all these areas and many more, solid agreements encourage democracies to depend on each other in the widest sense. They also provide a more stable framework for private citizens and firms to travel, trade, provide services, and invest abroad. This intermeshing of civil societies is an important by-product of the democratic peace and doubtless reinforces it. The development of global corporations, for example, makes domestic politics more visible to outsiders and raises the costs of breaking economic ties. But forming such links in the first place requires an overarching political framework based on interstate arrangements, a framework that establishes rules for commerce as well as institutions for dispute resolution. To capture these gains from specialization and interdependence, states must not only make agreements, they must be able to rely on them. Otherwise, the agreements would be hollow, little more than empty promises to ensnare the naı¨ve and trap the ill informed. Genuine gains from trade come only when agreements can be counted upon with some confidence. It is this reliance that induces real changes in actors’ behavior and a more efficient allocation of resources. The problem, of course, is that it is often risky to rely on international agreements. To rely on others is to become more vulnerable to their choices, including the chance that some partners will simply break their bargains when it suits them. Since there are no arrangements to compensate innocent parties (as there usually are in domestic contracts), states must take care or suffer the consequences. States constantly struggle to protect themselves from these risks while still capturing the benefits from international cooperation. They do both simultaneously, and our theories need to recognize this duality. In a world where states are interconnected strategically and economically, they have twin concerns: self-protection and self-improvement. Self-improvement, mainly greater economic prosperity, is valued in its own right by modern states. Elected leaders are held accountable for its successful pursuit, just as they are held accountable for national security. In a world where technological advances are central to military effectiveness, prosperity can be a source of security. But this does not mean economic growth is merely the handmaiden of military effectiveness or is sought mainly for that purpose. It is equally true that states value self-protection so that they can pursue greater wealth and freedom, and other aspects of self-improvement. Sometimes self-improvement is purely a domestic affair, but often it is not. It requires some measure of international cooperation and coordination, which

14

CHAPTER 1

supports a wide range of profitable exchanges. Even under the most difficult circumstances, states try to figure out ways to capture these gains from exchange if only they can adequately contain the risks. These risks never completely vanish, no matter what kinds of states are involved. On the other hand, the risks are not constant either. They are significantly lower when well-established constitutional democracies deal with each other. The reason—to reiterate the main argument of the book—is that these states are more reliable partners because of four basic features of their state structures: 1. high transparency, which allows outsiders as well as interested citizens to observe policy choices, grand strategies, and major regime discontinuities in a timely way, as well as to see the sources and intensity of support and opposition to specific commitments; 3. continuity of governance, based on clear rules for selecting senior officials and ensuring orderly succession, so policies evolve within the framework of stable domestic regimes; 3. high audience costs, which political leaders must pay if they break important promises to a mass electorate; and 4. constitutional governance, in which settled arrangements define and limit the powers of public officials and ensure due process in official acts, including government commitments to its own citizens and to other states. These complementary attributes raise the confidence of partners and reduce the dangers of surprise defections.27 They are basic—and unique—elements of constitutional democratic governance. Taken together, they form the contracting advantage of democracies. Democracies do have disagreements and conflicts of interest, sometimes serious ones. But they are also more reliable partners, better equipped to settle disputes among themselves by durable agreements short of war. These advantages cumulate in two ways. First, they are reflected in multiple bargains and extensive relationships among democracies, especially in economic and security affairs. These bargains not only provide them with more information about each other; they give them multiple tools to ensure compliance. Their individual bargains do not stand alone. They are embedded in a dense network of profitable relationships. Democracies are not bound together by one thick rope but by Velcro, with its innumerable strands. Second, as these relationships develop over time, partners learn more about each other. If their relations are successful and survive temptations to defect for short-term gains, they build mutual confidence. These multiple bargains, and democracies’ capacity to learn from them, give them unique capacities to manage relationships, avoid the costly escalation of disputes, and settle even serious conflicts with other democracies by durable agreements rather than by deadly fighting.

THE ARGUMENT IN A NUTSHELL

15

Figure 1.1 Contracting Basis of the Democratic Peace

This explanation is summarized in figure 1.1.

A Roadmap for the Book That is the argument in a nutshell. The remainder of the book explains the causal mechanisms, evaluates them in light of the empirical evidence, and considers alternative explanations. After that, I extend the argument by showing how secure contracting affects numerous relationships among democracies beyond war and peace. Before delving into all this, however, we need to be sure we are explaining something real and important. In the chapter that follows, I ask, “Is There Really Peace among Democracies?” Some scholars have protested that peace among democracies is an illusion, that the findings are statistically insignificant. Before trying to explain the democratic peace, I need to show that it is real and not a statistical quirk. In chapter 2, I briefly review the empirical evidence and consider its most important critiques. I reach two main conclusions. First, democracies really do fight each other less often. A wide range of empirical studies show that there is something real here to be explained. Second, existing explanations do not adequately capture the political logic of the democratic peace. In many cases, democracies have threatened each other with force and sometimes only narrowly averted war. Critics have examined some of these “close calls” and shown that the diplomacy looked very different from benign visions of the democratic peace. From that, they infer that the democratic peace is a mirage, that it is just Realpolitik plus dumb luck. That is wrong, I think, because it cannot explain why war between democracies is systematically less frequent. Still, the critics do raise a legitimate challenge. A convincing explanation not only needs to predict the outcomes accurately, it needs to explain the causes in a way that tracks and clarifies the diplomatic record and makes sense out of states’ actions.

16

CHAPTER 1

Chapter 3, “A Contracting Theory of the Democratic Peace,” outlines my own explanation, beginning with a rationalist theory of war. It shows why, according to such a theory, constitutional democracies should fight each other less often. The essence of the rationalist theory is that war stems from private (hidden) information and commitment problems, the inability to project promises dependably into the future.28 Stable constitutional democracies have systematic advantages in both areas, which should allow them to avoid war by reliable bargains. Getting to these bargains, however, may not be an easy ride. Although democracies have advantages, they still have some hidden information, some lingering doubts about compliance, and a normal desire to win the largest share of any joint gains. That means democracies will bargain hard and, in the process, could well threaten each other. On the other hand, they are better equipped than other states to settle these disputes without incurring the grave costs and risks of war. In chapters 4 and 5, I analyze the sources of this contracting advantage in greater depth and discuss the general problem of self-enforcing international bargains. These chapters develop the causal logic and show how democracies use their contracting advantages to forge dependable partnerships. This explanation incorporates some features of existing accounts of the democratic peace, which stress either democratic values or institutions, and folds them into a more comprehensive explanation. Chapter 4 is entitled “Why Democratic Bargains Are Reliable: Constitutions, Open Politics, and the Electorate.” Chapter 5 is “Leadership Succession as a Cause of War.” This explanation, as I have already mentioned, predicts that democracies will have other distinctive relationships beyond the fundamental achievement of peace. The contracting advantages of democracies should yield a web of agreements, covering a wide range of mutually profitable exchanges. Chapter 6 evaluates these predictions as well as the limits of the democratic peace. Chapter 7 is the conclusion, “Reliable Partners and Reliable Peace.” It quickly reviews the findings of the book, presents them in a straightforward table, and discusses the implications for world politics and for policymakers in democratic states. Our first task, however, is to make sure we are explaining something real. In the next chapter we ask: Is the democratic peace a meaningful finding or a statistical fluke?

2 Is There Really Peace among Democracies?

BEFORE TRYING to explain what causes the democratic peace, we need to be confident it is real. Democracies do seem to fight less frequently, but we need to know if the difference is statistically meaningful and can withstand the scrutiny of detailed case studies. That is no simple matter. Although wars are commonplace in history, there is only a slim chance that any two countries will fight in a particular year or even a particular decade. The problem, then, is to determine whether the rarity of wars between democracies is statistically different from the rarity of wars in general. Even if we do find a difference, we have to consider whether it might be caused by some factors unrelated to democracy. Perhaps these countries are simply richer or lie farther apart. Variables like these affect the likelihood of all wars, regardless of what kinds of states are involved. Most studies control for several key variables known to affect the likelihood of war: the distance between two countries, their differences in military capabilities, their status as neighbors or allies, and the possibility that either is a Great Power.1 With controls like these in place, we can concentrate on the distinctive impact of democracy on war and peace. The empirical debate has been heated. That is perfectly understandable because the democratic peace has far-reaching implications for policy and theory. Yet the debate has also been useful and productive. As it has continued, the quality and richness of the analysis has deepened. The prominence of the issues has led to more intense inspection of the data, sharper specification of the models, and more sustained engagement over difficult conceptual problems. The result is a growing body of serious research. My goal in this chapter is to summarize this work briefly, explore how the debates have expanded beyond the central issue of war and peace, and ensure that any explanations I offer are focused on meaningful outcomes.

Key Questions about the Democratic Peace The democratic peace was discovered by accident, doubted by those who found it, and overlooked by almost everybody else. The first empirical observation was in 1964 by Dean Babst, who published a brief article in a regional sociology journal.2 His conclusion was overlooked until 1976, when it was

18

CHAPTER 2

cited in another obscure journal by J. David Singer and Melvin Small, two of the most prominent quantitative students of war. Singer and Small dismissed the finding quickly—and prematurely—as an artifact of confounding variables, such as the greater distance between democracies.3 The issue was raised for a wider audience by Michael Doyle, writing in the early 1980s. In a series of major articles, he concluded that the democratic peace was both genuine and politically important.4 Doyle’s work is significant for two reasons. First, his was the most prominent early finding that liberal democracies do not fight each other. According to his data, there has been no such war since 1815—none at all. Second, drawing on the work of Immanuel Kant, he develops philosophical grounds for thinking this unusual outcome is no accident. His analysis is grounded in Kant’s 1795 essay, Perpetual Peace, which argues that monarchs are more willing to fight each other because they are safely removed from the awful costs. Citizens, on the other hand, are far less likely to take up arms since they would bear the costs themselves.5 Kant’s analysis, however, is not driven by costs alone. He also emphasizes constitutional restraints and democratic respect for individual rights, as well as commercial ties. Doyle concurs. He extends the analysis by stressing democracies’ respect for the moral autonomy of citizens and their legitimate right to choose leaders and policies. For Doyle, as for Kant, it is not simply voting that prevents war among democracies. What really matters is a wider set of liberal values, firmly embedded in stable political institutions. These rights include freedom of speech and conscience, political participation, and a cluster of economic rights, beginning with secure property ownership and freedom of contract. For Doyle, then, this is not so much a democratic peace as a liberal democratic peace. This important point has been embraced and extended by John Owen and others. When Doyle wrote, most international relations scholars were not especially interested in peace among democracies, liberal or otherwise. They were much more concerned about the dangers of war between capitalist democracies and communist dictatorships. That changed in 1989, when Mikhail Gorbachev released the Soviet stranglehold over Eastern Europe. Two years later the Soviet Union itself collapsed. These vast changes spelled the end of bipolarity and unleashed a wave of democratization in Central and Eastern Europe. These were the most far-reaching changes in international politics in fifty years, and they raised a whole series of basic questions. Would the new democracies take root? What would replace the bipolar structure of global politics? Would the close ties between the United States and Western Europe weaken now that the Soviet threat had disappeared? Would the spread of new democracies to Central Europe (and earlier to Latin America) portend a new era of peace? There have been essentially two kinds of answers: one triumphalist, the other skeptical. One celebrates the success not only of America’s military containment, but also the victory of a superior political organization (democracy),

IS THERE PEACE AMONG DEMOCRACIES?

19

a more humane society (personal liberty), and a more productive economy (capitalism). The mood is captured in the title of Francis Fukayama’s popular book The End of History and the Last Man,6 which sees the spread of liberal democracy as the culmination of a profound, Hegelian process of historical development. Its counterpart, for students of international politics, is a renewed interest in Woodrow Wilson’s vision of a cooperative, democratic world. The hope was (and still is) that these states might enjoy peaceful relations, founded on common norms of individual freedom, judicial settlement of disputes, and shared commitment to international institutions. The growing literature on the democratic peace fits snugly into this vision. Not everyone is convinced. Skeptics point to the familiar clash of national interests, which (they say) will drive a wedge between old allies now that old enemies are gone. International structure trumps domestic norms and institutions. Forms of government do not matter much. What really matters is the global balance of power and hard-nosed calculations of national advantage. As Lord Palmerston said over a century ago, his country had no permanent allies, only permanent interests.7 That is the essence of Realpolitik, and modern students have readily applied its lessons to the post–Cold War world. Former allies will soon begin to balance against each other. States with great potential power, notably Germany and Japan, will again play major, independent roles in world politics, sometimes at odds with the United States. That is the meaning of John Mearsheimer’s evocative title, “Back to the Future: Instability in Europe after the Cold War.”8 The future, according to this view, will look very much like the multipolar past, with one crucial exception: nuclear weapons. Their vast destructive potential makes stable deterrence possible and war unlikely, at least among Great Powers with secure nuclear arsenals. But peace among nuclear powers, based on a balance of terror, is hardly the same thing as peace among democracies, based on shared norms and representative institutions. These analytic differences inform the larger debate over the democratic peace, and we will return to them after considering the data. Actually, there are at least three major disputes about the data and its interpretation. One deals with findings from large data sets.9 The key question here is whether the rarity of wars among democracies is really different from chance. A second, related debate involves the classification of specific military disputes. Does fighting in a particular conflict rise to the level of a war, and are the states involved really democracies? Since wars are rare, the reclassification of a few disputes could affect the statistical significance of the overall findings. The third dispute focuses on conflicts among democracies that are resolved short of war. The question here is why? Do the specific cases show that democratic diplomacy really is different? Or do they reveal the familiar terrain of power politics, filled with threats, bluffs, and sharp calculations?

20

CHAPTER 2

These are not the only debates, but they are the most important ones. 1. Is the democratic peace statistically significant? 2. How do we classify hard cases? 3. Is there something distinctive about the way democracies resolve serious conflicts with each other? Beyond these three debates lies an overarching theoretical concern. If there really is a democratic peace, then our theory should not only predict this outcome, it should illuminate the underlying causal processes. The best check on that is to compare our theoretical account with the actual diplomacy of war and peace, to look at the processes by which peace is maintained.

Common Data and Definitions Fortunately, all these empirical debates have been fought out on common ground. Studies of the democratic peace—pro and con—rely on the same basic data. Figures about wars and casualties are drawn from the Correlates of War Project. Data on lower-level conflicts is generally taken from the collection on Militarized Interstate Disputes (MIDs).10 The characteristics of domestic regimes are taken from the Polity data sets. All these data sets have been refined through frequent use. Although war and democracy are elastic terms, there are some agreed-upon definitions in the empirical literature. War is an organized military conflict among states, a definition drawn from Clausewitz and other classical theorists. To exclude smaller skirmishes and the possibility that force might be used without central state authority, quantitative analysts generally limit themselves to conflicts involving one thousand or more battlefield deaths (although there is nothing magic about this round number). These definitions are widely used and predate any analysis of the democratic peace. They were not specially crafted to help proponents (or opponents) test this particular proposition. To classify the combatants, scholars turn to the Polity III data set, constructed by Keith Jaggers and Ted Robert Gurr, which ranks states along a continuum from democracy to dictatorship.11 Democracies score high in three basic categories: open, competitive participation in the political process; open recruitment of leaders; and constraints on chief executives. Jaggers and Gurr actually construct two inversely correlated scales, one to measure democratic features and another to measure autocratic. Ten on the democratic scale is a full, competitive democracy. Zero is Dante’s lowest circle of hell.12 Most scholars choose a threshold of six or seven to classify a regime as “democratic.”13 Still others, beginning with Doyle, impose one more limitation on the data. They concentrate on established democracies, where the regime has been in place for several years and power has been transferred peacefully by competitive elections.14

IS THERE PEACE AMONG DEMOCRACIES?

21

The Main Empirical Findings Using this common data, numerous scholars have studied the conjecture that democracies do not make war on each other. When Jack Levy surveyed the field over a decade ago, he famously concluded that no finding came as close to an empirical law. Since then, this topic has been the most thoroughly studied in quantitative international relations, by a wide margin. Bruce Russett, James Lee Ray, Nils Petter Gleditsch, and dozens of other sophisticated scholars have conducted countless studies, some reaching back as far as 1815, others covering only the period since World War II. Levy’s verdict still stands—and it now stands on much stronger foundations. Established democracies almost never go to war with each other, and that finding is significantly different from chance. The main empirical debates are whether this finding applies equally well to the nineteenth century and interwar periods, whether it covers all regions and all kinds of disputes (including disputes short of war), and whether it is reinforced by economic interdependence.15 Another debate is whether the phenomenon refers only to the interaction of democracies or whether it also applies to the foreign policy of any single democracy. Are democracies inherently more peaceful toward all other states, as R. J. Rummel argues?16 The growing consensus is that democracies are slightly more peaceful toward all states, but they still fight frequently against nondemocracies. That is not exactly news in Hanoi, Kabul, Belgrade, or Baghdad. In any case, the truly striking finding— conceptually and empirically—is not about individual democracies but about their interaction: wars between them are extremely rare, even though democracies fight other states frequently. The quantitative studies proving this point use a remarkable variety of statistical methods. They range from logit to probit to ordinary least squares, from generalized additive models to general estimating equations, from logistic regressions to hazard models. Their common focus is on pairs of states (dyads) and the likelihood that any given pair will go to war in a particular year. The central question is whether this likelihood is systematically lower if both states are democracies. There are vigorous skirmishes over how to code particular conflicts, how to classify the regimes of warring states, and whether to include military conflicts short of war. There are differences about the best statistical techniques, the right control variables, and the right way to handle specific pairs of states in the midst of large wars. Despite these differences and the wide variety of methods employed, quantitative analysts are agreed on the central finding: the democratic peace is real, not a statistical artifact or spurious result.17 Levy reiterates that conclusion in a recent review: “This empirical regularly cannot be explained by the geographic separation of democratic states, by extensive trade among democratic dyads,

22

CHAPTER 2

by the role of American hegemonic power in suppressing potential conflicts between democracies in the period since the Second World War, or by other economic or geopolitical factors correlated with democracy. . . . There is a growing consensus that the pacifying effects of joint democracy are real.”18 After nearly two decades of intensive research, Levy concludes that “no one has identified a stronger empirical regularly, and many make the law-like claim that joint democracy is a sufficient condition for peace.”19 The proponents’ strongest claim—that there have been no wars among democracies—is probably not correct. Using standard definitions, there have been one or two wars, although one combatant is almost always a marginal democracy. Nor do democracies confine themselves to polite diplomacy when dealing with each other. They sometimes threaten violence and occasionally back up their threats with troop mobilization and even a shot across the bow. Such disputes, however, seem to be less common among democracies than among other states. Still, the crux of the debate is over actual warfare among democracies. The answer to that is now well established. As Brian Lai and Dan Reiter put it, “A mountain of evidence has demonstrated that democracies do not fight each other.”20 This mountain is not a mirage, seen only by dreamy idealists. The democratic peace is real, and it extends well back into the nineteenth century. The point is to explain it.

Challenges to the Consensus: Gowa’s Numbers The most serious and thoughtful challenge to this empirical conclusion comes from Joanne Gowa and Henry Farber, in a series of coauthored articles and in Gowa’s subsequent book, Ballots and Bullets. They conclude that the democratic peace is not statistically meaningful before the 1940s. It does exist (and passes statistical muster) after World War II, but they argue it is the product of U.S. power and national interests, not democratic governance.21 Their argument is powerful for two reasons: it is empirically sophisticated, and it does not rely on questionable cases such as the Russo-Finnish War of 1939–40. What Gowa and Farber do is divide the historical record into several time periods, marked off by world wars, and look separately at each. Within each time period, they reduce the number of cases by counting only the first year of each conflict. Their argument is that subsequent years are obviously related to earlier fighting and so should not be analyzed with regression statistics designed for independent events. For much the same reasons, they choose not to count all conceivable pairs of opposing states in large, multilateral wars. They argue, with considerable justification, that these large wars have different dynamics from bilateral conflicts. For that reason, they simply exclude both world wars from any consideration of the democratic peace. That is obviously a controversial choice, but one happy effect is to exclude the war between Finland

IS THERE PEACE AMONG DEMOCRACIES?

23

and the Soviet Union, which some critics of the democratic peace include. Its coding is disputed because the Soviet Union, though a brutal dictatorship, later formed a wartime alliance with democracies. Does that mean we should code these democracies as fighting against democratic Finland? Some rules for coding multilateral conflicts say yes. They count every state on one side of a multilateral conflict as being “at war” with every state on the other side. Since that is misleading in the Russo-Finnish case and distorts its diplomatic history, its coding is controversial. Gowa and Farber avoid this tangle entirely by excluding all aspects of World War II. But if Gowa and Farber limit the actual number of wars, they add many more cases by including military disputes short of war. In these militarized interstate disputes (MIDs), states threaten or display force but do not escalate into full-scale war. Their argument is that the democratic peace should extend equally to these lower-level displays of force. Perhaps, but that methodological choice is controversial. The counterargument is that one important element of peace among democracies is their ability to limit escalation. That is especially valuable when quarrels are so serious that military force has actually been threatened. Using these methods to pare the data, count wars, and divide time periods, Gowa and Farber reach two major conclusions.22 First, the democratic peace is not statistically significant before World War II. Second, in the years since then, there is a peace among democratic states, but it misleading to call it that. It is really a Pax Americana, based on traditional calculations of national interest in a world suffused with American power. Although these arguments are thoughtful and sophisticated, they are less compelling than they first appear. To begin with, Gowa looks back as far as 1815 and finds only one war between democracies prior to World War II, the Spanish-American War of 1898. But because she has sliced this long duration into several periods, excluded so many cases, and included so many lesser threats, she cannot be sure if this rarity is statistically different from chance in each period. That even includes the 1920s and 1930s, when she finds no actual wars between democracies.23 Nor does she report any consolidated results for the period 1816–1945 (or 1816–1939), even though her most important conclusion is that the democratic peace does not hold prior to World War II. When we come to the great world wars, Gowa’s treatment raises still more questions. Whatever the technical reasons for excluding World War I and World War II, the effect is to sweep aside one of the central features of major war in the twentieth century. In these titanic struggles for survival and world domination, all the democracies lined up on the same side—twice. According to their own statements, that was no accident. They stressed shared democracy as crucial to their alliance ties and postwar goals. For the years since World War II, Gowa acknowledges that there really is a peace among democracies and that it is statistically meaningful. Her finding

24

CHAPTER 2

is a familiar one. Indeed, it goes unchallenged even by skeptics of the democratic peace. But Gowa downplays its democratic elements and offers a different interpretation. She points to traditional calculations of national interest rather than to anything distinctive among democracies. That is a reasonable conjecture, but she presents little compelling evidence for it. There is no real definition of interests beyond the assertion that there are no good quantitative measures for interests and that the best proxy is probably alliances. There is no effort to cope with the very real possibility that democracies share some common interests precisely because they are democracies— that their common interests are part of the democratic peace, not an alternative to it.24 In any case, she provides no affirmative account explaining how these interests are then translated into a stable peace. Most important of all, Gowa does not explain why this Pax Americana should be limited to democracies, when the United States also has numerous nondemocratic allies. The result is an intriguing set of arguments, a useful reminder that national interests are crucial to any convincing explanation, and a note of caution about the period before World War II. Gowa also reminds us that most democracies in the world are of recent vintage and that they came of age during a period of extraordinary American power. Even so, all this still does not add up to a fundamental challenge to a large, cumulative body of findings, which repeatedly shows that • there is a peace among democracies; • it is statistically meaningful when measured in a variety of ways; and • it is not limited to the Cold War years.

Challenges to the Consensus: Layne’s Realpolitik A second major challenge to the democratic peace comes from scholars who look closely at individual cases. Christopher Layne’s work is particularly important.25 In several detailed studies, Layne and others make two larger points. First, they argue that some contested cases, such as the Spanish-American War, are rightly counted as wars between democracies and that still others, such as the Anglo-Boer War and the U.S. Civil War, should be. That undermines the absolutist argument of those who think that democracies have never gone to war with each other, and it weakens the statistical case of those who think such wars are very rare. The real importance of these case studies, however, lies in their exploration of the political decisions about whether to fight wars or avoid them. The main point of this process tracing is to show that democracies are just like other states: when they draw near the edge of war, even war against each other, their calculations are dominated by traditional Realpolitik concerns. They look less like Kant and more like that other Prussian, Bismarck.

IS THERE PEACE AMONG DEMOCRACIES?

25

Take the Anglo-French conflict over possession of the upper Nile Valley. In 1898, their colonial forces confronted each other at the remote river town of Fashoda. There the commanders paused, invited each other to tea, and awaited instructions from Paris and London.26 When Paris eventually backed down, war between the two democracies was averted, and Britain extended its imperial control southward from Egypt. This outcome is consistent with predictions by democratic peace theorists. But Layne argues that considerations of peace among democracies had nothing to do with the outcome. What really counted were calculations of military strength and national interest. Britain had more troops on the ground at Fashoda and could send reinforcements from Egypt, if necessary. The French garrison was small and badly isolated. If a battle were fought in that remote location, Captain Jean-Baptiste Marchand and his small contingent would certainly lose. Britain was also better equipped for a wider war, should it come to that. Most important of all, France had to worry about its security within Europe, where it faced a formidable German neighbor. Germany had humiliated France in 1870–71, seized the French provinces of Alsace and Lorraine, and continued to build its own industrial and military power, backed by a growing population. War with Britain would make France more vulnerable to Germany and detract from its truly pressing security concerns along its eastern border. So France made the prudent decision and backed down at Fashoda. Friendship between democracies, international trade, and respect for human rights—the stuff of Kant’s democratic peace—had nothing to do with it. Fashoda is one of several cases where democracies came close to war. Layne and others highlight them, filling out the picture that democracies definitely engage in serious military disputes with each other. Their crucial point, however, deals with democratic decisionmaking about war and peace. Democracies avoid fighting for conventional reasons: they make standard prudential judgments, grounded in Realpolitik. Specific conflicts like Fashoda may not escalate to war, but that is only because of military deterrence and diplomatic caution. According to their argument, there is nothing new or distinctive about these policy processes. They are broadly consistent with realism and inconsistent with existing theories of the democratic peace. These objections to the democratic peace should be taken seriously on two levels. They invite us to make sure that individual cases are coded correctly, that fighting is rightly classified as a war (or something less), and that national governments are rightly classified as democracies (or something less). They also ask us to consider the processes by which war is chosen or avoided and to compare them with causal processes suggested by the democratic peace. Several students of the democratic peace—Bruce Russett, James Lee Ray, John Owen, and others—have responded effectively to the first part of this challenge. Looking closely at the most controversial cases, matching historical

26

CHAPTER 2

evidence to reasonable and consistent definitions, they have shown that virtually none of the controversial cases should be counted as wars between democracies. But we also need to take the second objection seriously, to understand the ways democracies manage to avoid war. Our theories should not only predict outcomes well, they should do so in ways that conform to real-world political processes and illuminate them.

Causality in Two Hard Cases: The Spanish-American War and the Russo-Finnish War Those who doubt the democratic peace frequently point to two wars as their best evidence: the Spanish-American War of 1898 and the Russo-Finnish War of 1939–40. Upon closer examination, however, neither offers strong support to the critics, to put it mildly. In the late-nineteenth century, an expansive U.S. democracy fought Spain, a decaying monarchy with some democratic elements. Antonio Ca´ novas del Castillo, Pra´ xedes Mateo Sagasta, and other Spanish leaders made some progress modernizing their country after the Second Bourbon Restoration (1875), but they were painfully aware their regime was still far from democratic or liberal. Formally, the government was a constitutional monarchy. In practice, it was an oligarchy. The two main parties shared power, rotating in office periodically through fraudulent elections, a system known as the turno pacı´fico. Government ministers were responsible solely to the crown. The judiciary, which was not independent, did nothing to curb these abuses or the pervasive corruption and clientalism that defined local politics.27 The whole system depended on rigged elections, arranged by local officials and the Ministry of the Interior.28 Results were often published before voting, and the parties rotated in office without regard to the actual vote.29 The designated winner would then “make” its majority in the Corte´ s, inverting the democratic practice where parliamentary winners form a new government. “No party could expect that [public] opinion should endow it with political power by electoral triumph,” says Raymond Carr in his authoritative history. “In these conditions Ca´ novas saw that no party could be allowed to monopolize crown favour in what was still an oligarchy. The oligarchs must rotate in office, like Aristotle’s citizens, in order to give the political nation the illusion of selfgovernment.”30 Even by the standards of the day, this was not a real democracy, much less an established constitutional one. Indeed, one consequence of Spain’s rapid defeat in 1898 was a renewed push to create truly democratic institutions in that country.31 The other “strong” case is the Russo-Finnish War, fought during the opening stages of World War II. Shortly after Stalin concluded his 1939 pact with Hitler dividing eastern Europe, the Soviet Union demanded territory and military

IS THERE PEACE AMONG DEMOCRACIES?

27

bases from Finland, its small democratic neighbor. Only a few decades earlier, the area had been part of Imperial Russia. Now Stalin was determined to regain as much as he could. According to Gerhard Weinberg, the Finns “had relied on Soviet adherence to their mutual treaties.”32 When Stalin demanded their territory in spite of these treaties, the Finns realized just how worthless their bargains with the Soviets actually were. To stave off invasion, Finland offered some territorial concessions, but not enough for Stalin, who launched a surprise invasion. Finland’s Arctic troops fought with extraordinary tenacity and considerable success against a mammoth Soviet army of forty-five divisions and three-thousand tanks. The Red Army lost more than 200,000 men during the winter campaign, before it finally overran the Karelian Isthmus, north of Leningrad.33 Stalin feared Western involvement if the war dragged on, so he decided to end it on compromise terms in the spring of 1940. He kept the territory he had conquered, most of it along the Baltic Sea, but he stopped fighting and dismantled the communist puppet government he had established to rule the country. The Finns had saved their independence by hard fighting against a formidable foe. Amid this life-and-death battle, Finnish diplomacy was very careful. They took special pains to avoid any conflict with their fellow democracies. Other democracies reciprocated. None declared war on Finland and none attacked it, militarily or even verbally. “The Soviet invasion of Finland outraged the West,” Williamson Murray and Allan R. Millett observe in their history of the Second World War. “The League of Nations expelled the Soviet Union, while Britain and France began inept planning for intervention on Finland’s behalf.”34 That planning was swept aside by the prospect that Nazi forces would soon attack Britain and France, with whom they were already at war. The Soviet Union remained neutral in this larger conflict. But Stalin’s recent agreement with Hitler worried the Western democracies, and with good reason. As Winston Churchill wrote a friend, “I still hope war with Russia may be avoided, and it is my policy to try to avoid it.”35 With these practical considerations in mind, neither Britain nor France wanted to provoke the Soviet Union by funneling military aid to Finland (although Stalin feared that they might do so). No democracy aided the Soviet Union during this brief but brutal war, and none was allied with it at the time. Depending on how multilateral wars are counted, Finland’s war against the Soviet Union still might be counted as a war against Britain and France, even though they did not declare war, did not fight directly, did not ally with Finland’s adversary, and never issued threats against the small country. Coding rules should be applied evenly, not bent in ad hoc ways to handle inconvenient cases. But coding rules for large quantitative studies should not blind scholars to the hard facts of individual cases. Realists should understand this especially well. After all, their theory was built in opposition to the empty forms of international legalism so prominent in the interwar years. Realism gains its

28

CHAPTER 2

considerable analytic power by seeing through this veil of formalism. It is inconsistent (and unworthy) to try to save a central Realist tenet—that all states behave alike under similar circumstances—by pretending that Finland’s national resistance to a unilateral Soviet attack is somehow a violation of the democratic peace. It is not.

What the Critics Have Right Another ten or fifteen possible exceptions to the democratic peace have been mentioned in the literature, some only in passing. Most are easily excluded by standard definitions of war and democracy and normal counting rules. A few involve new governments that are probably not democracies and certainly not well established. Most others are civil wars or efforts to secede, where one party is not a sovereign state. Civil wars raise important questions about the mechanisms of war and peace, but they do not challenge the basic empirical findings about the democratic peace. Moreover, for those who wish to count civil wars, the only proper procedure is to include all such wars. The results are hardly surprising and certainly not encouraging to critics of the democratic peace. Civil wars are common enough, but they overwhelmingly involve nondemocracies.36 They are rare among any democracies, and extremely rare among well-established ones, where power has already been transferred peacefully across party lines. The best quantitative studies show that the fewest civil wars occur at the two extremes of the political spectrum: established democracies and harsh, authoritarian regimes. Intermediate regimes are much more conflict-prone. They are open enough to permit displays of disaffection but are unwilling or unable to suppress it.37 Although democracies and harsh autocracies are roughly equal in avoiding civil wars, the autocracies are less stable than democracies and more vulnerable to civil wars if they liberalize.38 Based on data from 1816–1992, Ha˚ vard Hegre and his coauthors conclude that “the most reliable path to stable domestic peace in the long run is to democratize as much as possible. A change in that direction ensures the strongest ratchet effect in terms of consolidating political institutions and makes it less likely that the country will slide back into a state in which it is more prone to civil war. . . . There is a democratic civil peace.”39 A more profound challenge lies in understanding the diplomatic process surrounding wars between sovereign states, a point raised sharply by Layne and other critics. By choosing crises where war was averted, they are singling out prominent cases that the democratic peace should explain. Why, they ask, should we be convinced by gentle stories about shared norms and representative institutions when the real decisions are made by cold-eyed politicians, using a logic that Cardinal Richelieu would understand?

IS THERE PEACE AMONG DEMOCRACIES?

29

This is a critical objection because it implies that the democratic peace is nothing more than a correlation, an unexplained and possibly accidental statistical relationship. Democracies might have avoided war so far, they say, but that is essentially a lucky outcome. These states operate according to the same harsh diplomatic logic as all others, and they are all constrained by the same international structures. Defenders of the democratic peace have not rebutted this important claim. One reason may be that proponents of the democratic peace have diverse theoretical conceptions, some based on norms, some on institutions. But if either correctly specifies the causal process, we ought to find telling traces of it in the case studies. If we do not, then we need to reconceive our theories or abandon them. William Dixon is right when he says that the controversy “is no longer about history, but about theory. It is a controversy about historical contingency versus causal mechanism and the impact of democratic norms and institutions in securing and maintaining peaceful diplomatic relations.”40 The basic problem, I think, is that existing theories wrongly downplay the role of interests and prudential calculations in forging the democratic peace. That is why Gowa, Layne, and other critics can pose them as alternative explanations. They say, quite correctly, that current theories of the democratic peace concentrate exclusively on domestic political institutions and shared norms, ignoring the pursuit of national interests. “Democratic states, like others, have interests and experience conflicts,” observes Kenneth Waltz, the founding theorist of neo-Realism.41 He is right, of course. Omitting these mixed national interests is a profound theoretical gap in any understanding of war and peace, including the democratic peace. It is simply impossible to understand how states interact without building on the foundation stone of modern international relations theory, that states seek self-interested goals, as they understand them, and try to maximize their outcomes in a risky and uncertain international environment. That understanding is not only shared by Realists of all stripes, it is also the heart of modern, game-theoretic models of cooperation and thus central to institutionalism. This means our understanding of the democratic peace should assume that states are self-interested, value-maximizing actors. That is as true of democracies as it is of monarchies, dictatorships, and other nondemocratic states. What differs across these states is how they define their self-interest, how they inform others about it, and how they pursue it in their interactions. What substantive goals do they embrace beyond their obvious concern with physical security and territorial integrity? How do domestic actors and institutional arrangements shape this assessment of interests, and what role do norms play? In exploring these issues, we need to remember that the basic problem to be explained is not one state’s policy but a distinctive kind of interaction among democracies. That is the central finding of the quantitative literature. Individual democracies may be slightly more peaceful than other states, but the demo-

30

CHAPTER 2

cratic peace is fundamentally a process of interaction. It is not enough to say that democracies hold values different from other states and make policies in different ways. These normative and institutional elements must be connected to the interactive element of the democratic peace. Our explanation must go beyond simply predicting correlations. It should illuminate the logic by which democracies sustain a separate peace, and, if possible, show the limits of such cooperative action. These processes are not limited to the immediate moments in which war is chosen or avoided. They also include the longer period during which serious threats are perceived and conflicts simmer. That is exactly what Hobbes underscores when he speaks of a “state of warre,” a time of foreboding and military preparation that surrounds actual fighting.42

Hard Cases: Fashoda Let us look, then, at this period surrounding some conflicts mentioned by critics of the democratic peace. Fashoda is a good place to start because it is a hard case and we have already begun the discussion. The French provoked the crisis by sending a small group of military explorers across the nearly impenetrable jungles of central Africa to the Nile Valley. When the British learned of this threat to their control of Egypt, including the water route to India, they took immediate action. Prime Minister Salisbury, a cool and experienced realist, told Paris that his government fully intended to hold the Nile. He sent General Herbert Kitchener’s army south with five large gunboats to back up the claim. The French temporized and equivocated—it was all a misunderstanding, said the foreign minister who had actually sponsored the expedition—and then they wisely backed down. In one sense, this is a straightforward story of power politics. But that it is not the entire story. If we look at the events before and after Fashoda, we can see how democracies often manage to contain even the most serious conflicts. The face-off at Fashoda was serious not only because it came so close to war but because it was so deeply rooted in the basic policies of the two great imperial powers of the day. Britain’s African colonies ran along a north-south axis from Alexandria to Cape Town. In Cecil Rhodes’s grandiose vision, they should soon form a solid line, linked by rail, from the Nile delta to the Cape of Good Hope. The French, already well established on the Atlantic coast of Africa, were now extending their own colonies along an east-west axis. These two lines of imperial policy were bound to clash. They did at Fashoda. In the background was a long-simmering dispute over Egypt. Nominally, Egypt was part of the Ottoman Empire. In practice, France and Britain had established informal joint control by the mid-nineteenth century, working in tandem with weak local rulers. French engineers designed the Suez Canal, and

IS THERE PEACE AMONG DEMOCRACIES?

31

French financiers bankrolled it. Britain actually opposed building the canal, fearing it would leave France entrenched along the short route to India. When the waterway was completed in 1869, it immediately became the prime commercial artery between Europe’s Great Powers and their colonies in the East. It was ideal for the new steam-powered, propeller-driven ships, which could glide smoothly down its narrow corridor. Its importance, and that of Egypt, was heightened by the fragility of Ottoman power in Constantinople. Although the Suez Canal was working well, Egypt itself was not. Among its worst problems was a huge foreign debt, secured by tax revenues. When Egypt could no longer sustain its interest payments, its finances were taken over by European bondholders, led by Britain and France. This system of joint financial control soon foundered on two issues: ownership of the canal and maintenance of public order. The English took control of both, edging out the French. The English gained control of the canal in 1875–76, after Egypt’s ruler, the khedive, had to sell his stake to pay mounting debts. Prime Minister Benjamin Disraeli moved swiftly. When the Bank of England lacked enough ready cash to purchase the shares, he secured financing from the Rothschilds.43 The French were shocked to find their great achievement slipping into English hands. Worse was yet to come. In 1882, rioting broke out in Alexandria, killing fifty Europeans and threatening all foreign interests. The British and French, still the dominant foreign powers in Egypt, planned a combined intervention and moved their gunboats into position. At the last minute, however, the French shied away from going ashore and left the British alone to suppress the uprising. Britain’s success left it the dominant power on the Nile, a position it refused to share with France.44 London always claimed its rule was temporary, but it never seemed to end. Paris was humiliated and for more than a decade grumbled loudly about London’s domination of Egypt.45 Indeed, that is one reason why Marchand was sent across Africa, to reestablish French claims to the Nile. These concerns resonated with a wider French public, which supported colonial expansion, especially in North Africa. When French papers reported that Marchand’s forces had finally reached the Nile and now faced Kitchener’s army, there was considerable support for war. What had begun as a low-cost probe of British intentions and an assertion of French rights now became a national issue. These popular sentiments were fueled by the militarism and xenophobia that flared during the retrial of Captain Alfred Dreyfus for treason. Yet none of this swayed French policymakers when they contemplated a costly war with Britain over the inhospitable reaches of the upper Nile. What slowly dawned on French leaders was that Salisbury and his cabinet meant business. They were resolutely committed to holding the region and would not defer to French claims. London backed its claim by sending a large contingent of troops and was prepared to send more. True, the territory was remote and a war would wound Anglo-French relations, which had been im-

32

CHAPTER 2

proving steadily. It might even topple France’s fragile democracy, which was now convulsed by the Dreyfus trial. None of these considerations budged Salisbury, who indicated (quite credibly) that he would rather fight than concede. This information about British resolve was new to French policymakers, and, according to Kenneth Schultz, it was vital to the conflict’s outcome.46 British policy sent a series of costly signals, which clearly communicated their intent to hold the upper Nile. French leaders read these signals accurately and backed down before Britain’s superior military capabilities and its willingness to use them. French prudence and clearheaded calculation at Fashoda is not the end of the story. The aftermath is just as important, and it is crucial to our understanding of how the democratic peace works. France’s embarrassing withdrawal might well have fueled growing hatred and mistrust. Britain might well have emerged from its victory with swaggering contempt for France’s unrealized colonial ambitions and its vulnerable position in Europe. The war they so narrowly averted could have become grist for future conflicts. In fact, the opposite occurred. Neither government pursued this dangerous path of escalating rivalry. Rather, they built upon their peaceful outcome at Fashoda to reach a broader accommodation over colonial policy in North Africa. This more inclusive bargain had far-reaching consequences for their overall relationship. On France’s side, this policy of reconciliation was led by The´ ophile Delcasse´ , the same minister who had originally authorized the Marchand expedition to Fashoda. Britain sealed the bargain by supporting a dominant role for France in Morocco, effectively compensating them for their loss in Egypt.47 This required British confidence in their French partner because Morocco guarded the entrance to the Mediterranean, an area vital to British commercial, military, and colonial interests.48 In other words, both states sent costly signals expressing confidence in their new relationship. That, in turn, paved the way for further collaboration. Never again would the two largest empires come close to war over their colonial ambitions. Their settlement was engraved in the Entente Cordiale of 1904. It immediately produced cooperation in the far reaches of empire. In 1906, for example, French and British officers worked together to draw frontier lines from Niger to Lake Chad. As Martin Gilbert observes, “Two European Powers, that a decade earlier would certainly have clashed, and possibly even fought, over their competing claims to an African hinterland, were working in tandem, and in amity.”49 Although the Entente was initially limited to colonial affairs, it quickly developed a wider significance. To implement the bargain, French and British military leaders began to meet frequently, coordinating their force deployments and strategic plans. Their diplomatic links were progressively strengthened over the next decade.

IS THERE PEACE AMONG DEMOCRACIES?

33

Fashoda, then, marked the low point of Anglo-French conflict over empire. It was to be an inflection point. Its successful resolution was no mere tactical withdrawal. Rather, it led to a fundamental and enduring diplomatic adjustment that benefited both France and Britain. In the decade before World War I, these two democracies sharply curtailed their rivalry, surmounted a long history of mistrust, and forged increasingly close and reliable ties in the face of a growing German threat. There is one more twist to the story, which shows just how important it is to manage rivalry and assuage fears, and just how difficult. This epilogue concerns Germany’s effort to break the new ties between Britain and France. Germany had its own designs on Morocco and opposed French dominance there. Not only did Berlin want a larger colonial sphere of its own, it wanted to increase diplomatic friction between Paris and London. Trying to isolate France was nothing new for Germany. That had been Berlin’s steady policy since 1870. What was new was Wilhelm II’s ambitious Weltpolitik, a vague design to increase Germany’s global power. That project required new colonies and a blue-water navy to reach them. The drive for colonies marked a major change for Germany. Bismarck had avoided such diversions in the 1870s and 1880s, when others were scrambling to divide Africa. He concentrated instead on the more pressing tasks of European diplomacy and building a new German state. When an explorer tried to persuade him to acquire colonies, he responded, “On my map, Africa is located in Europe. Here is Russia, here is France, and we are situated in the center; that is my map of Africa.”50 Wilhelm II had a larger map but a smaller mind to read it. He was bent on a new course, one designed to make his mark in the world and catch up in the race for empire. He did not mean to be excluded from significant, uncolonized areas like Morocco, to which Germany surely had as much right as France. Wilhelm and his advisers also had a bigger goal in mind. They hoped to use colonial disputes like this—indeed, to provoke them—to drive a wedge between their old enemy, France, and Britain. As a bonus, the quarrel might force Britain to draw closer to Germany, a natural ally in any dispute with France. Wilhelm would play this card several times, always losing and continually doubling his bet. Indeed, he played the same card twice in Morocco, first in 1904 and then, with more military bluster, in 1911. The results were disastrous. They did not simply fail. They actually strengthened the alliance between Britain and France. London saw Wilhelm’s clumsy, belligerent policies as still more evidence that Germany was expansive, erratic, and militaristic—a dangerous combination for a country with Germany’s powerful economy, growing population, and fearsome military. By 1904, the British were already worried about Admiral Tirpitz’s plan to build a major navy. By 1911, the Royal Navy was truly alarmed and was busy building a new generation of Dreadnoughts to stay ahead of Wilhelm and Tirpitz. The Ger-

34

CHAPTER 2

mans were rapidly building battleships of their own. This reciprocal process quickly degenerated into a full-scale naval arms race. Not only were the Germans laying more keels; they were widening the Kiel Canal to move their new ships easily between the Baltic and the North Sea. This, too, posed a threat to Britain, and was understood as such. One senior British official privately predicted (accurately, as it turned out) that Germany would launch a European war as soon as it finished widening the canal. Together, these policies projected German power into the North Sea, which Britain traditionally considered its home waters and a protective moat against invasion. Now, all England looked across these narrow waters to the growing German battle fleet at Wilhelmshaven, and they shuddered. In this tense atmosphere, even German overtures for partnership with Britain were seen as veiled threats. Unlike France at Fashoda, Germany continually and obtusely misread the depth of British resolve. British policymakers called the German threats “blackmail” because they believed Kaiser Wilhelm was using an ominous military buildup to demand an alliance on his own terms.51 London considered Germany’s record and saw a belligerent and underhanded adversary, not a reliable partner. That was true even before the controversies over Morocco and the Ottoman Empire. In 1901, Francis Bertie of the Foreign Office warned, “In considering offers of alliance from Germany it is necessary to remember the history of Prussia as regards alliances, and the conduct of the Bismarck Government in making a Treaty with Russia concerning and behind the back of Austria, the ally of Germany.” Bertie goes on to describe the pervasive mistrust created by Germany’s military expansion and its successive betrayal of partners: Germany is in a dangerous situation in Europe. She is surrounded by Governments who distrust her and peoples who dislike or at all events do not like her. She is constantly in a state of Tariff war with Russia. She has beaten and robbed Denmark, and for that purpose she took as partner Austria and then turned round on her confederate and drove her out of Germany, eventually making her a rather humble ally. She has beaten and taken money and territory from France. She covets the seaboard of Holland and the Dutch know it, and, as the Belgians are well aware, she has designs on the Belgian Congo.52 As German power grew over the next decade, so did fear and mistrust among nearby states, for perfectly justifiable reasons. The German Empire, as they well knew, had been forged by blood and iron, built by wars it had deliberately provoked (in 1864, 1866, and 1870). It now bristled with the world’s most powerful army and a growing navy. Its diplomacy was a tissue of opportunism. Germany had grown by tactical alliances, discarded as soon as they had achieved their immediate aims. This combination of military power, menacing purpose, and diplomatic opportunism posed an acute danger to Germany’s continental neighbors and a rising threat to England’s home islands.

IS THERE PEACE AMONG DEMOCRACIES?

35

Given Germany’s new structural position, only the most carefully framed diplomacy and grand strategy could limit the threats that other Powers reasonably perceived. Such a policy would have avoided a major naval buildup, sought better relations with Russia (which Germany had undermined in the 1890s), and sent clear, credible signals to England that it would not challenge her vital commercial and strategic interests. The tragedy is that Germany could have done all that without jeopardizing its own security. Instead, the Kaiser and his policy advisers blundered forward, magnifying the inherent dangers. To assert Germany’s interest in Morocco, they provocatively sent a battleship. To develop economic ties with the Ottoman Empire, they began building a major rail line to Baghdad. Predictably, the British saw the railway as a direct challenge to their dominance in the Persian Gulf, long an area of British interest and, after 1903, the Royal Navy’s source of fuel.53 Nor was this the only evidence of Germany’s troubling new interest in the eastern Mediterranean. Berlin sent senior military advisers to modernize the Ottoman army, and the Kaiser himself traveled to the region, where Germany had never before shown any interest. In all these sensitive areas, from Constantinople to Agadir, the more Germany asserted its military power, the more it endangered British, French, and Russian interests. Its aggressive and threatening policies were self-defeating. They promoted a countervailing coalition directed at Berlin. As a result, between 1902 and 1907 Bismarck’s old network of alliances dissolved, and an increasingly isolated Germany faced a new Triple Entente. “This shift was partly the response to the growth of German power,” concludes Stephen Walt, “but it was also driven by the growing conviction that Wilhelmine Germany harboured unusually aggressive intentions.”54 None of these crises escalated to war, just as Fashoda had not. The difference is that the democracies—but not Germany—ultimately resolved their crises by • • • •

clearly and credibly communicating what they considered vital interests, mollifying each other’s legitimate fears, adhering to the bargains they struck, and ultimately incorporating individual bargains into more durable partnerships.

Democracies moved, fitfully at first, from wary deterrence to strategic cooperation. Their cooperation was firmly rooted in national interests, to be sure, but it was produced by policy mechanisms that involved essential features of democratic governance. This democratic component was not the direct voice of the electorate. In both France and England, the public actually seemed to favor war over Fashoda. It was not human rights, which played only a minor role, despite shared norms. It was not economic ties; Britain did more business with Germany than with France. It was, instead, the capacity to strike durable, cooperative bargains, based on their recognition of some joint interests and their ability to signal resolve credibly.

36

CHAPTER 2

Democracies were well placed to strike these bargains and sustain them. One important element was the continuity of governance in England and France, across changes of leadership. Another was transparency. Each government could watch as its partner’s choices were debated in parliament and the press. These institutional arrangements disclosed the depth of support (and opposition) to policies, reduced the amount of private information held by each government, and helped them identify areas where they could benefit by exchanging commitments. By building extensive lines of communication and adhering to earlier bargains, they reassured each other in critical moments, such as the Moroccan crises. Their success went beyond resolving specific conflicts. They progressively transformed a serious, long-term rivalry into stable, peaceful cooperation, much of it directed at the rising German threat.

Hard Cases: Anglo-American Relations in the Nineteenth Century Critics of the democratic peace tell us, with some justification, that this was not the only case where established democracies came close to war. They point to Anglo-American relations in the early phases of the Civil War. In November 1861, U.S. sailors boarded a British mail packet, the Trent, and seized two Confederate diplomats. This seizure rode roughshod over international law, and Britain was furious. Their maritime rights as a neutral power had been violated, and they protested vigorously. Lord Palmerston and his cabinet demanded an apology and return of the captives. They underscored the point by dispatching troop reinforcements to Canada. Now it was Washington’s turn to be angry, enraged that Britain was intruding into America’s internal struggle. Public feelings ran high, and some hotheads demanded war with England. Critics of the democratic peace, notably Layne, have correctly identified this as a serious crisis and highlighted the Realpolitik reasons why it did not escalate to war. This critique can actually be broadened to encompass a series of AngloAmerican disputes, where diplomacy was sometimes accompanied by military threats. Over the course of the nineteenth century, the United States and United Kingdom confronted each other over the Oregon Territory, Canada’s boundaries, fishing rights off both Newfoundland and Alaska, the right to build a canal across Central America, the Union blockade during the American Civil War, Confederate purchase of warships built in England, American demands to be compensated for damages caused by those warships, and, at century’s end, boundary disputes in South America and renewed questions about a Central American canal. Several quarrels led to military threats, and some, like the Trent affair, came close to blows.

IS THERE PEACE AMONG DEMOCRACIES?

37

Although the United States and Britain avoided any fighting after 1814, the long record of discord and threats does raise important questions. The fundamental issue, once again, is whether these disputes were resolved simply because of military deterrence and prudent diplomacy, based on familiar strategic calculations. Was this merely caution born of fear? Was there anything truly “democratic” about the peace that ensued? To answer these questions, it is helpful to consider the long arc of conflict between the United States and Great Britain during the nineteenth century. Their many conflicts stemmed from two basic sources: (1) the international dimensions of the Civil War and (2) the United States’ growing regional power and relentless expansion across North America. Expansion generated conflicts over Canada’s integrity, territorial control along the Pacific Coast, and spheres of influence in the Caribbean. The disputes were especially serious because Canada remained a British possession and because the United States still nursed ideas of annexation, which it had failed to accomplish in 1775 and 1812. London was well aware that Canada was nearly impossible to defend. The lines of communication were too long, especially to the new western territories, and the border ran for thousands of miles. There were numerous points of attack from the south. Canada was, in effect, hostage to the United States.55 Some of the most difficult moments in Anglo-American relations came during the Civil War. A conflict so vast, bloody, and protracted was bound to affect the world’s greatest trading country and maritime power. Britain had significant stakes in the American economy and divided loyalties in the secession conflict. For more than fifty years, it had been the world’s leading foe of slavery, with warships stationed off west Africa to halt the trade in human cargo.56 At the same time, its industries relied on long-staple Southern cotton and staunchly opposed protectionism by Northern manufacturers. Southerners believed (and Northerners feared) that these commercial ties would ultimately win out and that England would recognize the rebels, to be followed quickly by France.57 Of more immediate concern to London was the Northern blockade of Southern ports, which grew increasingly effective. That injured British commerce, raised naval insurance rates, and caused massive unemployment in the textile industry, which had no good alternative to raw cotton from Southern states. Britain was affected in other ways as well. As the world’s largest capital market, it was caught up in Confederate efforts to sell bonds to foreign investors. As the world’s largest shipbuilder, it was the target of Confederate efforts to buy and outfit warships. Because of Britain’s extensive trade with the South and its history of conflicts with Washington, tensions ran high in the North, especially during the Trent affair. But the popular mood did not sway Lincoln, who had no intention of fighting Britain. “One war at a time,” he said, leaving no doubt as to which war was more important. Although he damned Britain’s policies as unaccept-

38

CHAPTER 2

able interference, his focus was squarely on sustaining the Union against the Confederate rebellion. The Union had no resources to spare for a war with Britain and only a faint hope of victory. “With such a civil war upon our hands,” observed Attorney General Edward Bates, “we cannot hope for success in a . . . war with England, backed by the assent and countenance of France. We must evade it—with as little damage to our own honor and pride as possible.”58 Moreover, Washington knew that the outbreak of actual fighting would lead Britain to recognize the Confederacy. France would likely follow suit. Preventing such recognition was the centerpiece of America’s foreign policy during the Civil War. That was why its navy had seized the Confederate diplomats in the first place, to prevent them from lobbying Britain and France for recognition and weapons. Under these difficult circumstances, Secretary of State William Seward, Senator Charles Sumner, and other Union leaders advised Lincoln to meet British demands and return the diplomats. Lincoln finally conceded the point, and Seward issued an apology noting that America had always supported the maritime rights of neutral states.59 The bare bones of this crisis bargaining are strikingly similar to Fashoda. When the Union seized the Southern diplomats, it had serious purposes (to block European aid to the Confederacy) and reasonable hopes it would succeed without major opposition. (That is not so different from France’s position when it sent Marchand’s probe across Africa.) What Union leaders did not know— what they could not know in advance—was how London would react: very strenuously indeed, as it turned out. The London papers violently attacked Washington, claiming their country had been insulted and demanding satisfaction. The prime minister, Lord Palmerston, convened an emergency session of the cabinet by declaring, “I don’t know whether you are going to stand this, but I’ll be damned if I do!”60 These angry protests, in the streets and in government, were backed by clear, costly signals of resolve. The Royal Navy was put on alert and military supply operations prepared. Eleven-thousand troops were immediately dispatched to Canada, leaving port as an English band played “Dixie.”61 The Union quickly recognized the folly of its policy. It was simply too costly and too risky to continue. It raised the likelihood England and France would support the South, and it could well lead to outright war with England. Worst of all, it imperiled Washington’s overriding goal: winning the war to preserve the Union. There was no point in trying to bluff. The Union’s vulnerability was common knowledge. Once Lincoln and his advisers concluded that England would fight over the issue, they wisely backed down. As the Trent crisis unfolded, it produced important new information, which shaped the outcome. Before the diplomats were seized, Americans were uncertain about Britain’s position. London’s public stance was guarded and ambivalent, torn between economic dependence on Southern cotton and moral disgust

IS THERE PEACE AMONG DEMOCRACIES?

39

for the slave system that produced it. Once the diplomats had been seized, however, London made its position clear with blunt threats and troop deployments. Washington read the signals and conceded the issue. Even after the Trent affair was settled, other Anglo-American disputes arose and lingered. The thorniest was over shipping destroyed by two Confederate vessels, the Alabama and the Florida, which had been built and outfitted in England. Washington demanded satisfaction. These demands did not cease with Lee’s surrender at Appomattox. Successive presidents pressed London for compensation. Some senators went further, claiming that these warships had prolonged the Civil War and that Britain should pay for that, too. These issues, a volatile mix of national pride and serious money, festered for more than a decade. The United States refused to let the matter drop or settle it for a nominal sum. How did the United States and Great Britain handle this long series of disputes? With some bluster and occasional threats, but also with considerable tact about using force and a willingness to find common ground, as they did in 1842, 1846, and 1871. While London would surely have preferred divided power in North America—with several states sharing the continent south of Canada—it avoided provocation. It did not try to break the Union or block America’s westward movement. Washington showed similar restraint. Although the United States hoped to acquire Canadian territory, it made no effort to seize any after the War of 1812. It was loath to fight another war with Britain, which would have resisted any incursion. Washington also realized, with some consternation, that Canadians did not wish to join the United States. There would be no internal uprising to greet an American invasion. Conquest and domestication would be a matter of brute force. Canada later reinforced the point by building a transcontinental railroad and encouraging migration to the western prairies, partly for economic reasons, partly to head off migration by Americans who might seek union with the United States, as they had done in Texas.62 By the same token, Britain and Canada eased relations with the United States by making a number of concessions, including some border adjustments. They carefully avoided challenging the United States militarily. Canada’s arms were modest and purely defensive, posing no threat. The United States chose not to force the issue. The result was a series of self-enforcing agreements that ended the naval arms race (the Rush-Bagot Agreement of 1817–18) and subsequently limited fortifications along the border. That was expanded several decades later to become a completely demilitarized border. The Canadian-American relationship that began with war and then deterrence was gradually transformed into a secure relationship and eventually a security community where no one contemplated war. During the tense years of the American Civil War, Britain tried to maintain its neutrality. It refused to recognize the Confederacy unless it could establish effective sovereign control, which it never managed to do. Once the war had

40

CHAPTER 2

begun, Britain stood aloof and tried to prevent its arms from slipping into Southern hands. (Not diligently enough, according to the Union.) Any chance Britain would actually aid the Confederacy died on the battlefields of Antietam, Gettysburg, and Vicksburg. Of course, neutrality did not mean Britain would abandon its essential interests as the world’s leading maritime power, including its rights on the high seas. After the war, it was willing to negotiate over damages caused by Britishbuilt ships, but only for the harm they caused directly. It refused even to consider the opened-ended costs of “prolonging the Civil War.”63 Senior American negotiators quietly told their British counterparts that they understood and would not press the point, which they had no hope of winning. These calibrated policies produced a succession of agreements, which held up well without cheating, coerced changes, or the need for armed enforcement. The most protracted dispute, over the Confederate warships, was finally solved by submitting it to arbitration at Geneva. The settlement of Civil War claims by treaty and arbitration signaled a wider resolution of Anglo-American disputes in North America. Britain had finally made her peace with America’s westward expansion. As one eminent British historian put it, London gradually “dropp[ed] the idea of the United States as a potential enemy.”64 For its part, the United States repeatedly demonstrated that it no longer threatened Canada, despite America’s great military advantages. These policies of mutual forbearance, political accommodation, and reliable performance gradually built up confidence and overcame a legacy of fear and mistrust. As Lord Russell told an American diplomat early in the Civil War, “Mutual distrust had produced mischief. England and America seemed each to suspect the other of hostile intentions, while it was possible that both were quite mistaken.”65 Overcoming that distrust and avoiding still more mistakes was a learning process, and that was bound to take time. This process of accommodation and confidence building was largely completed by the Washington Treaty of 1871 and the Geneva arbitrations, which together settled the Civil War claims. With that, the last, desultory work on Canadian fortifications ended. The promise of an undefended border was finally realized, a truly historic achievement. Before Britain and America could achieve a full-scale rapprochement, however, they had to resolve one more dispute beyond their borders. It arose in the 1890s as America began to assert its newfound power in the Caribbean. The specific problem was an ill-defined border between British Guiana and Venezuela in the remote Orinoco River valley. The underlying problem was Washington’s effort to establish its regional hegemony and make the Caribbean its lake. Such a bold policy was bound to create frictions with Britain, which had long been the dominant foreign power in Latin America. As far as Venezuela and British Guiana were concerned, no one knew the exact boundaries, and no one much cared until gold was discovered. Then

IS THERE PEACE AMONG DEMOCRACIES?

41

they cared. The United States was mainly worried that Britain would become interested in the Orinoco valley and use it to project imperial control deep into South America. That is why the United States was eager to back Venezuela’s claim to the area. In fact, Britain had no intention of expanding its empire in South America, but it did assert its own counterclaim. The U.S. position grew increasingly shrill, unilateral, and public. Communications between London and Washington on this issue were often confused and testy, but the two powers never came close to armed conflict. The dispute began in 1895, when U.S. Secretary of State Richard Olney wrote his British counterpart that Washington was interested in the boundary dispute. He stated America’s position with a puffed chest and a memorable phrase. “The United States,” he said, “is practically sovereign on this continent, and its fiat is law upon the subjects to which it confines its interposition.”66 Despite this harangue, Olney and President Grover Cleveland did not try to impose a specific outcome. They simply demanded arbitration. The British took little notice but eventually responded with a bland call for further talks. Before that response arrived, however, President Cleveland mentioned the issue in his State of the Union address. His public pronouncement raised the temperature, though well short of the boiling point. Still, Cleveland’s speech raised audience costs and clearly showed American resolve, backed (as it was) by rising economic and naval power. After more letters and posturing, Britain accepted arbitration, effectively recognizing the United States’ rising interest in the region. Ironically, Britain won the arbitration. Here, as at Fashoda, the aftermath is critical to understanding relations between democracies. What Britain now understood was that the United States fully intended to play a dominant role in the Caribbean. U.S. intentions and resolve were clear. That was acceptable to Britain because the United States also made clear it fully accepted Britain’s existing colonies and its stake in trading and investing freely in the region. Those were the essential terms of a new and enduring regional relationship, which both sides were determined to sustain. Two years later, when the United States and Spain went to war over Cuba, Britain again signaled its recognition and acceptance of the new U.S. role. Despite Britain’s close ties to Spain, London immediately told Washington it had no interest in the matter and would stand aside. In fact, it offered the United States informal support during the war and won considerable appreciation in return. Britain also recognized U.S. regional hegemony in another, more concrete way. It did not object to U.S. plans to act alone in building and fortifying a Central American canal. That represented an important change in British policy, reversing a mid-nineteenth-century agreement that the United States and Britain would jointly control any waterway. Now, at the beginning of the twentieth century, the British negotiated a new arrangement, bowing to the U.S.

42

CHAPTER 2

position. The United States, for its part, wanted an amicable new agreement, not a unilateral abrogation of its earlier treaty commitment. During this power transition, the United States was often blunt and assertive; the British were sometimes resistant. But this major transfer of regional power took place without any serious danger of a diplomatic break between the United States and Britain, much less a war. Part of the explanation lies in Britain’s other global interests, including its security interests close to home. But much of the answer lies in the ability of the two countries to signal their positions effectively and work out a durable accommodation. That settlement acknowledged America’s new power while protecting Britain’s economic stake in the region. Based on their successful record of prior agreements, Britain and America had growing confidence in their ability to manage their evolving relationship. They were fully convinced they could strike bargains that both parties would uphold. This Anglo-America achievement bears on the democratic peace in several ways. One reason the United States did not try to invade Canada after 1812 was its failure to spark a popular uprising. Canadians made it clear that they were not eager to join the United States. The United States grudgingly respected that, refusing to incorporate them by force.67 Second, in the disputes over Civil War damages and the Venezuelan boundary, the United States and the United Kingdom were willing to submit their differences to binding arbitration. Some explanations of the democratic peace emphasize this willingness to rely on disinterested legal procedures, which parallels the domestic legal system of liberal democracies. Submission of serious interstate disputes to arbitration or international courts is rare, even among democracies, but it does occasionally offer a way out of disputes without the need for states to reach a bilateral settlement. In my explanation, it is one more element in the varied governance arrangements democracies make with each other. Third, almost all the disputes ended with explicit agreements, which both democracies then adhered to without cheating or opportunism. What made these bargains feasible was both sides’ ability to signal effectively and then to recognize and accommodate each other’s vital interests in regional trade, investment, and security. Naturally, both states pressed for individual advantage, but they also searched for mutually acceptable bargains and then kept them. Not every country seeks out such common ground, especially rising countries bent on asserting their regional hegemony. When Japan expanded into Korea, Manchuria, and China in the late-nineteenth and early-twentieth centuries, it initially accommodated the other imperial powers, partly out of weakness. By the late 1920s, however, a stronger and more nationalist Japan began excluding Western powers from trade and investment in its growing empire. Within a decade, it had unilaterally overturned a whole series of major agreements restraining Japan’s role in Asia. The result was an increasingly fractious relationship between the Japanese and other foreign powers in the

IS THERE PEACE AMONG DEMOCRACIES?

43

area: the British, Dutch, Americans, and Soviets. In little more than a decade, Japan was at war with all of them.68 Democracies do better, but their accommodation is not instantaneous. It was certainly not instantaneous between the United States and the United Kingdom. It had to be built by a combination of learning and confidence building. It took decades to reach encompassing agreements and overcome deeply rooted mistrust. It was not enough for states simply to recognize that they were both liberal constitutional democracies. (Here I disagree with Owen.) At the beginning of the nineteenth century, Anglo-American relations were often troubled and sometimes verged on war. Neither side had fulfilled all its major obligations under the Treaty of Paris (1783), which ended the Revolutionary War. The United States promised to repay British creditors and compensate Loyalists who lost properties when they fled. In fact, the new nation delayed its payments. By the same token, Britain refused to evacuate the Northwest Territory or abandon its string of forts within the newly independent United States, from Lake Champlain to Lake Michigan.69 Each charged the other with bad faith, and the disputes lasted for over a decade.70 Other disagreements, rooted in the treaty’s ambiguous language, lasted even longer. Britain flatly rejected America’s interpretation of its fishing rights off Newfoundland’s Grand Banks or the exact boundaries of maritime Canada. According to one leading historian, these differences over treaty obligations “remained for decades to bedevil American-British relations.”71 Nor did the Treaty of Ghent lead to a comprehensive Anglo-American settlement, though it did end the War of 1812. Until the 1870s, the two states managed to avoid war mainly by deterrence, caution, and sensitivity to the costs of fighting, what might be called a “cold democratic peace.” Still, they built a record of signaling their interests and resolve, settling their disputes without war, and then sticking to their bargains. That laid the basis for a deeper cooperative relationship, which emerged after 1871. What began as deterrence grew into a stable, reliable partnership—a warm democratic peace—by the turn of the century, once they had settled the Venezuelan boundary dispute and agreed that the United States would control any Central American canal. This deepening cooperation is partly explained by shifting interests. Britain’s security concerns in North America had declined, while its regional investments and trade had risen. Canada was no longer a colony, the United States had expanded across the entire continent, and contentious boundary issues like the Oregon Territory had been settled long before. Moreover, both Canada and the United States were vigorous, growing economies, with very profitable ties to Britain. As a truly global power, Britain also faced other pressures that diverted its gaze from North America. Its empire had expanded rapidly after 1870, requiring considerable attention and frequent use of armed force, most recently in the Boer War. By the early 1900s, Britain faced a growing German threat across the North Sea. These changes raised the cost of any

44

CHAPTER 2

conflict with the United States, reduced the potential gains, and opened wider possibilities for profitable collaboration. But shifting interests are not the complete explanation for the developing Anglo-American partnership. Their new relationship cannot be explained, for example, by saying that Britain was diverted by more pressing security issues close to home. In 1812, the United States and Britain had gone to war, even though Britain faced a much more urgent and daunting threat: Napoleonic hegemony in Europe. In the 1890s, when the United States and Britain completed their rapprochement, Britain faced no serious dangers in Europe. At the time, it was still uncertain whether its main rival was France, Germany, or Russia. Its focus on Imperial Germany did not come until a decade later, after Germany had begun building a major fleet and Russia’s power had receded after its defeat by Japan. By then, the Anglo-American partnership was already firmly in place.72 Nor did changing interests swiftly produce the stable partnership that finally emerged. It took decades for both countries to overcome mistrust, to learn from their success in resolving crises and sustaining bargains. In the process, they established realistic grounds for confidence and worked out the basic lines of a regional bargain. In retrospect, the terms are obvious. The United States established an uncontested sphere of influence in North America and the Caribbean, while Britain won substantial U.S. protection for its vital economic interests, including the right to expand its trade and investments. The United States, its continental expansion complete, asserted informal control in the Caribbean. That included the sole right to build and fortify a canal in Central America. By the same token, the United States made clear (by word and deed) that it would leave Canada alone, protect foreign investors throughout the region, and respect Britain’s right to trade freely. It was, in short, a partnership built on mutual interests encoded in reliable bargains. This was not a tactical or fleeting arrangement. It was an increasingly strong, durable relationship. It encompassed two democracies that had forged a series of successful agreements, managed them well, and now saw each other as trustworthy partners.

Conclusion Concentration on specific disputes, like Fashoda and the Trent, can be useful because it spotlights the process by which war is sometimes avoided. Critics of the democratic peace reasonably ask its proponents to examine such cases and explain how their general mechanisms of causation work in specific cases like these. As I have tried to show, there are genuinely democratic elements at work in these cases, alongside the traditional pulling and hauling of national interests that realists rightly emphasize. My explanation does not try to downplay these interests or struggles. Quite the contrary; they are fundamental to

IS THERE PEACE AMONG DEMOCRACIES?

45

our understanding of international politics and must be integrated into our understanding of the democratic peace. My explanation is designed to show how interests, prior beliefs, and information can work among democracies to produce stable agreements that avert war. Part of the answer lies in democracies’ better access to information about each other. With less private information, it is easier for them to establish mutual confidence, based on a better understanding of others’ preferences and behavior. Still, even in relatively open states, some information remains hidden. Here, too, democracies have an advantage: their superior capacity to signal intentions and reveal private information credibly. Of course, all states try to signal, but democracies have better tools and more of them. Their leaders can pledge themselves both to the electorate and to other elected officials. In the United States, for example, senior officials often commit the administration to specific policies in testimony to the Senate Foreign Relations Committee. Such commitments and public disclosures make it harder for democracies to bluff and easier for them to make reliable promises. In both the Anglo-French and Anglo-American cases, we saw how beliefs about partners can change in ways that open wider possibilities for agreement. To understand these changes, it is important to look beyond a single crisis and its resolution. We need to consider the evolution of bilateral relations over time, especially the ways beliefs and relationships change after a conflict. After the tense face-off at Fashoda, Britain and France worked out an entente regarding their colonial disputes, which deepened into general security cooperation (though still short of alliance commitments). After differing about who should pay for shipping damaged during the Civil War, the United States and Britain worked out a basic agreement over North America. After the Venezuelan boundary dispute, they extended that agreement to the Caribbean in ways that protected British interests and accommodated the United States’ rapidly rising power in the region. Before concluding our discussion of the Anglo-American and Anglo-French cases, we should recognize that they are unusual in two important ways. First, they are the only long-running disputes between established democracies. There are many such rivalries among nondemocratic states, according to an extensive quantitative literature. But these are the only two enduring rivalries among democracies, and they were resolved more than a century ago. Democracies proliferated in the twentieth century, especially in the last half, but there have been no long-running disputes among them. Second, in nearly all long-running rivalries, the protagonists fight multiple wars against each other. These democratic rivals, as we have seen, fought none at all. With the resolution of the Anglo-French rivalry after Fashoda, and the Anglo-American rivalry after the Alabama shipping claims (or perhaps the Venezuelan boundary dispute), there have been no enduring rivalries between established constitutional democracies, and no wars between them.

46

CHAPTER 2

What gives democracies this edge in settling disputes and avoiding war? My answer is that democracies have a contracting advantage, a unique capacity to diminish the risks of agreement with each other and forge stable, mutually profitable bargains. In the next chapter, I will show how, in theory, this contracting advantage should reduce the risks of war. After that, I will explore the sources of this democratic advantage in more depth.

3 A Contracting Theory of the Democratic Peace and Its Alternatives

THE LAST CHAPTER showed that the democratic peace is real. This chapter sets out a simple theory to explain it, based on constitutional democracies’ superior ability to forge reliable bargains with each other. We will then compare the contracting theory to other major approaches. For this contracting explanation to work, we need to show that • reliable bargains could, in theory, lower the risks of war; and • constitutional democracies are better equipped, in theory and in practice, to make such bargains. The first task confronts one of the most difficult issues in social theory: how to explain the onset of war. Most systematic work on the subject, including Realism, can be termed rationalist. That is, it assumes that the actors (usually simplified to mean states and their leaders) choose policies in purposeful, strategic ways. They do their best to advance their own interests, as they understand them, given the resources available to them and the information and beliefs they have about their adversaries and partners. This strategic interaction sometimes leads to war. Our main question, within this larger account, is how the actors’ capacity to strike reliable bargains might lower the chances of war. If reliable bargains really can avert war, then we need to know whether some states are better placed than others to make such bargains. My answer is that constitutional democracies have systematic contracting advantages. These advantages are the fortunate by-product of institutional arrangements designed to give citizens control over their leaders. They are the unintended consequences of structures devised for domestic purposes. Democratic policy processes, for example, are open to public view, subject to challenges from opposition parties, and scrutinized by a free press and other branches of government. In a world where news travels instantly, giving this information to voters, elected officials, and journalists also gives it to other states. That makes democracies inherently more transparent than dictatorships, traditional monarchies, or one-party states. In the language of information economics, democracies simply possess less “private information.” That can be a disadvantage at the bargaining table because it makes bluffs, deceit, and hyperbole less effective. But it is a major asset when it comes to making bargains reliable. Democratic threats and promises are simply more credible.

48

CHAPTER 3

Democracies also have significant advantages in making and sustaining long-term commitments. Their edge stems from two sources: elections and constitutions. Candidates make promises to voters, deliberately creating costs for themselves if they abandon them. These “audience costs” are a kind of imperfect bond, making commitments more credible and trustworthy.1 Elections are a source of continuity in another way, as well. Winning candidates typically concentrate on the center of the electorate (the median voter), where preferences are fairly stable.2 Elected leaders are constrained by these realities, which lends continuity across successive administrations. Besides elections, democracies are characterized by constitutional arrangements to prevent rulers from acting arbitrarily. These, too, are imperfect, but they are designed to ensure that states keep their fundamental commitments to citizens, such as respecting their civil rights, property, and transactions with the government itself. Profligate and headstrong kings regularly renounce such commitments, as the founders of modern democracy understood all too well. Democracies’ superior ability to share information and make credible, durable commitments, along with the continuity of their domestic regimes, gives them enormous advantages in building stable relationships with similar partners. They are better equipped to identify mutually profitable bargains and then depend on each other’s undertakings. When the regimes themselves are stable, as established democracies are, then these commitments can endure over very long periods. It is these unique institutional advantages that allow constitutional democracies to make more reliable international bargains, often grounded in multifaceted, long-term relationships, and thus avoid the dead-weight costs of war.

Rationalist Theories of War There are countless explanations for why states go to war. Some emphasize irrational ideologies, mistaken beliefs, or nonrational motives such as the quest for honor or glory. Most social scientists, however, fold even these motives into a broadly rationalist account. That is, they see war as the product of purposeful actors pursuing their goals (as they understand them), using efficient means to reach their ends, and interacting with others who do the same thing. Of course, the actors may well have different preferences, military capabilities, and information. They may face very different choices. The analytic problem is to understand how their capabilities, preferences, prior beliefs, new information, and choices interact, sometimes producing wars and sometimes peaceful solutions. A terrible paradox overshadows this explanatory approach. How can it possibly be rational to engage in combat when it leaves so much devastation and costs so much—in lives cut short, in soldiers and civilians maimed, in property

A CONTRACTING THEORY

49

destroyed, and in human effort devoted to making bombs and bullets instead of food, clothing, and music? The answer is that it is not rational after the fact.3 Unfortunately, it is sometimes quite rational before the fact, as each state sees it. Perhaps both states think they will win in a two-party confrontation.4 Perhaps one fears a preemptive attack. Maybe it misunderstands just how important a particular issue is to another state. Or maybe it mistakenly thinks its adversary will back down, only to find itself embroiled in a battle to the death. Such misperceptions and errors of judgment are not only possible, they are commonplace.5 But that stops short of a full understanding of war causation, as James Fearon points out in his penetrating analysis of the rationalist literature.6 Fearon focuses on a question that must be central to any rational analysis: Why can’t these misperceptions and errors be cleared away and war avoided by prior agreement? After all, the costs and dangers of war give the actors strong incentives to sort out their misperceptions and reach profitable bargains, if any are to be had. The short answer is: either private information prevents them from finding an agreement or commitment problems raise doubts about whether it will be carried out. That is, they cannot agree or they cannot be sure another state will actually do what it promises. If they are unsure whether another state will keep its promises, then they will either avoid agreements in the first place, refuse to rely on them, or restrict their compass and hedge them with costly devices for monitoring and enforcement. Hypothetically at least, an agreement could yield more to everyone because it would eliminate the costs and risks of war. Both winners and losers would receive whatever they would get in the gamble of war, plus some share of the resources that would otherwise be wasted in fighting. Everyone would be better off, a Pareto-superior solution.7 If such a solution is available, then why do we sometimes fail to grasp it? Labor economists ask virtually the same question about strikes, and their answer is instructive.8 Strikes, like pale versions of war, are wasteful means of reaching final settlements. “When the parties have reasonably good information about the costs and the likely outcome of the strike, it is irrational to strike,” concludes George Borjas. “The firm and the union can agree to the strike outcome in advance, save the cost associated with the strike, share the savings between them, and both parties will be better off.”9 John Hicks anticipated this point long ago in The Theory of Wages (1932).10 More recently, it has been formalized in game-theoretic models of employment contracts with incomplete information. The basic finding is that strikes are driven by gaps in the parties’ information. “Strikes can occur when information is asymmetric,” according to Ronald Ehrenberg and Robert Smith, “either because one party mistakes the other’s true position, or because only by striking can one party acquire a signal about the other’s actual ‘needs.’ ”11

50

CHAPTER 3

Since this point is directly relevant to our analysis of war, it is worth clarifying with a simple example. Consider a corporation that tells its union it is willing to increase wages, but only if it expects high profits. If it expects poor market conditions and low profits, then it simply cannot afford a pay hike. At least that is what the company says. Who knows if it is telling the truth? To prepare for bargaining, both the firm and union make their best forecasts about market conditions and profit potential. The firm, however, is in a much better position to forecast accurately because it knows much more about its new products, customers, competitors, financing, and myriad other aspects of its business. Possessing such sensitive information gives it a marked advantage in estimating future profits. Even if it wants to share this confidential data with the union, it may not be able to do so credibly. Naturally, if the company says the news is great, then the union will readily believe that. Since the company would then have to pay high wages, why would it say so if it were not true?12 But what if the firm says market conditions are gloomy and it can afford only low wages? The union has good reasons to be wary about such self-serving declarations. Maybe the company is truly sharing its private information and honestly giving its best forecast. Or maybe it is just lying to keep wages low. Under these circumstances, the skeptical union may strike, because if the company is lying, workers could win higher wages. On the other side of the table, management is privately willing to accept a strike if (and only if) it expects low profits and needs to keep wages down. In this stylized example, there would be a strike if the union expects high profits while the company expects low profits. In fact, the firm’s willingness to pay the price of a strike may be the only way it can effectively signal the union it really does expect tough times and cannot afford a pay hike. This signal is more credible than words alone because it is costly. That is, it is unlikely to be faked. As a result, it sends the union a strong, believable signal about the firm’s forecast of bad market conditions. Note that whatever the final settlement is—high wages or low—both parties could have done better without a strike. But they cannot find their way to such a settlement because of private information.13 That is true even if the union is entirely transparent and only the company has private information. The same analysis applies to all kinds of bargaining with private information, such as pretrial negotiations in civil litigation.14 The rational analysis of war proceeds along similar lines. Given the high stakes involved, there are powerful incentives for states to look closely at each other’s capabilities and intentions. They spend heavily to do just that, using a combination of analysts, diplomats, and spies. Working through diplomatic channels and aided by espionage agents and surveillance technologies, they hope to learn which areas are vital to their friends and foes, which policies please or trouble them, and which issues are good prospects for agreement. “Ambassadors are the eyes and ears of states,” as Guicciardini

51

A CONTRACTING THEORY 15

said of his fellow Renaissance diplomats. The same can be said today. But vigilant as they are, they are necessarily imperfect eyes and ears. Diplomatic talks, espionage agents, spy satellites, and policy analysts may still not be enough to separate fact from fiction or to anticipate another country’s policies with confidence. The problem is that governments often exaggerate and mislead about what matters to them and what they intend to do about it. Talk is cheap, in diplomacy as in wage negotiations. In practice, it may be impossible to distinguish between bluffs and sincere claims that an issue is vital or that force will be used in certain circumstances. States also hide their military capabilities, sometimes exaggerating and sometimes downplaying specific weapons. Take the Soviet Union (please). It often held parades to show off its massive arsenal, marching troops with their latest weapons through Red Square. To impress NATO observers, some major weapons were quickly repainted with new serial numbers and paraded through the square again. In 1955, in the days before satellite reconnaissance, the Soviets played the same trick with a new generation of long-range bombers. As squadron after squadron flew over Moscow, U.S. personnel counted and secretly snapped pictures, stunned by how rapidly the Soviets were building these new aircraft.16 Actually, it was only one squadron—one very busy squadron. It kept circling the city, adding a few extra planes each time while the Americans kept adding up the numbers. The elaborate hoax worked, obscuring Soviet weakness and stimulating American fears of a “bomber gap.”17 Such deception is routine in international affairs, repeated because it sometimes works. The result is that when states analyze military data and evaluate diplomatic communications, they may well miscalculate their relative power or misjudge how important an issue is to the other side.18 The underlying problem is that states have private information about their own capabilities, intentions, and resolve, and they may have compelling reasons to keep their own counsel and perhaps to deceive. As Lord Macaulay says admiringly of William of Orange, “He knew how to keep secrets, how to baffle curiosity by dry and guarded answers.”19 Machiavelli goes further, instructing princes not to flinch from cunning and deceit to protect their rule against all challengers. His chief example, both positive and negative, is Cesare Borgia, known as Duke Valentino. The duke’s rise shows how vital deception can be in securing a prince’s rule. His fall shows how failing to guard against treachery can be catastrophic, even for a skillful ruler. Duke Valentino’s great mistake was not to block the election of Pope Julius, whom he had previously injured.20 The Renaissance papacy was a powerful position, capable of destroying local rulers. Valentino could have prevented Julius’s election, but he was lulled into a false sense of security by the aspiring cardinal, who masked his hatred.21 As soon as Julius was elected and no longer needed Valentino’s help, the mask came off. The new pope

52

CHAPTER 3

swiftly revealed his full fury, abandoned his promises to Valentino, and ruined the duke and his kingdom.22 Machiavelli does not blame Julius for his duplicity. He considers it an effective tool of statecraft, if used properly. He blames Valentino, a master of dishonesty himself, for his own credulity. Machiavelli later makes the same point in a more general way. “Experience in our time shows that those princes have done great things who have valued their promises little, and who have understood how to addle the brains of men with trickery; and in the end they have vanquished those who have stood upon their honesty.”23 Granted, Machiavelli’s own time was unusually treacherous, and his profound insights into it are laced with worldly cynicism. Still, he describes profound features of politics across the ages: (a) private information, (b) incentives to mislead, and (c) no effective way of precommitting to fulfill promises. This combination has far-reaching implications for relations among states. It stands squarely in the path of profitable bargains that could avert serious conflict and outright war. The path to cooperation is open only if the participants can penetrate the fog of ignorance and deceit, identify a range of mutually acceptable solutions, settle upon one, and then ensure it is carried out faithfully. Can they find a solution better for everyone than the costly gamble of fighting, and can they secure it for the uncertain future? Theorists of liberal internationalism remind us of the many benefits. Machiavelli reminds us of the awful perils. In this simple schema, states act rationally at every stage, but they lack vital information and the ability to commit themselves reliably to future actions. “The fact that states have incentives to misrepresent their positions is crucial here,” concludes Fearon, “explaining on rationalist terms why diplomacy may not allow rational states to clarify disagreements about relative power or to avoid the miscalculation of resolve.”24 Even costly signals may not solve the problem because they may not be not perfectly informative. A very common signal, such as mobilizing troops, could well deter an adversary. But there is another possibility. The adversary could consider the mobilization a bluff, designed simply to win more at the bargaining table. It might be a feint, not a real indication that force will be used. If so, then counterforce and coercive action could force the bluff and defeat it without war. The bluffer would back down. The problem comes if the mobilization is a true sign of intent, a symbolic statement that the issue is vitally important and worth fighting over. In that case, it is not a bluff and mistaking it for one would lead to war. In this world of shadows and deceit, joint benefits can slip away—and wars may well begin—because the range of mutually profitable solutions and the actors’ fidelity to their promises are known only in hindsight, if at all. As Shakespeare puts it: “Time shall unfold what plaited cunning hides.”25 By then, unfortunately, time may have run out.

A CONTRACTING THEORY

53

A Rationalist Theory of the Democratic Peace That is the crux of the rationalist analysis of war, and it has powerful implications for understanding the democratic peace. Wars are, at bottom, caused by problems of information and commitment, which block states from resolving serious differences. Constitutional democracies are simply better placed to solve these information and commitment problems. They are inherently more transparent, and they have unique institutional mechanisms to project promises in ways that are more credible and trustworthy. These are fundamental aspects of constitutional, democratic governance, deeply rooted in the connection between citizens and the state. Together, they lower the chances that democracies will exploit each other or betray their agreements opportunistically for shortterm gains.26 That, in turn, gives them solid grounds for greater confidence in other democracies and opens wider opportunities to capture long-term gains from cooperation. Equally important, it allows them to avoid joint losses, including the costs of fighting. Partners that are trustworthy and transparent have much better chances of resolving differences peacefully by striking reliable bargains short of war. These bargains are rarely single treaties, standing alone. As important as individual treaties are, they are usually embedded in a richer set of bilateral relationships. States, especially neighbors or great powers, typically have many agreements with each other and a variety of long-term relationships. This combination gives them multiple sources of information, multiple tools to sanction violations, and multiple chances to learn their partners’ preferences, behavior, and dependability, for better or worse. In other words, individual deals are usually aspects of larger relational contracts, where both parties project their exchange into the future and secure it within their larger relationship. In settings like these, where the parties expect to continue dealing with each other and information is incomplete, reputations are crucial. Are they trustworthy partners, willing and able to keep their bargains? Do they cheat when they are poorly monitored? Do they wriggle out of deals to win short-term advantages? These are vital questions for potential partners in relational bargains. But reputation also has its limits. It is a kind of “surety bond,” and, like any other bond, it is worth forfeiting when the stakes are high enough. In a military crisis, for instance, the immediate stakes may be very high indeed, and the future sharply discounted. Prevailing in such a crisis or avoiding outright defeat may be more important than maintaining a general reputation for fair dealing or sustaining long-term relationships. What is at stake right now can— and sometimes does—dwarf the future. How do these difficult conditions affect democracies and war? First, democracies generally avoid crises in the first place, at least in dealings with each other. They develop profitable, durable relationships and manage low-level

54

CHAPTER 3

disputes as they arise, well before military force is threatened. Their relationships are generally stable and their disputes are managed effectively, in part because their promises are reliable. Still, as we saw in the preceding chapter, even well-established democracies do occasionally confront each other with military force. Britain and the United States came close to blows over the Trent affair; France and Britain came close at Fashoda. These crises were settled by prudent concessions. The stronger power demonstrated its resolve; the weaker one recognized its vulnerability. Machiavelli or Thucydides would hardly be surprised by such old-fashioned power politics. What makes military threats different among democracies (aside from their rarity) is that these states are less likely to bluff and more likely to carry out their threats. As a result, when democracies demonstrate their resolve in crises, their threats are more believable, just as their promises are. The transparency of democracies makes both threats and promises more credible, lessening the chances that misperceptions will lead to war. Finally, in the aftermath of these crises, democracies learn. Their bargains are not mere stopgap measures to settle crises. They seldom unravel or become sources for yet another political showdown. Having clarified their stakes in particular issues, they typically move beyond them and reconcile in mutually profitable ways. They develop stronger long-term relations, building a more durable peace over time. These are not the only ways democracies are better placed to pursue a separate peace. Their shared norms open a wider range for mutually acceptable solutions and may induce greater confidence in other democracies, as Owen says. Their economic interdependence (a common feature of capitalist states, and especially developed democracies) raises the costs of conflict and probably increases the bargaining range, as Russett and his colleagues have shown. Their interdependence also gives them still more tools to signal their resolve. In liberal democracies, the civil protection given to ethnic, religious, and racial minorities lessens the chances of conflict with neighboring states linked to these groups. That is especially true if the neighbor is also a liberal democracy and does not pursue its own policy of ethnic separatism and minority oppression. These issues of race and ethnicity rarely surface in diplomacy among democracies.27 Finally, there is some evidence that democracies can more accurately judge the costs and risks of conflict, mainly because they openly debate policy issues and must confront alternative perspectives. In the process, they gain a clearer understanding of the bargaining range both sides might prefer to war. All these systematic differences can be understood within the larger framework of reliable contracting.28 They either expand the range of mutually profitable bargains or clarify that range for the participants. There are several ways, then, in which democracies are likely to be more reliable partners. Of these, two stand out because they are so fundamental to democratic governance and so far-reaching in their impact on other states: democracies’ transparency

A CONTRACTING THEORY

55

and their capacity to project promises (and threats) effectively in a difficult, anarchic environment. In later chapters, we will discuss these contracting advantages in detail. But first, having laid out the basics of our contracting theory, we can usefully compare it to other explanations and other theoretical approaches.

International Relations Theory and the Democratic Peace How does this contracting theory of the democratic peace differ from other explanations? Contracting is a variant of the rationalist theory of war and, as such, shares its strengths and weaknesses. For example, it treats underlying state preferences as stable and exogenous. Although that makes the theory less comprehensive, it is not a gaping hole for at least two reasons. First, nearly all IR theories take preferences as given, rather than as determined endogenously. (Constructivism is the exception.) Second, the assumptions we make about democratic preferences are not unusual or controversial. Of particular importance: we do not assume that democratic preferences are especially peaceful. Thus, we are not hiding a set of powerful preferences just offstage, letting them do the hard work of explanation. In fact, we can simply use the same basic assumptions as neo-Realists, namely, that states primarily seek their own security.29 They need not be especially peace loving or determined to transcend warfare as an element of interstate relations. Of course, they may develop such goals—indeed, they do within the stable environment they have gradually forged with other democracies. But we need not assume that democracies have especially peaceful preferences to generate the predictions of the contracting theory. Security seeking is a common assumption in IR scholarship and underpins Kenneth Waltz’s foundational work on structural Realism. “In anarchy, security is the highest end,” as he succinctly puts it.30 Martin Wight, whose work on international society began the English School, characterizes state motives in much the same way. “By fear we mean, not an unreasoning emotion, but a rational apprehension of future evil, and this is the prime motive of international politics,” he concludes. “For all powers at all times are concerned primarily with their security, and most powers at most times find their security threatened.”31 This emphasis on security seeking is a useful and moderate assumption for a contracting theory of the democratic peace. The search for security, as Robert Jervis and others have noted, leads states to focus carefully on the motives of surrounding states.32 In a formal model of this process, Andrew Kydd derives the important conclusion that fears can be allayed and conflicts avoided if security-seeking states engage in reciprocal, costly signaling. That is a crucial building block of the contracting approach

56

CHAPTER 3

because, as we have argued, democracies are particularly well suited to signal their intentions credibly. Kydd agrees. He notes that while all states can signal, “democracies find this especially easy because their policy making processes are so open.”33 These costly signals may—or may not—be needed to convey what democracies want and what they intend to do. In either case, their transparency makes their preferences more comprehensible to partners. With this conventional assumption about preferences plus our simple schema of basic democratic features, the contracting theory produces the range of results we have already discussed. That is, democracies have a superior capacity to exchange enduring, self-binding commitments, devices to lessen the chances of surprise defections, and the capacity to govern their bilateral bargains as they are carried out. When they face crises, they signal resolve or its absence more clearly than other states. It is these capacities, taken together, that generate the prediction that established, constitutional democracies will not fight wars against each other, as well as the other conjectures we made about the distinctiveness of democratic dyads. (They are listed in chapter 1, analyzed in chapter 6, and summarized in chapter 7.) What about other well-known theoretical approaches to IR? Do they explain the democratic peace and go beyond it? One indication that they do not is the long silence among leading theorists. For more than a decade now, evidence has been accumulating that the democratic peace is real and that it needs a convincing explanation. The prominence of this empirical finding invites scholars from different theoretical perspectives to offer their explanations and contrast them to others. Yet leading theorists with well-defined analytic approaches have said little. They have not claimed much for their own theories and have said little about the shortcomings of others. Realists, to their credit, have directly engaged the issue. Their approach has been two-pronged, as we discussed in the last chapter. First, they deny that there really is peace among democracies. Second, they highlight threats and military confrontations among democracies, showing that these states behave just as others do. Democracies or not, states doggedly pursue their interests and sometimes coerce each other to get their way—nothing new under the sun. The chief empirical criticism—that there is no democratic peace—fails. Studies of large data sets, using a variety of statistical methods, have shown convincingly that democracies fight each other significantly less often and engage in fewer militarized disputes than do nondemocracies. These results are strongest for the second half of the twentieth century, but they hold for nearly two centuries, since the fall of Napoleon. On the other hand, Realists are surely correct when they say democracies have edged close to war. Christopher Layne and others properly demand that any causal explanations of the democratic peace make sense of these close calls. Purely normative theories, for instance, cannot account for the uncom-

A CONTRACTING THEORY

57

fortable fact that democracies sometimes threaten each other and even brandish military force. If normative explanations were right, then democracies should treat each other in benign ways because they share peaceful values, respect each other’s popular governance, and prefer to settle disputes through courts rather than through bare-knuckled power. Realist critics are also correct to point out that democracies actively pursue their own interests, maintain sizable military forces, and engage in hard-nosed diplomacy with other democracies. The difficult task is to explain both why war is so rare between democracies and why they nevertheless threaten each other occasionally (and actually go to war frequently against nondemocracies). The Realist critique of the democratic peace has problems beyond its failure to explain the statistical findings. While Realists are right to say that values and beliefs cannot explain the democratic peace, excluding them altogether masks deep-seated differences that affect interstate relationships. Values inform and shape interests, though surely not with infinite plasticity, and they shape beliefs about other states. Indeed, national interests cannot be understood without some reference to underlying values.34 Territorial and physical security are primary values, to be sure, but they are not the sole values of states and citizens. They take on that dominant role only when threats are high and imminent. In modern states, wealth is highly prized, both for its own sake and as an indispensable source of long-term military strength. That is especially true in democracies, where ordinary citizens have a stake in prosperity and a say about its importance as a policy priority. This political stake in prosperity, and its pursuit through international trade and investment, has strengthened ties among modern democracies.35 Some Realists also assume fear and mistrust are constant features of international politics.36 They are critically important, but they are not omnipresent. They vary among partners, and they vary over time, for reasons that go beyond raw military capabilities. To take a well-known example: the United States fears nuclear weapons in Iraq, Iran, or North Korea, but not in Britain or France, which are far more powerful.37 Here, too, prior beliefs and learning are crucial. The more general point is that established democracies have managed to assuage one another’s fears. Their mutual confidence has grown as their bilateral relationships have matured. It is a central element of the democratic peace. Policymakers are well aware that threats vary dramatically. For them, the key questions are: Which states pose threats? What kinds of threats are they? Which challenges are most troubling? and What should we do about them? States spend considerable time and effort trying to answer these questions. They carefully assess sources and levels of threats, not only because it is so costly to respond but also because their own actions may threaten others, who take countermeasures.38 Excessive fears carry another hidden cost: they block mutually profitable bargains because they are deemed too risky.39 That is why

58

CHAPTER 3

states try so hard to differentiate among potential partners and adversaries. Which states are trustworthy? Which are threatening, and which are simply ill defined and poorly understood? Theorists need to ask a set a parallel questions: Why do fear and mistrust vary? and What difference does it make?40 While Realists have been fighting this rearguard action, that most difficult of maneuvers, other theorists have enjoyed watching them squirm. Realists, they say, are missing one of the two or three most important—and compelling—findings in modern international relations. They denigrate the findings only because the findings denigrate their theory. After all, Realism in all its forms stresses that all states are alike, fearful for their security in an anarchic world—harsh realities unchanged since the Peloponnesian Wars. Democracies are no different, according to Realist theories, and they ought to behave no differently, either in their individual foreign policies or in their relations with each other. They should be profoundly wary of all other states. That was the main thrust of the vigorous debate over “relative gains.” Would states really forgo direct benefits simply because others might gain more? At the time, U.S. economic growth was sluggish, and prominent Realists argued that America had good reasons to be troubled by faster growth in Europe and Japan.41 True, they were friends now, but they could become adversaries if intentions changed. Capabilities, by contrast, change only gradually and visibly, making them a more reliable guide to potential danger. If European and Japanese growth rates remained higher than America’s over the long term, their latent military capabilities would also grow and pose a rising menace, or so many Realists argued. Today’s allies could convert their prosperity into tomorrow’s military power and endanger the United States. “The major effect of anarchy on states stressed in realist theory,” says Joseph Grieco in his study of U.S.-European trade relations, is “the recognition by states that others might seek to destroy or enslave them”42—friends today, enemies tomorrow. There is no question that relative power poses a recurrent risk in international affairs. States must be attentive to it, especially when it involves persistent foes and plausible threats. The real question is whether it is even remotely applicable to U.S.-European negotiations over nontariff barriers in the 1980s, the subject of Grieco’s book. His explanation is that fear about relative gains is pervasive and shapes bargaining even among longtime allies and trading partners.43 In this view, what really matters are relative capabilities. Intentions—and alliances—can change in an eye-blink. The Realists’ point, in the relative-gains debate and elsewhere, is that democratic alliances are just like all others. They are provisional responses to common threats, not lasting partnerships. Many alliances are exactly that. The United States and Soviet Union, for instance, saw their wartime alliance that way. As Assistant Secretary of State Adolf Berle wrote in his diary, “We are much better off if we treat the Russian situation for what it is, namely, a temporary confluence of interest.”44 Realists generalize that point. Indeed, they over-

A CONTRACTING THEORY

59

generalize it, viewing all alliances as temporary confluences of interest. Their rationale is that threats can emerge unexpectedly from any quarter, mainly because relative military capabilities can shift and intentions change.45 Capabilities and intentions certainly can change, and states must be watchful. What is missing here, however, is a more sustained consideration of the vital role played by incomplete information.46 Stabilizing expectations, reputation, signaling, and beliefs play no part in this analysis. The impact of material resources is not refracted through any wider set of common intentions or shared values. Beliefs are not updated by learning. Nor are they shaped by institutions, which foster continuing relationships and promote information flows.47 As a result, it is hard to understand how democracies build reputations, diminish fears, and establish long-term relationships with other democracies. But they do. Beliefs are crucial when states assess threats and choose strategies. To take just one prominent example: modern Germany and Japan have deliberately avoided building military capabilities to match their powerful economies. Wary Realists correctly note that they could always change course; true, but that misses a major part of the story. After all, these are two of the richest nations in history, and they have deliberately chosen not to build large militaries to defend themselves and assert their power. Unlike China, they have not pursued a basic strategy of balancing against America. Presumably, these rich and potentially powerful states are confident that American military force can and will be used to defend them, and that America will not use its overwhelming power for predation. Their steady refusal to convert their latent capabilities into actual military force sends neighbors and partners a powerful signal of their own benign intentions. It allows others to identify them as unthreatening partners. Their transparency makes that signal even clearer and gives others advance notice of any changes and the reasons behind them. The stubborn fact is that democracies do behave differently. And they achieve different outcomes. That embarrasses Realists on the most prominent terrain of their theory, security affairs, and on a core conclusion of their theory, that all states behave the same way in similar circumstances. In fact, there is more than enough squirming to go around. No major theoretical approach has offered a truly convincing explanation of the democratic peace. Few have seriously tried. Of course, no theories can be expected to explain everything. Even strong theories, which explain crucial features of international relations, may not be relevant to this particular issue or explain it well. That is not a fatal weakness, but it is a weakness. It is a more serious one if the theory focuses on security issues. Since the findings on the democratic peace are relatively recent, nearly all relevant IR theories were developed long before the evidence began to accumulate. Scholars working in these varied traditions have simply sidestepped the new empirical findings and avoided grappling with the them. The exception

60

CHAPTER 3

(aside from Realism, which denies the findings) is constructivism, which developed at the same time as the democratic peace literature. Not surprisingly, its scholars have been more concerned with offering an explanation. Still, the most remarkable feature of the theoretical landscape is not the strength or weakness of explanations grounded in different traditions, but their near-total absence. Amid so much cumulating empirical work, what is truly striking is the theoretical silence. Let me try to fill that silence. I want to capture, very briefly, how these other theories might explain the democratic peace, based on their general analytic approaches. In presenting them succinctly I inevitably conflate differences within each approach. My aim, however, is not to caricature them but to capture their defining elements. I want to show how they might explain the democratic peace, point out their strengths and weaknesses, and then compare them to my own contracting explanation. I begin with constructivists. The heart of the Constructivist explanation is the importance of shared ideas, particularly norms about core values, appropriate behavior, and fair procedures. This approach overlaps substantially with other normative approaches. The main difference is that Constructivists make even larger claims about the role of norms and ideas, arguing that they do more than affect national policies and facilitate cooperation. They are central elements constituting the international arena itself and, in many analyses, they are even more important than material resources. The basic Constructivist explanation of the democratic peace is that these states do not fight because they form a pacific community based on shared ideals and identities. These intersubjective ideas precede structure or, in Alexander Wendt’s version, actually define structure.48 It is this pacific community, constituted by shared understandings, that produces the democratic peace. Democracies “perceive each other as peaceful because of the democratic norms governing their domestic decision-making processes,” says Thomas Risse, summarizing the Constructivist argument. “For the same reason, they form pluralistic security communities of shared values. Because they perceive each other as peaceful and express a sense of community, they are likely to overcome obstacles against international cooperation and to form international institutions such as alliances. The norms regulating interactions in such institutions are expected to reflect the shared democratic values and to resemble the domestic decision-making norms.”49 Some versions go beyond shared norms to stress an “intersubjective liberal identity.” That identity transcends national borders and, according to Emanuel Adler, becomes a “marker and indicator of reciprocal peaceful intentions.”50 The strength of this explanation is that norms and prior beliefs do play a crucial role in the social construction of trust and confidence. Trust and confidence are, in turn, essential elements of any deeper peace. There are three problems, however, with making that the entire explanation. The first is its poor

A CONTRACTING THEORY

61

fit with the historical record, especially before World War II. In the nineteenth century, as we have seen, there were serious conflicts and near wars between established democracies. The United States, France, and Britain averted war, but they were clearly not part of a security community at the time. The second problem is that democracies’ sense of a security community is as much a result as a cause. It is the product of their success in resolving earlier conflicts short of war and then drawing appropriate inferences from them. Their prior beliefs count, but so does learning. Once developed, this sense of community can— and does—foster continued peaceful relations. The point is that mutual confidence develops gradually, based on experience, much as any reputation does. Finally, although shared norms and beliefs are important, they do not resolve all serious conflicts among democracies. That sometimes depends on deterrence, based on military resources and credible threats. This cold, deterrent peace is especially likely when two democracies have had little experience dealing with each other and are wary of each other’s intentions, especially if they are backed by formidable capabilities. That was true in the nineteenth century, and it is true among new democracies today. Even in more peaceful settings, where democracies have already developed good reasons for confidence in each other, they must depend on their institutional capacity to reassure partners and project promises well into the future. Their transparency, for example, allows them to discover where they can strike bargains based on shared preferences. Institutional arrangements also matter when they deal with recurrent problems of dividing joint gains and filling gaps in their incomplete agreements. In these distributional struggles, their immediate interests are often opposed, yet they still manage to govern their relations without escalating to war. Shared norms certainly help, but they are not the whole story. Democratic leaders calculate risks and rewards, just as other leaders do, and they cast a wary eye over others’ incentives, intentions, and capabilities. The confidence they develop in others as trustworthy partners (or untrustworthy ones) is based on their many diplomatic encounters, their surveillance of others’ policies at home and abroad, and their ongoing assessment of partners’ capabilities, intentions, and reputations. They continually update their beliefs based on experience. If that experience is sufficiently encouraging, these partners may embark on a succession of cooperative, but sometimes risky, steps that could ultimately forge a real sense of community. They typically move forward in small steps to minimize the risks. This sense of community is undoubtedly fostered by shared ideals and understandings, and it is partly constituted by them. But these understandings are not the entire story, either as cause or effect, and they are certainly not the entire causal story. Even communities with shared identities and shared ideas about the most profound existential issues may not be peaceful. It is worth remembering that medieval battlefields were filled with cavalry, archers, and foot soldiers who shared membership in the church universal and believed in its ideals of a

62

CHAPTER 3

peaceful Christian community. Its knights and higher nobility also held common ideals of chivalry, suffused with religious obligations. In deference to these shared ideals, they avoided slaughtering each other on Sundays. Their treaties fared little better. Christian princes pledged sacred oaths when signing their agreements but often ignored their obligations. “Such should be the good faith of Princes in the fulfilment of their obligations, that their simple promise is more sacred than any oath by another,” Erasmus lamented. “How base then not to abide by a treaty accompanied by the utmost solemnity among Christians! Yet we see this happening every day.”51 There are undoubtedly Muslims today who share this lament as they look on broken promises among fellow Muslims. Another approach emphasizing shared norms is that of Martin Wight, Hedley Bull, and the English School.52 Their central concept is international society. This society of states exists within the larger setting of international anarchy, but it is more than simply a system of interacting states.53 They also share cultural and historical roots and self-consciously adhere to some international rules and norms. “A society of states (or international society) exists,” according to Bull, “when a group of states, conscious of certain common interests and common values, form a society in the sense that they conceive themselves to be bound by a common set of rules in their relations with one another, and share in the working of common institutions.”54 War itself is one of these social institutions.55 Bull sees it as a permanent element of international relations and a recurrent problem for states. States have never seriously pursued Kant’s goal of universal peace, he says, but they do pursue a more limited “goal of peace.” That goal is “the absence of war among member states of international society as the normal condition of their relationship, to be breached only in special circumstances and according to principles that are generally accepted.”56 It is unclear whether these generally accepted principles include both major world wars, the wars of national unification in Germany and Italy, and numerous colonial wars, all fought by the European society of states over the past 150 years. In Bull’s analysis, war has a dual meaning. It is both a threat to be contained and a useful instrument of international society, important for enforcing laws, maintaining the balance of power, and promoting just causes.57 Wight agrees, but his tone is more pessimistic, more Hobbesian. The European state system meets all his requirements for an international society and has since the sixteenth century, yet it has been at war again and again. The main protagonists are also the main sources of Europe’s cultural heritage, not peripheral members of this society of states. Wight concludes that “diplomacy can do a little, perhaps, to mitigate the social conditions of war; it can circumvent the occasions of war; but the causes of war, like the need for diplomacy itself, will remain so long as a multiplicity of governments are not reduced to one government

63

A CONTRACTING THEORY 58

and international politics transformed into domestic politics.” Even nuclear weapons, he says, cannot overcome this Hobbesian dilemma.59 The English School does not offer a clear theory of war (or of international politics, for that matter), so it is difficult to be confident about its implications for the democratic peace.60 Wight’s views suggest that even common norms, history, and rules are not enough to ensure peace among democracies. Bull is slightly more optimistic, since he thinks war has been less useful to states since 1945. Two traditional objects of war, he says, are now less likely to be achieved by force: economic gains from seizing territory and the promotion of ideology or religion. As a result, “states now are reluctant to embark upon war except to achieve objectives of security.”61 That is really a description of national policy objectives, and it is not related to international society. In fact, Bull sees few long-term changes in war’s function within either the states system or international society.62 There is no implicit explanation here of the democratic peace at either the social or systemic levels. This discussion of the English School and Constructivism underscores several problems in using norms or shared ideals as the sole explanation for the democratic peace. First, it is unclear why shared perceptions of “sameness” do not work to prevent wars among fellow Christians, Muslims, or other groups with shared norms and identities. Among co-religionists, at least, peace is a prominent and oft-repeated value. Why do pacific norms fail there but work among democracies? To answer that, we need a much better understanding of how democracies translate internal norms into foreign policies and why they apply their values differently to some states than to others. One common value, the preference for judicial settlements, does little to explain the democratic peace. The United States and Britain used it twice to settle conflicts: shipping damages after the Civil War and the boundary conflict between Venezuela and British Guiana. But there is no systematic evidence that democracies use international courts frequently to settle major political disputes. They avoid military escalation and settle most disputes with each other through normal diplomacy, just like other states, only more successfully. Generally, when they do use courts or arbitration, the partners already have well-established, peaceful relations. By turning to neutral forums, they can insulate specific disputes from ordinary bilateral relations. Democracies’ common values and interests contribute to their cooperative relationships, but they are not sufficient to sustain it, at least not by themselves. That is one simple, but important lesson of the Prisoners’ Dilemma. Identical players, with common interests and values, still do not cooperate if the game is finite or if the future is sharply discounted. Democracies’ most important shared belief is their confidence in each other as reliable partners. That belief has to be earned, since misplaced trust can be fatal. It is not present at the creation, when new democracies begin dealing with each other. It is not a simple matter of recognizing that another state is a

64

CHAPTER 3

fellow democracy, just as it was not enough in Erasmus’s day to recognize that another prince was a fellow Catholic. Confidence in other sovereign powers can emerge only over time. It stems from successful experience: forging cooperative bargains, settling disputes effectively, avoiding opportunism, and sticking to promises even when they are difficult and inconvenient. By doing that, some democracies have managed to move beyond deterrence, signal their fitness as reliable partners, and forge genuine security communities. Their beliefs that other democracies are trustworthy helps sustain these arrangements. But it is misleading to look at the modern North Atlantic security community, observe that its members share values, and say that these beliefs caused the community. Their confidence in each other is as much a result as a cause, emerging as relationships mature and mutual reliance deepens. Institutionalists, by contrast, downplay these social and normative considerations and emphasize instead the role of information, interests, and strategic choice. Their long-standing focus on international cooperation should be directly relevant to the democratic peace, which is, after all, a prominent form of cooperation. Their explanations underscore the role of international institutions in facilitating cooperation on numerous issues. If that explains the democratic peace, then institutions should play a central role in resolving conflicts among democracies without war or armed threats. In fact, international institutions have played only a secondary role in promoting peace among democracies. Democracies have resolved their conflicts mainly through bilateral diplomacy, sometimes assisted by institutions to facilitate information exchanges. The one clear exception is postwar Europe, where security and economic institutions did play a large role. NATO was vitally important, both as an organization and as a symbol of America’s commitment to regional stability. As NATO members, Europe’s postwar democracies developed high levels of cooperation, including military integration, under the protective shield of U.S. power. Its success was complemented and reinforced by the European Union and its predecessors. Beginning in the early 1950s, they promoted economic integration and later encouraged cooperation on a broad range of social, political, and legal issues. The success of the European Coal and Steel Community led them to sign the 1957 Treaty of Rome. The success of the European Economic Community eventually led them to the Treaties of Maastricht, Amsterdam, and beyond.63 This economic integration plus the military security provided by the United States and NATO resolved the century-old conflict between France and Germany. That is an enormous achievement, one that paved the way for a peaceful relationship between Europe and North America. Beyond that, international institutions contributed to the democratic peace in two other ways: third-party dispute resolution and signaling intentions through membership. As we have noted, democracies have used arbitration to diffuse interstate conflicts in at least two instances and possibly more.64 Although help-

A CONTRACTING THEORY

65

ful in these cases, it has not become a standard practice for resolving major disputes among democracies. States have also used membership in international institutions to signal their benign intentions and their commitment to democratic government. That is one reason why newly democratizing states in Europe have been so eager to join NATO and the EU. The tangible benefits are obvious; less obvious is the way membership itself conveys useful information. By meeting tough criteria for inclusion, states broadcast their achievements and intentions, including their commitment to democracy. The most important such signal in recent years has come from Germany, newly unified after the Cold War. By indicating clearly and unequivocally that it would remain inside NATO and the EU, Germany effectively tied its own hands, limiting its independent economic and military role. That is a costly and reassuring signal to anxious neighbors.65 States can also inform others by refusing membership and its obligations. Iraq, North Korea, Libya, and other rogue states have refused to join arms control agreements that require thorough inspections, or they have joined and then refused to permit inspections. That, too, sends a signal about their behavior and intentions, a chilling one. The signals states send are an important element of the democratic peace, but the screening function of international institutions (such as meeting NATO’s membership requirements) plays only a supporting role. Although international institutions have the capacity to inform and adjudicate, their role in maintaining the democratic peace has been secondary.66 Outside of postwar Europe, peaceful relations among democracies preexist their membership in common institutions. When disputes arise, they generally settle them bilaterally, although multilateral institutions are sometimes helpful. It is impossible, then, to explain the breadth and duration of the democratic peace by attributing it to national participation in effective international institutions.67 What the institutionalist perspective contributes is its analysis of the ways that uncertainty, transparency, and signaling can affect strategic choices.68 Good information is essential to maintaining durable, peaceful relations among democracies. They have acquired it, however, without too much help from multilateral institutions. The outsized role of the United States suggests that hegemony might play a vital role in the democratic peace. It is surely important in the post–World War II era, when most democracies have been American allies. For some, the U.S. security guarantee provided a conducive setting to develop peaceful relationships. That was certainly true in Western Europe in the 1950s and 1960s, and it may be happening now in Central Europe. Even so, hegemonic stability theory cannot serve as a comprehensive explanation of the democratic peace. First, it exaggerates the U.S. role in promoting peace around the world. Second, the peace itself predates American hegemony.

66

CHAPTER 3

Third, before World War II, there was no alternative hegemon, although the United Kingdom did dominate the world economy until 1914. Since World War II, the United States has used its hegemonic power to promote democratization around the world and help existing democracies (at least when that did not conflict with America’s staunch anticommunism).69 But support for democracies does not imply the United States manages their bilateral relations or even tightly controls them, though it sometimes tries. One way of understanding these limits is to look at the global range of U.S. interests and diplomatic involvement. America has supported and maintained close ties with many nondemocratic states—states that are not part of any wider peace. If hegemony explained peace, then they, too, should be included. In fact, they have often been at war, and, on rare occasions, even at war with each other (for example, Jordan versus Israel in 1967 or Turkey’s 1974 invasion of Cyprus, overthrowing a government closely allied with Greece and seeking to unite with it). There are also regions, particularly the Middle East, where the United States has vital interests and can project military power but has been unable to prevent interstate warfare. In still other regions, the United States has been unwilling to act decisively, either because it lacks any vital interest (for example, the low-level fighting between Peru and Ecuador over the Amazon Basin) or because it is unwilling to pay the price of deeper engagement (India-Pakistan during the Cold War era). In short, American hegemony is not analogous to the kind of tight control the Soviet Union maintained within the Warsaw Pact, which did prevent interstate warfare.70 Moscow’s dominance of East European political leaders, its control of local communist parties, and its continuing military presence prevented warfare among Soviet allies. The United States never maintained that kind of hegemonic control among its allies and was unable or unwilling to suppress conflicts among neutral states around the world. Second, the democratic peace long precedes the era of U.S. dominance. That cannot be explained away by saying that the United Kingdom wielded hegemonic power during the earlier period. From the end of the Napoleonic Wars until World War II, no single state dominated international affairs. Yet, according to the best empirical evidence, peace prevailed among democracies. What about the large literature claiming Britain was a hegemonic power during the mid- and late-nineteenth century? That work is concerned with trade, money, and foreign investment, all issues where Britain was the world’s leading power.71 But British dominance was limited to the international economy and its own empire, formal and informal. It was not a military hegemon except for that empire and some littoral regions of Africa, Asia, and Latin America, where it could easily project naval power. Within its informal empire, it made only limited efforts to prevent local conflicts, though it was sometimes drawn into the fray, mainly because it was determined to prevent chaos and attacks on foreign economic interests.72 In continental Europe, the cockpit of nineteenth-

A CONTRACTING THEORY

67

century world politics, Britain was by no means a hegemon or even the most powerful state militarily. Thus, the theory of hegemonic stability cannot explain the long history of democratic peace and, except for the post–World War II era, does little to illuminate it. Marxist theories have little to say about the democratic peace and cannot explain it. Marx himself wrote sparingly about international affairs, although he did argue optimistically that “a new society is springing up, whose International rule will be Peace, because its national ruler will be everywhere the same—Labour.”73 Elsewhere, he speaks ironically of national war as “the highest heroic effort of which the old society is still capable.”74 Marx’s analysis of the Paris Commune makes clear that he considers war between states, like the state itself, to be the product of class struggle. Marx gives only a sketchy explanation of these international issues, but Lenin develops them in his famous booklet Imperialism, The Highest Stage of Capitalism.75 For Lenin and subsequent Marxists, the crucial point is that war is endemic to monopoly capitalism.76 States, controlled by finance capital, compete aggressively for foreign markets. Since the world is already divided among imperial powers, the only way each one can acquire more markets, raw materials, and investment outlets is to displace the others by force. That powerful motivation, says Lenin, makes war among advanced capitalist states inevitable. For both Marx and Lenin, the only way to end this cycle of war is to end class domination through proletarian victory. Their deepest claim is that only a communist world, free from class struggles, will be at peace. History has not been kind to these claims. In 1914, powerful socialist parties all across Europe backed their own countries’ war efforts instead of banding together against capitalist warmongering. Their members patriotically marched off to the killing fields. After World War II, when the Soviet Union was finally joined by another major communist state, the People’s Republic of China, the two powers quickly moved from a thin veneer of fraternal solidarity to outright hostility and balancing against each other.77 Their long border was the site of considerable military tension, large force deployments, and persistent killing. At the same time, they were jockeying for regional and ideological influence.78 That was particularly evident in Southeast Asia after America’s defeat, when they backed opposing sides in fierce local wars.79 The overall pattern is not good news for any Marxist-Leninist theory of peace. What is more important to our inquiry, however, is what Marxists say about capitalist democracies in a world where some states are communist. Would they intensify their fight over scarce resources or make common cause? Either is conceivable. No Marxist theory of alliances is available to provide their answer. If capitalist states were expected to make common cause, that might account for peace in the North Atlantic during the Cold War, when the Soviet threat was greatest. Beyond that, however, this class-based analysis has no explanation for why democracies—all of them capitalist—have remained at

68

CHAPTER 3

peace for so long, or why they have remained at peace in regions where revolutionary threats against them have been low. The classical liberal argument is almost the exact opposite of the Marxists’. Liberals see commerce bringing countries together, not driving them apart. Seeking to repeal the protectionist Corn Laws in the 1840s, Richard Cobden, John Bright, and other Victorian free traders expected to bring economic and political benefits to their country and its trading partners.80 As they saw it, the quest for trade would lessen the dangers of war for a very practical reason. Serious conflicts were bound to disrupt commerce. The greater the trading profits, the greater the incentives to avoid hostilities. Cobden summarized this worldview: “I believe the progress of freedom depends more upon the maintenance of peace, the spread of commerce, and the diffusion of education, than upon the labours of cabinets and foreign offices.”81 As the preeminent voice of Manchester liberalism, he fervently believed that progress, peace, and commerce went together, reinforcing each other. The simplest version of this argument, that extensive commerce prevents war, is not true. The years just before World War I marked a peak in economic interdependence, one unequaled until recent years. Imperial Germany and the United Kingdom maintained extraordinarily high levels of bilateral trade even as their political relationship descended into fear, hostility, arms races, and ultimately war. The major and minor European powers that fought in 1914 had been bound together by unprecedented levels of trade and finance. Nor is this the only case. Britain was America’s largest trading partner when they went to war in 1812. Japan depended heavily on American natural resources, especially oil, when it attacked Pearl Harbor in 1941. That attack was designed to protect its flank as it launched a military drive to acquire its own resources in Southeast Asia and the Dutch East Indies (modern Indonesia), while retaining its conquests in China.82 Although simple versions of the classical liberal argument are wrong, more sophisticated versions may well be right. In the post–World War II era, strong trade ties do seem contribute to peace. Since these ties are strongest among democracies, they may a source of peaceful relations. That point is confirmed in studies by Bruce Russett and John Oneal, Charles Anderton and John Carter, Solomon Polachek and his coauthors, Erik Gartzke and his coauthors, and others.83 According to Susan McMillan’s survey, most studies show that commercial ties lessen the chances of violent conflict. Still, as McMillan notes, there are a number of unresolved questions.84 There are on-going debates about whether trade has the same impact on major and minor powers, large and small states. There is another debate about whether trade is disrupted more than temporarily by conflict or war.85 The biggest questions, still unresolved, are whether the relationship between trade interdependence and conflict is spurious and, if it is not, whether the relationship is a general one that holds for periods before and after the Cold War.

A CONTRACTING THEORY

69

In a formal analysis of these issues, James Morrow raises fundamental questions and offers intriguing answers. Because losing trade is assumed to be costly, it should reduce both states’ willingness to fight. That is the classical liberal point. But, as Morrow shrewdly observes, reducing both sides’ willingness to fight produces indeterminate results in a crisis, where trading partners test each other’s resolve. Even if one side is more dependent and therefore vulnerable to economic pressure, the stronger partner is only tempted to push for more concessions and initiate more crises, hoping to gain still more in their distributional struggles. As a result, higher trade flows should not inhibit the onset of crises or their escalation. Morrow concludes that trade interdependence has only indirect effects on conflict behavior. The indirect effects are that economic links give states a wider range of costly signals to use. That allows them to show resolve more efficiently, increasing the chances they will reach peaceful settlements.86 This use of costly signaling to reach stable bargains is readily understandable from a contracting perspective, but it is a far different mechanism than that envisioned by Cobden, Bright, and other classical liberals. An alternative version of liberal international theory emphasizes domestic interests and their impact on state policy. Andrew Moravcsik, who has done the most to develop this theory, argues that there are several distinct but overlapping strands.87 One stresses social identities; another, economic interests; yet another, domestic institutions that aggregate political demands. These yield somewhat different perspectives on the democratic peace. The social-identities perspective, for instance, refers to “a group of developed countries that share democratic norms of legitimate dispute resolution—a plausible explanation for the ’democratic peace’ phenomenon.”88 That is virtually identical to a constructivist account. The economic perspective, by contrast, underscores domestic struggles between those benefiting from transnational ties and those suffering adjustment costs. What unites the various strands is their common emphasis on the domestic origins of foreign-policy preferences.89 Preferences are grounded in national political life; the interaction of these national preferences then determines international outcomes, including war and peace. “The liberal claim,” says Moravcsik, is “that the pattern of interdependence among state preferences is a primary determinant not just of individual foreign policies, but of systemic outcomes.”90 What this liberal theory says, in effect, is, “Democracies are at peace because their preferences are more harmonious.” This is an important claim, and there is some support for it. Preferences probably are more harmonious among well-established, economically advanced democracies.91 Moreover, we must refer to these preferences—to states’ expected gains and losses and the weight they put on them—in order to explain any long-term cooperation, including peace. That requires understanding what national preferences are, how they intersect, and which strategies are optimal.92 Some configurations of prefer-

70

CHAPTER 3

ences make cooperation hard, even impossible, although there are still joint gains to be had from avoiding the costs of war itself. As we have already explained, a simple, widely accepted notion of national preferences—that states seek security—is enough to yield peace under the right conditions. More extensively shared interests and norms may help, but they are not essential, and they may not be enough. They may not be enough if participants do not know much about each other and have not developed mutual confidence. They may not be enough if states are unable commit themselves to particular courses of action or are unsure if their bargains will hold. Even among states that share important goals, preferences differ, and there are bound to be distributional struggles. The question is how these struggles play out at the international level. One answer, stressed in the liberal account, is that democracies settle them peacefully because they adhere to norms of judicial settlement, based on their domestic experience. This exaggerates the role of international law and judicial settlement, as we have noted. Courts are rarely used to solve major differences among states, including democracies. Most settlements are negotiated, sometimes under the shadow of coercion as well as the shadow of law. Among pairs of democracies, the learning pattern is evident. Threats and military displays are sometimes used early in the relationship and become less important over time. The “close calls” that Layne and others highlight occurred more than a century ago, among democracies that went on to develop peaceful relations. What about close calls more recently? They involve new and still unstable regimes, with only a short history of working with fellow democracies. Democracies do occasionally use arbitration, and, in the highly institutionalized context of the European Union, they use courts extensively. Indeed, the European Court of Justice and deference to it by members’ national courts have spurred the integration process.93 But outside the EU, public international law and judicial settlement are much less important. At the interstate level, differences are still settled mainly by diplomatic bargains. In recent years, democracies have built on their stable bargains and strong relationships to create a new set of quasi-judicial institutions: dispute settlement mechanisms for international trade. They are now integral aspects of the World Trade Organization and NAFTA.94 Designed to govern transactions among private, transnational actors, they give firms a more predictable framework for rising levels of commerce. The democratic peace has opened the door to this proliferating legalization. If learning is as important as we have said, perhaps learning theories can help explain the democratic peace. John Mueller’s well-known work on the obsolescence of war certainly suggests such an explanation.95 To explore it, we need to distinguish between two kinds of learning. One, emphasized by Mueller, is a change in broad social understandings and conventions. The other is more narrowly focused on individual relationships and reputations. Both matter.

A CONTRACTING THEORY

71

Mueller’s chief claim is that states are less likely to fight major wars today because of the macrolessons they learned from deadly conflicts in the twentieth century. In his view, it is these lessons, not nuclear deterrence, that kept the peace between the United States and the Soviet Union. Just as dueling became outmoded and archaic, so the social convention of war is becoming obsolete. Social views about war have indeed changed, especially in Europe, but we must be cautious about inferring what these changes are and what their impact is. Democracies have learned several profound and costly lessons about war over the past century. What Mueller emphasizes might be called “the lesson of the Somme” from World War I.96 The line of trenches through northern France and Flanders and the ensuing war of attrition taught the survivors that war can be far more deadly, costly, and uncertain than its planners anticipate. That democracies learned this lesson so well is one reason why they were so vulnerable to Hitler’s expansion in the 1930s. They dreaded the high costs of fighting another horrific land war in Europe. They deluded themselves that Hitler would also recoil from the costs and pursue only limited goals. Britain’s craven—and futile—effort to appease Hitler taught powerful lessons of its own. The “failure of Munich” resonated through the Cold War. The Munich conference was held in 1938, with the German army poised to invade Czechoslovakia. To avoid war, British and French leaders accepted Mussolini’s proposal at the conference, giving Bohemia to the Nazis.97 In return, Hitler promised to stop his expansion after absorbing the Germanspeaking areas of Czechoslovakia. Within months, Hitler discarded his written promise, drove the Wehrmacht into Prague, and seized the entire country. What Western democracies learned from Munich and from the catastrophic failure of appeasement is that dictators are often aggressively expansive and always opportunistic, that their promises to the contrary are empty, and that they must be stopped early—by force if necessary—before they wreck more havoc and gain more resources to continue. No lesson carried more importance to America and its NATO allies during the early Cold War, when they had to cope with another tyrant, Josef Stalin, and feared another expansive, opportunistic dictatorship. The West’s grand strategy of containment was grounded in the lessons of Munich. Surprise attacks in 1941 taught still more costly lessons. The Japanese bombing of Pearl Harbor and the Nazi invasion of the Soviet Union (Operation Barbarossa) reinforced the old lesson that vigilance and military preparedness are essential. That is especially true in dealing with closed dictatorial states, which can hide their strategic choices and launch devastating surprise attacks, as both Imperial Japan and Nazi Germany did.98 That point was driven home again in June 1950, when North Korea launched a surprise invasion of the south. In short, there has been social learning about warfare, but it is not limited to the horrors of war. True, democracies are increasingly sensitive to the human

72

CHAPTER 3

costs of war, the very point Kant made with such foresight in Perpetual Peace. NATO forces, led by the United States, are reluctant to suffer casualties or inflict them on innocent civilians. That has reinforced the long-standing trend to saturate military force with high technology, substituting capital for labor, brains for sweat. This does not imply, however, that war is obsolete. For democracies, the larger lesson is that their own governments are fundamentally different from dictatorships, which pose continuing threats of expansion, oppression, and war. To pretend otherwise, as Britain and America did in the 1930s, does not prevent war. It only postpones the day of reckoning and makes it worse. This large lesson about the distinctiveness of democracies is complemented by their learning in relationships with each other. There is mounting evidence (presented in chapter 6) that democracies are gradually able to convince each other they are trustworthy partners. This bilateral learning is closely related to how states perceive each other, issues explored by Robert Jervis, Deborah Welch Larson, and others.99 Larson, for instance, shows how the lack of trust between the United States and Soviet Union repeatedly prevented them from making bargains that could have benefited both sides.100 According to Larson, their mistrust was rooted in misperception. Greater transparency and a willingness to take small, risky steps would have produced better outcomes. More generally, the psychological explanation of war attributes a major causal role to misperception and the mistrust that grows from it. That is easily transformed into an explanation of the democratic peace. Because democracies have more accurate perceptions of each other, they are better able to cooperate, build trust, and avoid war.101 That seems right on several counts. Democracies are more transparent, maintain closer contacts, have many more sources of information, and probably have more accurate perceptions of each other, at least according to our indirect evidence. Moreover, misperceptions can add to the dangers of war, primarily by lessening or even eliminating the perceived range of mutually acceptable solutions short of war. Jervis, who has written most extensively on the role of misperception in security affairs, never asserts it is a comprehensive explanation of war, only a contributory factor. He is particularly careful, for instance, to show that escalation processes may—or may not—be the result of excessive fears or misattributed motives. They are sometimes grounded in realistic assessments of others’ expansive goals and the firm steps needed to stop them. He also stresses that states can be too credulous as well as too fearful and that, even in retrospect, it may be impossible to divine an adversary’s motives. Jervis notes that states may go to war without any misconceptions, even if they expect to lose. They might be pressured by powerful domestic groups, or they might value fighting for its own sake.102 What is more likely, however, is

A CONTRACTING THEORY

73

that they take a rational long shot. They have a low but accurate expectation of winning, which is offset by the high value of the prize at stake. That alone is not enough to produce war. They also need to know if the expected value for fighting is higher than any alternative strategy (including a negotiated solution), adjusted for their risk tolerance. Would they prefer a less attractive but more certain status quo ante bellum? Jervis also underscores the value of transparency and the difficulties of achieving it. “Statesmen,” he says, “might be disabused of their misperceptions if they could listen in on their adversary’s deliberations.”103 They certainly make every effort. The difference is that democracies make it easy. They are systematically more open. Their most important deliberations are often public, from congressional hearings to parliamentary debates, and many that are closed (or even secret) are soon reported in independent news media. That not only clarifies motives; it lessens the possibility of surprise attack and the understandable fears that stem from it.104 Of course, what states want to know about each other goes well beyond their motivations and resolve. They also want to understand others’ capabilities, their adherence to existing commitments, the institutional limits on their future policy choices, and the likelihood they will continue particular policies or change courses radically. Some risk always surrounds these issues, but democratic transparency reduces it, systematically and significantly. In relations among democracies, this mutual transparency—and these states’ willingness to negotiate still more openness because it does not threaten their domestic political control—substantially reduces misperceptions on all sides. That lowers the risks of betrayal, builds confidence, and opens up a wider range of bargaining solutions. To make these solutions stick, however, requires institutional capacity and continuity. In the following table (3.1), I have summarized these varied theoretical explanations of the democratic peace, stating the main lines of each argument and its strengths and weaknesses.

Conclusion Few of the theories discussed here advance their own comprehensive explanations for the democratic peace. Several, however, provide valuable insights into the logic of relations among democracies. Shared norms and broad social learning, for example, help explain why democracies avoid conflicts over certain issues and why they are willing to consider bargains requiring some confidence in each other. On the other hand, as Wight and Bull make clear, shared values and common membership in a “society of states” do not in themselves block war. They have not prevented member states from repeatedly launching major wars against each other to advance their own interests.

There is no democratic peace; close calls indicate role of prudence, luck Shared understandings and norms among democracies

Realism

Internationalist interest groups seeking peaceful cooperation and economic integration dominate foreign policy

Liberalism

No evidence that institutions play a significant causal role in democratic peace

Fails to account for key evidence on wars and militarized disputes Shared understandings have followed, not led, the movement from deterrence to deeper peace

Weaknesses

Free trade and economic integration do seem to No evidence that democratic peace is driven by be associated with peace, at least since World interest groups or that they play a disproporWar II, although their causal role is uncertain tionate role in crises

Institutions play a supplementary role in sustaining relations among democracies; theory shows importance of information in international cooperation

Emphasis on prudential calculations based on national interests; attention to risks, dangers Prior beliefs and understandings increase trust, facilitate compromise solutions short of war

Strengths (as explanations of democratic peace)

High costs of war in twentieth century have taught states to avoid it

Does not exist because of rivalry among capitalist states

Democracies perceive each other more accurately U.S. dominance

Learning

Marxism

Psychological explanations

Hegemony

Many states seem to have learned nothing; democracies have learned to stop aggression by war; willing to use force against nondemocracies

Democratic peace strongest during post–World War II era, period of U.S. hegemony

Evidence that democracies do perceive each other more accurately

Democratic peace transcends periods of U.S. and British hegemony

Even completely transparent states with accurate perceptions could fight

Advanced capitalist states were most cooperative Capitalist democracies (which is to say, all in security affairs when challenged by comdemocracies) have remained at peace with munist power each other; working class has played no role in peace

Democracies do appear to be very sensitive to costs of war for their own citizens

International Society Civilized states form an international society Core states of democratic peace have been in the States integral to Western civilization have (English School) with norms limiting certain types of war North Atlantic fought each other often; its leading theorists do not think social or systemic reasons for war have declined

Joint participation in international institutions

Institutionalism

Constructivism

Basic Explanation

Theory

TABLE 3.1 Alternative Theories of the Democratic Peace: Summary Table

A CONTRACTING THEORY

75

Well aware of the risks, states are cautious about making bargains that leave them vulnerable. That is why they move so slowly and carefully to dismantle the costly armor of self-protection, even when dealing with familiar states. Realism offers the most profound insights into this prudential reasoning. It provides few answers, though, about how states peacefully resolve these security problems beyond the traditional emphasis on military deterrence and balancing alliances. Institutionalism underscores the vital role of information in making cooperative strategic choices, although multilateral institutions themselves have played only a limited role in the democratic peace. Psychological theories of international relations also concentrate on these information issues and suggest ways that transparency and learning make some bargains feasible. The configuration of national interests is obviously vital and is emphasized in nearly all theories. Liberalism is one of several that underscore the point. Some versions of liberalism also highlight the contributions of economic interdependence to lessening conflict. That is a promising argument, but its generality and theoretical grounding are still unproven. So far, there is no evidence to support one of liberal theory’s strongest claims: that domestic sources account for the configuration of national interests that produce the democratic peace. On the other hand, the basic institutions of constitutional democracy are vitally important for making commitments and allowing others to judge their integrity and durability. The contracting theory of the democratic peace incorporates a number of these theoretical insights in a systematic way, making them part of a coherent and testable explanation. It makes modest, conventional assumptions about state preferences: that democracies seek security just as other states do. All states are understandably worried about potential threats, especially from nearby states with substantial military capabilities. The contracting theory argues that democracies are better equipped to resolve disputes among themselves and make mutually profitable bargains, including those that avoid the manifold costs of war. One reason is that democracies have unique capacities to make binding, long-term promises—commitments that are hard to reverse and that transcend current leaders. Their transparency helps, too, since it allows partners to judge the level of commitment on a continuous basis. Equally important, it virtually eliminates the risks of surprise betrayal. Democracies simply have less private information than dictatorships do. Even so, they must bargain with incomplete information about each other. That means they must rely on their prior beliefs, which they update through experience. Shared norms help form these beliefs, and they may be important in what states learn from their subsequent interactions. They may also play a role in identifying salient solutions, a major contribution when multiple equilibriums are possible.

76

CHAPTER 3

None of this implies the democratic peace is everlasting or that democracies cannot fight each other. In an anarchical world, war is always possible. But the chances that constitutional democracies will fight each other are lower than for other states. They are even lower when democracies are well established and have gained experience dealing with each other. Among such pairs, the democratic peace is deep and durable.

4 Why Democratic Bargains Are Reliable: Constitutions, Open Politics, and the Electorate

IN PROJECTING promises, democracies have two major advantages that stem from their basic government structure. 1. Their constitutional procedures make promises more reliable. They are more likely to be carried out initially, harder to reverse later, and therefore more credible. 2. Their public debates, relatively open decisionmaking, and free press allow partners to make more confident estimates about how faithfully democracies will execute their promises, now and in the future. (This same open discourse probably allows democracies to make systematically better estimates of others’ intentions, as well.)1 These procedures make democracies more trustworthy and allow partners to gauge the depth of support for policies and promises on a continuous basis. They can see the strength and character of opposition and can reasonably judge the chances that opponents will come to power. That is the heart of what we mean by transparency: the visibility of debates, decisions, policy processes, and implementation to outside observers, from local citizens to foreign diplomats.2 Finally, partners can see and evaluate the procedural rules that constrain leaders—to a greater or lesser degree—from reversing their course. In other words, effective constitutional arrangements serve both to constrain and inform.3

Confidence and Reliance: The Role of Constitutional Practice Constraints differ, even within a single political system, because states have devised various ways of making promises. Some are deliberately more equivocal than others. These gradations of intent are reflected in how commitments are framed, especially their formality and publicness. A ratified treaty, for instance, is a more serious commitment than a joint communique´. An exchange of letters between heads of state is more serious than the same exchange between foreign ministers. Political leaders are well aware of these subtle differences, just as they are aware of such differences among threats. This simple hierarchy does not always apply; sometimes it is necessary to make crucial agreements quickly or secretly, as the United States and Soviet

78

CHAPTER 4

Union did to end the Cuban Missile Crisis. But it is usually true, and for a good reason. To highlight their deeper commitments, states need to use costly signals. Public, formal commitments are an efficient way to do that because they are very visible (and therefore costly) ways to pledge a state’s reputation as hostage to its performance.4 As matters of law, judges at the International Court of Justice conflate these differences. As long as the bargains are written and meant to be binding, they constitute treaty commitments as understood in international law.5 But prudent politicians and diplomats take a more nuanced view. For them, these gradations are meaningful. They convey different levels of commitment toward keeping particular promises, given the contingencies and changed circumstances that might arise.6 It is important to evaluate these commitments with a cold eye because it is the parties themselves, not judges at The Hague, who must make the promises binding. Why, then, do I refer specifically to the advantages of well-established democracies? After all, many new democracies possess similar written constitutions. Do they have the same promising advantages as their older cousins? Generally speaking, no. To begin with, constitutional arrangements must be reflected in ongoing political practice. They must effectively constrain rulers. As Charles Howard McIlwain once wrote of medieval constitutionalism, “The fundamental weakness . . . lay in its failure to enforce any penalty, except the threat or the exercise of revolutionary force, against a prince who actually trampled under foot those rights of his subjects which undoubtedly lay beyond the scope of his legitimate authority.”7 Modern democratic constitutions are designed to overcome this fatal flaw. They answer Marc Bloch’s powerful question, Who shall bind the ruler?8 Even in the modern era, formal democratic constitutions are no guarantee of actual democratic practices. Remember that the Soviet Union’s allies in Eastern Europe considered themselves “democratic” and printed up impressive rules and constitutions to underscore the point. East Germany even called itself the German Democratic Republic, without apparent irony. In those mercifully departed days, one sure way to tell an authoritarian regime was to find the words democratic or people’s in its name. The propaganda began when its very name was spoken. The new regimes of Central Europe are quite different from that. They are serious social projects to construct democratic political life and enduring institutions. But they are still far from complete. Their future is hopeful but far from certain. For their potential partners, there is also a trade-off between the turbulence of political change and the emerging practices of open politics and genuinely constraining (and empowering) constitutions. This turbulence is unavoidable. After all, these societies are in the midst of thoroughly transforming their politics and economics. Historic changes like these are bound to increase uncertainty. On the other hand, their new constitutionalism and greater trans-

RELIABLE BARGAINS

79

parency work to reduce it. We cannot say a priori which is the greater effect: the turbulence of social change or the limits imposed by new institutions. We can be confident, however, that uncertainty will inexorably diminish as the regimes are stabilized, if they are stabilized. If the regimes are not stabilized, if political turbulence returns and grows, as it did in Weimar Germany, then their reliability as long-term partners will decline as social disorder rises.9 What matters here are the connections between formal rules and political practice over time. For the argument I am making, the important question is whether the central features of democratic political practice are firmly in place and whether they are respected as constant, limiting features of domestic political life. Politics in constitutional democracies is confined within strict boundaries, in several senses. One is that governmental intrusion into personal life is restricted. Its counterpart in the public sphere is the right to autonomous political expression. Another is that political processes are commonly known and accessible, forming a shared institutional framework for decentralized strategic action among citizens. Even though specific political outcomes may be beyond prediction, they do fall within well-defined ranges. Because institutions are stable and political processes are open, it is possible to attach reasonable probabilities to likely outcomes. “The fact that uncertainty is inherent in democracy does not mean everything is possible or nothing is predictable,” says Adam Przeworski. “Actors know what is possible, since the possible outcomes are entailed by the institutional framework; they know what is likely to happen, because the probability of particular outcomes is determined jointly by the institutional framework and the resources that the different political forces bring to the competition. What they do not know is which particular outcome will occur. . . . Hence, democracy is a system of ruled open-endedness, or organized uncertainty.”10 This “organized uncertainty” can sometimes be squeezed down to a very small range of possibilities. Democracies have powerful—uniquely powerful—mechanisms for reducing doubts about major decisions and bargains that have been agreed to, even if they are not yet fully implemented. This mechanism utilizes the dispersion of powers that is fundamental to constitutional democracy. Because of this dispersion, decisionmaking requires extensive consultation and approval by independent decisionmakers within the government. That makes it hard for democracies to act, but it also makes it hard for them to reverse their actions or abandon their promises. Once democracies have made major decisions, including commitments to other countries, it is hard to overturn them. It is almost impossible to do so secretly, to take partners by surprise.11 This inertia is built deeply into the structure of constitutional democracy. To avoid autocratic rule, these states require periodic elections, and they limit the power of elected leaders. How they do so depends on the type of democracy.12

80

CHAPTER 4

Presidential systems, like those in the United States and France, require power sharing between the executive and a representative assembly. Their constitutions divide government authority, grant some autonomy to the legislature and executive, and make collusion between them difficult. In the U.S. system, the legislature itself is divided into two houses; central government authority is also shared with federal states.13 While this fragmentation impedes government action, it also blocks the accumulation of coercive, monopoly powers. Parliamentary systems are more unified, at least in principle, since the prime minister heads both the legislature and the executive branch. But their leaders are also constrained in several ways: by the constitution (written or customary), by the need to maintain an electoral majority, and by the independent power of senior parliamentarians. These constraints are obvious when no party holds an outright majority in parliament. That often happens in proportional voting systems, which make it easier for small parties to win some seats and deny the largest an outright majority. To form a government then requires cobbling together a coalition with minority parties. Their price is some control over key policies and some seats in the cabinet. But even when a single party controls parliament, prime ministers do not control government by themselves. Margaret Thatcher was a strong leader with a clear parliamentary majority, but she still had to share power with senior ministers, who were elected independently and led important factions of the ruling party. In Japan, factional leaders are even more powerful. They can and do veto government initiatives, even when one party controls parliament, as Japan’s Liberal Democrats have for years. These limitations on centralized control are reinforced by competitive political parties. The limitations are strongest when parties regularly rotate in power—in the executive, legislature, courts, and regional governments. Rotation is important because it gives each party strong incentives to respect prior decisions and adhere to well-established political processes. The parties are, in effect, playing one equilibrium strategy in an iterated game. They are cooperating to ensure the independence of other branches and agencies in anticipation that others will do the same, and all will benefit. If they actually rotate, then they have credible threats to retaliate if others do not reciprocate. (This also suggests why it is so much harder to maintain an independent judiciary in one-party states, where there are few incentives for such forbearance.)14 Some important decisions may require acceptance or constitutional approval by independent branches of government. The results are often bemoaned. These institutional arrangements do impede swift policymaking and encourage stalemate. They sometimes confuse partners with more centralized governments. But they also mean that decisions, once firmly taken, are not lightly reversed— again, because independent decisionmakers must give their consent, usually in a public process. The requirement for legislative supermajorities adds further

RELIABLE BARGAINS

81

stability to promises, at least for promises that democracies wish to make very certain, such as treaty commitments or payment of national debts. Democracies have one more important advantage in projecting promises: elections. The public’s essential role means that democratic leaders must appeal to large constituencies and retain their confidence. They are punished for ineffective policies or abandoning promises to voters. Not only does the public hold its leaders accountable directly, it does so indirectly as well, through its other elected representatives. These other representatives amplify their constituents’ complaints when promises are broken and policies fail. These “audience costs” serve as an informal bond for serious public commitments. They are an important source of credibility for democracies’ threats and promises. James Fearon initially proposed this major role for audience costs.15 Since then, he and several other scholars have probed this role for democratic audiences and tested the empirical effects.16 Testing is difficult because if leaders are truly concerned about audience costs, they will avoid confrontations they are likely to lose, fearing the electorate (or its other representatives) would punish them for duplicity or bumbling. The crises we observe are therefore a biased sample. Despite these methodological difficulties, several tests show that democratic threats are more credible and less likely to be bluffs. These tests confirm, at least tentatively, that audience costs do play an important role in democratic foreign policy, conveying important information to partners and adversaries. These results apply directly to our argument. Audience costs are just as valuable for promises as they are for threats. Audience costs are still a new idea, and several criticisms have been leveled. Critics say there is no proof that audience costs exist or can be measured. Second, they may do little to constrain democratic leaders, who have powerful tools to shape public opinion. Finally, even if audience costs do exist and do constrain, there is no evidence that they are higher for democracies than for other governments, or more constraining. After all, even dictators need the support of some key audiences, usually army generals, business leaders, and party bosses. Recent empirical tests, though not definitive, answer these criticisms. They indicate that audience costs are more stringent in democracies and do constrain elected leaders. Looking at more than four-hundred crises between 1918 and 1994, Christopher F. Gelpi and Michael Griesdorf test several hypotheses about democratic political structures. Their findings are clear-cut. Democratic governance is vitally important because it generates high audience costs and encourages leaders to select conflicts they can win. They observe, as we have, that “the openness of democratic societies makes their bargaining tactics credible to opponents. This credibility helps them prevail when they are willing to use force but prevents them from successfully bluffing resolve.”17 Jean-Se´ bastien Rioux reaches the same conclusion in his large empirical study.18 Rioux, Gelpi, and Griesdorf all confirm Fearon’s insight that audience costs can reveal

82

CHAPTER 4

a democracy’s true preferences, making their signals more credible than those of nondemocratic states.19 In another study (this one using data on militarized disputes rather than crises), Peter J. Partell and Glenn Palmer confirm that audience costs strongly influence crisis outcomes. States that can generate higher audience costs tend to prevail.20 That favors democracies, of course, but it also favors a few nondemocracies where leaders are constrained by strong ruling parties.21 Kenneth A. Schultz is more skeptical of measuring audience costs in nondemocracies and therefore is unwilling to say they are systematically lower. Yet he still concludes that democracies have marked advantages in conveying information and binding themselves. The key is “open political competition [that] creates conditions that are highly favorable for revealing information to both domestic and foreign audiences.”22 His empirical work confirms the nowfamiliar conclusion: democracies are less able to bluff, more selective in issuing threats, more likely to carry out the threats they do make, and more likely to prevail in crises. This democratic combination of credibility, constancy, and boundedness matter because promises are all about the future.23 Rules and practices must constrain through time, and others must be confident that they will. This means not only that the rules must be effective, but that they must be lasting. That is, the domestic regime must be stable so rules and processes remain firmly in place. If they do, then promises and threats will outlast today’s rulers. They will be credible over the long haul.

Reputation and Regime Continuity The regime’s durability matters for another reason as well. Because information is always incomplete, states understandably look to their partners’ reputations as rough measures of reliability. Reputations for promise keeping can be built up only over time, as experience accumulates. “Confidence is a plant of slow growth in an aged bosom,” observed William Pitt, the Elder.24 This sense of confidence—this judgment that a partner is trustworthy—can be destroyed swiftly, sometimes with only a single betrayal or opportunistic act. It may take years to recover.25 Reputations are informed guesses about how a particular actor will behave in particular circumstances, perhaps in a wide range of circumstances (such as a personal reputation for honesty). As with all probabalistic exercises, we want to know (1) What is our best estimate? and (2) How confident are we that we have it right? Of course, if we really had complete information about an actor’s preferences and incentives, and if we really knew what circumstances an actor would face, then we would not have to rely on reputation. We would know. Since we

RELIABLE BARGAINS

83

seldom have such complete information, we do the best we can and rely on the inferences we reasonably develop. Although we often focus on whether a reputation is “good” or “bad,” we also need to figure out whether we are “highly confident” or “unsure.” The problem for new democracies is not that they have bad reputations, but rather that they do not have much reputation at all. They are a mystery, one that will be revealed only over time. A new democracy faces two problems in building a reputation for trustworthiness. Neither can be solved quickly or easily. First, other states have not seen it make choices, execute policies, and live up to commitments in a variety of circumstances. As a result, whatever their best estimate is, it is bound to be surrounded by a wide band of uncertainty. It is very useful, for instance, to see how a state treats its old commitments when they are no longer so advantageous, when it is tempting to invent excuses and abandon them.26 Second, new democracies are less stable, their futures more doubtful, than their old and often-tested counterparts in Western Europe and North America. To take a very simple example: if a revolution is likely, then the reputation of today’s regime says almost nothing about tomorrow’s. Think of Spain during the tumultuous years after General Franco died in 1975 and his handpicked successor, Carrero Blanco, was assassinated. The democracy that emerged by fits and starts was fragile, its survival tenuous. Spain’s partners faced both kinds of uncertainty. They knew a great deal about the old Franco regime, but hardly anything about the new one. They had little basis for making solid inferences. They were also bound to wonder if democracy would survive in a country that had been ruled for so long by the military and where antidemocratic forces were still powerful. Yet they could also find many encouraging signs. Although Spain had its share of authoritarians and regional separatists, it also had a large, strong group of reformists who had already led Spain toward electoral democracy.27 Its neighbors were rich, stable democracies, eager to assist the transition. The only exception was Portugal, which was itself becoming a democracy. Spain’s external relationships were affected by these same divisions and uncertainties. Franco, with his fascist background, had led a staunchly anticommunist regime. It was a firm ally of the West, even if its unsavory government was unacceptable within NATO itself. After Franco’s death, doubts about Spain’s value to Western security rose as communists and socialists became major electoral forces.28 With the passage of time, however, these doubts have been resolved in Spain’s favor. Its politics are stable and democratic, and its neighbors know it. It has held several elections, and its ruling parties have rotated peacefully, including a long stretch of socialist rule. Its constitutional monarchy is firmly established, and its head, King Juan Carlos, played a pivotal role in sustaining democracy against internal threats. Trade, tourism, and investment ties with

84

CHAPTER 4

Western Europe have all grown dramatically. Spain’s admission to the European Union accelerated its economic growth and helped sustain it. International agreements, especially those with the EU and NATO, have had a direct impact on Spain, but their indirect impact is at least as great. They send a powerful signal around the world that Spain is now part of the European mainstream, domestically and internationally. After all, it met the stringent criteria for admittance to NATO (1982), the European Community (1986), and the Western European Union (1988). Spain’s success in managing its electoral transitions and its acceptance into these key institutions clearly proclaimed its new status as a stable Western democracy. As Madrid’s institutions and international status developed, its public and private partners built up a wealth of information about successive Spanish administrations. They now have well-placed confidence in the regime’s continuity as a democracy, and a much better sense of its commitments and likely policies. The uncertainties that surrounded Spain in the mid-1970s now surround many former communist states in Central Europe and former authoritarian states in Latin America and Asia. They—and their partners—must cope with the aftershocks of social and political upheaval. As in Spain, the uncertainties linger because the upheavals were not root-and-branch efforts to destroy the old order. Nor was the old order wiped away by military defeat. As a result, significant elements of the ancien re´ gime linger on, a continuing threat to democratic and economic transformation. Antiquated, state-run enterprises continue to operate in some states, sopping up capital resources and fighting against market reforms. The old ruling parties, though no longer in power, still operate under new names and remain an electoral force. In some countries, old leaders have become godfathers in corrupt new underground economies and patrons of government officials.29 The military, police, and spies may not yet be under secure civilian control. Outside of Central Europe, a number of new democracies have also faced difficult transitions, and some have failed entirely. Zimbabwe’s first elected leader, Robert Mugabe, gradually monopolized power, jailed his critics, confiscated their land, unleashed gangs to kill anyone who objected, and then stole the ensuing elections.30 In Venezuela, left-wing president Hugo Chavez was briefly deposed in 2002. The military failed to complete the coup, unlike Chile’s generals, who killed their elected president, Salvador Allende, three decades earlier. Pakistan has had even more difficulty maintaining democracy. Its history, unlike Venezuela’s or Chile’s, is mostly one of military rule, interrupted by brief periods of elected government, notable for its corruption and incompetence. The list of failures and hesitant transitions is a long one. There is no guarantee that new democratic regimes will succeed, as students of Weimar Germany know.

RELIABLE BARGAINS

85

For our argument, the sources of uncertainty are less important than their looming presence. When partners do not have much experience dealing with a regime, and when they are unsure whether it will survive (or survive without major changes), they are bound to be confused and uncertain. These problems may well diminish with time, but until they do, it is hard to tell whether new regimes will be truly reliable partners. Any guess—optimistic or pessimistic— is filled with doubts. That obviously impedes strong, durable bargains. If serious agreements are to be made, they must be well braced against such perceived risks. For all these reasons, then, the continuity of established democratic regimes adds significantly to their partners’ confidence and lowers the costs of securing agreements. Individual leaders come and go, but democratic institutions stay. Their constraining and empowering rules stay with them.31 The regime’s stability makes its major policy commitments more stable and enduring, especially those that enjoy support from the loyal opposition. In other words, in constitutional democracies, regime stability, audience costs, and entrenched procedures all work in the same direction, to reduce the expected variation in future performance.32 What if the opposition is unalterably opposed to some policy or commitment? In that case, too, the transparency of democracies is important. It allows potential partners to adjust to the expected difficulties, often well in advance. They can add safeguards or diminish their own reliance if they think opposition parties will come to power and eventually change policies.33 Likewise, transparency sharply reduces the dangers of surprise defection from agreements. If a current commitment is in serious doubt, the opposition will typically broadcast its position during the political campaign, before its election and long before it can act. If anything, the opposition’s policy differences are exaggerated for electoral effect. Once in office, they may adopt a more conservative stance favoring continuity. If their objections are to are believed, they will diminish others’ reliance on democratic promises, especially by risk-averse partners. If they are histrionic exaggerations, or if they are sincere but policies cannot be changed because of divided government and constitutional rules, then any lesser reliance by partners is simply a needless, expensive precaution. Dictatorships have the opposite qualities: public expression of opposing views is suppressed. Instead of independent newspapers, they have large posters praising the glorious leader and his superhuman achievements. These leaders hang on for as long as they can, but their tenure (and life expectancy) is always in doubt, and so is the regime’s ability to survive their death without sharp discontinuities. As one headline put it, “In Zaire, They Finally Ask, Who Follows Mobutu?”34 The real question, however, was not just who would follow Mobutu, but what would follow him. At the time, the former general had been in power for more than three decades, since his successful coup. No one had any idea what kind of government would follow him, or how stable it

86

CHAPTER 4

would be. Indeed, his countrymen could only raise these delicate questions once “Citizen President-Founder” Mobutu Sese Seko became seriously ill and had to leave the country for cancer treatment. As in all strongman dictatorships, the very question of succession is forbidden as long as the leader and his henchmen can enforce public silence. In such a regime, one of the most dangerous and short-lived positions is that of “likely successor.” As it turned out, succession in Zaire was unusually chaotic and deadly. Mobutu, near death, was finally forced into exile. Laurent Kabila, the leading rebel and Mobutu’s longtime enemy, seized power in Kinshasa. He renamed the country “the Democratic Republic of Congo” and proclaimed a new era of “national salvation.” He failed to mention democratic elections, which he had demanded so often while fighting. It was a telling omission. In fact, Kabila immediately gave himself unlimited powers and proved to be just another bloodthirsty thug. Within months, a UN human rights report condemned him as no better than Mobutu.35 The civil war continued, and Kabila was soon fighting other guerrillas and local warlords, some linked to the genocidal ethnic conflict in neighboring Rwanda and Burundi. There was a direct connection between Kabila and these conflicts, since he had seized power with vital support from the Tutsi in Rwanda. Before Kabila was shot dead in the presidential palace, Congo’s civil war had exploded into a vast regional conflagration, as half a dozen neighboring states stepped in to support contending rebel factions. This battle for influence, territory, and plunder—often called Africa’s first world war—has left more than two-and-a-half million dead and another million homeless.36 In the next chapter, I will discuss succession issues like these, since they are frequently linked to war. What is important to underscore here is that unless the regime has deep and stable institutional roots, the succession question is not just about which individual will take power, but about whether there will be radical breaks in policy or even an entirely new political regime. Will there be major discontinuities that affect international relationships, including commitments to other countries and foreign companies? As Brantly Womack has observed about the People’s Republic of China, “It is unclear how long [the current] regime will last, what might succeed it, and how it will attempt to resolve its contradictory commitments to repressive recentralization and to continuing modernization and ’openness.’ ”37 Who knows? There could well be continuity, with the existing regime sustaining its dual policies of political closure and economic openness. But China’s partners cannot count on it; nor can they count on a dramatic change of regime. There is a wide range of possible outcomes, and it is hard to be confident about any prediction. The same could be said for Cuba, Libya, North Korea, and a host of others. Beyond these questions of succession and regime continuity, nondemocratic states have inherently more secretive policymaking, often coupled with tight restrictions on public disclosure and discussion, as in Saudi Arabia, Syria, or

RELIABLE BARGAINS

87

Iraq. This secrecy and the absence of public opposition make it hard to tell whether a particular foreign commitment commands broad or narrow support, or even whether it is a personal commitment by the ruler, a commitment by the party in power, or a truly national commitment that can withstand large political changes. That leaves partners with irreducible doubts.

Secrecy, Openness, and Information Technology Why are nondemocratic states so secret? If secrecy creates such international problems for them, why don’t they simply open up political communications and reap the rewards?38 The answer is straightforward. They use secrecy and regulated information to maintain domestic control over politics, society, and the economy. Of course, secrecy also gives them some advantages in international negotiations. A state’s ability to bluff can hide weakness, exaggerate the minimum bargain it will accept, and thus yield some distributional gains, even though it also makes such states less desirable partners. But those are mostly side benefits and costs. Nondemocracies maintain secrecy mainly for domestic reasons, to cloak the mechanisms of domination. They care much more about securing their own power than about securing gains from international cooperation. When the two are at odds, the choice is easy.39 Secrecy and the suppression of public discourse help nondemocratic regimes maintain their political power because they block groups (inside the government and outside) from discovering their common grievances and organizing around them. Secrecy divides and isolates. It keeps information fragmented and opens a chasm between private preferences and their public expression. In countries where the public space is filled with smoke and mirrors, with armed force behind them, it is hard to know who is really embittered and who is a spy, how many are truly alienated, and whether any are ready to act on their complaints and aspirations. With the secret police lurking, it is dangerous to talk of such matters, and often just as dangerous to listen. The state not only prevents independent communication through the mass media, it jails, exiles, or kills those who express their opposition in other ways.40 On top of that, state-controlled media spew a continual stream of distortions, omissions, and outright lies.41 These stratagems of disinformation, isolation, and punishment have a cumulative effect. They encourage nonparticipation and social anomie, and they weaken incentives to join the regime’s opponents. That is exactly why dictators use them. Some techniques, like ubiquitous state-organized propaganda, are relatively new. They were perfected in the age of mass media. Others, like political censorship, are truly ancient. What is novel historically is not the suppression of the press, but its freedom and independence. Its rise is closely linked to the rise of democracy itself, both conceptually and historically. The conceptual

88

CHAPTER 4

link is that democracy requires an informed citizenry, one that is interested in public affairs and able to participate in them. That requires open lines of communication between citizens and elected officials.42 The historical link is equally clear. Mass-circulation newspapers, independent of state controls, arose during the nineteenth century as the electorate expanded and adult literacy spread in Europe and North America.43 Before the rise of modern democracy, restrictions on the press were commonplace. In 1815, when all Europe’s large states were controlled by monarchies or aristocracies, none had meaningful press freedom.44 All imposed significant restrictions, especially on the publication of political news and comments. Not only were newspapers and pamphlets regulated directly by the government; they also had to pay high stamp taxes, buy special licenses, and post expensive libel bonds. These combined measures stifled the growth of the press, limited criticisms of the government, and effectively prohibited the poor from publishing their own papers.45 That was exactly their intention. Over the course of the nineteenth century, most of these press controls were relaxed in tandem with the growth of voting rights.46 Scholars are just beginning to explore the relationship between the two, but the timing alone suggests a strong connection. By midcentury, the United Kingdom, Scandinavia, and the Low Countries had all developed relatively free presses. They were also the first to expand voting eligibility. On the eve of World War I, the only European states with severe press restrictions were Germany, Austria-Hungary, and Russia. All of them were repressive monarchies with narrow public spheres.47 The reasons for such press restrictions are plain enough. They prevent public criticism and curb political participation. Unpopular regimes are especially eager to hide information about opposition movements, since that limits their recruitment and effectiveness. Above all, political leaders want to retain their hold on power and maintain the quiescent order that serves them so well. Why wouldn’t they? In the nineteenth century, press controls promoted these regressive aims. They still do today. That is why it is such a perilous business for nondemocratic governments to relax their information controls. Even a seemingly minor opening can have disproportionate effects, and it is difficult to know in advance what the effects might be. The reason is that the articulation of political or economic demands by some groups informs others, who can then revise their own information accordingly. Onlookers, now armed with better information, may then decide to reveal more of their own antigovernment views publicly. Again, others revise and update their information, and the cycle continues. The sources of public disclosure can be quite varied: public demonstrations, commercial boycotts, “bread lines” for scarce products, or negative voting in official plebiscites.48 In a closed, dictatorial world, where citizens are starved for information, all these sources reveal significant new information about the

RELIABLE BARGAINS

89

scale and depth of opposition, the presence of organized resistance, and the risks the regime’s opponents are willing to take. Whatever the source of this disclosure, it can produce a cascade of increasingly accurate knowledge and revealed preferences, culminating in mass resistance to the regime itself.49 Unless direct repression stems this cycle of revelation and resistance, the regime itself will unravel. The communist regimes of the old Warsaw Pact, following Gorbachev’s policy of perestroika, learned this too late. They were old hands at suppressing free speech, neophytes at allowing it. They attempted to learn by doing, and they were undone. The elderly revolutionaries of China were watching closely and have strived desperately to avoid the same fate.50 They face a complicated task, since they not only want to strangle dissent, but they also want to encourage economic growth. They have learned the hard way that economic growth requires private incentives and decentralized decisionmaking. There’s the rub. As growth proceeds, it creates powerful and knowledgeable private actors, who have their own networks of contacts and communications—networks the state does not always mediate.51 The state cannot choke off these networks without damaging the economic growth it seeks. Moreover, China, like all modern states, finds it increasingly difficult to control the newest technologies of communications. Electronic networks are becoming cheaper and more accessible, as well as easier to use. Because personal computers and Internet connections are so widespread, and because sensitive communications can be encrypted, political authorities find them much harder to suppress or control than television, radio, and printing presses, which are much more centralized. Even older media like television are changing quickly and dramatically. Television sets, once expensive and rare, are now cheap and commonplace. The cameras that captured the news in the 1950s and 1960s were bulky and expensive, used only by professionals. Today, small video cameras and tiny “spy cams” are widely available, highly portable, and used by amateurs to record all kinds of public events, including some the police would rather keep private. Not long ago, broadcast signals came from large towers with a limited range and few channels. Now they come from fiber-optic cables and direct broadcast satellites, covering whole regions with hundreds of channels. A modern-day Goebbels would have many more vehicles to spread his messages, but he would also find it much harder to monopolize the channels of communications. Foreign broadcasters are increasingly able to send their signals across borders without interference. They did so initially by cable, which the state can regulate without difficulty. Now, they depend on satellites and the Internet, which are much harder to inhibit. These new technologies permit much more decentralized communication, and they are much more difficult for repressive states to control.

90

CHAPTER 4

There is a volte-face to these technologies. They also give the state wider powers of surveillance. An intrusive state can now collect and analyze billions of bytes of information on citizens from all kinds of electronic sources. The political impact of all these new technologies, and their complicated interplay, is just beginning to be felt.52 Control over public information, from newspapers to fax machines to Web sites, is a common feature of all dictatorial regimes. That is no accident. The goal of secrecy and state-disseminated information is to maintain political power, and that means stifling opposition. One effective way to do that is to prevent the aggregation of grievances. Since the aggregation process is public and interactive, nondemocratic regimes typically try to block individuals from openly expressing their preferences and criticisms. The state’s methods may be systematic or haphazard, depending on the regime’s efficiency, but their aim is always the same: stopping public criticism. The full arsenal is directed at the task—suppression, punishment, shaming, delegitimation, denial of media access, denial of other resources, assaults on family members, blackmail tactics, removal of suspected dissidents and their relatives from sensitive posts, tighter controls over public assembly, and so on. Political killings are part of the arsenal. “Assassination,” as George Bernard Shaw once remarked, “is the extreme form of censorship.” The empty public spaces are filled with praise for the regime and its accomplishments. It is a game for high stakes. Even if dissidents are brave enough to voice their complaints and skilled enough to evade the stringent controls, they are still blocked from disseminating their message widely as long as the state itself controls the broadcast media. Indeed, maintaining control over public information is standard operating procedure in authoritarian countries and illiberal democracies. This secrecy is especially tight around military and security affairs, which are central elements in the regime’s repressive apparatus of political and social control. With dissent suppressed in broadcast media and newspapers, public criticism must be expressed in other ways, if it is to be expressed at all. That is why public demonstrations so often play an important mobilizing role. Modern examples include the Monday demonstrations in Leipzig,53, the Mothers’ Marches in Buenos Aires,54 and the unsuccessful demonstrations for democracy in Tiananmen Square.55 Democracies, it might be objected, are also filled with demonstrations. Indeed, they are. They are commonplace and often large. They, too, inform fellow citizens, but in ways that reflect the different social and political context of democratic participation. Mass demonstrations like the Civil Rights March on Washington or antiapartheid demonstrations in London gain most of their impact because they are broadcast freely to the wider world. In one respect, they are like demonstrations in authoritarian states: they permit observers to update their information. In fact, they reach many more observers because they

RELIABLE BARGAINS

91

are freely advertised and broadcast. That is where the crucial difference lies. In nondemocratic states, where the media are controlled, demonstrations are not broadcast. They are actually a substitute form of communication. With the public airways blocked, demonstrations are an alternative way to inform their compatriots that an opposition exists, that it has specific grievances, that its proponents form a nucleus, and that they are sufficiently courageous and committed to face the regime’s wrath. It is a message rich with information. In democracies, on the other hand, public expression takes less courage and is therefore less informative. Demonstrations are usually meant to raise the salience of issues, promote specific causes, and inform political leaders about the intensity of supporters. (To find out the breadth of support, politicians simply read the omnipresent polls.) Occasionally, as in the 1968 Paris riots, they are used to express more fundamental opposition to the regime itself, often through violence. Truly revolutionary movements like this are quite rare in Western democracies and, in any case, the press is free to report on them and on the state’s countervailing efforts. Much more common in democracies is the blurring of lines between civil society and state officials, between private demonstrators and elected officials who speak at the rallies. No major demonstration on human rights, the environment, or foreign policy would be complete without several prominent elected officials among the speakers and marchers. Their participation not only shows their own political stance, it reminds us that public demonstrations are integral parts of democratic life and legitimate forms of political participation. Even if democratic governments wanted to prevent such demonstrations, they could not. There are sharp limits on any state efforts to control speech— public or private; political, artistic, or commercial; spoken words, acts, gestures, or signs. All are protected within a liberal, constitutional order, in theory and to a great extent in practice as well.56 Aside from widespread public support, democratic rights to free speech are guaranteed by statutes, court decisions, and the constitution itself.57 Why so many protections for speech? For three basic reasons: First and most fundamentally, free speech is valued in its own right, as an expression of individual freedom and moral autonomy.58 Second, the right to criticize government officials is one way of holding them responsible; conversely, it allows officials to gauge the depth of public views and adapt their policies accordingly.59 Finally, free speech is needed so citizens can vote on an informed basis, exercising their ultimate sovereignty within a democracy. There are some limits to free speech, even in highly democratic states, and one should not romanticize the actual practices.60 All governments, including democracies, work strenuously to sell their policies to the public, using the taxpayers’ own money to convince them. All have their own privileged access to broadcast media. All governments, including democracies, use official secrets to protect themselves from embarrassment, as well as to protect the public

92

CHAPTER 4

from real harm. All put some limits on morally objectionable speech and strict limits on dangerous speech, especially during wartime. Despite these limitations, the rules governing political speech in established democracies are distinctive in (1) their limited extent; (2) their disputed legitimacy; and (3) their open contestation by strong private groups, led by opposition parties and independent news organizations. As a result, in liberal, constitutional democracies, none of these restrictions creates a world of pervasive secrecy, silence, and disinformation. Nor do they create a reliance on one or two sources of information, much less on statecontrolled media. For nondemocracies, on the other hand, especially for the most repressive and dictatorial, tight control over information is a key element in maintaining their domestic power. The same is largely true for illiberal democracies, where one ethnic group or religion uses its majority voting power to marginalize and silence opposition views and, if possible, extinguish them.61 The impact of all this secrecy and disinformation on foreign relations is not so much a desired goal as it is the by-product of illiberal and nondemocratic state structures. These restrictions on information sometimes give individual states an edge, but they harm the chances for international cooperation and raise the possibility of military escalation and war. They do so because they undermine the ability of nondemocratic states to project critical promises and overcome fundamental commitment problems. Partners of nondemocratic states face three major sources of uncertainty that are largely absent when they deal with established democracies: 1. considerable secrecy surrounding the policymaking process 2. greater uncertainty about whether any commitment will effectively bind the leader and regime that made it, given the lack of internal controls over the ruler’s powers 3. greater uncertainty about what kind of leader and regime will emerge in the future and whether it will keep its predecessor’s commitments This unhappy combination is bound to erode confidence in all promises, especially longer-term ones that transcend the current leader. Some nondemocratic governments, such as strong monarchies and oneparty states, lie between the two poles of open democracy and secretive dictatorships. They have some continuity across individual leaders and perhaps some organized constraints on policymaking. Even so, their promises are less reliable than those of established democracies for two reasons. They are less transparent, and they have fewer ways to limit their leaders’ power, now and in the future. Constitutional democracies do have ways to limit that power, and they have meaningful procedures for making different kinds of commitments. These are not mere formalities. They are integral features of state structure, and they are invoked regularly.

RELIABLE BARGAINS

93

The institutional constraints and informational advantages of democratic government are thus profoundly important. Thanks to their constitutions, democracies can effectively encode some commitments as long-term pledges. That makes them very hard to reverse or abandon as long as the constitution itself remains in place. Constitutions also serve as additional protection for free political speech, which does so much to make the democratic polity transparent and its promises credible and reassuring. Transparency makes the depth of support readily apparent to outsiders and makes deception more difficult. Electoral accountability imposes a cost on leaders who wantonly abandon their commitments.

Self-Binding Democratic promises are more credible because they are the products of effective (if imperfect) institutional arrangements for self-disclosure and self-binding. That gives democracies a special capacity to make long-term commitments, to make promises that bind future rulers, as well as the present ones. This self-binding is possible because individual promises are made within a constitutional setting. They are the products of fundamental rules and procedures designed to shape and limit policies. This larger ensemble of constitutional arrangements includes basic protections for property and contractual rights, clear rules for ratifying and interpreting treaties and other international commitments, fair legal procedures (which can handle disputes involving the government itself), and special requirements that limit changes in key rules and procedures by requiring supermajorities. Because rules differ, even among established democracies, one might expect some differences in their ability to reverse commitments. Parliamentary systems, for example, are more flexible than presidential systems with divided powers. They can act more quickly and change policies more readily because the chief executive is also head of the legislature, where he commands a working majority. On the positive side, this consolidated power makes it easier to adapt policies to new conditions. On the negative side, it means that old commitments can be discarded more easily, although the system is still transparent and the prime minister is typically hemmed in by other senior parliamentarians and by the constitution itself. Democracies with many potential veto points are probably more rigid, and therefore more reliable, because they can block potential changes more easily.62 Judicial systems, like legislatures, can become veto points in democracies. Even though courts are reluctant to intervene in foreign policy questions and provide only imperfect protection, they offer some support for properly made commitments, some resistance to short-term political pressures.63 As with legislatures, their key features vary considerably across democratic govern-

94

CHAPTER 4

ments—with some impact on the reliability of international agreements. Some states, like Japan, lack truly independent courts. That may well be the result of decades of one-party rule, which left courts with little protection from political interference.64 Others, like the United Kingdom, have independent courts, but they cannot decide constitutional questions, such as the government’s right to abrogate treaties. This kind of constitutional competence is important because it can prevent arbitrary changes in policy, including foreign-policy commitments.65 U.S. courts are both competent to decide constitutional questions and truly independent, in black-letter law and in political practice. Their role sometimes confuses others, since it means yet another voice speaking for the government, but it is also an important anchor for long-term promises. These cross-national differences in courts and legislatures are important, and they deserve further exploration. But the differences should not obscure what is common—and vital—to democratic governments: the capacity of all constitutional democracies to solidify their pledges by making them hard to reverse. This common feature, deeply embedded in democratic state structure, helps them sustain long-term promises and differentiates them from nondemocracies. The Panama Canal Treaty offers a very simple example of how difficult it is for new democratic leaders to reverse old promises, once they have been locked into place by constitutional devices in a government of separated powers. The Canal Treaty was negotiated under Presidents Gerald Ford and Jimmy Carter and signed by Carter in 1977. It promised to turn the Canal over to Panama in several stages, ending in 2000.66 An agreement like this had to be ratified by the Senate, where a two-thirds vote was needed to pass it. When candidate Ronald Reagan ran for the presidency, he dismissed Panama’s nationalist grievances, calling them “blackmail.” He vigorously attacked President Carter for promising to “give away” the Canal. Still, Carter was the incumbent, and he submitted the treaty to Congress. After a grueling debate, it passed narrowly.67 With the treaty’s ratification, the issue was closed—firmly closed. By the time Ronald Reagan assumed the presidency, the ink was dry. What makes this finality so remarkable is that under the treaty’s terms, U.S. withdrawal from the Canal Zone was still years away. The withdrawal had not even begun when Reagan took office. In this case, however, the promise was irreversible, even though no action had occurred on the ground in Panama. Whether he wanted to or not, Reagan could do nothing to overturn the agreement or even ignore it after he took office. Even if he had faced no international pressures, he could not sidestep the constitutional restraints at home.68 The courts would not let it be interpreted away; the Congress would not let a president junk a valid treaty on his own authority. Almost two decades later, the United States turned over the last of its twelve military bases in the Canal Zone, leaving an area it had held for ninety-six years. A month later, at noon on 31 December

RELIABLE BARGAINS

95

1999, it transferred the Canal to Panama, exactly as it had promised over two decades earlier. The contrast between the Canal Treaty and the Law of the Sea is instructive. The Law of the Sea was a major multilateral initiative in the 1970s designed to update and modify ocean law. All the traditional issues were considered, along with several new ones such as seabed mining, which was becoming economically feasible. The negotiations took years, and U.S. participation was important to every facet of them. The Carter administration endorsed the compromise bargain, just as it supported most international organizations. It was particularly eager to see rules in place for seabed mining, since the administration believed a clear regulatory framework would encourage economic activity in a sector where U.S. firms dominated. The agreement was nearing completion but had not yet opened for ratification as the 1980 U.S. presidential election heated up. Candidate Reagan staked out a very strong position, just as he had on the Panama Canal Treaty. In contrast to President Carter, Reagan was firmly opposed to the treaty’s supranational organization. Its main effect, he thought, would be to tax American corporations, the world’s leaders in seabed mining technology.69 What was really needed, in his opinion, was the freedom to get on with business, not overregulation by some petty Third World bureaucrats. In this case, Carter’s years of multilateral negotiations were overturned when Reagan won. The United States might pay a diplomatic price for reversing course, but President Reagan had ample authority to do so because the United States had not yet ratified the treaty or made any binding commitments. The U.S. Constitution is quite clear on the point. The United States had never promised to adopt, without ratification, the new Law of the Sea. Indeed, Reagan had no obligation even to submit it to a Senate vote. Regardless of President Carter’s negotiations, regardless of what our partners hoped, the United States had not committed itself. Others could argue, quite reasonably, that the United States had negotiated in bad faith. Their frustration was palpable. But the United States had never made a national commitment, formal or informal, and in the end, it chose not to do so.70 The fate of the Kyoto Protocol on Global Warming was virtually identical. Negotiated in 1997, it was never submitted to the Senate for ratification. President Bill Clinton signed the treaty in 1999, mainly as a gesture of support, but he indicated he would not seek ratification unless major developing countries were also required to reduce their greenhouse gas emissions. Clinton’s successor went further. George W. Bush openly opposed the treaty and, soon after taking office, decided that he would never submit it to the Senate. Pleas from Japan and Europe were fruitless. The United States refused to ratify the treaty and refused to be bound by its provisions. Negotiating partners are nonplused by episodes like this, and understandably so. But that’s life in a democracy. Policy ideas are openly debated; private

96

CHAPTER 4

interests are openly organized and promoted. Public officials advance their own preferences and pursue their bureaucratic interests, just as they do everywhere, but in democracies their actions are open to wider scrutiny, both by the public at large and by other officials performing their duties. The chief executive, far from rising above the fray, is sometimes unable to commit the state to a policy he favors and may even have negotiated. That is true even when the stakes are very high, even when democratic leaders have a clear sense of direction. It was certainly true for Woodrow Wilson and the League of Nations. It was equally true for Franklin Roosevelt in the early days of World War II in Europe, when the United States was not yet a belligerent. Roosevelt was well aware that American help was essential to stop the Nazi advance in the late 1930s. At the very least, he wanted to extend generous financial assistance to anti-Nazi governments.71 He managed to arrange some Lend-Lease aid, but it was modest under the urgent conditions. The United States edged toward helping its would-be allies, but its policies did not change fundamentally until Japan attacked the U.S. Pacific fleet at Pearl Harbor, and Hitler recklessly declared war on the United States. Until then, Roosevelt had faced clear limits. Public opinion was lukewarm, at best, about greater involvement in Europe’s war. Isolationists were well organized and prominent, and there were numerous constitutional obstacles to giving more aid without congressional authorization.72 Roosevelt packaged and sold the Lend-Lease program brilliantly, skirting the edges of the law while trying to supply Britain as it fought for its life.73 Roosevelt certainly understood the national security issues, but he did not dictate to the people; he led.74 His campaign to aid England gained special urgency after France fell swiftly to the German Army in May 1940, leaving Hitler master of continental Europe. Against this darkening sky, Roosevelt built political support for Lend-Lease in the country and on Capitol Hill, overcame isolationist objections, and managed to pass the bill in March 1941.75 It was backed by rising public approval, which Roosevelt himself had done much to foster.76 The whole episode shows the gaping difference between Roosevelt’s democratic political leadership and the autonomous power wielded by Hitler, Stalin, and Mussolini. To take just one example: Hitler was able to abandon, in total stealth, his treaty with Stalin dividing Central Europe. The Kremlin was just as secretive. When the Molotov-Ribbentrop Pact was originally negotiated in August 1939, the only Soviet leaders who knew about it were Stalin and Molotov. Even the most senior members of the government and the communist party—the Soviet Politburo and executive committee of the Comintern—were kept in the dark. For years, their regime had been Hitler’s most vocal opponent. These senior officials learned that their most basic foreign policy had been completely reversed when they read it in the newspaper.77

RELIABLE BARGAINS

97

By then, the pact was already signed and needed only pro forma ratification in the two dictatorships. Hitler always viewed cooperation with the Soviet Union as a purely tactical maneuver. It was designed to hand him a large chunk of Poland with Stalin’s active help instead of his bitter opposition. A quick consolidation of these gains would free Hitler to fight on the western front, his next goal.78 That, too, was part of a larger plan of conquest. First a limited strike at Poland, then wheel quickly through the Low Countries and France, and finally turn eastward again toward the great prize, unlimited territorial expansion across central Europe and into Russia. The idea was to fight successive wars against one opponent at a time, an updated version of Germany’s Schlieffen Plan for World War I. To accomplish these megalomaniacal goals, the Wehrmacht had to defeat its adversaries in the west quickly and then turn east before Stalin was ready to fight.79 Better yet, they could lull him into thinking he might never have to fight the Germans and then strike him unprepared. The whole idea was to launch a series of one-front wars of maneuver, designed to overrun all Europe in a series of lightning attacks. Hitler’s planned sequence of expansion made short-term cooperation with the Russians desirable—indeed, essential if a two-front war was to be avoided—but it left no joint interests to pursue beyond that.80 Quite the contrary: there would soon be a direct clash of fundamental interests. Hitler certainly knew it, and possibly Stalin did as well.81 Accordingly, Hitler began thinking about attacking Moscow soon after he signed the Molotov-Ribbentrop Pact.82 The war against Russia, Operation Barbarossa, would be Hitler’s fatal mistake, but it was his mistake to make. Treaty or no treaty, he launched the attack as soon as he thought the Nazi war machine could beat the Red Army. The onset of full-scale war on the eastern front was hastened by a grim symmetry. If Hitler was to move eastward at all, he had to move quickly because Stalin seemed to be planning his own westward expansion in Finland and Turkey.83 In fact, the Red Army had already moved across the agreed demarcation line in Lithuania.84 As one historian of the period concludes, both fascists and communists saw “treaties and agreements [as] tactical devices in an unending contest.”85 Amazingly, Stalin was actually deceived by Hitler’s oily promises. He was completely surprised by Barbarossa. The British had urgently warned him an attack was coming, but Stalin refused to believe them, fearing they were trying to trick him into joining their war against the Nazis. A few days later, the Wehrmacht launched a full-scale assault. Stalin, still enthralled by his agreement with Hitler, refused to believe his own army’s initial reports of the devastating attack. It is worth remembering that these disasters were not accidents, somehow beyond the scope of human control. Although the outcomes and costs were not foreseen, they were the product of deliberate strategic choices made with

98

CHAPTER 4

patchy information. They were coherent policies, developed by autonomous leaders who spoke authoritatively for their countries. It was their interaction, sparked by Hitler’s limitless ambition, that produced World War II. Thirty years earlier, another combination of deliberate state choices and their catastrophic interaction had produced World War I, sparked by ethnic dilemmas within the Austro-Hungarian Empire, Germany’s exaggerated fears of growing Russian power, and the Reich’s goal of acquiring substantially more territory in eastern Europe. Foreign policy mandarins complain, quite predictably, that no state can act responsibly unless its leaders speak for it with one coherent voice.86 Likewise, foreign partners are often troubled when policies change after an election, or even in anticipation of it. As it happens, elections are actually designed to permit peaceful changes, if the public wants them. That is the whole point of elections. Sometimes the “outs” win, and public policies are modified or even reversed. Inevitably, electoral competition produces some erratic policies. New leaders with new ideas take control, and they may be inexperienced. So the criticisms are true, up to a point. Yet they miss something much more fundamental. Although nondemocratic governments may speak with a more consistent voice, they have no way of ensuring that their successors will speak with the same voice or fulfill their predecessors’ promises.87 Indeed, they have no way of ensuring existing leaders will later fulfill their own promises rather than discard them.88 This kind of assurance is precisely what is needed for stable, long-term bargains. It is sufficient when combined with the presence of joint gains and some agreement about how to divide them. Although assurance may be necessary for long-term bargains, nondemocracies cannot offer it because they will brook no effective institutional limits to their own domestic power. Without such limits, they cannot possibly bind themselves, much less their successors, to the promises they make. Without that assurance, who will want to rely on their promises without ample precautions? Partners of nondemocratic states are thrown much more to their own devices—to the self-protection that so often incites fear in others, the security dilemma. When they try to escape by cooperating, they find the path is arduous and uncertain. They must devise signals of their benign intentions that cannot be easily faked, and hope that others can do the same. They must count heavily on the stability of others’ national interests over time, bolstered by intricate and costly safeguards on the agreements themselves. All these troubles are self-inflicted. All too often, even these elaborate efforts are not enough, especially when combined with states’ fears about their neighbors and a veil of secrecy shrouding their own capabilities and strategic choices. Wars grow out of this nexus of fears, furtiveness, and failed bargains. The fears are understandable. So is the desire to hide vital information about national capabilities and intentions. In a dangerous world, there are powerful incentives to conceal and deceive.89

RELIABLE BARGAINS

99

Even if states could contract around these problems—and they often try— they still find it difficult to strike stable bargains that go significantly beyond deterrence and actually lessen the costs and risks of war. Democracies are much better equipped to make such bargains, and they are much more successful. Almost all wars are the result of deliberate choices made with very incomplete information, not only about the other side’s resources, but also about the risks involved and their adversary’s likely choices. Until relatively recently, states had very little solid information about others’ capabilities and intentions. What little they had was hard-won. That was not just a consequence of slow communications and few specialized officials to collect and analyze foreign data. It was also a natural consequence of state secrecy in a political environment where military conflict was commonplace. Foreigners were blocked at every turn from collecting information that might give them a candid glimpse of others’ intentions or capabilities. Throughout early-modern Europe, diplomats were admitted to conduct interstate business, but they were typically regarded as spies. They were watched carefully, severely restricted from meeting with local merchants or officeholders, and sometimes guarded by troops. In eastern Europe, they were frequently kept under house arrest.90 In Venice, local officials were forbidden from having any unauthorized contacts with foreigners regarding state business.91 Besides these restrictions, espionage was largely an ad hoc affair—a spy here, a bribe there, rather than the extensive, specialized intelligence agencies used by modern states. Before the telegraph, reporting any discoveries was also a slow and difficult task. The result was that early-modern states had only a murky picture of their adversaries’ aims and options. When they undertook major foreign policy initiatives, including wars, they did so with little understanding of others’ interests, strategies, and capabilities—and, therefore, little grasp of the dangers. Charles Wilson says exactly that about the world-shaping wars of the sixteenth and seventeenth centuries: “Great military projects—like Spain’s ‘enterprise of England’ of 1588, England’s intervention in the Netherlands in 1585, Denmark’s invasion of Germany in 1625, even Richelieu’s entry into the Thirty Years War in 1635–6—were launched in elementary ignorance of the risks they faced.”92 States are less ignorant today. They have developed much better military intelligence. They have sophisticated technologies for spying, as well as large, specialized bureaucracies for collecting and analyzing foreign information. They have a clearer grasp of each other’s resources and perhaps their intentions. They may have a better feel for strategic interaction, although it would be hard to top the subtle calculations of Renaissance princes and absolutist monarchs. These advances are not trivial, but they still do not overcome the basic information and commitment problems that are faced by all states and are particularly acute for nondemocracies. Dictators have compelling reasons to hide their capabilities and intentions, to confuse and deceive, not only to

100

CHAPTER 4

give themselves an edge in interstate conflicts, but to secure their domestic governance. Even when they see the need to reassure their neighbors, they have precious few tools at their disposal. A story from Hitler’s inner circle is instructive. To celebrate Foreign Minister Joachim von Ribbentrop’s fiftieth birthday in 1943, several aides gave him a beautiful box, decorated with semiprecious stones. They had planned to fill it with copies of all the important treaties and agreements he had concluded. They finally chose to leave the box empty because the Nazis had trampled every agreement Ribbentrop had made. As one aide told Hitler at supper, “We were thrown into great embarrassment. . . . There were only a few treaties that we hadn’t broken.” Hitler’s eyes filled with tears of laughter.93 Lenin had the same viewpoint. He saw bargains as mere tactical tools, to be used and cast aside. “Promises are like pie crusts,” he said, “made to be broken.”94 It was more than a quip. It was a deep-seated attitude that Soviet officials took to heart. Decades later, when a U.S. ambassador told the long-serving Soviet diplomat Andrei Gromyko that he believed both countries were sincerely seeking an agreement, Gromyko replied, “You may be sincere but governments are never sincere.”95 Actually, some are, and it makes a difference. Not all states try to ensnare others in this frayed web of broken promises and unreliable commitments, a web that traps not only the naı¨ve but finally traps its own maker. It blocks bargains when they really need them, just as “crying wolf” eventually obscures real cries of distress and blocks assistance. Constitutional democracies have been able to escape this tragic bind, not because they have better, wiser leaders but because they have better, wiser, more open governance structures. It is these structures that directly address and largely resolve basic problems of disclosure and reassurance.

Making Long-Term Promises Credible: Disclosure, Continuity, and Constraints Democracies have advantages beyond disclosure and reassurance. They are better equipped to make commitments for the long haul, not just the immediate future. That gives them a major advantage in forging durable cooperation. This point is often overlooked because discussions of credibility usually deal with threats, which focus on the here and now. Can they prevent a military attack, for example? The key question is whether threats are effective deterrents today and in the near future. With promises, the central question is different. Are they reliable over time? Treaties and informal agreements, like all contracts, are designed to project exchange into the future. Will they still work well tomorrow, when one side (or both) still needs to carry out its promises? Most international agreements are not “spot transactions,” where an exchange

RELIABLE BARGAINS

101

is made and completed at the same time. Rather, they are commitments that stretch well into the future, covering the level of tariffs, the shape of borders, or a state’s military arsenal. Well-established democracies have a marked advantage in making these medium- and long-term commitments and exchanging them with others that can reciprocate. In fact, it is quite rare for democracies to overturn explicit policy commitments made by prior governments, particularly significant interstate agreements, even when successors have deep reservations about their wisdom. One reason, as we have already noted, is that democracies can encode certain promises in ways that deliberately make them hard for themselves to break. They have strong precommitment devices, requiring concurrence from numerous officials who are independently elected. This could mean that approval requires a supermajority, or a positive vote on several “readings,” or passage by more than one branch of government. There are also less legalistic reasons. Even when one party controls a unified parliament, major policy reversals are difficult to achieve. To survive a floor debate, they first need broad support from senior cabinet ministers. Paradoxically, the rotation of national leadership is also important in sustaining clear national commitments. It gives the party in power real incentives to uphold the precedent that “our country keeps its bargains.” Besides the immediate effects on reputation, current leaders gain from strengthening the presumption that their own bargains will stand through future governments, whether they are led by their own party or the opposition. As long as both parties expect to share power and have a long time horizon, they can sustain an equilibrium in which they respect foreign commitments already in place. J. Mark Ramseyer has shown that similar calculations play an important role in keeping the judiciary independent in the United States, where parties rotate in power, but not in Japan, where one party has dominated for decades.96 The same mechanism of reciprocity-through-time works to strengthen the practice of keeping foreign commitments. The continuity of democratic regimes is also important. Changes of leaders, whether they are presidents, prime ministers, monarchs, or dictators, always raise questions about whether predecessors’ commitments will be kept. Democracies have two advantages in projecting their commitments across changes of leadership. One is that constitutional democracies, especially in the North Atlantic, are quite stable and have well-articulated procedures for undertaking and rescinding promises they have made to other states. Democracies can, if they wish, fix their promises in laws or treaties, which are themselves grounded in established constitutional mechanisms. These nested procedures govern the passage, repeal, and interpretation of national commitments and ensure that they are not discarded lightly or arbitrarily. The machinery of democratic governance is designed to make long-term commitments that are readily discernable, hard to revoke, and therefore credi-

102

CHAPTER 4

ble to partners. Douglass North and Barry Weingast make exactly this point about the rise of parliamentary authority in seventeenth-century England.97 By making it virtually impossible for the Crown to renounce debts, the Glorious Revolution significantly lowered the risks of government borrowing, and therefore its costs. The ability to make credible commitments for long periods transformed the government’s creditworthiness, which, in turn, spurred the rise of London’s capital market.98 I am making a very similar argument about democracies and their treaty commitments. The same kind of institutional machinery that makes commitments credible to the capital markets, transcending specific rulers and blocking arbitrary renunciation, also allows states to make effective long-term promises about other obligations, as well. Second, given the openness of democratic procedures, it is easier for outsiders to determine if changes of leaders really constitute more far-reaching changes of regimes. Do the changes threaten earlier relationships or even the constitutional order itself? By contrast, it is much harder to tell at the time whether Stalin’s successors, or Mao’s, or Mobutu’s, represent radical breaks in governance. It is much easier to make that kind of assessment about democracies. Not only is their policymaking more open; they have open, regular procedures to select their leaders. That helps outsiders determine if new presidents or prime ministers constitute fundamental changes of regime that could jeopardize prior commitments. Because partners can readily observe democratic policymaking and leadership rotation, they have more time to prepare for possible breaches and therefore less reason to fear the consequences when they originally make agreements. As far back as Machiavelli, observers of “republican” regimes have noted their slow movement in foreign policy matters.99 While this sluggishness poses strategic problems for democratic military planners, it has some important diplomatic advantages in reassuring partners. “In cases where there is urgent danger,” according to Machiavelli, “some stability will be found more in republics than in princes. For although republics have the same intent and the same wish as a prince, their slow motion will make them always have more trouble in resolving than the prince, and because of this have more trouble in breaking faith than he.”100 He goes on to argue that republics “can be trusted more than the prince” and have traditionally held to their agreements, despite temptations to break them for immediate gains.101 These are important insights. Five centuries later they still stand. Even if Machiavelli’s republics and today’s democracies were not especially trustworthy, their typical slowness to act and their openness in changing course would lessen their temptation to initiate a violation in the first place. The reason is simple: they would lose the windfall gains from surprise. This slowness and visibility offer partners considerable protection. It gives them an opportunity to anticipate possible breaches and prepare for them. Indeed, it may give them confidence that they can effectively preempt an expected breach.

RELIABLE BARGAINS

103

This cushion of time for self-protection should encourage reciprocal reliance by democratic regimes. By contrast, in the closed world of nondemocratic states, it is far harder to discern policy shifts in the making, far harder to assess the meaning of new leadership, and consequently much easier to be caught off guard.102 Autocratic regimes can adopt new policies swiftly and secretly. Their stealth gives them real military advantages.103 But it also poses inherent risks for any potential partner. It discourages trust. It encourages wariness and military precaution, the generative grammar of the security dilemma.104 For all these reasons, democracies have crucial advantages in projecting their promises effectively into the future, exchanging these promises with each other, and actually relying on them. The form of the state thus becomes an integral part of the technology of self-binding. If democracies’ promises are generally more credible, then they ought to be more trustworthy allies and more reliable treaty partners. They ought to be able to reassure each other better than nondemocracies can and therefore they should be more willing to exchange promises. By offering rational grounds for trustworthiness, they lower the risks that agreements will be broken opportunistically. Facing these lower risks, they can capture more gains from exchange in both economic and security issues. They should be better at sustaining interdependent economic relationships, for example. Why? Because partners in such relationships must rely heavily on each other for export markets, raw materials, investment outlets, and so forth. This dependence makes trading states (and their workers, consumers, and firms) vulnerable to the closure of foreign markets. That is why they need solid assurances or ready alternative markets before making any dedicated investments.105 A simple example is the vast automobile trade between the United States and Canada. That trade was fostered by interstate agreements, some of them decades old, allowing firms to ship cars and parts easily across the border. Without such agreements, auto companies would hesitate to make huge capital investments in Canadian factories designed specifically to produce for the U.S. market.106 To sustain such interdependence, sovereign states must be able to make credible commitments about their future behavior and develop some trust to deal with incomplete agreements and inadequate bonding arrangements. Similar contracting issues arise in national security affairs, where states may be willing to make long-term commitments, such as limiting certain missiles or ships, but only if they are confident their partners will reciprocate over the course of the agreement. In cases like these, the capacity of constitutional democracies to exchange convincing promises, along with their greater openness to scrutiny, should dampen the security dilemma, where mistrust and protective weaponry build upon each other in a kind of echo chamber. Reliable bargains are especially important in security affairs, since the consequences of betrayal can be so devastating.107

104

CHAPTER 4

Thus, the contracting advantage of democracies is not only the key to the democratic peace; it is the key to a wider range of exchange relationships. Democracies’ capacity to assure each other that they will uphold agreements fosters a nexus of commitments, trust, and reliance. It encourages gains from specialization, which are now quite extensive among advanced democracies. These intergovernmental relations are complemented by private firms that profit from close transnational ties and from the development of transnational civil society, where strong ties can grow across borders that are open to travel and communication. This analysis suggests that reliable promises not only lower the immediate risks of war among democracies; they also lay the basis for durable security communities, where mutual fears recede, where states no longer see each other as military rivals, and where the prospects of war seem increasingly remote, perhaps even unthinkable.108

Caveats: Some Limitations on Democratic Promises Of course, not all democratic promises are gold plated. The contracting approach suggests why. If the most credible promises come from sturdy regimes, with constrained leaders and transparent procedures, then the least credible democratic promises should come from their polar opposites: 1. regimes that are new or unstable, 2. leaders with few constitutional restraints, or 3. issues where policymaking is most secretive. New or shaky regimes may not survive long enough to fulfill their promises, or at least that is what their partners fear. That does not matter if bargains are meant to be brief or if the future is deeply discounted. But if the future really matters, then partners of new democracies will worry about their stability, their political trajectory, and their lack of reputation. Against this background of doubts, potential partners are reluctant to rely on untested, unknown governments or to make large investments now in hopes of future gains. If they do, they will undoubtedly want to take special precautions against breach.109 Even if democracies are stable, their leaders are not always constrained by strong constitutional rules or by effective independent branches of government. Even in well-ordered democracies, the executive may accumulate vast powers, or a single party may win control for long stretches. Both outcomes generate the same problem: relatively weak restraints on executive autonomy. We have already discussed how nondemocratic states face severe contracting problems because they fail to circumscribe the head of state. Democracies without effectively shared powers or rotation in office face similar problems. Japan’s partners worried about exactly these issues during the Liberal Democratic Party’s long tenure.

RELIABLE BARGAINS

105

A third major problem is that some democracies are more transparent than others. Some cross-national differences lead to misunderstandings and costly self-protection. Again, the most frequent target of complaint is Japan, which is a kind of “opaque democracy” because its policy is so often formulated behind the closed doors of its bureaucracies. That is certainly true of its economic policy, especially its industrial policy. To make matters worse, government bureaucracies are closely linked to consortiums of major Japanese firms, which are the prime beneficiaries of their policies. Outsiders simply cannot penetrate these vital links. Predictably, this opacity leads to grave suspicions of bias and bad faith when interstate agreements fail to produce the anticipated results. Since it is impossible for outsiders to gain meaningful access to bureaucratic decisionmaking in Japan, they rely instead on blunt instruments to specify desired outcomes. Take trade in semiconductors. A normal interstate agreement might specify some procedural rules to restrict nontariff barriers. That should open up the market and produce positive results for U.S. chip exporters. The rules are agreed to, and—surprise—the results in Japan fall far short of expectations. Since the same rules work well in other export markets, American companies cry foul. The rules are not being administered fairly, they say. The Japanese government and chip manufacturers, on the other hand, attribute the shortfall to outside factors such as adverse exchange rates or the cultural myopia of U.S. firms. With the agreement failing to affect transactions as anticipated, U.S. trade negotiators consider a backup strategy. Why not just set a target for semiconductor sales, probably as a percentage of the market, and then agree on that? I call this a proxy agreement, since it uses easily measured targets as a surrogate for what the parties really want: fair market access. In general, proxies are used to lower the costs of transactions when the underlying variables— what the parties would really like to agree upon—are simply too hard to measure reliably. Quota agreements are frequently used as proxies for processes that are filled with private information and produce results outsiders do not trust. These opaque processes range from semiconductor imports to college admissions. Unfortunately, all proxy agreements are saddled with the same generic difficulty. The proxy variables may not track the underlying variables (the ones the parties really care about), or there may be sharp disagreements about whether they do. Either can cause problems. Consider, for example, an agreement that gives U.S. producers 20 percent of the Japanese semiconductor market. If fair transparent arrangements would give them only 15 percent of the market, then the quota creates significant rents. Inefficient U.S. producers are the beneficiaries. (If the quota were set at 10 percent, then inefficient Japanese producers would be the beneficiaries since they would have a larger share than they could win in fair market competition.) Once these quota arrangements are in place, however, they are very hard to dislodge. Not only will the

106

CHAPTER 4

beneficiaries fight to keep them; they can fund their fight with the very rents they receive. All too often, the result is a self-perpetuating partnership between the recipients of these special benefits and the agencies that so generously provide them. Proxy agreements like this tend to arise when it is hard to measure the desired underlying variables, but where some close approximation might permit a mutually profitable exchange.110

When Procedures Are Opaque Opaque procedures and closed institutions pose two overriding problems for international agreements. First, they foster suspicion and impede reassurance. That not only produces compliance problems through miscommunication, but it strongly colors the interpretation of difficulties when they do arise. It also leads to rough-and-ready solutions like proxy agreements, which permit exchanges to go forward but are riddled with problems of their own. Second, they allow commitments to be overturned without warning, and possibly without wide support. That heightens fears among partners and prompts them either to avoid agreements altogether, to buttress them in costly ways (if that is possible), or to violate them preemptively before their partner does. Even in thoroughgoing constitutional democracies, some decisionmaking is hidden from public view and insulated from public debate. But what is rare and exceptional in democracies is commonplace and fundamental in dictatorships, absolutist monarchies, and other nondemocratic states. This difference is central to my explanation of the democratic peace. For states that want to reassure each other, transparency is vital. In open settings, partners can observe pulling and hauling before policy shifts take place. They can see whether opposition parties support a policy or oppose it, and whether they would try to overturn a settled agreement. Given some advance warning, these partners can take measures to protect themselves. Equally important, they are likely to consider full disclosure as prima facie evidence of benign intentions. As a result, when problems crop up, as they always do, partners in more open settings are likely to view them as honest mistakes or garbled communications rather than as fraud or opportunism. In other words, the task of reassurance and dispute resolution is aided by the credibility of communications.111 When procedures are closed, on the other hand, outsiders gain little sense of why other states make commitments or keep them. They are unsure who supports them and how much. In a dangerous world, where there are often gaps between promises and performance, such secrecy fosters suspicion. There may be no obvious way to rule out deceit or bad faith, and there are often incentives to lie or misrepresent.112 That is exactly how Shakespeare begins one of his political histories, with the character of Rumour proudly saying that he stuffs “the ears of men with false reports; I speak of peace, while covert

107

RELIABLE BARGAINS 113

enmity under the smile of safety wounds the world.” In such a world, states seeking security may fear the worst and react accordingly. The dangers of defection—real and perceived—are compounded if communications are slow or surveillance is poor. In either case, a breach could go unseen and unpunished for some time, with mounting (but still hidden) costs to the innocent party. The possibility of such hidden breaches raises the risks of making agreements in the first place. Thus, if we have correctly specified the mechanism underlying the democratic peace—a nexus of reliable commitments that democracies are especially well suited to project and exchange— then we ought to find stronger, more stable agreements among democracies than among nondemocracies. Of course, even within well-ordered constitutional democracies, not all issues are equally transparent. In some areas, policymaking is secretive and crucial data classified. It may be impossible to tell what decisions have been made until the policies themselves have been carried out, and perhaps not even then. Democracies openly debate the big issues of national strategy, force structures, and military budgets, but the operational details are often hidden from public view. This veil of secrecy is readily understandable. It obviously covers tactical military operations, but it sometimes extends well beyond that. During the Cold War, for example, the size and scale of U.S. intelligence agencies were highly classified secrets. Such secrecy is most common in national security issues, but it also affects some economic policymaking. Central banks are the most important example. Their decisionmaking is secret in all countries, even the most democratic. Key monetary policies, notably those affecting the money supply, interest rates, and currency values, are decided by small groups in closed rooms, well insulated from elective institutions. That is exactly how the United States ended its commitment to the dollar-gold exchange standard, the core of the Bretton Woods monetary agreement. Any advance hint the United States would change policies would have launched a chaotic run on gold reserves.114 America’s European partners anticipated it would “close the gold window,” but Washington could not actually tell them without undermining the policy itself. There may be good political and technical reasons for such secrecy, but it does affect the possibilities of international cooperation. It is hard for partners to cooperate when they cannot show each other—in timely, reliable ways— that they are committed to carrying out their bargains. The transparency of democracies gives them a generic advantage in projecting promises, but even the most open states have some secrets, and good reasons for keeping them. Issues with the least transparency are bound to pose special obstacles to cooperation. In monetary policy, for instance, major states have adapted by holding frequent international meetings and maintaining close contact among senior officials. That helps build a sense of trust, familiarity, and strong working relationships, despite the need for secrecy in some areas of foreign-exchange policy.

108

CHAPTER 4

Durable Agreements, in Spite of the Obstacles Regular, informal meetings among principals are one way of dealing with information problems in international relations, although that hardly eliminates incentives to conceal or mislead. Officials, confronted with incomplete information and the possibility of deception, must make hard choices. One is whether to bring in a neutral third party to collect data or verify disclosures. The most common strategy is to avoid that and increase transparency among the principals themselves. For example, they can exchange information directly, such as lists of deployed weapons. They can permit each other to conduct inspections to verify the data that has been exchanged as well as compliance with the agreement itself. Alternatively, they can use outsiders to verify performance. The International Atomic Energy Agency (IAEA) checks compliance with the Nuclear Non-proliferation Treaty. Likewise, the World Trade Organization agreed to monitor the 1996 U.S.-Japan agreement on semiconductors. Generally speaking, states are most willing to use third parties when the latter are technically competent, when their monitoring would facilitate agreements, and when mistakes carry only modest risks. When the stakes are very high, such as national survival, they are loath to use third parties. If an undetected violation could have devastating consequences, then vulnerable countries prefer to handle their own monitoring. The dangers are simply too great if the third party is careless or incompetent. The agent may shirk. The principal will not. The other basic choice is how best to contract around situations where more intensive inspections do not significantly improve information. The choice here is between proxy agreements, which we have already discussed, and limited or circumscribed agreements, which are restricted to a few readily measured variables. The idea behind circumscribed agreements is to economize on search costs and increase confidence levels by limiting their bargains to a few easily verified issues.115 The United States and Soviet Union, for instance, concentrated their arms control agreements on missile launchers, which were relatively easy to count. Their real concern was with nuclear warheads, not launchers, but it was impossible to count warheads without intrusive on-site inspections. Launchers were a reasonable proxy that could be monitored effectively by satellites. Yet another way to economize on information gathering is to ban certain activities altogether, rather than to regulate them. To preserve the world’s endangered elephants, for instance, all trading in ivory is prohibited. Although the prohibition is aimed squarely at poachers, it also harms some impoverished African states that have worked diligently to restore their elephant herds. The costs of preservation are high, the countries are poor, and the benefits are reaped by the world at large. Their conservation efforts would doubtless be encouraged

RELIABLE BARGAINS

109

by permitting them, but not poachers, to sell tusks. Unfortunately, the costs of monitoring such a complex arrangement are prohibitive. The global treaty to ban land mines works the same way, with no room for exceptions. While these information problems affect all countries, constitutional democracies cope with them better because of their transparency. What is more, they have found it relatively easy to accommodate each other’s requests for still more information. In security affairs, at least, they rarely abandon their right to verify compliance, even with trusted allies. They take the opposite tack, one that does not depend on blind trust. They agree without much strain to much higher levels of transparency. They agree to live in glass houses to prove they will not throw stones. Their negotiated solution reassures partners and minimizes the dangers of mistakes, misinterpretations, or outright cheating. It is a prudent middle way for partners to demonstrate once again that they are reliable. Doing so allows them capture gains and contain risks in a dangerous world. Not surprisingly, democratic states have led the global movement toward more transparent agreements. They have consistently supported on-site monitors, surprise inspections, and third-party verification. Their staunchest opponents have been dictatorships, initially led by the Soviet Union and later by military regimes in Africa, the Persian Gulf, and Asia. This pattern is clear in U.S.-Soviet arms control negotiation, where verification was one of the most contentious elements. Although both sides were wary of cheating, they were unable to resolve it for decades in the most obvious way, by effective on-site inspections. Moscow simply would not permit them. “Whatever the Russians may have said,” one of President Eisenhower’s senior advisers concluded, “they will not agree to what we consider adequate inspection.”116 They did not, and it blocked agreements for years. In the mid-1950s, Khrushchev rejected Eisenhower’s proposal for “open skies,” which would have allowed overflights to monitor each other’s military posture.117 Several years later, the Soviets rejected on-site visits to confirm foreign seismic data on nuclear tests. That “turned out to be the main—and ultimately insurmountable—obstacle to agreement on a comprehensive testban treaty,” according to Per Fredrik Ilsaas Pharo.118 Nuclear arms agreements were finally possible in the 1970s only because satellites could verify them without intruding on Soviet territory. Even then, the USSR was so sensitive it refused to mention satellite verification in the treaties themselves (obliquely referring to “national technical means”).119 When did the Soviet position finally change? When they liberalized their own political system under Gorbachev, permitting more transparency at home. As the Soviets embarked on perestroika and glasnost, they dropped objections to more intensive verification in international agreements.120 The final superpower treaties allowed inspectors to visit each other’s facilities and closely monitor compliance.121 Since then, verification has proved a much easier issue for Russia (which is still not a full-fledged democracy but is far less authoritar-

110

CHAPTER 4

ian than the Soviet Union). Their nuclear agreements with the United States have gone beyond old provisions for on-site inspections. These new, more intrusive inspection arrangements were approved without the quarrels and wariness that characterized superpower talks for decades. Russian democratization, though still incomplete, has made transparency easier to negotiate. That, in turn, has made agreements more feasible and more secure. Although democracies have been relatively successful in forging cooperation, they have faced some special problems of their own. The most important, which we touched on earlier, is the sheer complexity of the policymaking process and the large number of actors who participate. That may mean senior negotiators, even if they have full credentials, cannot authoritatively commit their governments to the bargains they have negotiated. Even the president cannot commit the United States if the agreement requires Senate ratification or congressional approval. As a result, it is sometimes unclear to partners whether democratic leaders can actually pledge their states. The key point is that in constitutional democracies, where popular sovereignty is coupled with established governing procedures, making credible external promises requires genuine internal support and state approval. A leader’s word may not be enough to pledge the country, especially in serious undertakings. Woodrow Wilson signed the Treaty of Versailles; Congress failed to ratify it. The Truman administration negotiated an International Trade Organization; Congress failed to give it much support, and the administration pulled the bill before it lost a floor vote.122 Jimmy Carter signed SALT II. He had to withdraw it after the Soviets invaded Afghanistan, and the Senate was sure to defeat it. Although presidents and senior diplomats may sign treaties, their signatures alone do not constitute effective state promises. The procedural rules of democracies are genuine constraints. Partners may find the rules irksome at times, but they also understand that treaties must be approved in contested political environments. Other democracies understand these limitations full well. Ratification procedures convert tentative promises into full-fledged agreements. Stamped by the imprimatur of state authority, interstate agreements are then backed by the state’s reputation abroad and its judicial institutions at home. Of course, not all democratic promises require this kind of ratification. Lesselaborate arrangements are needed for executive agreements, memorandums of understanding, or interagency bargains. Some may require simple enabling laws, others may require nothing more than executive orders. In fact, there is often ambiguity about which types of agreements require full-scale ratification. A dispute like this arose over the Panama Canal.123 Could the Carter administration relinquish U.S. control without a treaty? The Carter administration probed this question, legally and politically, but found it could not. Aside from any legal requirements, leaders have to make political choices about how they wish to encode their agreements. If speed or secrecy are required, informal agreements are best. That might mean an exchange of letters

RELIABLE BARGAINS

111

or a joint communique´ . At the other end of the spectrum are formal procedures for treaty ratification. Cumbersome as they are, they also carry more weight politically and are more stable through time. There is a trade-off, then, between simplicity of approval and gravity of commitment, between informal agreements and formal treaties. In effect, democracies have a wider range of meaningful procedural choices. They can use them to calibrate commitments. Because approval of serious agreements is not automatic, partners face some risks if they rely on pledges by individual democratic leaders or other kinds of informal agreements. In a domestic political system where power is fragmented, diplomats and political leaders may be promising more than they can deliver. That is not only why wider approval is sometimes required; that is why it is so valuable. It is exactly these established procedures for formal approval that make commitments visible, believable, and hard to reverse. John Jay, writing about the new American constitution, appreciated this point and considered it a major virtue. In the Federalist Papers, number 64, he argues that opponents of the constitution “would do well to reflect that a treaty is only another name for a bargain, and that it would be impossible to find a nation who would make any bargain with us, which should be binding on them absolutely, but on us only so long and so far as we may think proper to be bound by it.”124 Jay’s point is simple but profound. He went on to argue that treaties were serious undertakings that should not unilaterally renounced, unless that was permitted by the treaty itself. Jay’s basic insight still holds: stable bargains require strong, reciprocal commitments that cannot be cast aside easily or unilaterally. Democracies have tremendous advantages on exactly this point. Democracies do not make the agreement process easy. They make it hard— deliberately hard. To make their most serious commitments, democracies set the hurdle high. Agreements must pass through open, contested political processes in states where power is not unified in the head of government. Difficult as these political hurdles are, they give great stability to democratic promises. After all, if a leader’s word alone cannot pledge a state to some actions, neither can it easily break existing pledges. The elaborate techniques of democratic political approval make their self-binding more credible and more enduring. The result, as Stephen Holmes points out, is to hold open “possibilities that would otherwise lie beyond reach.”125

5 Leadership Succession as a Cause of War: The Structural Advantage of Democracies

THE LAST CHAPTER dealt with all four sources of the democratic contracting advantage: transparency, audience costs, constitutionalism, and continuity of regimes. This chapter continues the examination of regime continuity, focusing on leadership succession. Succession crises are a frequent source of wars. Democracies are systematically better at averting them and, hence, at averting the wars that stem from them. Because democracies have solved the succession problem internally, they avoid its most dangerous consequences externally. My main point, however, is not simply that succession crises produce wars. They obviously do. My aim is to show how they block the path to peace, mainly by creating vast uncertainties and obstructing long-term bargains that could prevent war.

Turbulent Transition versus Peaceful Succession One of the great achievements of democracy is its peaceful solution to the ageold problem of leadership transition. That is an extraordinary feat, yet it is so ordinary today that its historical rarity goes unnoticed. In other forms of government, succession is frequently contested, sometimes violently. It is a recurrent source of war. With the great prize of state power at stake, potential leaders often seek outside help to win it. Foreign powers have incentives to extend that help because, if they succeed, they will gain an ally and possibly some influence over its aims and policies. Or they might have defensive motives. They may fear that if they hesitate, others will act first, leaving them with a new enemy rather than a friend. At the very least, they hope their intervention, whatever its form, whatever its motives, will lead to warm relations, greater commerce, and perhaps military partnership. Often they want even more, a truly like-minded ruler who will forge deep and stable bonds between countries. What they mean by “like-minded” depends on the dominant sources of international (or regional) political conflict and coalition. Where religion is the major source of political cleavage, foreign powers may back a fellow Catholic or Lutheran, a Shiite or Sunni. Where ethnicity is the central dividing line, they want a Hutu or Tutsi, Serb or Croat, Slav or Teuton. Where dynastic lineage matters most, they want a fellow Habsburg or Bourbon. As late as

L E A D E R S H I P S U C C E S S I O N A N D WA R

113

1870, the French feared that the Spanish crown might go to a Hohenzollern, giving Prussia a new dynastic ally and France a military problem on two fronts instead of one. During the Cold War, when political orientation and economic ideology were intertwined, the superpowers wanted fellow communists or capitalists. These cleavages are part of the social construction of security and danger, of friend and foe. They have powerful self-reinforcing elements, since they are pivotal in choosing allies and assessing adversaries, as well as deciding when and how to intervene in foreign leadership struggles. Every political system must confront problems of succession, but established constitutional democracies have deep-seated advantages in solving them peacefully, without civil strife or international conflict. They select new leaders regularly by fair and open elections, using established rules and evenhanded procedures. The winners take power smoothly and are considered legitimate, even if their victory is narrow. Different democracies have their own rules about who can vote, how candidates are elected, and how election results are cumulated to produce national leaders. These rules have evolved considerably over the past century, notably with the gradual expansion of voting rights. Women, ethnic and racial minorities, and poor people have all been added to the electorate. Yet one basic feature has remained constant over time and across states: constitutional democracies determine political succession by regular, fair elections. Because the process is routine and legitimate, disruptive foreign intervention is unlikely. Losing candidates in fair elections do not seek it. Other democracies do not extend it. The succession process does not produce turbulence or confusion for outsiders to exploit. Even opportunistic rivals are unlikely to be tempted, since the voting country is not weakened by the election itself or the transition of power.1 There are only two exceptions to these placid transitions. One is if losers think the election was stolen, which sometimes happens in new or incomplete democracies (and constantly happens in sham democracies, by definition). The other is if they think the winner will abolish democracy itself after taking power. Both violate the basic terms of constitutional democracy and may be contested before a (potentially) corrupt winner can consolidate power. Free, fair elections are rightly considered the fundament of democratic life. To insure peaceful, orderly succession, the rules may be broadened to cover contingencies of death, illness, and incapacity. In parliamentary systems, when leaders die in office, a new leader is chosen the same way as the old one, by vote of the parliament’s elected members. That typically prompts a new election in the near future, although it need not. In presidential systems, the line of succession is a hierarchical list of government positions, usually beginning with the vice president.2 The United States also has laws to choose interim leaders when the president is seriously ill, and independent ways to determine if he is.3

114

CHAPTER 5

Non-democratic states, on the other hand, have many more problems with leadership transitions—problems that often lead to war. It is easy enough to understand why. Succession crises produce a fluid, confusing political environment. Aggressive neighbors may pounce, sensing that an old adversary is temporarily vulnerable. Or neighbors may fear that they themselves will soon be innocent victims of attack, that new foreign leaders will be more hostile and belligerent, ready to launch a war of aggression against them. They may choose to preempt instead. Stephen Walt makes similar arguments about the consequences of revolutionary transitions.4 Revolutions are especially disruptive and disorderly, but some of the same issues arise, in paler colors, whenever there is a turnover of monarchs, autocrats, or dictators. The flux and confusion of leadership struggles increase the likelihood of misperception on all sides. Even prudent neighbors may have exaggerated fears and lack enough solid evidence to reject them. They may fear—understandably enough—that others want to capture valuable real estate, establish more secure, defensible borders, or reclaim lost territory (depending on whether they are greedy, defensive, or revanchist).5 Their neighbors may well fear exactly the same thing, that others will choose this tempting moment to achieve important national objectives, either offensive or defensive. The uncertainties of succession can thus translate into diplomatic tension, reciprocal hostility, and mutual fears of preemptive war, perhaps based on misperceptions, perhaps not. Whatever the potential sources of conflict, they need not lead to fighting. The differences could always be resolved by agreement short of war or frozen in place by a deterrent stand-off. If states knew in advance how wars would turn out, then rational states would want to choose that outcome in advance and avoid the terrible costs of war itself. Unfortunately, antebellum agreements are hard to reach for two reasons, as we have already discussed.6 First, even if we assume that states are purely rational and unitary actors, they still must discover a bargain they mutually prefer to the gamble of war. Second, the chosen bargain must be stable and enforceable, or at least the parties must think so. These are always difficult hurdles.7 They are far higher during uncertain transitions. Information problems are serious because states know much more about their own capabilities and resolve than others do, and they often have strong incentives to keep that knowledge to themselves. Even if they wish to share it in order to reach agreements, they may not be able to do so convincingly. These problems are surely much harder when • the rival’s political environment is turbulent and changing rapidly, and • its capabilities and preferences are opaque. When the two combine, as they do when closed political systems face leadership struggles, then information problems become severe. Combined with

L E A D E R S H I P S U C C E S S I O N A N D WA R

115

problems of making reliable, long-term commitments, they are major obstacles to pre-war settlements. In case of truly large-scale social change such as revolutions, it may be hard to divine even the most basic features of the emergent regime. Take its military capabilities. Can the new regime adequately fund its armed forces, maintain its equipment, recruit qualified soldiers, renew its officer corps, and sustain morale? Similar doubts plague any political assessment. Will the new leaders’ rhetoric change after they take power? Will their major policy goals, foreign and domestic, change dramatically from those of previous rulers or, indeed, from what they said themselves before they took power? What about their attitudes toward foreign investment and international trade, which have a significant impact on other countries?8 The larger these political changes are and the faster they occur, the more difficult it is to make accurate, confident predictions. To compound the difficulties, it may be necessary to make these judgments swiftly, before a rival in transition has consolidated its power and becomes a more formidable threat. A country swept up in such leadership changes often faces its own severe problems in communicating effectively. Entangled in internal struggles, it may pay too little attention to foreign affairs and underestimate just how threatening it appears to others. It may have no time or inclination to send costly, credible signals to assuage others’ fears. One obvious reason is that such signals could easily be interpreted as weakness, proclaiming a country’s vulnerability and inviting others to demand concessions or launch an attack. As a result, a state in turmoil may well face stiff foreign challenges before it fully grasps the problems confronting it or the gravity of the threats it presents to others. The opacity of nondemocratic states adds markedly to this overall assessment problem. Even if rival states could overcome these informational problems, even if they could locate mutually profitable agreements, they would still have to make them stick. That is often difficult for nondemocratic regimes, as we have seen, but it is especially problematic during rocky transitions. That is when doubts are greatest that they can bind themselves for the long haul. Their preferences, beliefs, and capabilities are most confusing and least certain to others. It is hard to judge what kinds of the internal constraints they face. It may be hard to tell whether they will remain in power and, if they do, which contending factions will rise to the top. The Jacobins, Girondins, or Montagnards? Bolsheviks, Mensheviks, liberals, or reactionaries? Such moments of leadership transition offer insights into the contracting advantages of democracies and the ways they inhibit international conflict. Well-established democratic institutions achieve such transitions smoothly, without civil discord or international incidents. Nondemocracies sometimes do as well, but they fail more often than established democracies do. If the path of succession is murky, then nondemocracies can easily sink into life-or-

116

CHAPTER 5

death political struggles involving elites or wider social groupings, often drawing in foreign powers. To explore these issues, I will focus first on the problems of succession within fledgling democracies. After that, I will consider succession in several different kinds of nondemocratic regimes. The most common are hereditary monarchies. I will also consider others, including “strongman dictatorships” and powerful ruling-party systems, such as the People’s Republic of China and the Soviet Union. In examining these varied nondemocratic forms, I want to emphasize three important dimensions of governance and succession. The first is what kinds of situations make nondemocratic transitions harder or easier. Which leadership changes pose the greatest problems? How are they related to the regime’s chief characteristics? The second is the degree to which state institutions limit the new ruler’s powers. In particular, are there strong constraints binding new rulers to old agreements? Third, I want to consider systematic differences in state transparency—the degree to which governments can view each other’s transitions, policy processes, and choices clearly and promptly. These three dimensions are important because they bear directly on trust, reassurance, and peaceful cooperation. Democracies are well structured to cope with all three. Nondemocracies are not, although some manage better than others.

The Perils of Fledgling Democracies To contest divisive issues passionately but peacefully is a daunting achievement for any political system. So much is at stake: the nation’s security, individual fortunes, deeply held moral values, and perhaps personal safety and political freedom. To do battle over major issues time after time without renting the social fabric or jailing the losers is one of the great virtues of established constitutional democracy. Unfortunately, not all democracies are so stable or so successful. Not all of them can tolerate a loyal opposition, one that argues freely and organizes voters. Likewise, not all of them are certain their leaders will give way to successors in a peaceful, legal, and democratic way, especially if those successors are political opponents rather than hand-picked prote´ ge´ s. Many newer democracies have not yet solved these problems. They have not developed all the components of electoral government or made the essential social accommodations, such as political toleration and acceptance of multiple, independent sources of power and wealth. One of the key components of constitutional democratic government is rotation in office by regular, free elections. The smoothness of this rotation is an important marker for how effective the democratic process actually is. The first election in a newly established democracy is crucial, but it is only a first

L E A D E R S H I P S U C C E S S I O N A N D WA R

117

step. The most important election may actually be the second or third, particularly if it portends regular changes between opposing parties. A democracy is generally established after power has changed peacefully at least twice. It is clearest to partners, however, when all major parties contending for power have actually ceded their governing positions voluntarily. It is all too easy for the first elected leader, entrusted with considerable power and often revered, to subvert democratic procedures before they become entrenched. One of George Washington’s greatest achievements was relinquishing his presidency voluntarily after two terms, proving he was no King George. Democratic chief executives were a novel concept at the time, and Washington’s departure showed that they did not stay in power until death. As with so much of Washington’s life, it was a lesson written in deeds. Just as that lesson mattered to the young American Republic, it matters even now to young, shaky democracies. Today’s new democracies are vulnerable to coups, not just by disgruntled colonels but by their own elected leaders. Elected leaders are frequently dissatisfied with their limited powers and irate about leaving office. They may fear (reasonably or not) that their successors will overthrow democracy and perhaps kill them in the process. Launching a coup is one way to sweep aside all these restrictions and fears. It is worth remembering that Hitler initially won office at the ballot box. His Nazi party won a plurality of the German vote in July 1932, amassing the largest vote total in German history and forcing the aged president, General Paul von Hindenburg, to appoint him chancellor.9 Over the next year, Hitler managed to seize total power, but his seizure came after his electoral success, not before. Alan Bullock argues that Hitler’s electoral strategy was a lesson learned from the disastrous Beer Hall Putsch of 1923.10 After this early coup attempt failed, Hitler worked assiduously to gain office by constitutional means. It was a thoroughly cynical strategy, but it reassured traditional German elites and the military while still appealing to disaffected workers, shopkeepers, and youth.11 Once in office, however, Hitler quickly grabbed control of state ministries and abolished all elements of democratic life, from free speech to free elections, along with all traces of Weimar liberalism and constitutional protections.12 Hitler’s example is a grim one, but it suggests a more general problem that lives on: democracies can be cut down by coup, and the perpetrators may come from leaders of the elected government as well as outsiders. Such coups have plagued unsteady new democracies throughout the modern era. Napole´ on III, whose adventures in the Ottoman Empire did so much to destroy the Concert of Europe, was elected president before he crowned himself emperor.13 He took elective office in the portentous year of 1848, after a spreading rebellion destroyed the Orleanist monarchy. Louis Napole´ on’s name was put forward by the new Party of Order, made up mostly of royalists and Catholics, who considered him popular but pliable. As the nephew of Napole´ on

118

CHAPTER 5

Bonaparte and the family’s leader, his name carried weight throughout France. He won the election with wide support. Once in office, his democratic trappings fell away. He moved quickly to circumvent his own party, stocking the army and administration with his personal supporters. Within months, he had chosen a cabinet beholden solely to him, not to his party or the National Assembly. He also began to lay the groundwork for a coup, advertising himself as a strong leader who could solve France’s economic problems and restore her former glory.14 If he was to remain in power, he would have to seize it because the constitution expressly forbid the reelection of presidents. To revise this prohibition legally, he would need a supermajority in the Assembly, which he lacked. Facing the certain loss of elected office, Louis Napole´ on carried out his coup after three years as president. He immediately declared a new constitution, more congenial to his needs. The following year, he held a plebiscite to ratify the coup and then another one to crown himself emperor. Over the next two decades, he ruled an increasingly prosperous France as a moderate, if somewhat erratic, authoritarian. He was overthrown in 1870 when Prussia handily defeated the French army at Sedan and marched on to Paris.15 The dangers of coups d’e´ tat by elected officials are not merely historical curiosities. They still hang over many vulnerable new democracies in Latin America, Africa, and Asia. Alberto Fujimori was elected president of Peru in 1990; two years later he abolished the legislature and suspended legal freedoms. In neighboring Ecuador, the elected president, Jamil Mahuad, was unable to cope with economic crisis and political protests. The army stepped in, overthrew him in January 2000, installed the former vice president in his place, and then stepped back out. The list of such examples is a long one. Although many developing countries have struggled successfully to stabilize their democracies, others have held abortive elections, moved forward (or backward) by fits and starts, and sometimes abandoned the process entirely. Haiti is yet another example of these still unresolved social-political struggles. It remains a fragile polity, somewhere between a democracy and a dictatorship. Desperately poor, it has major problems of political representation, free debate, safety for regime opponents, and leadership succession. Despite considerable outside support, it faces a troubled, uncertain future. In 1990, Jean-Bertrand Aristide won the presidency in Haiti’s first democratic election. It was hotly contested, often violent and corrupt, but no one could doubt that Aristide had overwhelming popular support, especially among the urban poor. He had equally strong opposition from Haiti’s tiny elite and their allies in the government and military. They were not “loyal opponents” in any democratic sense. They were his deadly enemies, and vice versa. Aristide’s promise of fundamental social and economic changes was a threat to these long-entrenched oligarchs, and he was swiftly ousted by their institutional defender, the military.

L E A D E R S H I P S U C C E S S I O N A N D WA R

119

Eventually, the United States managed to reinstall Aristide at gunpoint, but not before extracting a public promise that he would leave after his first term. The United States wanted that promise for good reasons. Aristide may have been popularly elected, but his own commitment to democracy was questionable. Death squads did not disappear after he took office, and some of his political opponents disappeared forever. Aristide’s promise to leave office was given reluctantly after months of U.S. pressure. It could not be relied upon. Holding him to it required considerable diplomatic effort, wielded by a extremely powerful neighbor. Many of the same problems arose under his successor, Rene´ Preval. Aristide’s subsequent return to office, surrounded by many of the trappings of one-man rule, was less the triumph of democratic rotation in office and more a sign that Haiti was again falling into dictatorship. In Africa, too, there are numerous examples of young or putative democracies facing similar succession problems and uncertain futures. Zambia is a case in point. The man who led the country to independence from Britain, Kenneth Kaunda, became its dictator in the 1970s and 1980s. When he finally allowed a election, he was crushingly defeated by Frederick Chiluba and his Movement for Multi-Party Democracy. The party’s name was unintentionally ironic. Chiluba had absolutely no interest in multi-party democracy, as he quickly proved once in office. One of his first laws barred Kaunda from ever running for office again (on the extraordinary grounds that Kaunda’s parents had been born outside Zambia, which was then under colonial rule).16 Farther south, in Zimbabwe, President Robert Mugabe destroyed a young democracy to maintain himself in power. The reason was familiar. Political opponents had surfaced to oppose his corrupt, violent, and incompetent rule. The problems of nascent democracies are not limited to Africa, South America, and the Caribbean. In 1999, Pakistan’s elected prime minister, Nawaz Sharif, was ousted and imprisoned hours after he fired the head of the army. In the Philippines, Indonesia, and many other developing countries, there has been fierce street-fighting over democratic succession. Their future as democracies, even as marginal democracies, is anything but certain. There are also question marks in Central and Eastern Europe. For the past two centuries, this area between Russia and Germany has been the battleground of Europe’s largest wars. It still holds considerable strategic importance to the United States, Russia, and Western Europe. Following the collapse of the Soviet Union and its alliance system, the region’s formerly communist countries began the difficult process of developing democratic institutions and market economies. Some, including Poland, Hungary, and the Czech Republic, have been relatively successful. Others have not. The importance of these democratizing efforts is not lost on the major states of the North Atlantic. They have consistently offered tangible support and rhetorical backing to democracy and economic transformation.

120

CHAPTER 5

These outside powers have had a very positive impact on democratization in the region.17 They have pushed for honest elections, monitored by neutral observers. They have supported the development of other key democratic institutions, from a free press to an independent judiciary. They have backed market reforms, designed to encourage economic growth and decentralization of power. To signal this support and enhance the chances of success, western states have not only favored economic ties—aid, trade, and investment—they have provided political and military assistance. The institutional cornerstones are NATO and the European Union, both of which have expanded into Central Europe. For NATO’s key members, the military advantages of enlargement are clearly secondary, if they exist at all. Its real aim is to provide strong, visible support to the new democracies of Central Europe, just as NATO’s original aim was to bind nascent German democracy to the West, while deterring Moscow’s military threat. The hope now is to reward and aid democratic reforms while drawing NATO’s new members firmly into the Western orbit. NATO’s expansion is likely to achieve these important political goals. The hard question is whether it can do so without simultaneously threatening Moscow, fueling nationalism and xenophobia, and undermining Russia’s own tenuous steps toward democracy. Can the West strengthen the new democracies of Central Europe without simultaneously threatening Russia? Expansion by the European Union complements NATO’s security efforts. EU expansion is designed to promote economic growth, ensure a coherent framework for micro- and macroeconomic policy, and enlarge Europe’s internal market. The International Monetary Fund, World Bank, and World Trade Organization reinforce these efforts at economic transformation in Central Europe. The message is clear. Major outside powers are willing to use their money, power, and control over international institutions to support constitutional democracy in a region where it has many followers but shallow roots.18 Earlier democracies faced an even more uphill road, since they had almost no outside help and few models to follow. The newly formed United States had none at all, except for French military aid, when it fought its war of independence and then sought to establish democratic rule.19 America’s democratic foundation was firmly established by a written constitution and gifted early leaders. These revolutionary leaders were keenly aware of the hard tasks they confronted and the difficulties of sustaining popular government. Indeed, America’s first constitution, the Articles of Confederation, was so riddled with problems it had to be replaced. The new constitution had more centralized powers, especially in defense and foreign affairs. Still, this form of government was entirely new—there had never been a democracy larger than a city-state— and no one knew if it would survive or prosper. The uncertainties of the moment were nicely captured at the end of the Constitutional Convention, when

L E A D E R S H I P S U C C E S S I O N A N D WA R

121

a woman approached Benjamin Franklin and asked, “Well, Doctor, what have we got, a republic or a monarchy?” “A republic,” Franklin replied, “if you can keep it.”20 Keep it they did, largely because they worked out practical arrangements for sharing power between the federal states and the central government and between the president and a strong legislature, along the lines envisioned by the Constitution. Equally important, senior officials took office and left it the same way, by regular elections.21 After Washington’s retirement, all national elections were vigorously contested, but not the constitutional process of succession in either the executive or legislature. George Washington refused to become a monarch, in name or in practice. The second president, John Adams, gave way without protest to his elected successor, Thomas Jefferson, even though the two had been intense rivals for years. The night before Jefferson’s inauguration, Adams quietly left the White House and returned to his farm in Massachusetts.22 The pattern was set. There is no surer mark of stable democracy than this peaceful alternation between deep-seated opponents. In modern Israel, for example, the Likud and Labor parties have battled over fundamental issues, exchanging bitter public invectives, but they have also alternated regularly in office, with no question that the loser would make way for the winner.23 In Algeria, by contrast, the military staged a bloody coup in 1992 when Islamic fundamentalists won office. The Algerian military and their supporters in France not only resisted the winners’ policies, they charged (with good reason) that the fundamentalists were part of an armed insurgency, dogmatically opposed to secularism and open discourse. There was little hope they would sustain democracy if they ever took power. That was not just the military’s self-serving view. It was also voiced by a small but articulate group of Algerian intellectuals. They noted that fundamentalists had violently attacked independent newspapers, repeatedly threatened to kill journalists, and actually murdered dozens of reporters and editors. “Those who fight us with the word,” declared the Islamic Salvation Front, “shall die by the sword.”24 These are the illiberal ideals of true believers. They are reflected in wide-ranging attacks on behavior and cultural expression and translated into political practice by a guerrilla army deeply opposed to secularism, toleration, and the other tenets of liberal, constitutionally ordered democracy. Once in office, by arms or ballot, they would not permit their views to be contested freely or overturned by vote.25 On the spectrum between Algeria and Israel lie a wide range of intermediate cases, where one party dominates, where voting is tainted by some corruption, and where elected officials are not always the real political powers. There are usually some constraints on opposing parties, a tilted field of electoral competition, and some uncertainty about whether the opposition party could actually take power if it ever won. Whatever one calls these governments, they fall well short of constitutional democracy.

122

CHAPTER 5

These limitations on electoral competition underscore one of the great achievements of genuine democracy: the uneventful transition of political leadership, using fair, regular electoral procedures. Indeed, the resolution of succession problems by free, fair elections is a large part of what we mean by modern democracy. (The other part is that elected leaders actually wield power over vital political issues—lawmaking, national security, taxes, expenditures, police, and the like—and they wield their power by established legal means.) This stable, orderly process of democratic succession has an important impact on the initiation of wars. First, orderly transitions lower the dangers of confusion, uncertainty, and misperception abroad. These informational advantages complement the overall transparency of democracies, where policy choices are open to considerable public scrutiny. This openness is an integral part of the democratic process. Numerous elected officials have a legitimate role in policy formation. They need information to make their decisions, and they communicate it to build support for the choices they make. Second, electoral rotation in office is part of the regime’s overall continuity. It gives confidence to partners that they are sealing solid bargains with continuous regimes, not ephemeral ones with today’s ruler. Together with constitutional restraints on leaders’ autonomy, this continuity of political regimes gives stable democracies a unique advantage. They have the capacity to make credible long-term promises. Most regimes, including fragile democracies, simply cannot. They may not survive without major changes in the form of government, and even if they do, their new leaders may not be prevented from overturning old policies. This uncertainty about the present and future helps explain why new democracies are more war-prone than established ones.26 If the democratic peace is ultimately explicable because of durable, reassuring agreements, then shaky new regimes will have far more difficulty achieving them, whether they are electoral democracies or not. Established constitutional democracies have more ballast. They have the institutional capacity—and thus the political opportunity—to project commitments further into the future. Since other democracies can reciprocate, their structural advantages establish the contractual foundations for peace among democracies. Since nondemocracies cannot reciprocate fully (because their regimes lack continuity, transparency, and domestic institutions for self-binding), the relations between democracies and nondemocracies will be more contentious, as will relations among nondemocracies themselves. The best way to see the distinctiveness of constitutional democracies is to contrast them with alternative political systems where (1) the process of succession itself is contested and (2) leaders have greater autonomy in policymaking. Strong monarchies are a good place to begin because they have both characteristics and have been central to international politics since the rise of nation-states in the seventeenth century.

L E A D E R S H I P S U C C E S S I O N A N D WA R

123

Dynastic Succession and War Royal rulers inherit their power and with it their political legitimacy. Although the exact terms vary across cultures, dynastic rule is probably the most common form of government. Until relatively recently, the diplomatic history of nations could be written along dynastic lines. As late as the eighteenth century, it was nearly universal in Europe. Except for Britain (after 1689) and the Dutch Republic, all the major powers were effectively ruled by hereditary monarchs, born of long royal lines.27 On the eve of World War I, most Great Powers were still governed that way. The Hohenzollerns of Germany, Habsburgs of Austria, and Romanovs of Russia were all strong dynasties that governed with few constitutional fetters. The rules of succession differed a bit from country to country. In some, females could inherit the throne; in others, they could not. In some, only Catholics could rule; in others, only Protestants. But in each country, rules of succession were reasonably well articulated, widely understood, and accepted in both theory and practice. Most had been established over centuries of use. Even so, eighteenth-century Europe was repeatedly convulsed by wars of dynastic succession. The most powerful states in the world were entangled in the War of Spanish Succession, the War of Polish Succession, the War of Austrian Succession, and others.28 All were major wars fought for high stakes. In a world where kings actually ruled, and where states were allied mainly by blood and marriage, royal succession truly mattered in international politics. A friendly monarch in a neighboring country could mean a strategic alliance, often sealed by marriage. By the same token, an unfriendly monarch could mean a warlike foe, or at least a persistent problem. Dynastic ties did not always produce these results, but they often did, and they always mattered a great deal. In a world where political authority was personal and familial, kinship was the most useful bond available.29 As a result, the Great Powers all strived to see their favorites, and especially their close relatives, succeed to neighboring thrones. Time and again, France, Spain, Austria, Russia, and England connived to achieve this goal across the continent. Each also suffered its own succession crises, typically violent ones that led to interstate war. If the rules of inheritance were reasonably clear, how did succession crises arise? With a dull similarity: a ruler died without a direct heir, usually a son. Strong rival claims could then be made for brothers, daughters, nephews, or cousins—or, more likely, for each of them by different supporters at home and abroad. The specific issues differed from case to case and were not repeated often enough for a country’s laws and customs to evolve clear answers to idiosyncratic cases. That is, the established rules of succession were incomplete. They did not determine a unique, legitimate heir in all cases. This ambi-

124

CHAPTER 5

guity carried a heavy price. Since Europe’s royal families were intermarried, each prospective heir was sure to be closely linked (by blood, marriage, and religion) to some foreign power, which naturally backed that heir. This kind of dispute, which began whenever a clear line of inheritance ran out, might be called an interdynastic dispute, since it pitted new family lines against each other, each hoping to begin its own royal house, often with foreign help. On the other hand, when a direct heir to the throne was available (usually a male, but not always), we can speak of intradynastic succession, or inheritance within an existing dynastic line.30 Succession within a dynasty was generally well specified in law and practice, or, rather, by each country’s own laws and practices. By the late Middle Ages, when government authority was becoming solidified in Western Europe, direct royal inheritance was rarely contested. Even when the new king was a small child, which required a regent to rule in his name, direct inheritance seldom led to civil wars or engulfed other kingdoms in succession quarrels, although the new leader’s policies might well provoke conflict later.31 Interdynastic rivalries, on the other hand, could not be settled so easily. They frequently led to major conflicts. Often, the problem could be foreseen years in advance but still not settled. Either an aged ruler had no surviving male heirs or the ones he had were sickly and unlikely to survive, often because of inbreeding. Because succession problems like these were so disruptive, one of the ruler’s principal tasks was to produce a healthy “heir and a spare,” and more if possible.32 Spares were useful because life spans were short. Younger sons frequently inherited the title. To understand how interdynastic quarrels could lead to war, consider royal inheritance in medieval France. The period is remote, but the same issues come up again and again, at least through the eighteenth century. The French royal dynasty was founded by Hugh Capet in the tenth century. For more than three hundred years, his lineal descendants ruled the country. Sovereign authority passed down to the eldest surviving son in a direct male line. In their early days, Capetian rulers sometimes granted sizable territories to younger children, and there were occasional family squabbles over which child would inherit the crown.33 These early practices died away by the thirteenth century. By then, the king’s eldest son had a well-established right to inherit and rule an indivisible kingdom.34 The law of primogeniture was clear. It kept the kingdom whole and eliminated confusion about which of the king’s children would inherit. There was room for dispute only if there were no surviving male heirs. That is exactly what happened in the early fourteenth century. When Philip the Fair died in 1314, the French crown passed uneventfully to the oldest of his three sons. After a brief reign, the new king died without a son, so the crown passed smoothly to his oldest surviving brother. He also died quickly without an heir. That left only one brother, Charles the Fair. Like his brothers before him, he ruled briefly and produced no heirs. He was the

L E A D E R S H I P S U C C E S S I O N A N D WA R

125

last male in the Capetian line. When he died, the stage was set for a struggle between aspiring new dynasties (an interdynastic conflict). The winner would establish his descendants as rulers of France for generations to come, just as Hugh Capet had. It was a prize worth fighting for. One strong claimant to the throne was Edward III of England. His mother was sister to the three short-lived French kings, and Edward himself was the only grandson of their father, Philip the Fair. He had vast holdings in France, and even though his Plantagenet family had repeatedly fought the Capetians, they were also vassals of the French king. Not surprisingly, the French nobility strongly opposed Edward’s claim, which would likely herald an English ascendancy and weaken their own powers. They met and endorsed what they claimed was the traditional practice: the French crown could be passed down only through a direct male line (termed Salic Law, suggesting it was an ancient practice of the Salic Franks). The effect was to disinherit Edward and pass the contested crown to his distant cousin, Philip of Valois.35 He was a direct descendant of Hugh Capet through the male line, but his father had been a second son (the younger brother of Philip the Fair), not the eldest. With him began the Valois dynasty. But Edward III did not give up his claims easily, even though they came through his mother. He considered them strong, legitimate, and, of course, immensely valuable. His forceful reassertion of them began the Hundred Years’ War (1338–1453). One of the main conclusions of that exhausting contest was to end forever English claims to the French throne. Interdynastic struggles like these continued to plague Europe as late as the eighteenth century and often led to war. The onset of war was prompted by confusion over rights to the crown, the presence of several legitimate claimants, and the entanglement of these rivals with foreign powers, each with its own reasons for wanting a favorite on the throne. Needless to say, there was also very poor information about who might win a war and at what cost, and little confidence that states would keep their bargains. It was extremely difficult to settle these major conflicts by prior agreement. The War of Spanish Succession (1702–1713/14) is a good example of such wars in the Westphalian era of nation-states. It was characteristic of the many eighteenth-century wars of succession.36 It was also a true world war, engaging all Europe’s great powers. One striking feature was the persistent but unsuccessful effort to avert war by international negotiations over succession and inheritance. The problems centered on Spain’s King Charles II. His father, Philip IV, died when Charles was quite young, so regents ruled the country for a time. Charles himself was a sad wreck of a man, the sickly product of generations of intermarriage among Habsburg cousins. After finally assuming the throne, Charles married twice but remained childless. With no heir in sight, Spain’s Habsburg dynasty was ending and a succession crisis was beginning. All the Powers saw it coming. Most were surprised Charles lived as long as he did, and no one expected him to produce a heir. They were keenly aware

126

CHAPTER 5

his throne was a gigantic prize. With it came not only Spain itself, but also its valuable possessions throughout Europe, including Milan, Naples, and the Spanish Netherlands, plus its immense colonies in the New World. Also at stake were a number of important strategic issues. Spain was traditionally allied with Austria, ruled by its Habsburg cousins. That left France encircled by Habsburg power. The French had long chafed at this encirclement and repeatedly fought to break out of it. The rich and powerful Maritime States had a different concern. They were troubled by Spain’s control of the southern Netherlands, which bordered the Dutch Republic and lay opposite the Thames estuary and London. The Spanish themselves were mostly concerned about finding a king strong enough to hold on to their national wealth and power. They were eager to keep their profitable empire together, rather than see it parceled out among various claimants. These issues came to a head as Charles II lay mortally ill. Louis XIV of France saw a golden opportunity. He was already married to Charles’s sister. Because Spanish inheritance was not so clearly through the male line, his son or grandson could lay claim to the Spanish crown. If they inherited, Louis could end his long conflict with Spain and break his encirclement by the Habsburgs. Naturally, the Austrian emperor saw things differently. He had an equally strong claim to rule Spain, as did his son. Yet another claimant was the electoral prince of Bavaria. Though powerless in his own right, he was a great grandson of Charles’s father, Philip IV, and could legitimately assert a right to inherit. What he needed was political backing. The Bavarian claim actually helped the French and Austrians reach a compromise. In anticipation of Charles’s death, they agreed to divide Spain’s possessions into three parts. The Dutch and English ratified this compromise because it gave the Spanish Netherlands to the Bavarian elector, keeping them out of France’s powerful grasp. This was the First Treaty of Spanish Partition. The beneficiaries were reasonably happy, but the episode infuriated the dying Spanish king, who had not been consulted about the disposition of his own kingdom. Worse yet, the bargain did not last. It collapsed when the Bavarian prince died unexpectedly, before King Charles of Spain did. The French and Austrians, still eager to reach an agreement and avoid war, settled on a second partition, but it too collapsed. The problem this time was King Charles, who made a deathbed effort to hold the Spanish empire together. He did that by signing a last will, making Louis’s grandson (his own greatnephew) his chief heir. Charles’s bequest not only harmed the Austrians, it directly threatened the Dutch and English. By handing the Spanish Netherlands to France, it would make the strongest nation on the continent a powerful presence on the North Sea. The Maritime Powers considered this an unacceptable threat to their national security and commercial supremacy. Louis understood his dilemma well enough, but he had little room to maneuver. If he rejected Charles’s will, then the Spanish would offer their empire intact

L E A D E R S H I P S U C C E S S I O N A N D WA R

127

to the Austrians, who would doubtless accept it. That would leave France still encircled by Habsburgs, with no gains at all. Under these difficult circumstances, Louis decided to accept the bequest, making his grandson King Philip V of Spain, the first Spanish Bourbon. Louis then fought the Grand Alliance of England, the Dutch Republic, Austria, and a number of other German states, including Prussia, united to prevent this extension of French power. The War of Spanish Succession thus retraced the long-standing rivalry between France and the Habsburg powers, which had fought repeatedly since the Thirty Years’ War (1618–48). The details of each succession dispute—in Spain, Austria, Poland, Bavaria, and elsewhere—are mind-numbing. All too often, they turn on accidents of marriage, childhood illness, and longevity. But the idiosyncrasies are knit together by a powerful, recurrent theme. Even at the height of European absolutism, the rules of dynastic succession did not always ensure a smooth transition of royal power and did not avert interstate conflicts. To be precise, they did not ensure a smooth transition unless there was a male heir who inherited from a direct line of eldest males, or unless inheritance by or through females was already a well-settled matter. This point can be framed in more general terms. When the rules of leadership succession are uncertain, when states are strategically interdependent, and when they have opposing interests in prospective rulers, then succession disputes will inexorably become interstate disputes. Given the poverty of eighteenth-century information and communications and the absence of other effective cooperative institutions, it was hard to settle these disputes short of war. That is exactly what happened, time and again, in absolutist Europe. The ambiguities of succession and the high stakes repeatedly drew Great Powers into the vortex. As we have observed, this vortex need not mean war if states can arrange compromises (by dividing territory or making side payments) and if they are confident their partners will stick to their bargains. Side payments and territorial divisions were easy enough to arrange in early-modern Europe—far easier than they are today. In the days before nationalism, rulers could exchange fortified cities and whole territories with little objection from the inhabitants. The first Spanish partition is a good illustration. Enforcing these compromise agreements was another matter entirely. Treaties were broken with few second thoughts. Dynastic states were sure that others would break their bargains when the first opportunity arose, as they would themselves. Louis XIV’s decision to accept the Spanish inheritance, for instance, directly violated his prior agreement with England and the Dutch Republic. Paul W. Schroeder says that such arrangements were always intended to be brief and that the terms “promoted the evasions of alliance commitments and sudden reversals of alliances for which the eighteenth-century was famous. Alliance flexibility of this sort was considered normal, so that statesmen, when concluding alliances, regularly tried to calculate at what point their ally was

128

CHAPTER 5

likely to defect. Indeed, they considered it a necessary mechanism for maintaining an overall balance.”37 With cooperative arrangements so easily discarded, it was difficult to reach durable compromises on succession problems. All too often, they led to war.

Succession Disputes in Modern Dictatorships: The Contrasts with Democracies Violent disputes over succession are still an important feature of political life in nondemocratic states, as the tragic slaughter in Rwanda and Burundi testifies. Civil conflicts over leadership are likely to become international conflicts as rival claimants seek outside support, and as foreign powers push their favorite candidates. In dynastic states, as we have seen, foreigners have strong incentives to champion a relative. In states where ideology and religion matter more than lineage, outsiders try to install fellow believers. They have every reason to fear others will do the same. That was certainly true during the ideological contests of the Cold War. The United States and Soviet Union struggled countless times over whether communists or anticommunists would take power, from the Greek Civil War to Indochina to Afghanistan.38 More recently, the terrain has shifted to religious and cultural conflicts, with a different set of protagonists.39 From central Africa to southeast Asia, Islamic states have aggressively promoted their co-religionists abroad, often favoring fundamentalists who would impose shari’a law and strict Islamic government (al-hakimiyya).40 Within Islam itself, there have been serious conflicts between Sunni and Shi’a factions, reminiscent in some ways of Europe’s religious wars between Catholics and Protestants, though without a strong doctrinal element.41 There have also been pitched battles between Islamic fundamentalists and secular nationalists.42 The Islamic Republic of Iran has been especially active in both kinds of clashes, promoting Shiite power with warlike ferocity and demanding adherence to strict Islamic law. In doing so, it has repeatedly supported terrorism abroad and intervened in foreign civil wars. Although these succession struggles differ in important details, they mark a well-worn path to war. • They can produce major changes in power relationships and do so quite rapidly. • They muddy the assessment of regional military balances. It is hard to predict accurately the fighting capabilities of states torn by succession disputes. The analytic tasks are even more difficult after social upheavals, not only because the changes are greater but also because the new regime itself

L E A D E R S H I P S U C C E S S I O N A N D WA R

129

poses a murky (and perhaps nonexistent) “threat” that is hard to calculate. On the other side of the coin, states in turmoil have very poor capacities for assessing other states with care. The information problems are two-sided and serious. • Succession struggles put pressure on states to make decisions quickly on life-and-death issues. The vulnerability produced by threatening social changes may be only temporary. After a brief period, new leaders can consolidate their power. If they really pose a threat, then that it will only worsen as the new leaders pull together the latent capabilities of their military, state bureaucracy, and citizens. This time pressure puts additional strain on information and negotiation. In fact, there may be little time for negotiations and little opportunity to narrow the differences in perception. This limits the chances of finding common ground, if any exists. Asymmetries in time pressure may also complicate bargaining. The state in turmoil naturally hopes to delay. It gains by waiting, since its power will revive as its polity stabilizes. Its rivals, on the other hand, gain by swift action. They are more likely to win a conflict now rather than later. These differences are inauspicious for discovering a mutually desirable bargain prior to war, or for making it stick if they do (because it is difficult to make the resulting bargain time-consistent, that is, continuously desirable to both parties throughout the span of the agreement). The Iran-Iraq war shows how social turmoil, military uncertainty, and time pressures can intertwine to produce preemptive violence during a succession crisis. Iraq calculated that it had a unique and fleeting opportunity to seize vital territory from Iran while the larger country was torn by religious revolution and its army racked by social and political upheaval. Iran had always been the more powerful state. After the dust of its revolution settled, it was likely to regain that dominance.43 Moreover, its new government, based on the vigorous assertion of strict Shiite Muslim orthodoxy, posed a many-pronged threat to Iraq, a secular regime with its own restive Shiite minority. Beyond these serious conflicts of interest, the two states had radically different assessments of Iran’s fighting strength. Iraq thought the revolution had seriously weakened its large rival, opening a golden (but fleeting) opportunity for attack. Iran, on the other hand, thought its religious fervor would actually produce a better fighting force. Its ardent troops would more than compensate for lost officers and equipment. It would field an army of true believers, who would fight to the death for a transcendent cause.44 In the end, Iraq miscalculated. It launched a war for limited objectives, thinking it could win quickly without high casualties and then hold the valuable territory it had seized.45 It did win some initial victories, but the war proved to be far longer, far deadlier, and far less successful than Iraq had expected. The

130

CHAPTER 5

professionalism of Iraq’s army was a big advantage, but not big enough. The casualties on both sides were horrific. Iraq came close to winning in the early going, but the fighting then settled into a long war of attrition, a bloody stalemate that lasted years. Iraq’s miscalculation underscores a larger point. When national capabilities are shifting rapidly, as they do after social upheavals, they are very hard for others to assess accurately. That is true even if their leaders are wiser strategists than Iraq’s Saddam Hussein—a low hurdle, to be sure. There are genuine and perhaps unresolvable doubts about countries in turmoil—their ability to mobilize for war, tax their citizens, maintain their arsenals, defend vital targets, and stabilize domestic political power. Moreover, turbulent contests for state leadership, like all important social changes, create considerable confusion and uncertainty. That is true in modern dictatorships just as it was in earlier interdynastic struggles. Confusion over succession can be contained, even in nondemocracies. Established monarchies generally had little trouble passing the crown to eldest sons. Likewise, modern authoritarian regimes with strong party systems, like the People’s Republic of China, have managed succession smoothly, if secretly. Such regimes are far less common, however, than strongman dictatorships, where succession is an erratic and often deadly process. On average, then, the uncertainties surrounding succession are higher in nondemocratic states, with far-reaching implications for anxious neighbors.

Succession Differences among Nondemocratic States Succession in nondemocratic regimes varies widely, and so does the level of uncertainty associated with changes in their leadership. There are systematic reasons why uncertainty is higher in some cases than others. Once again, royal succession provides a useful starting point. Crises were far more common in early medieval Europe than in the high Middle Ages or early-modern period. Over time, the institutions of monarchical rule matured and their basic practices stabilized. In early medieval Europe, political institutions were primitive. They varied a great deal from place to place and year to year, remaining in almost constant flux. Power was personalized and linked to tribal loyalties rather than regions of settlement. Rulers were kings of the Franks, not kings of France. Basic rules of inheritance varied considerably and often idiosyncratically, depending on the king’s personal preferences. In fact, there were only vague distinctions between public and private rights to govern or inherit. Given the technologies of European war and transportation, organized violence was mostly local and decentralized, which meant that numerous feudal barons could contend for higher power. The idealized hierarchy of feudalism did not stop these conflicts. In the age before national states were consolidated, anarchy was often as much local as international.

L E A D E R S H I P S U C C E S S I O N A N D WA R

131

But what was commonplace in the ninth or tenth centuries was far less common in the fourteenth. Inheritance rules reveal these changes. They acquired considerably more precision and continuity. Not only did the rules become clearer and more elaborate, they became basic building blocks in the larger feudal order, covering all landed nobility. Inheritance by the king’s eldest son became more legitimate and less contested. One reason was that all other nobles inherited their power and authority in the same way, through primogeniture. This routinization did not eliminate wars of succession. What it did was sharply curtail their likelihood as long as there was a clear-cut, legitimate heir. When a dynastic line died out, however, and only distant relatives were available, royal succession rules were underspecified. That produced deadly serious contests to win the throne and establish a new dynasty. Few nondemocratic governments today have the well-ordered succession rules of absolutist monarchies. But there are similar distinctions to be drawn between those that have more established, or “corporate,” governance and others that have less institutionalized, more personalistic rule.46 Where strong ruling parties have held power for several generations, like the People’s Republic of China, they often have more settled procedures for succession, and greater continuity across leaders, than newer dictatorships or military juntas. During the interregnum, these established, party-ruled states are also more likely to have experienced, continuous bureaucracies to administer day-to-day affairs. The Soviet Union, for instance, managed succession by party leaders from Brezhnev to Andropov to Chernenko to Gorbachev.47 In terms of institutional continuity, then, some nondemocracies are better off than others. They have institutions that help them project promises into the future. Best off are those with effective limitations on one-man rule and with well-established governing arrangements, such as strong ruling parties. They can make plausible commitments—perhaps even credible commitments—that transcend specific leaders. Worst off are states facing massive social upheaval, which makes future governments unpredictable, or states ruled by powerful dictators, with few domestic constraints on their future actions. Their successors are apt to be equally powerful and unfettered, which makes any long-term agreements with them problematic. The next ruler (or even the same ruler) could overturn commitments and change course radically, without being contested internally and without advance notice to partners. Among nondemocratic regimes, the gravity of succession problems and their impact on international cooperation varies along four important dimensions. The first is the state’s level of social and political turmoil. High levels spread confusion and generate manifold threats to neighbors. The second is how regularized the succession process is. Is the selection of leaders controlled by a well-organized party or clan, or is it controlled by a single ruler and his clique? Strongmen and military juntas have no real rules of succession. Organized parties and dynasties at least try to establish them, though not al-

132

CHAPTER 5

ways successfully. Third, how transparent is the process? The task of replacing old leaders and selecting new ones may be more or less open to view by outsiders. Fourth, how constrained are new leaders after they are chosen? Are they free to act autonomously or are they sharply limited by indigenous institutions, usually strong party government? These differences among nondemocracies have an important impact on foreign powers and the dangers of war. Social turmoil, violent contests among elites, and poorly institutionalized succession procedures produce confusion and possibly discontinuous change. They may make it inherently difficult for leaders to make solid, long-term promises, even if they are sincere and highly motivated, since future rulers may be free to discard them. These commitment problems cast a cloud over many nondemocratic regimes, impeding their ability to cooperate with others. They are hung by a silken noose of their own making. Unfortunately, their own people and many others may be hung as well. Many nondemocratic states—probably most, except for monarchies—have no clear rules of succession at all. This is not a simple oversight. The last thing a dictator wants is a potential successor lurking in the wings. After all, the heir apparent and his supporters might not be so patient. Indeed, his mere presence could become a focal point for opponents of the current ruler. The Ottoman Empire solved this problem with awful simplicity. Through most of the empire’s history, new sultans were picked for their competence rather than their place in the birth order. After the new ruler had been chosen, all his brothers and half brothers were immediately strangled with wire cord. Problem solved. Actually, even these drastic measures did not entirely solve the problem. When the sultan produced no male heirs, there were still violent struggles for power. In the later years of the empire, male relatives were more likely to be imprisoned than killed. That, however, did not entirely eliminate them as threats. A common threat to all rulers comes from their deposed predecessors. They already know the levers of state power and have powerful political connections. That is one reason why so few are left living in nondemocracies. Subsequent rulers understand the threat and have good reasons to kill them. Nondemocratic leaders anticipate this dilemma while they are still in power and refuse to retire. They typically cling to office until they die, peacefully or brutally. In monarchies, heirs-in-waiting can easily become the hub of disgruntled elites. The Prince of Wales occasionally played this focal role in early-modern England, the dauphin in France. Dynasties are bound to have such heirs around, but dictatorships and military juntas are not, and they do not want them. They would rather have their countries suffer the deluge after they are gone than risk a shower themselves. Aging strongmen like Fidel Castro, Mohammar Khadaffi, and Yasser Arafat have no obvious successors, and none in training.48

L E A D E R S H I P S U C C E S S I O N A N D WA R

133

Others, like Saddam Hussein, have hoped to convert their personal rule into a dynasty, as Hafez al-Assad did in Syria. Stalin had no anointed successor, of course. As long as he lived, what was the life expectancy of an heir apparent?49 As a result, when Stalin died in March 1953, he left the Soviet Union with no successor and no way to choose one. That created an amorphous power struggle among the Communist Party elite. They were so anxious about the regime’s grip on power that they waited a full day after Stalin’s death before announcing it to the country. They wanted to be certain of the army’s loyalty and have troops stationed around the country in case of disturbances. The Politburo immediately began ruling collectively, an awkward process, while fighting strenuously among themselves to capture the top posts in the Community Party and government. It took years of scheming and political murder to produce a winner. The process was hidden from the Soviet public and outsiders, except when high party officials mysteriously disappeared. Within months, the Politburo arrested and killed Stalin’s henchman, Lavrentii Beria, head of the secret police. Over the next two years, a coalition managed to isolate and block Stalin’s most likely heir, Georgii Malenkov.50 The successful infighting was led by Malenkov’s main rival, Nikita Khrushchev, who gradually emerged as “first among equals.”51 Still, it took five years for Khrushchev to become supreme leader in his own right. In 1958, five years after Stalin’s death, he finally removed his last major opponent, Nikolai Bulganin. Khrushchev then took over the USSR Council of Ministers and appointed his own new government.52 Khrushchev may have succeeded Stalin, but he never managed to acquire his extraordinary powers. Senior party officials were determined to see that never happened again. It threatened their positions and their lives. Outsiders peered in without much success. Even the CIA never gained a clear understanding of exactly how much Khrushchev controlled or what decisionmaking was really like inside the Kremlin. That became known only after the Soviet Union fell and some of its archives were opened—not much use to Eisenhower or Kennedy then. The issue here is not just who will be chosen as successor—we do not know that in democracies either—but rather how he or she will be chosen and how large a change is entailed. Will the process be peaceful or violent? Will the new ruler (and the process of choice itself) sustain the current regime or produce sharp discontinuity? Will the selection process and contested issues be visible to outsiders, or hidden from view like those in the old Soviet Politburo or China today? In Cuba, Libya, Syria, and many other dictatorships, the process itself is not merely underspecified, it is hardly institutionalized at all. The same is obviously true of social revolutions where popular leaders, organizers, and military commanders arise during the violent struggle and then vie for control once the new regime is in power.

134

CHAPTER 5

Of course, some nondemocratic states are more organized than others. Their governance and leadership selection are more routinized. If they are monarchies, they have established rules of dynastic inheritance. If they are entrenched one-party states, they usually have some selection mechanisms within the ruling party. The methods are typically secret and highly centralized, but at least they do select. Mexico’s long-ruling party, the PRI, had such a centralized process, with the current president personally choosing his own successor. There was a firm rule against a president serving more than one term, but the incumbent controlled his party’s next nominee. That candidate always won, until Vicente Fox’s upset victory in 2000.53 These procedures, though hardly a model of free social choice, are far more regular and predictable than those of strongman dictatorships like Iraq or Syria or military juntas like Nigeria, Algeria, or Myanmar. Whether nondemocratic selection mechanisms are well developed or not, they pose greater risks of war than their democratic counterparts do for three reasons. First, they are typically underspecified, posing problems similar to interdynastic succession. Second, they are opaque to outside powers, adding to others’ uncertainty. Third, the ruler who emerges may lack wider political legitimacy, inside the government or outside, making his authority in office more tenuous. That opens the door to entangling alliances by the ruler himself and other contenders, as they scramble for support. These risks are compounded by the mechanics of governing in nondemocratic systems. Although the new ruler’s legitimacy and tenure may be uncertain, his powers in office are far wider than those of democratic leaders, who are limited by legal procedures, precedent, and countervailing centers of power. This difference is fundamental. It means, among other things, that the replacement of nondemocratic leaders is more likely to produce major, unpredictable transformations of policy. These are serious problems for interstate relations because they amplify uncertainty. That does not mean they must lead ineluctably to war. Other states, however greedy or opportunistic, may be deterred by military force. Or they may achieve compromise agreements. There was no war, for example, after Stalin’s death, even though Soviet leadership was uncertain and contested for five years. The power and continuity of the Soviet Community Party and the strength of the Red Army meant that the country was quiescent and its defense capacity remained high. The United States had no desire to launch a preemptive strike, partly because of its own values, partly because it would have been so foolhardy. After all, with or without Stalin, the Soviet Union now had the trump card of the atomic age: thermonuclear weapons. But when states are less well armed and have less institutional continuity, leadership struggles can draw outsiders into violent disputes, covert action, and sometimes war itself.54

L E A D E R S H I P S U C C E S S I O N A N D WA R

135

Conclusion: Peaceful Succession and Peace among Democracies The constitutional structure of democracies dampens uncertainty in several important ways. First of all, the selection process is regularized. Second, candidates articulate their policy preferences during public campaigns. Third, the powers of elected leaders are constitutionally limited and often must be shared with other elected officials and other elected bodies, as well as courts and administrative agencies, all operating under legal rules. Finally, the transparency of democracies makes it far easier and quicker for outsiders to spot turnover in leadership and major changes in policy. In nondemocratic states, by contrast, the range of uncertainty is far higher on all these dimensions. The process of choosing leaders is more opaque. There is a far greater chance of sharp, discontinuous changes at the top. It is much harder to spot the scope and direction of change quickly. It is also harder to forecast what the new leaders want. Because they emerge from the hidden recesses of military or state bureaucracies, under the strong hand of repressive leaders, it is much more difficult to tell what policies they actually prefer, or how large a change they represent.55 Mikhail Gorbachev is a prime example. He rose to the pinnacle of power through the Community Party and Soviet state apparatus. Only when he reached the top did his ideas about democratization and economic restructuring become public, along with his willingness to take great risks to reform his country. Even then, it took years for outsiders to understand the depth of his commitment, to recognize that he wanted genuine social reforms, not cosmetic changes. Khrushchev’s secret attack on Stalinism three decades earlier had been equally surprising, within the Kremlin and without, and it took some time to leak out to Western audiences. Yet it was one of the chief sources of the deepening split between the Soviet Union and the People’s Republic of China, which thought the attack on Stalin weakened the foundations of its own party rule.56 The continuity of the Soviet Communist Party and the sheer scale of modern bureaucracy ensured continuity at the level of day-to-day agreements with the USSR, such as those covering trade and debt payment. They probably encouraged stable performance on strategic arms agreements (that is, consistent low-level cheating by the Soviets). But future economic arrangements and the whole arena of high politics were far less predictable, just as they are now in China. The examples from Soviet leadership and absolutist monarchies are compelling because the regimes themselves were relatively stable and because their leadership selection was better articulated than in strongman dictatorships or military juntas.57 Yet even these relatively predictable arrangements generate

136

CHAPTER 5

far more confusion than democratic elections do. Faced with these greater uncertainties, foreign observers may well exaggerate the threats and underestimate the cooperative opportunities posed by new, nondemocratic leaders.58 Even if their estimates are unbiased, the variation around them is bound to be great because information is so sketchy. While democracies may be more transparent and more predictable, they, too, are open to outside influences when choosing leaders. Foreign powers can voice their preferences openly or give money secretly. They can help a party in power achieve diplomatic success, in hopes it will win the votes to stay in office.59 Or they can oppose the incumbent party, obstruct it diplomatically, and hope to cost it votes. If they are more daring, they can try to foster civil unrest or even sponsor assassination attempts. It is striking, however, that well-established constitutional democracies have generally avoided heavy-handed interference in each other’s elections, despite the openness of their political processes. Violent or illegitimate interventions by one democracy in another’s elections are rare. The United States did so secretly in early postwar Europe, where democracies were new and poorly established. In Washington’s view, electoral victories by communist parties would kill the new democracies and produce more allies for Moscow. The U.S. intervention against Salvador Allende in Chile is another prominent exception, prompted by Washington’s anticommunism and the anticipation that Chilean communists would eliminate democracy. These exceptions have a common theme. U.S. policy was ambivalent because it was pulled in different directions by its two great principles: let democracies decide their own fate, but halt communist expansion. U.S. policymakers managed to reconcile them by arguing that democracies should not be allowed to choose communism because the Community Party, once in power, would eradicate democracy itself.60 Opposition to fundamentalist Islam has not achieved the same totemic status or strategic importance as anti-communism, but victories by fundamentalist parties pose some of the same issues for policymakers. The military coup in Algeria, which France endorsed, is suggestive. It is too easy, however, to say that Western states are simply hypocritical, that they favor democratic outcomes only when they like the winners. The recent Turkish case suggests a more tolerant interpretation. When fundamentalists won a plurality in Turkey, no outsiders intervened to prevent their forming a government even though they won barely 20 percent of the vote and Turkey’s democracy is far from firmly established. In general, there have been few conflicts among democracies over the succession process. There are several reasons why. First, if an election is conducted fairly and the winner is committed to sustaining democracy, there is a strong normative presumption against direct interference and certainly against violent disruption.61 It is almost impossible to win support within any democ-

L E A D E R S H I P S U C C E S S I O N A N D WA R

137

racy for overturning another’s electoral decisions. There are some counterexamples, but they are revealing in several ways. They have been rare; they have been secret; they have mostly involved new and unstable democracies (where an antidemocratic winner might well overturn the regime itself) and, with the notable exception of Chile, they have been nonviolent. Although these normative considerations are important, there are also very practical reasons for leaving other democracies to their own devices. If these states remain democratic, then today’s leaders will become tomorrow’s opposition party. They will not stay in power forever. If it is costly to intervene, and if future outcomes are important (and not just current events), then it is wise to wait or to intervene only in low-cost ways, such as campaign donations, direct or indirect endorsements, or diplomacy designed to aid one candidate. This prudential restraint would change if (1) current events vastly outweighed future outcomes, or if (2) the new government seemed likely to convert its electoral mandate into permanent rule, whether democratic or not. In either case, waiting would become too costly. These calculations help explain the sequence of U.S. intervention in Chile’s democracy during the 1970s. The initial U.S. policy was a low-cost, low-risk effort to block Salvadore Allende, a democratic communist, from taking office. He had won a plurality of the popular vote but not an outright majority, so his election needed to be ratified by the Chilean legislature. Such ratification was routine in Chile’s long-standing democracy. Still, the United States made some relatively modest efforts to block his election at both stages, at the ballot box and in the legislature. They failed. Once Allende was in office, his increasingly close ties to Cuba and the Soviet Union convinced the Nixon administration that Chile had become a greater security threat, that Allende’s party was becoming more entrenched in power, and that it might convert its electoral mandate into permanent rule. This new assessment led to a more aggressive policy. Three years after Allende took office, the United States endorsed a right-wing coup, which deposed the Allende regime and killed Allende himself. In the controversies that followed, the U.S. government always contended that its role was minor and indirect and that Allende’s government was no longer democratic when it was overthrown.62 The public revelation of the CIA’s role in the coup caused a major scandal in both Chile and the United States, led to highly publicized hearings in Washington and eventually to significant reforms at the CIA itself, including new laws limiting such interventions. Nothing similar has occurred between other stable, constitutional democracies. A much more common form of intervention—and a much more legitimate one—is the outspoken support of one democratic leader for a candidate in another country. In 1996, for example, the Clinton Administration virtually endorsed Russia’s Boris Yeltsin and Israel’s Shimon Peres in their respective elections. Likewise, foreign firms can use their local subsidiaries to donate legally to specific parties and candidates. Democracies are inherently open to

138

CHAPTER 5

such influences. Their presence may produce diplomatic frictions, but they have been far from violent and so far have raised little real concern about sovereign autonomy and foreign intervention, much less been a cause of war. The overriding point is that succession crises are very rare within established democracies and, with the exception of Chile in 1973, have almost never involved a second democratic state. They have never led to an outright war between established democracies. By contrast, dynastic states, revolutionary regimes, and other nondemocracies face succession crises time and again. In the process, they often draw outside powers into violent conflicts and, worse, into full-scale war.

6 Extending the Argument: Implications of Secure Contracting among Constitutional Democracies

NOW THAT WE have examined the contracting advantages of democracies, we want to see if the theory does more than explain the democratic peace. A robust theory should offer insights into a range of related phenomena. Looking beyond the questions of war and militarized conflict, we want to see what else our contracting theory predicts and then check these predictions against established research findings. If the predictions hold up, we have additional grounds for thinking democracies are at peace because they are reliable partners. If secure contracting really does explain the democratic peace, then we should also expect the implications presented in table 6.1. These inferences cover a broad range of international relationships and, if borne out, they strengthen our confidence in the contracting explanation.1 They also meet two crucial tests proposed by Imre Lakatos and Mancur Olson. Lakatos says a theory is retrograde if it simply “predicts” previously known facts, facts that the theory itself was designed to cover.2 To be progressive, a theory must predict new data and uncover unexpected relationships. Olson adds that theories are stronger and more persuasive if they explain a diverse array of facts.3 Both tests are worth taking. The goal, however, is not simply to pile up more findings, however wideranging. The real goal is to cumulate these findings as part of a larger, integrated whole—one that gives us a better understanding of war and peace. Dina Zinnes calls this integrative cumulation, which she distinguishes from the mere addition of findings.4 We cannot achieve such integration by pure induction, piecing together results by trial and error, like a jigsaw puzzle. Rather, we must begin with a theoretical explanation and see if its predictions are supported by a wide range of empirical work. The theory itself then serves to integrate the findings, making them part of a coherent picture rather than disjointed pieces. Our contracting theory is just such a proposed explanation for the democratic peace, with larger implications for international politics. The hard questions are whether it predicts a diverse array of facts, highlights unexpected relationships, and serves to integrate them within a coherent and powerful explanation.

140

CHAPTER 6

TABLE 6.1 Table of Conjectures: Implications of Reliable Contracting between Democracies Self-protective mechanisms

Self-protective mechanisms are stronger and more extensive when agreements involve non-democracies

Alliance duration

Alliances are more durable when only democracies are members

Learning effects

Because of learning effects, longer periods of joint democracy should produce more stable peace

Conflicts early in relationship

Corollary on learning effects: More serious, militarized conflicts between democracies should occur early in their relationship and diminish over time

New democracies

New, weak, or marginal democracies should not be part of the democratic peace because such states cannot eliminate commitment problems

Enduring rivalry

Long-term rivalry among democracies should be very rare

Better at every level Democracies should have some militarized disputes, but there should be fewer of dispute at every level along the escalation ladder Near-misses

“Close calls” between democracies should involve fundamental, existential issues since less important matters can be resolved through agreement

Clear-sighted

Democracies should assess each other’s capabilities and intentions more accurately than nondemocracies do

Shocks

When democratic regimes are threatened by external shocks and instability, the democratic peace should be weaker

Leadership change

Changes of leadership pose greater concerns when bargains involve nondemocracies

Bilateralism

When issues are bilateral and within the control of two democracies, their bargains will be stronger and more reliable than if third parties are involved or the external environment poses a major risk

Are the Additional Predictions Right? Self-Protective Mechanisms Self-protective mechanisms are stronger and more extensive when agreements involve nondemocracies. When agreements carry high risks of nonfulfillment, the parties have three basic options. They can decline to make an agreement altogether; they can make an agreement but refuse to rely on it; or they can buttress the agreement with mechanisms designed to increase the likelihood of performance or lessen the impact of nonperformance. States use all three options to cope with risk, threats, and uncertainty. Our focus here is on the mechanisms used to buttress agreements, the topic of our conjecture. States employ a wide range of techniques to protect themselves when dealing with risky partners, where the dangers of breach are high. In addition, they monitor performance more intensively and lessen their reli-

EXTENDING THE ARGUMENT

141

ance on these bargains. These costly, imperfect arrangements seem to be more extensive and pervasive when democracies deal with nondemocratic states, although the evidence is not systematic. What exactly are these self-protective arrangements? One is to make sure that bargains are time consistent. Neither party should perform much more than another at any moment during the bargain. That way, if the bargain fails after a week, a month, or a year, neither party owes much to the other. Each has completed a roughly equal share of its obligations. In its most extreme form, each party performs its obligations simultaneously, typically in a series of small, matching steps. Hobbes recommends exactly this strategy in his harsh state of nature. Do not perform your obligations first, since your partner will not reciprocate later. For he that performeth first, has no assurance the other will perform after; because the bonds of words are too weak to bridle men’s ambition, avarice, anger, and other Passions, without the fear of some coercive Power.5 Nation-states coexist in a world of anarchy—that is, without an overarching global government—but they do not exist in Hobbes’s severe state of nature. For example, they have repeated contact and can identify the partners they are dealing with. That allows them to build reputations for trustworthiness. If individual states can benefit from joint action and if the future is important to them, iteration can support self-enforcing cooperation. However, when reputations are poor and the dangers of defection are high, then states are wise to take Hobbes’s advice and design bargains where neither party performs first. If these timing mechanisms are to work effectively, bargains must be decomposed into a series of small, discrete steps. Negotiators jokingly call this “salami slicing.” The aim is to ensure both parties fulfill their obligations in lockstep. When confidence is extremely low and long-term agreements are unlikely to be completed, these small steps may even be couched as separate agreements and new ones negotiated as the need arises. The advantages are clear from Hobbes’s description. Small steps, performed at the same time, sharply limit the damage any partner can inflict by unilateral defection. The downside is their high transaction cost. It is often difficult— and sometimes impossible—to decompose bargains so they are constantly balanced through time between two or more parties. Overseeing their performance moment by moment is a unceasing chore, and a difficult one. Intensive verification is a common element among risky bargains. States are unwilling to make bargains with adversaries (or potential adversaries) unless they can be monitored in a timely, effective way. The goal is to avoid prolonged, undetected cheating. When risks are high, then investments in verification will also be high, which lowers the net value of the agreement (compared to the value of an easily verified one). Without transparency to help, verification must be well planned, constant, and detailed.

142

CHAPTER 6

To make that feasible at reasonable costs, the bargain may be tailored to make verification simpler. This simplification is costly because it makes the underlying bargain less desirable (in the abstract). One obvious way of simplifying verification is to concentrate on easily measured obligations. Obligations that are hard to verify are simply excluded from the bargain. Again, that is costly because both parties could presumably profit from some exchange if only they could monitor it. Another possibility, which we discussed earlier, is to reframe the bargain around a set of proxy variables. Instead of exchanging X for Y, which the parties would really prefer, they decide to exchange X’ for Y’. X’ is a commitment or obligation that roughly approximates X but is easier to measure. For example, an exporter and importer might agree that a level playing field is the best solution for their commercial transactions, at least in principle. However, if either suspects market manipulation, they may settle on a much simpler quota system. The reasons stem from information problems. It is difficult to uncover market manipulations; they are well hidden for good reasons. Remedying them is another difficult, time-consuming task, if it can be done at all. If monitoring costs were zero, then quotas would be a decidedly inferior solution. In this case, though, quotas have the decisive advantage that compliance can be measured cheaply and easily. That can make them superior when all transactions costs are taken into account. Bonds, collateral, and hostages are another way of protecting innocent parties against default and nonperformance. Unfortunately, these devices are difficult to use in international diplomacy, although medieval kings did sometimes exchange personal hostages to guarantee treaty performance, with sporadic success. First, how do you ensure that the bond (or hostage or collateral) is returned when performance is completed? In domestic commerce, that problem is usually solved by having third parties hold assets in escrow. When the obligations are successfully completed, the holder is legally required to transfer the assets. Beyond this fiduciary duty, companies providing escrow services transfer funds reliably to protect their business reputations. That is essential if they are to run a profitable escrow business. While these techniques are available to secure international commercial transactions, they are beyond the reach of most interstate political bargains. A second major difficulty is that bonds, hostages, and collateral rarely match the scale of diplomatic promises. To see why, consider a commercial example. When a construction firm wins a new contract, it is often required to take out a performance bond matching the size of the job. If the company fails to meet its contractual requirements, it loses the bond, which is then used to compensate the innocent client. There is simply no way to do anything like that in major diplomatic bargains. What bond could Wilhelm II post to compensate France and England if he broke the 1839 treaty, signed by his state and his ruling family, guaranteeing Belgium’s neutrality?

EXTENDING THE ARGUMENT

143

True, a nation’s reputation does serve as a kind of bond, one that is forfeited when it fails to perform its obligations.6 But it is an imperfect bond in several ways. One is that the prospect of lost reputation may not be costly enough to prevent breach.7 When war and peace are at stake, it seldom is. In any case, one side’s loss of reputation does nothing to compensate the aggrieved party. It merely serves as a deterrent, for better or worse. To compound these problems, the riskiest countries—the ones where large bonds would be most useful—cannot use their reputations because they are worthless. If Britain ratifies a chemical weapons treaty and then violates it, it loses some of its valuable, hard-won reputation for reliable performance. If North Korea or Syria violate the same treaty, they lose nothing. So reputation is a useful bond for Britain, whose partners hardly need it. It is a meaningless pledge and worthless bond for North Korea and Syria, whose partners need it badly.8 Finally, agreements with risky, unreliable states are likely to be spelled out in greater detail. More contingencies will be covered, more obligations defined precisely. Why? Because if problems of definition or applicability arise later, their partners are uncertain they can resolve them. By contrast, reliable partners—partners that are confident about each other’s performance and good faith—are also more confident they can govern their bargains while they are being carried out, long after they are made. They can cooperate to fill in gaps, settle interpretive differences, and cope with changed circumstances. Their capacity to govern bargains down the line, as they are carried out, allows them to write simpler initial bargains, economizing on transaction costs. This is not an exhaustive list, but it does indicate how states can bolster their agreements to cope with risks of cheating and noncompliance. The question is whether democracies use fewer of these costly mechanisms in dealings with each other. There are no systematic studies comparing bargains between two democracies and those between nondemocracies or mixed pairs. No one has conducted either large empirical surveys or comparative case studies. At this point, the best we do is explore the conjecture by looking at two contrasting pairs of states dealing with the same dangerous problem. Are democracies better placed to resolve the issues? Do their bargains look markedly different because of differences in their domestic governance? One appropriate case is nuclear proliferation, which poses grave risks in bilateral rivalries. This issue arose during the 1980s and 1990s between North and South Korea and between Argentina and Brazil. All these states feared— with good reason—that their rival might be trying to develop a nuclear arsenal. Throughout the period, South Korea was a democracy; North Korea, a closed communist dictatorship. Argentina and Brazil underwent major political changes. Both began the period as military-run governments and then democratized, conducting several elections smoothly. These regime changes in South America produced significant changes in bilateral nuclear bargaining.

144

CHAPTER 6

Under their military governments, both Argentina and Brazil launched secret nuclear-weapons research programs. These programs were important because both had major nuclear facilities with significant military potential.9 There was a genuine possibility each would develop weapons and delivery systems, beginning an arms race between two historic rivals. Each certainly feared that possibility, which only spurred competitive efforts not to be outpaced. After democratization in the mid-1980s, however, Argentina and Brazil took a series of small steps, internally and externally, that gradually veered away from an arms race and into a solid bargain to avoid one. To begin with, each opened its once-secret programs to public scrutiny and criticism by legislators, political parties, and the press. At the same time, their elected leaders and senior officials began diplomatic discussions, which soon led to high-level visits to each other’s nuclear facilities. That opened the door to progress on the most politically sensitive issue: ongoing mutual inspections. The aim was to make sure nuclear facilities were devoted entirely to civilian uses and that no fissile materials were being diverted to weapons. The result was a 1991 joint agreement not to develop nuclear weapons, with monitoring by a new joint agency, the Brazilian-Argentine Agency for Accounting and Control of Nuclear Materials (ABACC). The ABACC inspections were the heart of the deal. They included the agency’s right to make surprise visits to any suspected facility.10 All information from the inspections was made available freely to both governments. That opened the way for still more inspections by the International Atomic Energy Agency, a prospect both countries had strenuously opposed for years. The entire arrangement, using bilateral and multilateral inspections, has worked exactly as promised. By May 1995, ABACC and the IAEA had verified the inventory of nuclear materials at all seventy Argentine and Brazilian nuclear installations.11 The arrangements have continued to work well. Democracy made this secure, negotiated outcome possible in several ways. According to Tom Zamora Collina and Fernando de Souza Barros, who know the issue intimately, “Major progress . . . was not made until democratic governments were elected by both countries. . . . Under democratic rule, scientific groups, citizens’ organizations, and newly empowered legislators were able to argue openly for restrictions on the nuclear programs. This helped to build a political climate more conducive to the implementation of bilateral and international inspections.” They add that “strong commitments by the Argentine and Brazilian presidents to transparency were essential.”12 Numerous policymakers in both countries told Michael Barletta that “concurrent democratization” was a major factor in improved bilateral relations and nuclear confidencebuilding.13 Another analysis, conducted by the Henry L. Stimson Center in Washington, reaches identical conclusions about the importance of democracy in resolving these nuclear issues in the Southern Cone.14

EXTENDING THE ARGUMENT

145

A direct, bilateral bargain was crucial to reassuring both Argentina and Brazil. Multilateral inspections were not enough. The IAEA disclosed only its methods, not its findings at specific facilities. By contrast, “bilateral inspections give these governments—and possibly their press and their citizens— direct access to information that they might otherwise not have,” according to Collina and Barros. They “can thus provide for more timely and possibly more credible assurance that materials are accounted for or that there may be a problem. For countries that are suspicious of each other, this access could build mutual trust.”15 They have done exactly that. Democracy helped in another, less direct way. Elected leaders thought nuclear safeguards would promote economic development by aiding the quest for more foreign trade and investment. Without recognized safeguards, nuclear programs were considered an impediment to foreign capital inflows. They indicated a country was politically risky, operating outside the mainstream of international understandings, and a possible target for sanctions. In particular, nuclear inspections and safeguards were essential if either country wanted to import high-technology goods from Europe and North America. Germany made that point emphatically in diplomatic discussions.16 These major exporters wanted reassurance from the multilateral inspection regime, and they got it.17 Meanwhile, Argentina and Brazil got the security assurance they needed from the bilateral regime. The story on the Korean peninsula is far different, filled with suspicions, unfulfilled bargains, and glacial progress. In 1994, this combustible mix almost led to war. It came perilously close. The solution was fragile and temporary. Only two years earlier, North and South Korea had been working toward the same solution as Argentina and Brazil. They were engaged in bilateral talks to develop a nuclear inspection regime for the peninsula. Under considerable pressure, North Korea had also decided to join the Nuclear Non-Proliferation Treaty (NPT) and accept inspections by the International Atomic Energy Agency. But the path to this auspicious moment had been tortuous, filled with deep political conflicts, profound mistrust, and near-total secrecy in North Korea. This unhappy combination soon blocked the path to a negotiated solution between North and South Korea, led to an unprecedented crisis between IAEA inspectors and the North Korean government, and produced an agreement between the United States and North Korea only after a high-stakes game of nuclear brinksmanship. Not surprisingly, the resulting agreement was heavily buttressed. It involved very small, tit-for-tat steps, with the United States rewarding North Korea (by providing electric power plants) only as long as the North met specific treaty obligations that could be confirmed independently. The theme, as one United States official put it, was “mistrust and verify.”18 The background of this mistrust was rooted in very different political regimes that had fought a major, inconclusive war in the early 1950s. Not only

146

CHAPTER 6

did the war end without a clear victor or even a peace treaty, the peninsula itself became a prominent site of the Cold War, where the dangers of renewed violence were palpable. The dangers were increased by the lack of information. North Korea’s political system was a black box, hermetically sealed against outside influences. The only exceptions were military and economic support from China and the Soviet Union. It was the Soviet Union that erected North Korea’s nuclear power plants, with fuel rods that could be diverted to military uses. When the USSR collapsed and China embarked on market reforms, North Korea was truly isolated. One of its few sources of diplomatic leverage was its embryonic nuclear program.19 It was unclear then—and even now—whether the nuclear program was designed mainly to increase its negotiating position or to produce weapons for domestic deployment and foreign sales. In either case, Washington and Seoul were alarmed as spy satellites turned up evidence North Korea was building plants that could reprocess “spent” nuclear fuel and produce weapons-grade material. North Korea indicated it was willing to discuss the matter, but only if American nuclear weapons were totally removed from South Korea. That could easily have frozen into a another deadlock, but it was resolved in the early 1990s when President Bush (the elder) decided to withdraw all America’s short-range and tactical nuclear weapons deployed overseas, including those in Korea. With the Cold War over, these weapons could be eliminated unilaterally, partly to lessen the dangers of misuse and partly to encourage Russia, Ukraine, and other former Soviet states to decommission their own nuclear weapons. With that impediment removed, North Korea was willing to negotiate nuclear issues bilaterally and multilaterally. The North-South negotiations began promisingly enough but quickly degenerated into name-calling. In March 1992, the two countries established a Joint Nuclear Control Commission to do what the ABACC had done for Argentina and Brazil. They even agreed to conduct their first inspections by June. That idea quickly foundered. Mitchell Reiss explains why: “Given that the two sides had absolutely no experience in creating and implementing any form of inspection regime, that the agreements they had signed provided little guidance, and their relationship was still characterized by mutual suspicion, if not rampant hostility, it was hardly surprising that discussions soon bogged down.”20 They never emerged from the bog of negotiations. The right to conduct surprise inspections remained a deadlocked issue, and a crucial obstacle to any effective bilateral bargain. Multilateral diplomacy and even “settled” bargains did not work much better. North Korea promised to join the Non-Proliferation Treaty but then balked for several years about actually signing. When it finally did join, Pyongyang turned over some essential data, as required, and permitted some inspections, but soon slammed the door on the required follow-ups. From the inspections already conducted, it was clear the North Koreans had simply lied about some nuclear facilities already built and others under construction. These facilities,

EXTENDING THE ARGUMENT

147

it was feared, could process weapons-grade materials. The United States discovered still more damning evidence through its spy satellites. When the head of the IAEA confronted North Korean officials, they replied that the United States had faked all the data, including the surveillance photographs. Of course, they refused to back up their assertion by admitting any inspectors. Equally troubling, the data North Korea originally supplied plus follow-up IAEA inspections showed that earlier U.S. intelligence had failed to uncover large elements of Pyongyang’s nuclear program. This discovery came after similar ones in Iraq, which had managed to hide a large weapons program under the nose of IAEA inspectors. Thus, in the early 1990s, North Korea’s nuclear program remained a troubling mystery, despite written promises to make it more transparent. What really worried the United States, the IAEA, and South Korea was that North Korea might be hiding nuclear reprocessing facilities, which would allow it to convert reactor fuel into bombs.21 Even arranging diplomatic meetings to discuss this matter was contentious. Meanwhile, the North’s nuclear program was proceeding. By 1993, Brent Scowcroft and Arnold Kantor, senior figures in previous administrations and experienced, cautious policymakers, had become so concerned that they publicly proposed a preemptive military strike if North Korea did not end its nuclear weapons program and permit inspections.22 Tensions reached a peak in June 1994, when North Korea announced it would withdraw from the NPT, presumably to begin building weapons. Across Korea, North and South, militaries were put on high alert. The region was closer to war than at any time since the early 1950s.23 Former President Jimmy Carter’s visit to North Korea reopened communications, undercut America’s threat to impose new sanctions, and diffused the immediate crisis. His visit was quickly followed by successful negotiations that produced the Agreed Framework of October 1994.24 This is not the place to trace the labyrinth of those negotiations. What matters for our analysis are a few salient points. First, any agreement had to cope with problems posed by North Korea’s closed polity and its highly secretive military. Even after it signed the NPT, North Korea balked at required inspections. If North Korea was to be incorporated effectively into the nonproliferation regime, it had to drop these objections. The inducements could be carrots or sticks. Any other security agreements with Pyongyang raised the same problem. In fact, North Korea’s leaders were so sensitive about these issues of autonomy and sovereign control that they objected to the very word inspections. They had to be called visits, a trivial point easily conceded. More seriously, Pyongyang repeatedly blocked international verification of its commitments and failed to meet crucial obligations, despite its clear pledges. Leon Sigal argues that the United States and IAEA also reneged on promises. In a tense, hostile relationship like this, it is difficult, and perhaps impossible, to figure out whether one party has unilater-

148

CHAPTER 6

ally broken its promise or is simply retaliating for another’s violation. In either case, the result is a shattered agreement that fails to achieve its original goals. That is why, when North Korea finally did agree to another deal—forgoing its nuclear reprocessing in return for subsidized electric power plants—the arrangement had to be structured very carefully to protect all parties against betrayal.25 It was.26 “The Agreed Framework could be considered a monument to the highest levels of mistrust between two nations,” according to America’s chief negotiator, Robert Gallucci.27 The alternative, he said, “is not trust . . . but a carefully crafted deal that exposes neither side to more harm than it would suffer absent an agreement, even if the other side does cheat, and that provides a basis for inspection to resolve concerns about cheating.” According to Gallucci, the combination of inspections and satellites gave the United States “high confidence” it could verify activities at the four or five North Korean facilities of greatest concern.28 The United States also chose to dole out its rewards for compliance in small increments, and then only as long as North Korea strictly adhered to its commitments. “It is a step-by-step program,” said Secretary of State Warren Christopher shortly after the agreement was concluded. “They take the first steps.”29 As he told the Senate Foreign Relations Committee, “The burden of upfront performance falls on North Korea, not the United States. [The agreement is structured so] we are not disadvantaged in any significant way if [North Korea] reneges on its commitments—at any time.”30 In other words, this agreement was loaded with devices to protect against breach and lessen the consequences if it occurred. All these devices were needed, as it turned out. In October 2002, North Korea admitted that its nuclear program had secretly continued despite the “Agreed Framework.” Pyongyang has been violating the agreement all along. This focused comparison between the Korean case and the Argentine-Brazilian closely tracks our conjecture. Self-protective mechanisms are stronger and more extensive when agreements involve nondemocracies. Without more extensive studies, however, we cannot be sure if this relationship holds across a wide variety of cases and is truly general.

Alliance Duration Alliances are more durable when only democracies are members. Several studies compare alliances that include different types of domestic regimes. They clearly demonstrate that alliances among democracies last longer. Kurt Gaubatz conducted the initial quantitative study in 1996.31 His basic conjecture—confirmed in his study—is that democratic alliances are more enduring. He attributes their longevity to democracies’ greater ability to make credible commitments and to the continuity of their foreign policies, thanks to more stable institutions and regularized changes of leadership. As he points

EXTENDING THE ARGUMENT

149

out, “the same factors that make it difficult for democratic states to enter into commitments also make it harder to get out of them. For example, the stronger role of domestic actors makes it more difficult to break or renegotiate commitments once the difficult barrier of ratification has been passed.”32 Quantitative studies of alliance duration must cope with a number of vexing methodological issues. There are familiar (but still difficult) questions about proper statistical controls, the most appropriate methods to estimate alliance survival, and how to handle nonlinear relationships. There are questions about how to deal with alliances involving many partners, the same problem confronting researchers on multipower wars. Should they be decomposed into a series of bilateral relationships or treated as a unified whole (and coded either all democracies, all autocracies, or mixed)? There is also a question about how to cope with alliances that do not begin or end within the years of the data set. Some alliances may already be several years old when the data start. Others are still operating when the data arbitrarily end. It is misleading, for example, to code an alliance as lasting only three years if it begins in 1962 and is still going in 1965, when Gaubatz’s data end. Finally, there is a substantive question about whether to distinguish alliances that entail higher levels of military commitment (which the Correlates of War codes as “defense pacts”). Thoughtful scholars are bound to differ about the best ways to resolve these complex issues. That should make us cautious about accepting any one finding as definitive. Gaubatz’s own findings are clear-cut: his model shows “a significant effect for the duration of alliances between liberal democracies. These effects are consistent in direction across all of the aggregations of the data and are statistically significant for the dual democracy coefficient in all of the models that use all alliances and for one of the dual democracy coefficients in the defense pact models.”33 Gaubatz also confirms findings by Randolph Siverson and Juliann Emmons that the effect is due to interactions among democracies, not the monadic properties of individual democracies.34 William Reed retests and confirms Gaubatz’s conjectures on the hazard rate of alliance decay. There are two main differences in the studies. Reed’s data runs longer (from 1815 until 1992, instead of 1815 to 1965), and he uses a different methodology. Reed uses discrete-time event history analysis (EHA), which is more appropriate, he argues, than Gaubatz’s use of continuous-time EHA. Despite these differences, Reed’s findings strongly validate those of Gaubatz. They show that “the addition of democratic members to an alliance has a significant negative impact on the probability of alliance failure. This relationship exists even when controlling for the number of members in the alliance, the type of alliance, the number of major powers in the alliance, and periods of world war.”35 D. Scott Bennett also confirms this finding, concluding that “alliances involving more liberal states are likely to last longer than alliances involving less

150

CHAPTER 6 36

liberal states.” Even after security variables are added as controls, democracy remains a vital determinant of alliance duration.37 The one cautionary note comes from Brian Lai and Dan Reiter, who test whether democracies are more likely to ally with each other.38 They conclude, somewhat surprisingly, that they are not. Their finding casts some doubt on the notion that democracies forge distinctive partnerships through credible commitments. Theirs is not a test of alliance duration, however, and it does not directly challenge the findings of Gaubatz, Reed, and Bennett. The findings on alliance durability are uniformly strong and uncontradicted. They confirm our contracting explanation: democracies make peace, just as they make long-standing alliances, because they are reliable partners.

Learning Effects and Conflicts Early in Relationships Because of learning effects, longer periods of joint democracy should produce more stable peace. Corollary: More serious, militarized conflicts between democracies should occur early in their relationship and diminish over time. There is no reason to expect long periods of joint democracy should produce harmony. But, according to our contracting explanation, pairs of democracies should develop better, cheaper, more reliable ways to manage differences when they do arise. They should fight fewer wars and threaten each other less as they learn to sort out conflicts and manage their necessarily incomplete agreements. Frictions should be highest early in these bilateral relationships, when states know less about their partners and have prudential reasons to be wary. Common norms may help, but they cannot overcome states’ normal calculus of interests or their reasonable assessment of risks. In other words, joint democracy produces a slow, hard path to peace, not an instant panacea based on new ideas and shared values. Interactions should improve steadily as democracies negotiate durable bargains with each other, diminish mutual fears, and gain confidence that they can defend their own interests and manage disputes without military threats or actual violence. This learning effect predicts, for example, that the unbroken relationship between the United States and United Kingdom should produce a stronger, deeper peace in the late 1800s than earlier in the century, and that their peaceful relations should be even stronger in twentieth and twenty-first centuries. Nor is this relationship unique. The same patterns should be evident in diplomacy among other democracies. The path leads from a cold, deterrent peace to a warmer, more durable one. With enough time and positive experience, it may even lead to a community of joint security interests, where war becomes unthinkable. Grounded in an increasingly dense network of mutually profitable relations and agreements, their norms and revised beliefs reinforce a growing

EXTENDING THE ARGUMENT

151

confidence that they can capture greater benefits at lower risks—a virtuous circle that strengthens the democratic peace. The clearest findings about these conjectures come from Lars-Erik Cederman, who estimates both static and dynamic logit models using Correlates of War and Militarized Interstate Disputes data. Because Cederman is interested in long-term learning processes, especially among democracies, he concentrates on “dispute propensities” rather than outright war. There are simply too few wars among democracies to allow him to examine any learning trends.39 With a static model, using conventional control variables, he confirms the standard finding: militarized disputes are significantly lower when both countries are democratic. What is new is Cederman’s dynamic model, which allows him to explore learning in relationships between different kinds of regimes. All kinds of pairings seem to learn something, but democracies learn more, faster. “Even in the presence of statistical control, democratic states still experience a noticeably faster learning rate among themselves than do other dyadic combinations.”40 These findings are noteworthy because Cederman controls for alliance ties and differences in national military capabilities, the Realist variables Joanne Gowa emphasizes. Because these results are sensitive to a few cases in the nineteenth century and because democratic dyads become “saturated” with learning after several decades, Cederman extends his model by looking at new dyads as they emerge and mature. The key question here is whether each additional year of common democratic history makes two states more peaceful toward each other. Once again, all dyads—democratic or not—show some learning, especially during the Cold War period. But democracies show far more than others. “The nondemocratic and mixed categories were considerably more conflict-prone than all-democratic dyads,” Cederman reports. For dyads where both countries have been democracies for more than fifty years, there is almost no conflict. Newer democracies, with a shorter history of relations with their new peers, start at much higher levels of mutual conflict but learn quickly. These newer democratic dyads show a steeply falling curve of dispute propensity.41 Cederman’s central conclusion strongly supports our conjecture that contracting among democracies grows stronger and more reliable as they draw on their experience working together and forge better mechanisms for governing their relationships. These findings suggest two other important inferences about the democratic peace. (1) As relationships between democracies mature, the basis for peace shifts from military deterrence to more stable, voluntary bargains. Military threats play a steadily decreasing role. (2) Democracies will take more active (and costly) steps to protect their bargains in the early stages of relationships with other democracies, and fewer such steps later in their relationships. As they learn more about each other and develop a good record of compliance rather than opportunism, they can relax some of these costly and unnecessary self-protective mechanisms.

152

CHAPTER 6

Early on, the absence of war between democracies is largely the consequence of military capabilities and deterrence, perhaps supplemented by thirdparty guarantees for smaller states. If they manage to avoid war and their relationship strengthens, then the role played by military capabilities and threats should recede markedly. Deterrence is replaced by durable agreements, grounded in the parties’ confidence that they can settle even serious differences without military threats, crises, or fighting. This conjecture is consistent with growing research that shows “the substantive and pacific impact of democracy on war increases over time.”42 The same confidence should allow democracies to design agreements that are less encumbered by fears of cheating, breach, or renunciation. This sequence of increasingly secure contracting is a plausible causal explanation for the progressive deepening of peace among democracies as their relationships lengthen. Emanuel Adler calls this deepening “seasons of peace,” echoing the biblical language of Ecclesiastes. It is a gradient running from absence of war to stable peace and, ultimately, to strong, pluralistic security communities in which war among the members is unthinkable.43 Do dictatorships, monarchies, and one-party states move along the same gradient? Only part of the way. They can—and often do—secure peace based on deterrence and military threats. So do mixed dyads such as the United States and Soviet Union, where deterrence became increasingly stable because of their mutual capacity for nuclear retaliation. In this case, as in so many others, deterrence becomes more stable as bilateral relations mature and participants learn by experience. But there are limits. Dyads that contain nondemocracies are far less likely to achieve stable peace because it is so much harder for them to reassure each other that they are trustworthy partners. They face greater risks of surprise defections and preemptive strikes because nondemocracies are more opaque. Their policy processes and often their underlying preferences are hidden from view. Their behavior, including broken promises, is also more likely to be hidden. Moreover, dictators and strongmen are not limited by constitutional processes, which democracies find so useful for self-binding. If such leaders break commitments, they face no audience costs for lying to the electorate. Yet for all their unfettered power, these same regimes are often brittle, more threatened by coups and revolutions than democracies are. No leader of modern Iraq, for example, has died of natural causes44—strong man today, dead man tomorrow. These structural features of autocratic politics make long-term bargains more difficult to secure. There is simply more risk that policies made in secret will catch partners by surprise. There is also greater risk that the regime itself will change abruptly and discontinuously. These are fundamental, cumulative barriers to a lasting peace. They block the path where deterrence can evolve into a true security community, where a peace based on mutual fear can become a deeper peace based on mutual confidence.

EXTENDING THE ARGUMENT

153

New Democracies (Democratization) New, weak, or marginal democracies should not be part of the democratic peace because such states cannot eliminate commitment problems. New democracies proliferated in the 1980s and 1990s, mainly in Latin America and Eastern Europe. International relations scholars naturally want to understand the implications for war and peace and a variety of other relationships. Will these countries’ newfound commitment to democratic norms and institutions mean that they, too, are incorporated in the democratic peace? Or will their instability and immaturity lead to more conflicts? This debate was launched in 1995 by Edward Mansfield and Jack Snyder, who studied democratization over the preceding two centuries.45 They find that states moving toward democracy are considerably more war-prone than the older, nondemocratic regimes they replace.46 The effects are most pronounced when countries make sharp transitions from full autocracy to full democracy.47 Mansfield and Snyder conclude that although “mature democratic states have virtually never fought wars against each other, promoting democracy may not promote peace because states are especially war-prone during the transition toward democracy.”48 Their findings are a warning to those who naively assume that spreading democracy eliminates war. The underlying problem, according to Mansfield and Snyder, is that many new regimes cannot cope with surging demands for political participation. Their analysis mirrors Samuel P. Huntington’s critique of rapid political development.49 For Mansfield and Snyder, as for Huntington, the governments of developing states tend to be weak and poorly institutionalized. They are ill equipped to meet strong, often nationalistic demands, which elites sometimes stimulate for their own purposes. Political coalitions, which help channel demands in mature democracies and mediate between the public and the government, are less effective here because they are still in formation and frequently shift. The cumulative result is greater political instability, which raises risks for surrounding states. Michael D. Ward and Kristian S. Gleditsch revisit several of these issues. Their conclusions are more optimistic and nuanced. The key, they find, is the smoothness of the transition to democracy. When the transition is steady, democratization lowers the risks of war. When it is rocky and occasionally reversed, then outcomes vary considerably and the risks of war may actually increase.50 Such uneven transitions are all too common, as Ward and Gleditsch note. This research poses important questions about how new regimes fit into the democratic peace. Most of it, however, is limited to individual states, not pairs of democracies, old or new. Mansfield and Snyder ask whether a new democracy is more likely than its predecessor to be involved in wars with any neighbor.51 Ward and Gleditsch also focus on single states, not dyads. As a result, their findings do not tell us whether new democracies are more or less likely

154

CHAPTER 6

than their predecessors to go to war with other democracies. To answer that, we must look directly at pairs where one or both countries are new democracies. Research on this issue is surprisingly sparse. Only Paul Senese has conducted studies directly on point, dealing with dyadic conflicts among new democracies. These new regimes, he concludes, are neither fully inside the democratic peace nor fully outside it. Because there are so few wars among democracies, Senese concentrates on Militarized Interstate Disputes. He uses data from 1816 to 1986 and runs additional tests on the post–World War II years. Unlike other studies, which look separately at democracy and regime maturity,52 Senese is primarily interested in their interaction. That is, he is concerned with relationships involving democracies (and nondemocracies) at different levels of maturity. His ordered logit models yield strong results for both historical periods. His central finding: “The effect of democracy on the hostility levels reached in disputes is conditional” upon the regimes’ maturity.53 These results point in the same direction as Cederman’s work on militarized disputes. As democratic dyads mature and gain experience dealing with each other, their relationships become more pacific, according to Cederman. The same trend is evident among nondemocratic states, although at higher levels of hostility and slower rates of learning. Cederman concludes that “democracies are indeed faster learners when interacting among themselves, and their relations become more peaceful with common experience of long duration.”54 Both these studies examine Militarized Interstate Disputes. It is also important to look directly at warfare. The MIDs data set that Senese uses counts only four wars among two democracies: the Spanish-American War (1898), the Lithuanian War of Independence (1920), the Second Kashmir War (1965), and the Cyprus War (1974).55 All have been discussed repeatedly in the literature, almost always to object that these are not really wars between democracies. I have done so myself in discussing the Spanish-American War, and I think there are serious objections to the others, as well. But there is another important point to be made here. Each of these wars involves a regime that is, at most, a new and still marginal democracy trying to establish its political footing. Take the Second Kashmir War, in which Pakistan is coded as a marginal democracy. Since the statistics are compiled on a yearly basis, they do not reflect that President Mohammad Ayub Khan seized greater power just before the war. In the next annual coding, Pakistan is clearly nondemocratic and is coded that way. My point here is not to debate whether this is really a war between democracies. It is rather to indicate that this war, like the SpanishAmerican, Lithuanian, and Cyprus wars, involved regimes that were young, unstable, and (at most) on the margins between democratic and nondemocratic. That is exactly where our contracting theory says the democratic peace should be weakest.

EXTENDING THE ARGUMENT

155

Enduring Rivalry Long-term rivalry among democracies should be very rare. Long-term rivalries produce nearly half of all wars, and there is now cumulating evidence that democracies are almost never engage in such rivalries with each other. Several scholars have conducted major research on long-term rivalries, using slightly different definitions and terminology. Gary Goertz, Paul F. Diehl, and William Thompson treat them as “enduring rivalries.”56 Jonathan Wilkenfeld and Michael Brecher call them “protracted conflicts.”57 What is common is the idea that pairs of states sometimes engage in repeated disputes, which are difficult to defuse, often violent, and frequently escalate to war. Goertz and Diehl have done the most extensive research. They say rivalries are “enduring” when a pair of states engages in six or more militarized disputes over two decades or more.58 Such long-term rivalries are rare, but when they do occur, they often lead to war. That is markedly different from isolated rivalries, characterized by only one or two episodes of militarized conflict. Since 1816, some 880 isolated rivalries have led to 14 wars. By contrast, there have been only 63 enduring disputes, but they have produced 39 wars. Within an isolated rivalry, there is at most a slim chance of war; within an enduring rivalry, the likelihood is about 60 percent.59 Put another way, only about 5 percent of all disputes are “enduring,” but they have generated half of all wars since 1816.60 These are striking findings. They suggest a strong nexus between long-running disputes, frequent displays of force, and actual fighting. Prolonged antagonisms feed suspicion, block settlements, and produce wars. Where do democracies fit into this? The findings are straightforward. They have almost no long-term rivalries with each other. There are only two exceptions: (1) the United States and United Kingdom until the 1860s; (2) France and the shaky Weimar Republic in the 1920s and 1930s, a continuation of the Franco-German rivalry dating back to the mid-nineteenth century. In other words, it has been 150 years since two stable democracies were long-term rivals. Moreover, two established democracies have never started a rivalry. The Anglo-American rivalry begins before 1816, when Goertz and Diehl’s data starts. It ends in 1861. There were no threats of force after that, although the negotiated conclusion to their rivalry came a decade later when they agreed to the Washington Treaty and Geneva Arbitrations, which resolved all remaining issues in North America. Later, when the United States expanded its influence in the Caribbean, the two countries disagreed about the boundary between Venezuela and British Guiana but resolved that without any display of force. Indeed, the resolution led to a peaceful transfer of hegemony in the Caribbean. The Franco-German rivalry long pre-dates democracy in either country. Goertz and Diehl say the rivalry began in 1850 and ended in 1945, with the destruction of Nazi rule. Still, there was a brief period in the 1920s and early

156

CHAPTER 6

1930s when Germany and France were both democratic. From the end of World War I until Hitler’s consolidation of power in 1933, Germany was a fragile democracy, continuously challenged by local antidemocratic forces. The failed treaties ending the Great War left Germany temporarily weakened and permanently aggrieved. They also produced a wobbly regime, threatened by fascists on the right and communists on the left. There were frequent bouts of rioting, street fighting between organized paramilitaries, and numerous coup attempts.61 France, a much more stable democracy, was understandably worried that neither the Versailles Treaty nor the Weimar Republic were adequate safeguards for its own security. Weimar was, by turns, unable and unwilling to meet its obligations under the Versailles Treaty. When Germany failed to pay war reparations in the early 1920s, France briefly occupied the Ruhr Valley to extract payments in kind, mainly coal. Diplomatic relations improved after the Locarno Agreements of 1926, but disintegrated entirely when Hitler took power and began his systematic expansion and remilitarization. Openly contemptuous of Versailles and Locarno, Hitler quickly revived the Germany army and sent it to reoccupy the Rhineland, bordering France, in direct violation of German treaty commitments. By word and deed, he made clear that Nazi Germany was not bound by any prior agreements. The interwar tensions between France and Germany closely match the logic of the contracting explanation for the democratic peace. Weimar was never a stable constitutional democracy and never demonstrated a durable commitment to a negotiated settlement with France. France, unwilling to rely on paper pledges from Germany, wanted to weaken it permanently and, failing that, to contain it diplomatically and militarily. In the chaotic years immediately after World War I, the fragile Weimar Republic refused to meet the international obligations imposed on it by the victors (which is why France occupied the Ruhr). The rivalry ebbed briefly in the mid-1920s, precisely when Weimar was strongest. It was then that Germany pledged itself to a comprehensive negotiated bargain with France (the “spirit of Locarno”).62 That soon buckled as the regime was undermined by fascists and, to a lesser extent, by communists. It collapsed entirely when Hitler established one-party Nazi rule. Other than the Franco-German and Anglo-American cases, all democratic rivalries have been brief and kept to low levels of violence. The record is markedly better than rivalries involving one or more nondemocratic states. Out of 1,166 total rivalries since 1816, only 62 involve two democracies. According to Goertz, Diehl, and Paul Hensel, most were brief and involved only a few disputes.63 Among states that became more or less democratic during the course of a rivalry, “militarized disputes were generally much less likely during periods of joint democracy. . . . Among rivalries that are transformed to joint democratic status . . . the conflict level tends to be lower when there is a democratic dyad, and the rivalry almost always ends shortly thereafter.”64 Scott Bennett confirms this point: joint democracy helps end rivalries.65

EXTENDING THE ARGUMENT

157

Better at Every Level of Dispute Democracies should have some militarized disputes, but there should be fewer at every level along the escalation ladder, as conflicts become more threatening and violent. There is strong evidence that pairs of democracies not only fight fewer wars; they engage in fewer militarized disputes than do nondemocracies or mixed pairs.66 In fact, the relationship is now so well established that “joint democracy” is used as a control when looking at other sources of militarized disputes.67 When disputes among democracies do occur, there is some evidence that they are less likely to escalate to full-scale war than disputes in other dyads. That, however, does not fully resolve the question of escalation short of war. After all, there are many levels of militarized disputes, ranging from verbal threats to displays of troops to blockades, raids, and border violations. What is now contested is whether democracies perform better at halting escalation within this middle range between threats and war fighting. In separate studies, Paul Senese and William Reed make the case that they do not. Both Senese and Reed replicate the conventional finding that democratic dyads do engage in fewer militarized disputes and fewer wars. But, according to Senese, once two democracies have actually begun a militarized dispute, they are as likely to escalate as nondemocracies are.68 In fact, once democracies have actually threatened to use force against each other, they may even be slightly more prone to escalation.69 Reed, using a different model, reaches similar conclusions. Democratic pairs are less likely to engage in militarized conflicts and less likely to go to war. But once militarized conflicts have actually begun, joint democracy has no pacifying effects on escalation.70 His conclusion is that “jointly democratic dyads avoid war because they rarely become involved in militarized disputes.”71 We made the same observation earlier in the section on learning: democracies rarely become involved in militarized disputes. In another study, Reed and Robert A. Hart explain that the “dyadic selection effect” is crucial, since pairs of democracies largely select themselves out of militarized disputes in the first place.72 William J. Dixon is more sanguine about escalation among democracies. His argument stresses shared norms of “bounded competition” in domestic politics. If foreign partners are willing to reciprocate, he hypothesizes, then democracies will extend these same norms of mutual restraint to their international dealings. The effect is that democracies have “a mutual basis for contingent consent,” so their disputes “should evolve somewhat differently than do disputes between states not sharing these norms (i.e., between democratic and nondemocratic states or between two nondemocratic states).”73 His study looks at conflicts between 1945 and 1974, trying to explain which of six sequential phases they reach. He uses probit models with a variety of controls, including

158

CHAPTER 6

alliances between the parties, previous military hostilities, and third-party mediation, and tries to purge the results of cross-phase dependence among the observations. His major conclusion is that democratic dyads are “very effective” in promoting settlements; few conflicts rise beyond the lowest phases.74 These results are stable across alternative specifications of the model.75 In a subsequent study, Dixon extends his data to 1984. Using binary logistic regressions, he again finds mutual democracy deters escalation and encourages more settlements by peaceful negotiations.76 Jonathan Wilkenfeld and Michael Brecher approach the same issue by studying crisis behavior. Their extensive research shows that “as the proportion of democracies among the actors in an international crisis increases, the likelihood of violence decreases.”77 In fact, violence is twice as likely in crises with few democracies, compared to those with many democracies.78 Moreover, when actors do use violence in crises, “the severity of violence lessens as the prevalence of democracies in the crisis increases. . . . These crisis-related findings add a significant new dimension to the theory of democratic peace.”79 Another study of crisis behavior, by Joe Eyerman and Robert A. Hart, finds that democratic dyads go through fewer stages of crisis escalation than do nondemocratic or mixed dyads.80 Manus I. Midlarsky summarizes this scholarship in his introduction to the Handbook of War Studies II. “Not only is there now strong evidence for the democratic peace between dyads, but also that even the escalation of low-level militarized disputes between democracies to higher levels is rare.”81 In fact, they may all be right. Escalation by democratic dyads is very rare, but that is because they manage to avoid serious military conflicts and longterm rivalries in the first place. There seem to be fewer disputes between democracies at each stage of the escalation ladder, as our original conjecture stated. Our rationale, however, is unproven and may be wrong. We posited that democracies avoided escalation because they were better at reaching agreements as conflicts became more costly. The evidence on that point is mixed. Dixon finds that pairs of democracies are better at containing escalation; Wilkenfeld and Brecher find they are also better at managing crises. But Senese and Reed find that they are no better than other states at halting escalation once they have actually begun a militarized dispute, and are probably worse. Their findings suggest that democracies continue to push for advantage and are willing to escalate as a way to signal their resolve, even against other democracies. One possible explanation is that democratic leaders face higher costs (the risk of losing elections) if they make public threats and then back down. The same audience costs that make democratic threats and promises credible might impel them to escalate, at least until they approach the high costs of war itself. In short, the evidence on escalation is ambiguous. What is clear is that democratic dyads rarely reach the highest rungs of violence, mainly because they avoid militarized disputes in the first place.

EXTENDING THE ARGUMENT

159

Near Misses Involve Fundamental Issues “Close calls” between democracies should involve fundamental, existential issues since less important matters can be resolved through agreement. The rationale here is that democracies, because they are better informed about each other, should engage in few conflicts because of poor communication or unrealistic fears. They should be better at finding compromises with each other, leaving only the most difficult issues unresolved by normal diplomacy. There are no comprehensive studies dealing with this general proposition. The best we can do is piece together some suggestive evidence. First, there is some support for the idea that democracies largely avoid miscommunication and escalating cycles of fear when dealing with each other. Joint democracies have entirely avoided arms races and worsening security dilemmas. For mature dyads, those are the fruits of their successful relationship. They enjoy high levels of confidence in each other. For less mature dyads, these findings may indicate they are successful in sending costly signals of benign intent, which democracies are better placed to send and receive (a point we take up in the next section). What about the “close calls” where democracies have narrowly avoided war with each other? The evidence here generally runs against the conjecture. All the U.K. cases concerned its colonies and, even then, were not central to the maintenance of its empire. Fashoda involved the protection and extension of its informal control in Egypt. The running disputes with the United States in the nineteenth century generally involved Canada and U.S. territorial expansion. The one crisis that did involve a major issue for Britain was the Trent affair, where the United States seized Confederate diplomats aboard a British ship. For Britain, this posed a challenge to its unmolested control of the seas, the heart of its security policy. For the United States, which prudently backed down, the Trent affair involved little more than preventing Confederate diplomats from reaching Europe. As in all prominent disputes, questions of national honor and reputation also arose. London did not challenge Washington over the vital issue of blockading Southern ports. Nor did London challenge the basic U.S. policy of geographical expansion across the continent in the nineteenth century. Britain slowly conceded these issues to the United States, as long as the United States respected Canada’s integrity and Britain’s economic interests. It made no effort to contain the United States in eastern North America or to foster multiple states south of Canada, a real possibility after Texas became an independent state and again during the early years of the Civil War. France’s close call at Fashoda was also a marginal colonial venture. On the other hand, its occupation of the Ruhr Valley in 1923 was sparked by a central security issue: enforcing provisions of the Versailles settlement designed to

160

CHAPTER 6

weaken Germany and compensate France for its war losses. For Weimar, the issue was equally important: keeping its territory free from foreign occupation. There is no clear pattern here indicating that conflicts between democracies are over vital issues. Some, like Fashoda, were marginal to both sides. Others, like U.S. expansion, were vital to one side but not the other. In these unbalanced cases, the winner was typically the democracy that considered the issue more consequential because it touched on central security matters.82 That may be indirect evidence that democracies communicate their vital interests clearly, both as senders and receivers of these messages.

Clear-Sighted Democracies should assess each other’s capabilities and intentions more accurately than nondemocracies do. There is mounting evidence that democracies are, in fact, more accurate at assessing their adversaries. They also appear to be better at sending clear, truthful messages to adversaries and partners. The key evidence here is that democracies are disproportionately successful in choosing conflicts they can win. The most extensive evidence is on wars. A number of studies have shown—and none have refuted—that democracies are more likely to win wars in which they participate.83 (There is also evidence that democracies fight shorter wars and suffer fewer casualties.) Other studies find democracies are similarly successful in their militarized disputes and international crises.84 Naturally, these studies control for confounding variables. The simplest explanation is that democracies are better than other states at assessing conflicts. Because democracies are better at assessment, they are also better at selecting which conflicts to engage in, either by initiating the conflict or by responding vigorously to others’ challenges. This is exactly what most political philosophers of democracy expect. John Stuart Mill is particularly forceful in arguing that democracies’ open debate, greater public information, and multiple viewpoints lead to systematically superior judgments on public issues. Modern political philosophers support that conclusion.85 It is possible, however, that even though democracies choose conflicts more carefully, that is not because they are more perceptive. They are simply more risk-averse. Unless they are very confident they will succeed, they refuse to gamble on war, or, for that matter, on militarized disputes or crises. A dictator could, in theory, have exactly the same judgment about the probabilities of success but choose to gamble with worse odds; perhaps he is not held accountable for losses but would gain special benefits from victory, such as rents for the state he controls and probably plunders.86

EXTENDING THE ARGUMENT

161

Risk aversion, if it really does drive democratic decisions, is important because it opens a wider range of mutually profitable bargains short of war. Conversely, two states that are risk-acceptant may prefer the costly gamble of war. That is, each would rather have a small chance of winning the jackpot than receive a sure, negotiated outcome, even though it has a higher expected value. For example, each would rather take a 40 percent chance of winning everything than accept a fifty–fifty split. Depending on how risk-acceptant states are, they might find no common grounds for agreement to avert conflict.87 As for sending messages, there is now ample evidence that democracies are less likely to bluff. The reason, as we have consistently observed, is that democracies have less opportunity to bluff because they have less private information. They play the game of diplomacy with more of their cards face up. Democratic leaders also confront greater penalties if they bluff and fail. Their domestic electorate holds them accountable, either for imprudence or outright incompetence. Using data on militarized disputes between 1816 and 1980, Kenneth A. Schultz shows that democracies engage in fewer bluffs than other states. Because their threats are more likely to be genuine, Schultz argues, their targets are more likely to concede.88 Schultz recognizes, though, that he cannot rule out another explanation for these concessions: democracies choose their targets more carefully. The evidence, then, generally supports our conjecture that democracies are more clear-sighted, at least in the sense that they are better at selecting winnable conflicts. They also appear to be better at sending credible signals to nondemocratic states. What is unclear is whether democracies’ success in selecting targets is due to better judgment, greater risk aversion, or both.

Shocks When democratic regimes are threatened by external shocks and instability, the democratic peace should be weaker. The rationale here is that shocks raise questions about whether democracies will remain in power, and that makes them less reliable long-term partners. The evidence on learning and newly formed regimes supports this rationale, but it is indirect. It confirms that the democratic peace is weakest when partners are new and their futures uncertain. That suggests our rationale is correct, but it does not deal directly with external shocks. More evidence comes from the immediate aftermath of world wars, when fledging democracies were established. In those difficult settings, when political shocks and aftershocks were enormous, the victors maintained occupation forces, trying to establish the new regimes and make sure they adhered to

162

CHAPTER 6

peace treaties. The victors were clearly unwilling to rely solely on agreements with other regimes—democracies or not—in these turbulent and uncertain settings. In such strong winds, there was too much danger that paper agreements would blow away. That is suggestive evidence, but it is far from conclusive. The extensive empirical literature on the democratic peace has not yet addressed the issue of external shocks, so we do not know if economic depressions or regional wars affect the stability of peace among democracies. Our conjecture therefore remains unconfirmed.

Leadership Change Changes of leadership pose greater concerns in bargains involving nondemocracies. Changes of leadership pose two major challenges for agreements. One is the possibility that new regimes will reverse policies and refuse to carry out their predecessor’s undertakings. The other is that changes raise the overall level of uncertainty at home and abroad. Both challenges are more serious when changes are discontinuous, as they are after revolutions. Neighbors often fear a new regime will launch a unilateral attack as it tries to build public support at home. Although such assaults are actually quite rare, as Stephen Walt and others have shown, the fears they engender are real, and so are the risks of war they pose.89 New regimes are more likely to do what Robespierre urged French revolutionaries to do in 1792: save their limited resources for more immediate challenges at home.90 Lenin made exactly the same argument to his fellow Russian communists in 1917 and 1918. Foreign powers were hardly reassured. They saw far more than Lenin’s prudence, and they were worried by what they saw. The new communist regime was busy overturning the most basic policies of Tsarist Russia. Their most stunning act was to stop fighting the Great War, which allowed the formidable German Army to concentrate its forces against the Allies in France and Belgium. At Brest-Litovsk, Leon Trotsky accepted the harsh peace terms dictated by Germany. At the same time, Trotsky declared that Russia’s new government would no longer conduct normal diplomatic relations. One concrete manifestation of this radical shift was the new government’s renunciation of the large debts Russia owed its foreign creditors.91 To make matters worse for the Allies, the Bolsheviks encouraged uprisings elsewhere, just as the French revolutionaries had. Although the communists offered little direct aid (they had none to give), their improbable success gave others hope of overthrowing their own rulers. The Bolsheviks’ very existence, in power against all odds, was a potent symbol to workers’ movements across Europe. To all the Great Powers, then, it was becoming painfully clear that

EXTENDING THE ARGUMENT

163

this new communist government was different not only from the Tsarist regime but from any European predecessor.92 It was simultaneously fragile, novel, and menacing. To cope with this menace, Britain, France, and America all fielded troops in the Russian Civil War (1917–20), siding with the counterrevolutionary White Russians. The Allies’ initial aim was to protect their troops and mate´ riel already in Russia for the Great War. They also hoped to bring their wayward ally back into the fight against Germany.93 When that failed, Britain stayed on (at Churchill’s urging) to stop the spread of communism.94 Japan, meanwhile, attacked around Vladivostok on the Pacific Coast, hoping to protect its existing colonial interests and win additional land in Siberia.95 The Russian Revolution was one of the profound events of the twentieth century, and it is misleading to compare ordinary leadership changes to it. It does, however, illustrate a characteristic sequence of troubling uncertainties and defensive actions that often arise when leadership changes could mean far-reaching alterations in politics, economics, and society. It is a sour moment for reaching diplomatic bargains or projecting them reliably into the future. Other states, threatened by their neighbor’s instability and perhaps by its new ideology, press their own counterclaims. They may also see their neighbor as temporarily weak and vulnerable to attack. That is exactly what happened after the Iran’s Islamic Revolution in 1979. Iraq’s secular regime was threatened by its neighbor’s religious revival and particularly by its impact on Iraq’s minority Shiite population. At the same time, Iraqi leaders and many other foreign observers concluded that the revolution had seriously damaged Iran’s military capabilities. After all, the army had been purged of its most experienced but politically unreliable officer corps, and the entire Iranian economy had been disrupted. Zealous amateurs were now running everything. Facing this combination of threats and opportunities, Iraq’s Saddam Hussein moved quickly to seize the Shat-el-Arab delta with its rich oil deposits. So began a horrific war that lasted nearly a decade. In ancien re´ gime Europe, revolutionary changes like these were rare, but wars of succession were commonplace. At stake was who ruled and whether he governed all the domains of his predecessor. Neighbors often pressed their own choices for monarch or tried to grab contested territories when their relative strength was greatest. Frederick the Great of Prussia did exactly that in Silesia, as young Maria Theresa of Austria struggled to assert her control over the sprawling Habsburg dominions. As a female, Maria Theresa could inherit the Habsburg crown. What was uncertain was whether she could consolidate control in German lands beyond her family’s holdings, as her male predecessors had done. Her father had anticipated the problem and spent most of his reign trying, with only partial success, to ensure her inheritance by making treaties with surrounding monarchs. When her father died, Frederick II of Prussia immediately moved

164

CHAPTER 6

south to capture Silesia, the valuable Habsburg territory bordering his state. It took him three major wars, involving all the Great Powers, to complete his conquest. His hard-won victories stripped Austria of one of Europe’s richest regions. It earned him the title of Frederick the Great and transformed Prussia into a Great Power. Another set of succession issues arises if a leader might be overthrown. If autocrats are failing politically but are unconstrained constitutionally, they may pursue high-risk strategies of conflict. They could initiate a war or militarized conflict as a desperate gamble to resurrect their political fortunes. Although the chances of victory may be poor, it is their last, best hope to remain in power.96 It is like a poor man betting his last dollar in the lottery. Democratic leaders, facing electoral defeat, have similar incentives to take long-shot gambles. But there is a crucial difference: they face formidable institutional barriers. Within constitutional democracies, divisions of political authority make arbitrary, high-risk choices less likely. Established constitutional procedures check the executive’s whims; independent politicians are positioned to obstruct them; and the free press is eager to expose a leader’s egotism and folly. No wars among democracies arise from this logic, nor, so far as I know, do any militarized disputes. In fact, what electoral challengers fear most in the weeks before an election is not the gamble of war but a self-serving diplomatic triumph by the incumbent. In the United States, it even has a name: the October Surprise, since it is likely to come immediately before the November elections. The award for best surprise goes to Henry Kissinger’s misleading press conference on the Vietnam peace talks. Just days before the Nixon-McGovern election in 1972, he dramatically told America that secret negotiations in Paris had produced excellent news. For the first time, he said, “peace is at hand.”97 There was no time to weigh his statement carefully before the election, much less evaluate the terms of any prospective accord. It was aimed squarely at an American electorate hungry for peace. Its real meaning, as a North Vietnamese negotiator later explained, was to tell voters “there is no more Vietnam question, so they can elect Nixon as a hero of peace.”98 Some hero, some peace. There have been few quantitative studies assessing the impact of leadership changes on war and peace. One, by Suzanne Werner, supports our general conjecture. “Unexpected, fundamental changes in the government of either belligerent tend to increase the risks of recurrent conflict.”99 According to Werner, that applies only to nonconstitutional leadership changes. Constitutional changes of leaders have little effect because they are both anticipated and limited in scope. They do not increase the chances of renewed conflict. The reason, says Werner, is that leadership changes affect the likelihood of conflict through “expectations and uncertainty. Small changes in government do not fundamentally alter expectations . . . [or] significantly increase levels of uncertainty.”100 Violent changes of leadership, on the other hand, dislodge expecta-

EXTENDING THE ARGUMENT

165

tions and raise uncertainty, and so increase the chances that peace will fail. Werner adds one important reservation: “This effect . . . holds only for recurrent conflicts short of war.”101 As we have already noted, there have been no recurrent conflicts between democracies since the early 1920s and none between established democracies since the 1890s. Almost all recurrent conflicts involve nondemocratic states. What Werner’s study indicates is that leadership changes in nondemocratic states should be a source of concern in negotiations to prevent war, especially when long-term rivals are involved. That offers some support for our conjecture. What is missing, however, is evidence (either from case studies or statistical analyses) that states recognize these risks and structure their bargains differently to cope with them.

Bilateralism When issues are bilateral and within the control of two democracies, their bargains will be stronger and more reliable than if third parties are involved or the external environment (beyond their control) poses a major risk. According to this conjecture, democracies should be most successful dealing directly with each other about issues they jointly control. They can forge more reliable bargains when they face each other directly across the table. They can resolve emerging problems before they arise (by well-planned contracts) or afterward (by effective governance mechanisms). In practice, they forge both kinds of bilateral arrangements and control them effectively. What about multilateral arrangements? Democracies should perform better than other states because they generally adhere to their commitments and are less able (and probably less willing) to hide any cheating. Even so, multilateral arrangements pose real problems for extending the democratic peace. When many countries interact, information problems become more complicated. It is often hard to figure out which states are creating difficulties, which are violating their obligations, and which are merely responding to others’ provocations. In addition, there is the thorny problem of sharing collective burdens. The issue is all too familiar: countries try to capture the benefits of collective action without paying their full share. It is always difficult to provide public goods. Finally, it is particularly hard to provide one essential public good: punishing violations and noncompliance. The most immediate problem is to penalize direct violations of rules and agreements (sanctioning first-order violations). Unfortunately, enforcement is a collective-goods problem in its own right. Meting out punishments is costly, and the benefits are reaped by all. If that problem is not solved, the entire system of sanctions will unravel. With no sanctions, some states are likely to cheat and others will eventually tire of being exploited. That means another layer of sanctions is needed, one that

166

CHAPTER 6

strengthens the primary system of sanctions by penalizing states that fail to join in the general task of punishment. Of course, that, too, is a collective action problem. So there are three fundamental problems transforming sound bilateral bargains into multilateral ones: • Information • Provision of public goods • Enforcement: providing sanctions for direct violations and for failing to punish violators These persistent problems mean that even though the democratic peace is real and important, it is based on a web of strong bilateral relationships and a penumbra of shared values. It falls well short of Kant’s vision of a “pacific union” of all democratic states. To transform it into one requires solutions to some of the most difficult and dangerous problems in international politics. The most obvious and persistent problem is collective action. It is difficult to act in unison because democracies, like all other states, have incentives to free ride. They are tempted to avoid paying for joint burdens. All states, including democracies, have incentives to shirk their collective responsibilities, pass the buck to others, and avoid the costly tasks of punishing those who violate their commitments. Perhaps the clearest example of these multilateral difficulties comes from NATO. This should be an “easy case” for democratic cooperation since it is one of the most durable and successful alliances in history. It is dominated by large, established, and wealthy democracies that have worked well together for more than fifty years. There is considerable trust among them, bilaterally and multilaterally. No alliance member threatens another, with the possible exception of Greece and Turkey. Nonetheless, NATO professionals have complained for decades that some members contribute too little to the alliance.102 Writing about NATO’s “burden-sharing problem” is a cottage industry of long standing. These problems are both relieved and compounded by the mechanics of multilateral decisionmaking. They are relieved because big contributors, notably the United States, maintain institutional control commensurate with their contributions.103 Washington gains some important advantages by underwriting a disproportionate share of collective burdens. These individual benefits, plus America’s gains from the joint activities themselves, give the United States strong inducements to pay up. In general, large contributors have a large voice in determining exactly which collective goods are provided. That aids the effort to provide them. Still, there are recurrent problems in multilateral action. One is that decisionmaking is awkward, cumbersome, and slow. NATO’s difficulties waging war by committee in Kosovo made that abundantly clear to members and other

EXTENDING THE ARGUMENT 104

167

observers. The most prominent reason for these high transaction costs is that decision after decision has to be hammered out among states with mixed interests. Another is the cost and difficulty of collecting essential information in multilateral settings. It is hard to tell who is contributing, who is cheating, and who needs to be watched very closely. The problem is not just “who will contribute” to desired public goods, but also who will pay the costs of sanctioning states that fail to comply. Providing these sanctions is critical if collective action is to work. The only alternative is for one or two major states to provide the enforcement and inducements themselves.105 That works only if they have ample resources and a very strong interest in providing the goods. Realists correctly emphasize these information problems in a multipolar world. Many of the same issues arise in multilateral institutions, compounded by problems of decisionmaking and sanctioning. In NATO’s case, the United States’ great size eases the burdens. It is by far the largest and wealthiest participant, with high stakes in solving collective problems and ample resources to do the job. That confirms a standard conclusion of collective action theory. It is much easier to provide public goods when a small group (or, better yet, a single state) benefits from providing them, even if others do not contribute.106 That fits a wide range of modern international institutions, including NATO, where the United States and a few European partners play outsized roles. These findings imply that multilateral action is more difficult when large numbers are involved, when no actors are dominant, and when sanctioning violators is a serious and costly problem. These are not abstract issues. They sank the League of Nations in the 1930s. The United States refused to participate at the outset. Many members were new, unstable democracies, impoverished by the war. Violations of the League Covenant were frequent and serious, going to the heart of members’ security concerns. Key participants were unwilling to pay the individual price—diplomatically, financially, or militarily— that was required to sanction violators. They certainly had no intention of sanctioning the vast majority of members that adhered to the rules themselves but refused to join in collective punishments. When a few members were mildly censured for unilateral aggression, an egregious violation of the League’s Covenant, they simply withdrew from the League entirely and suffered little cost for doing so. The results are infamous and devastating. The ideal of collective security during the interwar years was an illusion. It was a snare for any state that relied on it and a tragic misstep for the Allied victors who, at Woodrow Wilson’s insistence, made it the centerpiece of their peace settlement after World War I. There was little collective action. There was no security. Some of the same problems arose in smaller settings, outside the League of Nations, during the 1930s. Britain and France, the most established democracies in Europe, had peaceful, nonthreatening relations with each other. But Britain failed to help France confront Nazi Germany’s rising power in the mid-

168

CHAPTER 6

1930s. A decade earlier, as part of the Locarno Agreements, Britain and Italy had offered joint guarantees regarding France’s border with Germany. Yet neither France nor the two guarantors was willing to act when Hitler renounced the agreements and began remilitarizing the Rhineland. Italy, under Mussolini, had no intention of confronting Hitler over purely French issues. Britain still could have acted and, together with France, could probably have stopped Hitler at this early stage. To do so, however, was to risk fighting another war on the continent, which Britain was plainly unwilling to do. There was no domestic support for such a major continental commitment so soon after the Great War. Suffering through the Great Depression at the time, Britain’s leaders and voters were unwilling even to pay for rearmament to meet the German threat. Britain’s open discussion had alerted France to these problems well in advance. That was cold comfort to France, which was unwilling to act alone. The 1930s were a low, dishonest decade, but the problems of multilateral action are recurrent. They arise even in better times. They still bedevil Woodrow Wilson’s benign vision of a pacific league of democracies.107 Among themselves, democracies form an increasingly dense web of trustworthy bilateral relations. New democracies, though less-certain participants, are increasingly woven into this web, a task made easier because they are surrounded by so many established democracies willing to help. Unfortunately, when this diverse global collection of democracies is challenged by rogue states, it has real difficulties acting together effectively. These varied democracies have mixed interests and often prefer different strategies. Whatever choice they make will affect participants differently and raise distributional conflicts among them. With so many parties involved, the transaction costs are high, from collecting information to choosing specific strategies. To settle on common policies, including the use of force, they must surmount the hurdles of multilateral decisionmaking. They must also resolve the generic problems of paying for whatever collective action they choose and coping with free riders. All are formidable problems. These problems are why democracies have not been able to convert their pacific relationships into an omnibus alliance that can deter, sanction, and, if necessary, fight—an alliance that not only works internally to preserve peace among its members but also secures them externally against nondemocratic challengers. To work well against aggressive challengers, a union of democracies must have deterrent value, based on credible threats. The web of democracies lacks that capacity and has none in prospect. That leaves the United States, sometimes joined by NATO or ad hoc alliances, to confront the dangers . . . if it chooses to do so. Among themselves, democracies are pacific. Unfortunately, that does not solve the deep-rooted problems of transforming their peace into an effective pacific union or collective-security organization.

7 Conclusion: Reliable Partners and Reliable Peace LET’S QUICKLY REVIEW where we stand and then consider larger issues. The starting point for this study, and for so many others, is a striking correlation: when two countries are democracies, they very seldom fight wars against each other. The first task is to see whether this relationship might be mere coincidence or the product of unrelated factors. That question has been subject to the most sustained inquiry of any issue in modern international politics. Quantitative tests, using a wide range of statistical methodologies, show that the relationship between “joint democracy” and war withstands close scrutiny and stretches back well into the nineteenth century. It is real, and it is not limited to the Cold War era. So the correlation stands. In fact, it seems to be getting stronger historically. It certainly grows stronger as pairs of democratic states gain experience dealing with each other. There have been no wars between established democracies for well over a century and few serious militarized disputes. By the same token, critics have correctly pointed out a number of problems and reservations. One is that the democratic peace has not always been so peaceful. There have been several “near misses,” where democracies came very close to war, mostly in the nineteenth century. In these cases, democracies threatened force, mobilized troops, and avoided outright war only because of prudential calculations. Second, many of the world’s democracies are relatively young. That means our statistics covering pairs of democracies are drawn heavily from the period after 1945. As a result, we are somewhat less confident about the democratic peace before World War II simply because there are fewer cases to evaluate, statistically or qualitatively. Once again, we have to be wary of spurious causation. After all, the years since World War II have been the great age of American world dominance. These questions bear directly on the causes of the democratic peace. My answer is that constitutional democracies avoid war because they have a unique capacity to form effective partnerships that assuage fears, diminish risks, and capture joint benefits. That answer conforms to a rationalist theory of war itself, which stresses the importance of two factors: private information and commitment problems. Democracies do better on both counts. They have significantly less hidden information about their preferences, strategies, capabilities, and resolve. They also have a variety of institutions that make their promises more reliable. As a result, they ought to have a better chance of avoiding war and enjoying a peace dividend.

170

CHAPTER 7

Figure 7.1 Contracting Basis of the Democratic Peace

To evaluate this explanation, we pursue three paths. One is to make sure that, in theory, these factors yield this projected outcome. The second is to see if these causal factors actually appear in the historical cases. Do the theoretical causes match the diplomacy of hard cases where the democratic peace nearly collapsed? Finally, we want to see if the theory does more than explain familiar outcomes. The theory itself can be stated briefly. Democratic structures and procedures, by their very nature, disclose preferences rather than hide them. They make promises more credible and deceit less likely. Audience costs in a democracy also lessen the incentives to renege, since voters hold elected officials accountable for their promises. These promises transcend today’s leaders. Constitutional procedures, plus fairly stable voter preferences, help lock in promises from one president or prime minister to the next. Moreover, the regimes themselves are stable, unlikely to be overturned by revolution or sharp political discontinuities. That is why established democracies can make credible commitments for the long haul. Their continuity, transparency, and credibility all work to reassure partners and facilitate bargains with those who can reciprocate. These bargains between democracies do not stand alone. They are embedded in a dense network of political, social, and economic relationships. These multiple connections give them more diverse information about each other, more opportunities to discover profitable deals, and more ways to sanction broken promises or bad faith. Finally, democracies’ experience dealing with each other and their success upholding agreements creates a virtuous circle. They update their beliefs as they learn how reliable their partners really are. Taken together, these arrangements lessen the dangers of defection or surprise changes, which leave partners vulnerable and discourage them from making bargains in the first place. They lessen the need to bolster agreements against cheating or opportunism, or to avoid making them altogether. Instead,

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

171

they foster secure, long-term bargains between democracies, which allow them to resolve serious differences without war and build deeper, more peaceful relationships. This contracting explanation is summarized in figure 7.1 (from chapter 1). If this explanation is robust, it should offer insights into a range of related phenomena. Looking beyond the democratic peace, we want to see what else our contracting theory predicts and then check these predictions against established research findings. If the predictions hold up, we have additional grounds for thinking democracies are at peace because they are reliable partners. We begin by summarizing these other findings and then consider the larger implications of reliable partnerships among democracies (tables 7.1 and 7.2). Overall, the predictions stand up well. Across a wide range of related topics, they support inferences drawn from the contracting theory of the democratic peace. Democracies have built a separate peace because they are reliable partners. Our conjectures missed cases where democracies escalated their conflicts or came close to war over lesser issues. The evidence is instructive. It shows democracies are willing to dig in their heels. They do so even when they face another democracy, sometimes even when the stakes are modest. This finding needs to be combined with another: although democracies do better at avoiding militarized disputes with each other, they are willing to escalate those they begin. These findings are a problem for theories that heavily stress benign norms and shared commitments to judicial settlement. The same mechanisms that facilitate commitments can, at times, make confrontations worse. Constitutional procedures and audience costs are effective commitment devices, but they encourage democracies to lock horns as well as lock in agreements. These findings on near misses and escalation, although different from our initial expectations, are not fundamentally at variance with our contracting theory of the democratic peace. We expect struggles over distributional questions, as long as the participants are sensitive to the costs of the struggle itself and to the price of nonagreement. The silver lining is that democracies are clear-headed about avoiding excessive costs, better at locating solutions, and more reliable at carrying them out in good faith.

Implications from Contracting Theory With these basic findings in mind, we can return to the causal explanation and flesh out its mechanisms and larger implications. The contracting explanation of the democratic peace emphasizes four aspects of domestic governance that have far-reaching consequences for international affairs: transparency, regime continuity, audience costs, and constitutional procedures.

172

CHAPTER 7

TABLE 7.1 Evidence from the Conjectures Conjecture

Findings

Self-protective mechanisms are stronger and more extensive when agreements involve nondemocracies

Generally supported, although the evidence is not systematic

Alliance duration: Alliances are more durable when only democracies are members

Strongly supported in large empirical studies

Learning effects: Because of learning effects, longer periods of joint democracy should produce more stable peace

Strongly supported in large empirical studies

Conflicts in relationship: Corollary on learning effects: More serious, militarized conflicts between democracies should occur early in their relationship and diminish over time

Strongly supported in large empirical studies

New democracies: New, weak, or marginal democracies should not be part of the democratic peace because such states cannot eliminate commitment problmes

Mixed results; dyads with new democracies are more conflict-prone than older democratic dyads, but are less conflict-prone than nondemocracies; few studies directly on point

Enduring rivalry: Long-term rivalry among democracies should be very rare

Strongly supported in large empirical studies

Better at every level of dispute: Democracies Democracies have fewer militarized disputes with should have some militarized disputes, but each other, but, once involved in disputes, may there should be fewer at every level along the be more willing to escalate short of war escalation ladder Near misses: “Close calls” between democracies Conjecture rejected. Near wars among should involve fundamental, existential issues democracies sometimes involve less important since less important matters can be resolved issues as well as vital ones through agreement Clear sighted: Democracies should assess each other’s capabilities and intentions more accurately than nondemocracies do

Several large empirical studies show democracies are better at selecting winnable conflicts; not clear if that is due to better assessment or greater risk aversion

Shocks: When democratic regimes are Insufficient evidence threatened by external shocks and instability, the democratic peace should be weaker Leadership change: Changes of leadership pose Supported when dramatic changes are possible; greater concerns when bargains involve few studies of bargains involving large numbers nondemocracies of non-democracies Bilateralism: When issues are bilateral and within the control of two democracies, their bargains will be stronger and more reliable than if third parties are involved or the external environment poses a major risk

Clear support but evidence is not drawn from large empirical studies

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

173

This explanation is grounded in the rationalist theory of war, with its emphasis on imperfect information and credible commitments. Transparency, for example, is an information device. It diminishes the likelihood of bluffing, gives partners advance notice of policy changes, and lessens the dangers of surprise. That diminishes the windfall gains from betrayal and lowers the risks to innocent partners. If commitment devices were perfect, the risks of defection would be zero. In practice, they are imperfect for all states, which makes it important to control risks. Democracies have two other important commitment devices. Their constitutions are designed to publicize and lock in commitments. Making commitments difficult to reverse makes them more credible. In democracies, these commitments are also publicized in elections, parliamentary debates, mass media, and public discussion. That creates audience costs for elected leaders, who can be punished for abandoning their promises. The visibility of this process in a democracy allows foreign partners to gauge the depth and limits of commitments. It also allows democratic leaders to make some commitments more credible by deliberately raising the stakes, highlighting their importance to domestic audiences. The idea of “credible commitments” should not be understood narrowly. The usual focus is whether an actor will do what he promises or threatens. Obviously, that is crucial in our analysis. But we are dealing here with a composite (or corporate) actor, the state, so we also need to know if its identity will remain the same or change dramatically. France was a radically different polity in 1785, 1795, and 1805. Many fundamental commitments of the ancien re´ gime, however sincere, were swept aside by the Revolution; those, in turn, were swept aside by Napoleon’s empire. Not so in established democracies. A major element of their credibility lies in their regime continuity. Serious commitments undertaken by one president or prime minister transcend that office holder’s tenure and are not severed by discontinuous regimes changes. Regime continuity strengthens contracting in several ways. It allows states to develop reputations and allows partners to learn about their policy preferences and reliability, particularly in bilateral interactions. It gives partners confidence they can sustain long-term bargains, which will not be overturned by revolution, coup, or other discontinuous change. It also minimizes periods of turbulence and heightened uncertainty at times when the new regime’s preferences, capabilities, and goals are unclear and its very survival may be in doubt. As important as these issues are, our focus is not on the continuity, transparency, or credibility of a single actor, acting alone. It is on the ability of two or more actors to forge reliable agreements, using their institutional resources. Their ability to do so depends not only on their transparency and credibility, but also on their ability to manage incomplete agreements and the larger relationship in which they are embedded. Once again, the continuity of established

174

CHAPTER 7

democracies matters. In relationships between democracies, this continuity gives them time to learn about each other and encourages longer time-horizons. These features of democratic governance go a long way toward resolving the commitment and information problems that bedevil all international relationships.1 Overcoming these problems allows democracies to reassure each other, strike cooperative bargains, and avert costly conflicts. They can make more tenable long-term commitments and effectively signal that they do not pose security threats. Over time, that has allowed them to forge a durable peace. They can evade cycles of fear and reciprocal defensive armament and, instead, make mutually improving moves toward peaceful relationships. This reassurance is especially noteworthy because well-established democracies have the world’s most dynamic and advanced economies, which could readily translate into powerful military threats—directed at each other—if they chose. None of this implies that long-term cooperation is the only sustainable equilibrium. But it should be strongly favored on the general contracting grounds that parties will exhaust the possibilities of self-improving exchanges. The contracting advantages of democracies promote cooperation; they certainly do not produce harmony. Democracies still have mixed interests, some congruent, some conflicting. They still struggle, albeit peacefully, to get the lion’s share of the gains from cooperation. When they do finally settle on a bargain, they have the capacity to commit themselves effectively and then stick to what they have agreed upon. Their commitments are not perfect, but they are far better than those of most states. As a result, when democracies deal with each other, they do not have to begin their distributional struggles anew each day. Their old deals hold. They can profitably build upon them, rather than spend their days constructing bunkers of self-sufficiency, Maginot Lines to safeguard themselves against devastating betrayal. That does not mean they can ignore compliance issues or gaps in their agreements. Even solid bargains hit snags, encounter unexpected issues, face external shocks, and, as a consequence, may threaten to collapse. That is why partners need to construct governance mechanisms to monitor performance, handle disputes as they arise, and protect themselves from the worst default risks at reasonable costs. Again, transparency aids them. It allows partners to assess their promises and preferences more accurately. That is not a problem, since they generally enter agreements in good faith. They are seeking honest deals rather than opportune moments to defect and trap naı¨ve partners. Their principled commitment to open governance also makes it easier to negotiate still more transparency for specific agreements. Transparency does not challenge democracies’ sovereign control since their state authority is not based on secrecy. That is why democracies, but not dictatorships, are typically willing to add on-site inspections and information exchanges on a mutual basis. For them, it is simply a practical issue.

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

175

Democracies’ fundamental commitment to independent courts also makes it easier to govern international bargains. It is not a magic bullet (or magic gavel), but it is helpful. First, democracies have been willing to settle some difficult disputes by submitting them to arbitration. That is how the United States and Britain settled their festering dispute over ships damaged during the U.S. Civil War. That is also how boundaries were drawn in the Orinoco River valley, a three-cornered dispute involving Britain, Venezuela, and the United States. But ad hoc judicial settlements are only part of the story, and not the most significant one. More important in recent years has been the coordinated push by democracies to build dispute-settlement panels into their agreements. These are systematic efforts to govern incomplete bargains, and they extend to all but the most sensitive issues of national security. They are particularly useful when interstate agreements are complex and apply mainly to private actors. Complex agreements require extensive, detailed interpretation so that private actors—importers, exporters, and investors—can make their own plans. Useful as these dispute panels are, they remain controversial because they directly affect sovereign autonomy and sometimes limit domestic courts and legislatures in ways the states did not originally understand or intend. Still, the use of these panels is growing and creating a more orderly setting for transnational activity.2 Dispute panels and negotiated transparency are now common elements of international governance among democracies. They illustrate, once again, how similar state structures and some shared norms make it easier for democracies to manage bargains with each other and therefore to strike them in the first place. The result is a widening array of reliable bargains touching all aspects of international life. Increasingly, that is how constitutional democracies have organized their relationships with each other. This political framework has made possible countless cross-border ties among private partners operating within a secure international setting built by democratic bargains. These private ties depend on a wide range of practical governance arrangements: public and private; bilateral and multilateral; formal and informal, regional and global. They are overlapping and mutually reinforcing. All this would be much harder to accomplish and much less advanced if states could not forge stable, bilateral bargains among themselves and use them to build durable, peaceful relationships. One might still find extensive investments—foreign manufacturers in China and oil drillers in Nigeria, for example—but they would need to pay considerable (and costly) attention to selfprotection. That is essential when large, fixed investments are at stake and might be confiscated by predatory states or corrupt rulers. It is less important for investments that require little overhead, from fast-food restaurants to accounting firms.3

176

CHAPTER 7

By contrast, ties between Germany and Holland or between the United States and Canada are much more extensive and diverse, and much more secure. They involve substantial immobile assets dedicated to specialized tasks. Those investments would be extremely risky if political ties did not bind. But they do bind. That is because these constitutional democracies, thanks to their internal structures, have an unprecedented capacity to strike durable bargains. That allows them to capture the gains from cooperation without erecting expensive safeguards against remote dangers. States and private actors have strong incentives to pursue the opportunities that open up as a result. If there is well-founded confidence among the participants and a long shadow of the future, deals can be governed at much lower costs. That not only encourages more such deals, it promotes their evolution from transitory, arms-length transactions into sustained, self-governing relationships. Long-term bargains among states—established democracies all—form the framework for profitable ties of interdependence and specialization. Global business relationships have flourished because the risks are controlled, the economies are attractive, and the private actors are free to explore the ventures they choose. When an interstate political framework reliably spans borders and supports commerce, when the costs for communication and transport are low, then private actors can use their independence to create their own decentralized global networks. They have done exactly that. Constitutional democracies are well positioned to encourage these decentralized ties across borders. That is because they are characterized by strong civil societies, where citizens, firms, and voluntary organizations have their own resources and considerable freedom to act as they choose, independent of state control. This wide scope for private action is one of the characteristic features of Western success, economically and politically. Building on this social foundation and stable interstate cooperation, firms and nongovernmental organizations (NGOs) have spun rich, growing webs of transnational ties. They represent the early stages of a new transnational civil society. This nascent society is fostered by the openness of liberal societies and the fairness of their domestic legal systems, which can be used to reinforce private bargains. Transnational ties like these should continue to grow as long as interstate bargains hold firm, communication and transportation costs stay low, and domestic societies remain open and tolerant. This process is already well underway in Western Europe, North America, and East Asia. It is growing deeper as more transnational ties develop. It is spreading geographically as more states seek to share in their success. In these zones of democracy, interstate agreements have moved well beyond traditional nineteenth-century diplomacy. Traditional bargains were made between heads of state or foreign ministries. They dealt with the age-old topics of interstate relations: alliances, territorial adjustments, border tariffs, “friendship and commerce,” plus some housekeeping details like postal arrangements

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

177

and river navigation. States still make such agreements, both bilaterally and multilaterally. They are mundane but essential, the deep pilings on which this vast, private structure of transnational relations ultimately rests. These foundational agreements among states must be strong and stable to support the wide (and widening) range of cross-border relationships, which depend upon them, directly and indirectly. That is yet another reason why the democratic peace is so important. Not only has it dramatically lowered the risks of war and threats of violence, it has laid the basis for much more. It makes possible an extensive latticework of bargains spanning the world of constitutional democracies. These modern bargains look quite different from deals brokered by Richelieu, Metternich, or Bismarck. Those statesmen redrew maps, assembled alliances, and controlled small neighbors by subsidies and coercion. Diplomats perform many of the same tasks today, although nationalism has stopped any facile border changes. What is different now is that diplomats deal with a whole range of new issues alongside the old ones. To cope with a growing global economy and channel its development, they have devised new arrangements to encourage freer movement of people, goods, corporations, banks, ideas, planes, telephone calls, software, charities, religious organizations, professional associations, and on and on. With the architecture of interstate arrangements in place, transnational actors can make their own plans and pursue their own goals, whether they are profits, proselytizing, or preserving endangered species.

The Limits of My Argument What are the limits of the democratic peace, according to my argument? The most important is that it does not claim that established constitutional democracies must remain at peace with each other, now and forever, time without end. Instead, my argument stresses how structural features of democracies help to overcome, though never completely eliminate, the information and commitment problems that all states face in anarchy. The key point is that it is harder for democracies to reverse serious promises. It is especially difficult for them to betray others by surprise—harder, but not impossible. Democracies find it difficult to spring strategic surprises because they are inherently more transparent. But even the best is not perfectly transparent, and some, like modern Japan, are relatively opaque to outsiders. As a result, they are less firmly incorporated in the democratic peace, even though their refusal to build a large military sends a powerful and reassuring signal of their peaceful intentions. Another limitation is that even stable constitutional democracies may not last forever. They are subject to exogenous shocks and internal disruptions that could threaten changes of regime. All states must consider a partner’s stability and its possible impact on future performance. There are good reasons why

178

CHAPTER 7

these problems are less likely to occur among constitutional democracies, but that does not mean they are wholly absent or that trust completely crowds out suspicion. The early years of most democracies are shaky. Some fail, reverting to dictatorships. According to empirical studies, these uncertain transitions raise the risks of war and militarized disputes. The good news is that once democracies have passed these birthing pains, they are remarkably stable and successful political arrangements. A third limitation is that multilateral relationships are less secure than bilateral ones, often markedly so. Pairs of democracies can form secure bonds. Ties among large numbers of democracies are inherently less stable and reliable. Information problems are more difficult: who did what to whom? The crucial problem, however, is the familiar one of providing collective goods. When every state benefits, even if it does not pay the costs, there are incentives to free ride. That is a severe test for all states, including democracies. Nor is it an abstract concern. It has destroyed most historical efforts to provide collective security, including the League of Nations, which relied on Western democracies. The problems have not disappeared with time. They remain difficult when large groups are needed to sustain joint activities. The incentives to free ride generally grow in tandem with group size, straining the reciprocal ties that support individual contributions. Since World War II, the great size of the United States and its vast stake in international order has relieved many problems of collective action. Even so, burden sharing is a recurrent source of diplomatic friction (since the United States wants others to contribute more) and a looming problem for the future (since America’s relative size is likely to decrease). It remains a formidable obstacle to achieving Kant’s goal, a pacific union of democracies, or Woodrow Wilson’s goal of collective security. Fourth, democracies cannot blithely assume their peers will always perform their agreements as promised. They can be more confident in dealing with fellow democracies, but they cannot entirely ignore the risks of breach. That is particularly true if their agreements become unbalanced over time so that one party receives most of its benefits before the other. The state left waiting is vulnerable. Its best hope is that its partner expects to pay a high price for breaking the agreement. That price might be the loss of a profitable relationship, damage to its reputation, or direct sanctions. While breach and cheating are problems for all international relationships, democracies are better placed to solve them. They expect to maintain a wide range of profitable, long-term ties—economic, military, and social. These prospects lower the risks of breach and lessen the need for extensive safeguards, even if they do not eliminate them entirely. Finally, state preferences might change, unraveling old bargains and making new ones difficult. All rationalist theories, including mine, make the simplifying assumption that underlying state preferences are constant. In practice, states sometimes go through periods of heightened nationalism and greater demands for expansion, mostly for internal reasons rather than external fears.

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

179

Hans-Ulrich Wehler, for instance, makes such arguments about Wilhelmine Germany.4 In the United States during the nineteenth century, westward expansion was stimulated by powerful domestic sources, not external threats. It produced conflicts with Britain, Russia, and Spain, war with Mexico, and conquest of native Indian tribes.5 Expansionist goals like these are likely to be strongest and most contentious during periods of rapid state-building, when nationalism is often associated with popular mobilization and elite conflict. If democracies have such preferences, there will be fewer joint gains to be captured, narrowing the basis for democratic bargains. What all these limitations have in common is this: democracies must make their way in an international environment that is sometimes harsh and always requires attention to lurking dangers and divergent interests. Fortunately, democracies do not make these dangers worse in dealings with each other. Quite the contrary—they have deep-seated structural attributes that allow them to ease fears, pursue joint gains, and build confidence on solid footing.

Stable Contracts as a Source of International Order This ability to form stable, secure international relationships constitutes an important source of international order. The basic problem of social order—domestic or international—arises because decentralized cooperation is so difficult and tenuous. That is the starting point of Hobbes’s Leviathan, the most profound formulation of contractarian social theory. In the state of nature that he posits, individuals have no social context, no institutional supports, no expectation of long-term associations, no ground for trusting each other, and no reason expect honest dealings. In this arduous environment, individuals lack the capacity to create order through decentralized cooperation or shared social norms. As a result, they cannot provide for their own security, develop economic exchange, or sustain social relationships. Their lives are nasty, brutish, and short. To solve the problem of order, these individuals impose on themselves a repressive structure of state authority. Hobbes’s central conclusion is that individuals in such a state of nature would actually choose this repressive structure rather than continue to live in constant peril and grinding poverty. They willingly defer to a “common power to keep them all in awe.” It improves their lives. The “anarchy” of international relations theory is not quite so brutal as Hobbes’s state of nature. One reason is that states, even in isolation, can provide some of their own basic needs, particularly self-defense. That undercuts the need for strong international institutions analogous to the domestic state. In fact, international authority structures are relatively weak and fragmented. Among students of international politics, the dominant theme is that political order, such as it is, derives mainly from military power and deterrence. That usually means countervailing (balancing) alliances, although it could mean

180

CHAPTER 7

hegemonic power and bandwagoning alliances. Because states can defend themselves, alone or in alliances, they are more resilient than Hobbes’s individuals. The survival rate of states is impressive. But self-defense does not mean self-sufficiency, as modern states know. Almost none have chosen autarky, and many (especially the most developed) have chosen high levels of interdependence. That is particularly true in economic affairs, where they have promoted cross-border ties to increase economic efficiency and growth. These ties suggest another way that states, operating in anarchy, differ from Hobbes’s individuals in a state of nature. His individuals are essentially identical and have no identifiable long-term relationships. States, by contrast, interact frequently over long periods and can recognize their specific partners. Because states are distinguishable, they can and do build reputations. Because they have ongoing relationships and some common interests, they can often support cooperation, although that is not the only possible outcome. Building on this point, contracting theory identifies another fundamental source of international order: a web of durable agreements, embedded in longlasting relationships among stable constitutional democracies. Having learned to govern their relationships and contain the risks of opportunism and betrayal, they have captured important joint gains by multiple agreements. That point can be put another way. States avoid wars in two ways, through mutual fear or through mutual confidence. Democracies seem to have advantages in both. They are more effective at projecting credible threats and credible promises. Equally important, they have frequently made that rarest of political transitions: from fear of each other to confidence, avoiding war first by deterrence and then by building a deeper peace, convinced that their partner will not attack. Britain and the United States made exactly this transition in the late-nineteenth century; France and Britain did so in the early-twentieth century. Some newer democracies have done so as well, as the ArgentineBrazilian nuclear agreement shows. They, too, are proving that they can increase transparency, signal effectively, and develop reputations as reliable partners. Japan, South Korea, and Taiwan have traveled the same path, aided by America’s stabilizing military presence in northeast Asia. The real test will come after the United States draws down its troops. U.S. troops played the same role in postwar Europe, but those forces are no longer what keeps the peace between France and Germany. In case after case, established democracies (and some newer ones) have built a firm, peaceful order, based on a dense network of explicit agreements and informal relationships. That order has spurred the development of regional and global institutions, which, in turn, have strengthened commercial, social, and security relationships. These relationships do not mean that their interests are the same. Some are, some are not. It does mean that they have identified some important common interests (including the shared goal of avoiding war) and have devised ways

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

181

to realize them. They have also learned from their long-term relationships, establishing reputations for handling disputes as they arise and avoiding the temptations of short-term opportunism. As a result, among well-established democracies, what began as hegemonic dominance or military deterrence has gradually evolved into a voluntary security community of reliable partners. This community has no inherent limits. It is not build on opposition to outsiders. New states can join as they develop common interests and expectations and prove that they, too, are reliable partners.

History without End The spread of democracy and hopes for a more inclusive democratic peace have led some to argue that history is at an end. Following Hegel’s analysis, they mean that the motive forces of history have already played out their most important consequences. The modern version of this argument says that democratic liberalism and market institutions will soon cover the world, that the new democratic states will spell an end to conflict and war.6 This is the alleged “final stage” of history. Would that it were so. There are reasons for optimism but not for this kind of millenarianism. The reasons for some optimism are clear: democracy has spread significantly, and peace among stable democracies is real. These are historic changes of great moment, milestones in the quest for enduring peace. But they mark progress, not completion. The global movement toward democracy and market reforms has been strong for almost two decades, but there are no guarantees for the future. History is not unidirectional, written by Whigs. It is not moving with a sure majesty toward a certain conclusion: a world filled with liberal, constitutional democracies. Fortunately, we have not seen any long-standing democracies descend into authoritarianism. They appear to be remarkably stable. They have survived depressions and far-reaching economic transformations. They have extended the franchise in wave after wave with little disruption. But some democracies with new, shallow roots have failed, especially in bad economic times. The battered young democracy of Weimar gave way to the Third Reich. German voters gave Hitler an electoral plurality in 1933, which he quickly converted into a brutal dictatorship.7 He was not the first elected leader to do so, or the last. It is a recurrent problem in Africa, Asia, and Latin America, where new democracies are trying to gain their footing, and not always succeeding. Becoming democratic poses some international risks of its own. The required social and political changes are bound to produce uncertainty abroad and raise the chances of misperception. That creates a slew of familiar problems, especially if new democracies are not firmly established. Those states that do make a smooth transition seem to be less war-prone than the regimes

182

CHAPTER 7

they replace. Those that suffer major setbacks during democratization seem to be more war-prone. That is probably because their uncertain future undermines their ability to make credible, long-term commitments and sparks a combination of fear and opportunism in their neighbors.8 As the regimes stabilize (if they stabilize), one would expect these transitional effects to disappear. In the interim, however, the confusion and anxiety they arouse can be serious. States that become democratic, or remain that way, may still lack the essential features of constitutional order, including protections for civil society, toleration for diverse viewpoints, and limitations on state control. The contracting explanation applies only to constitutional democracies, not to all states that elect leaders. Illiberal democracies, lacking a variety of constitutional protections, may threaten peace in two basic ways: opaque procedures and expansionist goals that leave little room for agreement. Some use elections, perhaps plebiscites, to demonstrate public support for their government but include few other mechanisms of popular control. They give little protection to civil society; they permit no opposition parties and only the mildest dissent. Aside from voting, they have few features of constitutionally ordered states. The Islamic Republic of Iran under Ayatollah Khomeini is a good example. Other states permit freer voting but have procedures that are impenetrable to outsiders and offer little protection for their rights. These opaque states are usually dominated by a single political party and deny the opposition full access to information. Their domestic opponents are not strong enough to provide critical oversight to the government, gain access to sensitive documents, or affect key policy debates. Truly illiberal states, whether they have popular support or not, frequently promote expansionist claims that pay little heed to the rights of other states and peoples. If so, that narrows the room for compromise with other states. That space for agreement could shrink further if they are willing to take large, uncertain gambles—for that is what war often is. What is more important for our argument, however, is the opacity of illiberal states, democratic or not. The closure of their government’s decisionmaking weakens the possibility for durable agreements, raises the risks of surprise defection, and increases the chances that mistakes will be considered evidence of bad faith, malign intentions, and opportunism. This opacity is occasionally an issue among democracies. It often crops up in U.S.-Japanese relations, for instance, producing tougher diplomatic problems than those between the United States and its European partners. Over and over, commercial agreements have led to outcomes far less attractive than the United States envisioned. The resulting suspicions are not easily dispelled, given the closed, private character of Japanese policymaking, its long domination by one set of politicians, and the intimate ties between large firms and state planners. One unhappy consequence has been to push the United States

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

183

toward “proxy agreements,” using easily verified quotas instead of hard-toverify pledges of open markets. Japan’s liberalism can be doubted, but not its stability as a partner.9 Disputes between Japan and America or Europe are troubling because they suggest that stable democracies can still face persistent conflicts in their bilateral relations, not only over the distribution of gains from trade—that is to be expected—but also over whether they are keeping agreements fairly. When agreements deal with closed areas of government policymaking or with murky issues of policy process, misunderstanding can arise easily and lead to charges of bad faith. That need not be the precursor to war, but it is poor grounding for a deepening peace and a security community. This may sound bleak, but the overall picture is still a hopeful one for peace among democracies. The spread of democracies after the Cold War has been impressive. Many, especially in central Europe, have made significant strides toward transforming their rusting economies, establishing civic spheres, and creating democratic polities. These are historic undertakings. They cannot be accomplished overnight and in some regions, like the southern tier of the old Soviet Union, they may well fail entirely. Still, over wide areas, there has been an encouraging spread of democracy and some indications that its roots are taking hold. Meanwhile, older democracies in the West have proven extraordinarily stable and resilient. This spread of democracies has important implications for international relations because there are no inherent geographic limits to the democratic peace. It need not be confined to mature states of the north Atlantic. It is not a private gated community that preserves its safety by keeping the riff-raff forever outside. Its membership may be exclusive, but it is also open-ended. With time and effort, states can achieve the status of stable constitutional democracies. Doing so requires difficult changes with far-reaching effects on domestic life. They touch and transform basic features of social relations, political rights, and economic controls. Not all states try to make this transition, and not all who try succeed. This is not meant to imply that democracy is the only source of international stability or peace. The presence of nuclear weapons among major powers— or, to be more precise, the possession of effective second-strike deterrents— has significantly raised the risks of launching any major assault on them or their close allies. This stable nuclear standoff encouraged the long peace between the United States and Soviet Union, despite their deep differences, and it now does the same for the United States and China. No one doubts that nuclear warfare would be catastrophic, for the “winners” as well as the “losers.” These extraordinary risks clearly change the calculations about starting a war. Even if the risks of nuclear annihilation are very small, only the most reckless or irrational states would be willing to run them. That is why the

184

CHAPTER 7

limited spread of nuclear weapons has encouraged a kind of “cold peace” that extends beyond the democratic states. Nuclear weapons have been far less important in relations among democracies, which have not relied on mutual deterrence. That bears repeating: the richest and potentially most powerful nations in human history have not kept peace among themselves by military deterrence, whether nuclear or conventional. Still, the presence of nuclear weapons in France and the United Kingdom is not wholly irrelevant to relations among democracies. These small nuclear arsenals were designed mainly to strengthen the deterrence America provided Europe (by making clear to the Soviets that an attack on Britain or France would start a nuclear exchange, which would draw in the Americans). These weapons were not aimed at other democracies. They were built to establish a more stable strategic context during the Cold War, one that complemented other security bargains among the North Atlantic democracies.10

Policy Implications Does all this theory and evidence really matter to the men and women who make foreign policy? It should. At the beginning of the book, we mentioned an old academic joke: “We know it works in practice. Now we have to see if it works in theory.” Actually, knowing how and why the democratic peace works in theory points the way toward better policies that can make it work more effectively in practice. Why? Because different theories suggest different policies that should be pursued. Take the Realist critique, which says democratic governance has no impact on international affairs and the democratic peace is a mirage. From that perspective, promoting democracy is a fool’s errand, even if it succeeds. First, it diverts the United States and its major partners from calculating and pursuing their national interests in a prudent, sensible way. Instead of coolly determining American national interests in our relations with China, the Balkans, Haiti, or Sudan, we are mesmerized by human rights violations and dragged into difficult and often fruitless projects of democratization and nation-building. This not only diverts America’s limited resources to less important tasks, it may well produce regional turbulence and even war. Why? Because the new regimes America promotes are not especially stable, which can draw them into conflicts with neighbors, and because the United States itself may use its military force for idealistic goals such as overthrowing dictators and installing fledgling democracies. Calculations like these are why Realists have uniformly opposed NATO expansion in Central Europe. The new members add no military strength to the alliance, and they threaten Russia unnecessarily. The only reason to add them is if these costs are offset by the value of strengthening democracy in

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

185

Poland, Hungary, the Czech Republic, and elsewhere. For Realists, that value is low, so they oppose NATO expansion. Aside from Realists, most scholars think the democratic peace is supported by overwhelming empirical evidence. But they explain it in different ways, with different policy prescriptions. Take those who emphasize shared norms, particularly the importance of peaceful, judicial settlement. If that is what produces the democratic peace, then the United States, Western Europe, and Japan should work hard at spreading democratic ideals around the world. Their aim should be to inculcate liberal, democratic values—the missionary position. Since judicial settlement has a prominent place in this explanation, they should try to create more effective institutions to settle interstate disputes, from arbitration to the International Court of Justice to an International Criminal Court. Most institutionalists would support these recommendations and go further. They advocate a large role for UN peacekeeping missions and want to see new democracies swiftly added to major democratic organizations such as the European Union and NATO. Liberals, by contrast, think the democratic peace grows out of domestic social and political organization. Several policy recommendations follow naturally. One is to promote civil society: encourage a wide variety of independent, voluntary organizations and foster transnational links among them. Another variant of liberalism stresses economic interdependence, which creates profitable links across borders. The implication: more free trade and foreign investment are the path to peace. A third variant emphasizes the importance of strong domestic institutions, from democratic legislatures to a free press. They support nation-building activities as an important element of foreign policy. I have already discussed the strengths and weaknesses of these explanations. Their limitations suggest problems with their policy recommendations. However valuable those recommendations are in their own right, they are not the high road to a wider, deeper democratic peace. The contracting explanation suggests a different approach. Its central point is that the democratic peace is based on secure bargains, grounded in relationships that develop gradually. The policy focus, then, should be on promoting stable bilateral relationships. That is particularly important among new democracies, which have thin reputations and untested relationships with other democracies. It is these countries and these pairings where peace is most at risk. Which policies are most appropriate? First, where states are already on the road to democracy, the United States, Europe, and other established democracies should encourage it. Their policy should have two aims: making democratization irreversible and increasing transparency. Risks of local war are greatest when democratization is incomplete or politically contested or, worse, when it is actually reversed. There are limits to the inducements the United States, Europe, and Japan can provide, but they should be linked to fair elections, equal

186

CHAPTER 7

treatment for democratic opponents, and rotation in office. Established democracies should also seek to foster transparent institutions abroad: free speech for dissidents, an active media, and legislative oversight of the executive. Transparency should have a negotiated international dimension, too. Democracies are willing to negotiate transparency with each other because they already have open societies. That should be encouraged, especially among new democratic neighbors. A useful model is the cooperation between Argentina and Brazil on sensitive nuclear issues. They not only resolved a major source of tension and averted an arms race, they learned a great deal about each other and developed a more mature relationship in security issues. The United States should also be wary of identifying its policies too closely with the ruling party in semidemocratic states, where a single party has held power for years. True, such states are often stable. But they are also more opaque because the ruling party is not forced to disclose government secrets or defend its policies against highly placed and well-informed critics. The lack of transparency also promotes corruption, which drags down the economy and stifles the growth of a civil society independent of state patronage. While encouraging transparency, the United States and other established democracies should recognize its limits and pitfalls. Transparency can make a state vulnerable to its nondemocratic neighbors. That may not matter if a democracy is surrounded by similar states. But if it faces threatening nondemocratic foes, then there is an important role for outside protection and support. Outside support for national security is most helpful in regions where several new democracies are getting started. These new regimes will eventually develop peaceful relationships, but, in their early years, bilateral relations can be rocky, even threatening. International institutions can aid in this transition. One way is by creating a safer context for bilateral relations to mature. That, after all, was the essence of U.S. policy in Europe after World War II. Beginning with the Truman administration, the United States not only provided a security guarantee for Western Europe against outside forces; it provided an internal guarantee against military conflicts among the members themselves. In this stable setting, relations among European democracies could grow and develop, transcending the centuries of warfare that had characterized their relationships. The United States did much the same thing (without forming multilateral institutions) in Northeast Asia, allowing young democracies in Japan, South Korea, and Taiwan to forge peaceful relations despite their bitter history. Peacekeeping missions may also help, at least marginally, in some newly democratizing regions. But their role should be limited. They may (or may not) be able to stabilize internal politics; they surely cannot create a regional peace. At most, they can serve as a final layer of reassurance among the parties. A stable peace can grow only out of the states’ own preferences, their willingness to make costly commitments that entail some risks. What outside powers

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

187

TABLE 7.2 Causal Theories and Policy Recommendations Causal Explanation Realist

Policy Recommendation

Democratic peace does not exist Avoid promoting democracy at the expense of other, hard-nosed policy goals

Normative/ Shared norms Constructivist

Promote spread of democratic ideals and international law, including courts

Institutionalist

International institutions

Promote international institutions such as peacekeeping, arbitration, international courts

Liberal

Domestic society, political groups

Encourage development of civil society, economic interdependence

Contracting

Stable bilateral relations

Provide external support, if necessary, as pairs of new democracies gain experience

can do is lessen the risks and ease the difficulties of forming reliable agreements. The presence of peacekeepers can sometimes help by lessening fears of opportunism and betrayal. That makes bargains more feasible because it eases concerns about verification and lessens the need for simultaneous performance. But third parties should not be the dominant force in peacemaking. That task belongs to the principals. What third parties should do is contribute to a stable framework, one where protagonists can work directly with each other to resolve problems, find areas of agreement, and build a durable peace. International institutions can help in another way. Membership in them can send useful and reassuring signals, but only if joining is costly. Only then will membership inform others that a state is truly democratic and that it does not intend to pursue predatory policies. If the benefits of membership are great, then it is easier to set high standards as the price of admission. NATO and the EU have done exactly that. NATO, for instance, requires that prospective members create competitive democratic institutions and settle ethnic disputes before they can be admitted. The carrot of NATO membership has been a powerful inducement to resolve these long-simmering issues. These policy recommendations are summarized in table 7.2.

The Promise of Peace In their complementary work, Robert Putnam and Elinor Ostrom have taught important lessons about the social and organizational sinews of cooperation, both formal and informal.11 Their studies deal with domestic interactions, mostly on a small scale, but the lessons they teach are more general. What they say is that rationally grounded trust, easy access to good information, and continuing interactions can produce a wider ground for civic cooperation. This

188

CHAPTER 7

book argues that the same factors can produce international peace and cooperation, that they do so disproportionately among democracies, and that this cooperation can be developed beyond fleeting tactical adjustments to become enduring relationships. These relationships are stable because they are self-enforcing. That encourages states, firms, and citizens to build upon them, to forge their own transnational ties and self-governing relations. International cooperation is certainly not limited to established democratic states, even when important security issues are at stake. The Concert of Europe between 1815 and 1848 shows that conservative monarchies could adjust their policies and produce valuable collaborative arrangements, based on common understandings.12 True, the circumstances were favorable. After Napoleon’s defeat, the French lacked the power to wage another war for continental hegemony. No one else had the power either. But then no one aspired to that kind of dominance, and no one tried to build the necessary forces, which would have threatened other Great Powers. The only revisionist power was France, and its aims were relatively limited, thanks to the moderate peace imposed by England, Austria, Prussia, and Russia. Even so, France spent years pondering the risks and rewards of breaking out of the Vienna settlement. Such calculations are nothing new. States are continually evaluating their interests and the costs of pursuing them, alone or in groups, by diplomacy or by force. They did so during the Concert, just as they always have. But these calculations and the need for armed self-protection does not imply that force needs to be asserted directly or even threatened to create order. The order produced by the Concert was not based primarily on raw military power. Capabilities surely mattered—especially the crucial distinction between the Great Powers and all other states—and force was sometimes used against smaller states. The shadow of military deterrence was always present, hovering in the background of Great Power relations. But the Powers’ capabilities were tempered by their limited purposes and shared understandings. These moldy monarchies, as Thomas Hardy called them, were able to make effective accommodations, to strike informal bargains that held and worked. While it lasted, the Concert covered all Europe and produced a peace that lasted almost four decades, from Napoleon’s defeat at Waterloo to the Crimean War. Its success shows that democracies are not unique in their ability to forge cooperative arrangements. That is a hopeful sign for a world filled with nondemocratic regimes. But if nondemocratic states can sometimes produce lasting peace, their democratic counterparts do so almost all the time. They very rarely go to war against each other. They have never been involved in long-term diplomatic hostilities, permanent rivalries, or costly arms races, which are so often the precursors of war. The results are even more peaceful when the democracies are well established and have a long record of diplomatic relations. Among such states there have been no wars at all.

R E L I A B L E PA R T N E R S , R E L I A B L E P E A C E

189

Surely the most basic form of cooperation is to abstain from mutual injury.13 It is the precondition for further cooperation and peaceful competition.14 In fact, democracies have made important steps toward achieving this expanded cooperation. With time, they have even managed to forge security communities, in which states do not prepare for war against one another because such war is unthinkable. Without existential fears that other democracies might attack and destroy them, they can undertake a wide range of “conditional” cooperation—extending cooperative feelers in hopes they will be reciprocated, and then acting to sustain them when they are. This general environment of lower risks, longer time horizons, better information, and autonomy for private actors has fostered an extraordinarily deep, wide set of cross-border ties, incorporating both state agencies and civil societies. That is truly a peace dividend. Why so successful? Because stable, constitutional democracies have unique advantages that make cooperation easier, more likely, and more durable in their dealings with each other. They learn from their success, further strengthening their bargains and diminishing the threats they pose to each other. These are not simply policies of convenience or momentary choices. They are enduring policies, grounded in profound structural advantages. They are built into the warp and woof of democratic institutions in open, constitutionally ordered states. From these elegant strands they have woven a democratic peace.

Notes Chapter 1 The Argument in a Nutshell 1. The empirical literature is a burgeoning one, much of it relying on data from the Correlates of War Project. There are at least three fine syntheses: Bruce Russett with William Antholis, Carol R. Ember, Melvin Ember, and Zeev Maoz, Grasping the Democratic Peace: Principles for a Post–Cold War World (Princeton: Princeton University Press, 1993); James Lee Ray, Democracy and International Conflict: An Evaluation of the Democratic Peace Proposition (Columbia: University of South Carolina Press, 1995); and Spencer Weart, Never at War: Why Don’t Democracies Fight One Another? (New Haven: Yale University Press, 1998). There are also at least two valuable reviews of the literature, by Steve Chan and Miriam Fendius Elman. Steve Chan, “In Search of Democratic Peace: Problems and Promise,” Mershon International Studies Review 41, suppl. 1 (May 1997), pp. 59–91; Miriam Fendius Elman, “The Never-Ending Story: Democracy and Peace,” International Studies Review 1 (1999), pp. 87–103. 2. The other well-known regularity seems far less important, namely that distant states with little contact rarely fight. What do they have to fight about? The other side of that coin is much more interesting. Most wars take place between nearby states, usually those with common borders. This has a nontrivial implication: interdependence among states can be a source of conflict and war, just as it can be a source of common interests and peaceful ties. What is needed is a theoretical specification that separates these two opposite consequences of interdependence. 3. Jack S. Levy, “Domestic Politics and War,” Journal of Interdisciplinary History 18 (Spring 1988), p. 662. 4. Nils Petter Gleditsch, “Geography, Democracy, and Peace,” International Interactions 20 (1995), p. 297. 5. David Lake, “Powerful Pacifists: Democratic States and War,” American Political Science Review 86 (March 1992), pp. 24–37. 6. Edward Mansfield and Jack Snyder, “Democratization and War,” Foreign Affairs 74 (May/June 1995), pp. 79–97; Edward D. Mansfield and Jack Snyder, “Democratization and the Danger of War,” International Security 20 (Summer 1995), pp. 5–38; and Jack Snyder, From Voting to Violence: Democratization and Nationalist Conflict (New York: W. W. Norton, 2000). 7. Kenneth N. Waltz, Theory of International Politics (Reading, MA: Addison-Wesley, 1979). 8. Immanuel Kant, “Perpetual Peace, a Philosophical Sketch,” in Perpetual Peace and Other Essays, trans. Ted Humphrey (1795; reprint, Indianapolis, IN: Hackett Publishing Company, 1983). Kant’s views are important in formulations by Michael Doyle, Bruce Russett, John Owen, and others. 9. John Owen offers the clearest and most thorough presentation of the normative approach. John M. Owen IV, Liberal Peace, Liberal War (Ithaca: Cornell University Press, 1998).

192

N O T E S T O PA G E S 2 – 1 0

10. For the contrast between democratic and nondemocratic norms and the argument that democracies externalize their norms, see J. Joseph Hewitt and Jonathan Wilkenfeld, “Democracies in International Crisis,” International Interactions 22 (1996), pp. 123–42. 11. Charles W. Kegley, Jr., and Margaret G. Hermann, “Military Intervention and the Democratic Peace,” International Interactions 21 (1995), pp. 1–21; Kegley and Hermann, “The Political Psychology of Peace through Democratization,” Cooperation and Conflict 30 (1995), pp. 5–30; Kegley and Hermann, “How Democracies Use Intervention: A Neglected Dimension in Studies of the Democratic Peace,” Journal of Peace Research 33 (1996), pp. 309–22. 12. T. Clifton Morgan and Sally Howard Campbell, “Domestic Structure, Decisional Constraints, and War: So Why Kant Democracies Fight?” Journal of Conflict Resolution 35 (June 1991), pp. 187–211. 13. Presidential systems typically have more checks and balances, more locations where policy can be vetoed. Federal systems have more decentralization, checking central power by dispersion. For a general discussion, see David Collier and Stephen Levitsky, “Democracy with Adjectives: Conceptual Innovation in Comparative Research,” World Politics 49 (April 1997), pp. 430–51; the implications of these differences for the democratic peace are discussed in Miriam Fendius Elman, “Unpacking Democracy: Presidentialism, Parliamentarism, and Theories of Democratic Peace,” Security Studies 9 (Summer 2000), pp. 91–126. 14. Bruce Russett et al. Grasping the Democratic Peace, pp. 39–40. 15. The high cost to dictators who lose wars is examined in Henk Goemans, “Fighting for Survival: The Fate of Leaders and the Duration of War,” Journal of Conflict Resolution 44 (October 2000), pp. 555–79. 16. Hew Strachan, “Essay and Reflection: On Total War and Modern War,” International History Review 22 (June 2000), p. 342. 17. The idea of governing mechanisms is drawn from Oliver E. Williamson’s work on transaction costs. Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting (New York: Free Press, 1985); Williamson, The Mechanisms of Governance (Oxford: Oxford University Press, 1996). 18. Charles Lipson, “International Cooperation in Economic and Security Affairs,” World Politics 37 (October 1984), pp. 1–23. 19. Kenneth A. Schultz, Democracy and Coercive Diplomacy (Cambridge: Cambridge University Press, 2001). 20. Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” in Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowledge (Cambridge: Cambridge University Press, 1976), pp. 91–196. 21. We cannot infer that nondemocracies betray agreements more often, even if they are less reliable partners. The reason is that prudent states avoid agreements with unreliable partners in the first place. So there is a selection-bias problem in looking at what happens to existing agreements. 22. Christel Lane, “Introduction: Theories and Issues in the Study of Trust,” in Christel Lane and Reinhard Bachmann, eds., Trust within and between Organizations: Conceptual Issues and Empirical Applications (Oxford: Oxford University Press, 1998), p. 21.

N O T E S T O PA G E S 1 0 – 1 7

193

23. Robert O. Keohane and Joseph S. Nye, Power and Interdependence, 3d ed. (Boston: Addison-Wesley, 2000) is the classic work on various forms of interdependence and their policy manipulation. 24. Jon Hovi, Games, Threats and Treaties: Understanding Commitments in International Relations (London: Pinter, 1998); James D. Morrow, “How Could Trade Affect Conflict?” Journal of Peace Research 36 (July 1999), pp. 481–89. 25. Olson revives an old term, consilience, to refer to theories that explain a wide range of evidence. He attributes the term to William Whewell, a nineteenth-century philosopher of science. Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities (New Haven: Yale University Press, 1982), p. 13. The topic has been developed recently by E. O. Wilson, Consilience (New York: Knopf, 1998). 26. In addition to this contracting advantage, constitutional democracies (by their very nature) avoid two recurrent sources of international conflict. Because all property rights are secured by due process, at least during peacetime, foreigners do not face confiscation or arbitrary seizures, which could lead them to seek forcible help from home governments. Second, because civil rights and minority views are protected, at least in liberal democracies, there is little call for intervention from foreign protectors. These kinds of reciprocal protections undoubtedly strengthen ties among democracies. The absence of these basic protections elsewhere is a recurrent source of tension among nondemocratic states and between them and democracies. It may also be an important source of uncertainty about new democracies, whose neighbors cannot be sure whether foreign property or vulnerable minorities will be protected. 27. Of course, the parties still have to solve distributional issues, but these are also present in domestic bargaining. 28. The rationalist theory is explained and formalized in James D. Fearon, “Rationalist Explanations for War,” International Organization 49 (Summer 1995), pp. 379–414. Chapter 2 Is There Really Peace among Democracies? 1. These standard controls were first used by Stuart A. Bremer in “Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 1816–1865,” Journal of Conflict Resolution 36 (June 1992), pp. 309–42. They have been adopted (and adapted) by Bruce Russett, John Oneal, and other leading quantitative investigators of the democratic peace. Implementing these controls involves difficult choices. For example, when Bremer, Russett, Oneal, and others consider whether two countries are contiguous, they look not only at the home countries’ borders, but also at their colonies or dependencies. In coding just the how democratic the countries are, they rely on Polity II (or III) data sets and generally use the score for the less democratic of the two countries. Some of these controls are theoretically contested. For example, rationalist explanations generally conclude (counterintuitively) that differences in military capabilities will not foment wars because they are commonly known and therefore should lead the weaker state to make concessions before any war. That is Thucydides’ point in the Melian Dialogue.

194

N O T E S T O PA G E S 1 7 – 2 0

2. Dean Babst, “Elective Governments—A Force for Peace,” Wisconsin Sociologist 3 (1964), pp. 9–14; and Babst, “A Force for Peace,” Industrial Research (April 1972), pp. 55–58. 3. Singer and Small also published their findings in a small journal. Melvin Small and J. David Singer, “The War-Proneness of Democratic Regimes,” Jerusalem Journal of International Relations (1976), pp. 50–69. 4. Michael W. Doyle, “Kant, Liberal Legacies, and Foreign Affairs,” pts. 1 and 2, Philosophy and Public Affairs 12 (Summer 1983), pp. 205–35 and (Fall 1983), pp. 323–53; and Michael Doyle, “Liberalism and World Politics,” American Political Science Review 80 (December 1986), pp. 1151–69. 5. The observation about citizens’ reluctance to bear the costs of war appears in Immanuel Kant, Perpetual Peace, in Kant, On History, trans. Lewis White Beck (1795; Indianapolis: Library of the Liberal Arts, Bobbs-Merrill, 1963), pp. 94–95. 6. Francis Fukuyama, The End of History and the Last Man (New York: Free Press, 1992). 7. “We have no eternal allies, and we have no perpetual enemies,” Palmerston told the House of Commons on 1 March 1848, in defense of his foreign policy. “Our interests are eternal and perpetual, and those interests it is our duty to follow.” 8. John J. Mearsheimer, “Back to the Future: Instability in Europe after the Cold War.” International Security 15 (Summer 1990), pp. 5–56. This theme recurs in Mearsheimer’s book The Tragedy of Great Power Politics (New York: W. W. Norton, 2001). 9. Studies of the democratic peace—pro and con—rely on the same basic data: (1) data on wars and casualties from the Correlates of War Project and (2) data on the type of governments involved from the Polity data sets. Some studies also use data on lowerlevel disputes, generally from the collection on Militarized Interstate Disputes (MIDs). The common definition of a “war” is an organized military conflict between states that results in one thousand or more battlefield casualties. That definition predates the analysis of the democratic peace. The Polity data sets rank each government along a continuum, with ten being a full, competitive democracy. Different scholars have chosen different thresholds to qualify as a “democracy,” ranging upward from five. 10. The data on militarized disputes rank them by intensity. At the lowest level are explicit threats to use force. At the next level, states mobilize, deploy, or display military force. Beyond that is the actual use of force short of outright war. The MID data sets are limited to deliberate acts of state policy. 11. Keith Jaggers and Ted Gurr, “Polity III: Regime Change and Political Authority, 1800–1994” (computer file, 1996). 12. To score zero on the democracy scale, a state must suppress political competition, restrict political participation, select leaders in a noncompetitive process, which is closed or hereditary rather than elective, and place few (if any) limitations on the chief executive’s power. 13. Autocratic states are those that rank high on the Polity “autocratic scale,” usually five or higher (although the choice is up to individual scholars). Using these two scales, some states do not qualify as either democracies or autocracies. These intermediate forms of governance are awkwardly termed anocracies. Generally speaking, the democratic peace literature treats all these nondemocratic states alike. 14. Zeev Maoz and Bruce Russett, “Normative and Structural Causes of Democratic Peace, 1946–1986,” American Political Science Review 87 (September 1993), pp. 624–

N O T E S T O PA G E S 2 1 – 2 4

195

38; Paul D. Senese, “Democracy and Maturity: Deciphering Conditional Effects on Levels of Dispute Intensity,” International Studies Quarterly 43 (September 1999), pp. 483–502. 15. The relationship between economic interdependence and the democratic peace is analyzed in Bruce Russett and John R. Oneal, Triangulating Peace: Democracy, Interdependence, and International Organizations (New York: W. W. Norton, 2001). 16. The idea that pairs of democracies do not fight is called the dyadic proposition. By contrast, the monadic proposition, associated with the work of R. J. Rummel, is that individual democracies are inherently more peaceful, regardless of the type of state they interact with. 17. Jack S. Levy, “War and Peace,” in Walter Carlsnaes, Thomas Risse, and Beth A. Simmons, eds., Handbook of International Relations (London: Sage, 2002), pp. 350–68. 18. Levy, “War and Peace,” pp. 358–59. 19. Levy, “War and Peace,” pp. 359. 20. Brian Lai and Dan Reiter, “Democracy, Political Similarity, and International Alliances, 1816–1992,” Journal of Conflict Resolution 44 (April 2000), p. 205. 21. Joanne Gowa, Ballots and Bullets: The Elusive Democratic Peace (Princeton: Princeton University Press, 1999); Gowa, “Democratic States and International Disputes,” International Organization 49 (Summer 1995), pp. 511–22; Henry S. Farber and Joanne Gowa, “Polities and Peace,” International Security 20 (Fall 1995), pp. 123– 46; Henry S. Farber and Joanne Gowa, “Common Interests or Common Polities? Reinterpreting the Democratic Peace,” Journal of Politics 59 (May 1997), pp. 393–417. 22. Gowa also makes controversial choices in limiting her control variables. She drops “wealth” because she lacks good data before World War II and drops “alliance” because, in her view, it is a product of interstate interaction, much like war itself. She also argues that it is not joint democracy but rather common interests (for which alliances are her proxies) that explain the so-called democratic peace. In a critical review, Thomas Schwartz observes that by Gowa’s own account, alliance should be the best control variable but she chooses not to include it. Thomas Schwartz, “Review of Joanne Gowa, Ballots and Bullets: The Elusive Democratic Peace,” Comparative Political Studies 33 (June 2000), pp. 690–93. 23. The problems of statistical inference are compounded because there are far fewer democratic states before World War II. After World War II, when the number of democracies increases, Gowa’s results meet her standards of statistical significance. 24. Brett Ashley Leeds and David R. Davis, who find extensive cooperation among democracies beyond the democratic peace, make this point about common interests. “Some might argue that our findings could as easily support the idea that common interests account for the democratic peace. The fact that democracies have exhibited high levels of cooperation in their foreign policies could be interpreted to suggest that democracies have had little cause for conflict during this period [1953–78, the period of their data]. We have argued, however, that such a claim holds only if we also assume that democracies have held common interests by chance and that there is no causal relationship between domestic political structures and international interests. We argue instead that domestic political institutions play a role in creating common international interests among democracies.” “Beneath the Surface: Regime Type and International Interaction, 1953–1978,” Journal of Peace Research 36 (January 1999), p. 18.

196

N O T E S T O PA G E S 2 4 – 2 8

25. Christopher Layne, “Kant or Cant: The Myth of the Democratic Peace,” International Security 19 (Fall 1994), pp. 5–49; Christopher Layne, “On the Democratic Peace,” International Security 19 (Spring 1995), pp. 175–77. 26. Kitchener’s initial, secret instructions were to try to dislodge the French but without any shooting or killing. “No corpses,” as Salisbury put it. Darrell Bates, The Fashoda Incident (New York: Oxford University Press, 1984), p. 140. 27. Raymond Carr, Spain, 1808–1975, 2d ed. (Oxford: Clarendon Press, 1982), p. 373; Octavio Ruiz, “Spain on the Threshold of a New Century: Society and Politics before and after the Disaster of 1898,” Mediterranean Historical Review 13 (1998), pp. 7–27. 28. There was even a term, caciquismo, for these voting arrangements. Caciquismo referred to both falsified voting and the clientalist system of political influence it made possible. Carr, Spain, 1808–1975, p. 367. 29. Stephen Jacobson and Javier Moreno Luzo´ n, “The Political System of the Resto´ lvarez Junco and Adrian Shuration, 1875–1914: Political and Social Elites,” in Jose´ A bert, eds., Spanish History since 1808 (London: Arnold, with Oxford University Press, 2000), pp. 94–109. 30. Carr, Spain, 1808–1975, p. 357. 31. Jose Varela Ortega, “Aftermath of Splendid Disaster: Spanish Politics before and after the Spanish-American War of 1898,” Journal of Contemporary History, n.v. (1980), pp. 317–44; Ivan Musicant, Empire by Default: The Spanish-American War and Dawn of the American Century (New York: Henry Holt, 1998); John L. Offner, An Unwanted War: The Diplomacy of the United States and Spain over Cuba, 1895–1898 (Chapel Hill: University of North Carolina Press, 1992). 32. Gerhard L. Weinberg, A World at Arms: A Global History of World War II (Cambridge: Cambridge University Press, 1994), p. 103. 33. Robert Service, A History of Twentieth-Century Russia (Cambridge: Harvard University Press, 1998), p. 257. 34. Williamson Murray and Allan R. Millett, A War to Be Won: Fighting the Second World War (Cambridge: Harvard University Press, 2000), p. 56. 35. Martin Gilbert, Churchill: A Life (New York: Henry Holt, 1991), p. 630. 36. Matthew Krain and Marissa Edson Myers, “Democracy and Civil War: A Note on the Democratic Peace Proposition,” International Interactions 23 (1997), pp. 109– 18. 37. The risks of political upheaval, coups, and revolutions increase as harshly authoritarian regimes loosen their draconian rules. That, at least, is the current verdict of quantitative research. Most strongmen seem to agree, with devastating implications for political liberalization. Despite protests and unrest, dictators usually (but not always) calculate that their chances of survival will decline if they relax their controls. Ha˚ vard Hegre, Tanja Ellingsen, Scott Gates, and Nils Petter Gleditsch, “Toward a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816–1992,” American Political Science Review 95 (March 2001), p. 33; Ronald A. Francisco, “The Relationship between Coercion and Protest: An Empirical Evaluation in Three Coercive States,” Journal of Conflict Resolution 39 (1995), pp. 263–82. Edward N. Muller and Erich Weede, “Cross-National variations in Political Violence: A Rational Action Approach,” Journal of Conflict Resolution 34 (1990), pp. 624–51.

N O T E S T O PA G E S 2 8 – 3 2

197

38. Scott Gates, Ha˚ vard Hegre, Mark P. Jones, and Ha˚ vard Strand, “Institutional Inconsistency and Political Instability: Persistence and Change in Political Systems Revisited, 1800–1998,” paper presented at the American Political Science Association Convention, 2000. 39. In the short run, however, liberalizing an autocratic regime does produce more violence. Hegre et al. “Toward a Democratic Civil Peace?” p. 44. 40. William J. Dixon, “Dyads, Disputes, and the Democratic Peace,” in Murray Wolfson, ed., The Political Economy of War and Peace (Boston: Kluwer, 1998), p. 103. 41. Kenneth N. Waltz, “The Emerging Structure of International Politics.” International Security 18 (Fall 1993), p. 77. 42. Hobbes’s comment on the “state of warre” comes near his comment on the absence of a “common power to keep them all in awe,” quotes that have a totemic status in international relations theory. “Hereby it is manifest, that during the time men live without a common Power to keep them all in awe, they are in that condition which is called War; and such a warre, as is of every man, against every man. For War, consisteth not in Battle only, or the act of fighting; but in a tract of time, wherein the Will to contend by Battle is sufficiently known. . . . So the nature of War, consisteth not in actual fighting; but in the known disposition thereto, during all the time there is no assurance to the contrary. All other time is Peace.” Thomas Hobbes, “Of the Natural Condition of Mankind, as concerning their Felicity, and Misery,” in Leviathan (1651; reprint, New York: E.P. Dutton, 1950), pt. 1, chap. 13, p. 103. 43. Niall Ferguson, The House of Rothschild: The World’s Bankers, 1849–1999 (New York: Viking, 1999), pp. 295–303; David S. Landes, Bankers and Pashas (Cambridge: Harvard University Press, 1958). 44. The year before, France had done something similar (and more deliberate) in Tunis. Until then, France had worked alongside Britain, Italy, and other European bond holders in jointly supervising Tunis’s finances. Neighboring Algeria was already under French colonial rule. In 1881, France decided to launch a surprise attack on Tunis, seizing it with thirty-thousand troops. The Italians, who had their own designs for a colonial sphere in north Africa, were particularly aggrieved. This colonial dispute played an important part in Italy’s decision the next year to join Germany and Austria Hungary in the Triple Alliance (concluded on 20 May 1882). 45. Britain tried to assuage French concerns about Egypt in several ways. The most important was its guarantee of free navigation through the canal. Kenneth Bourne, The Foreign Policy of Victorian England, 1830–1902 (Oxford: Clarendon Press, 1970), p. 141. 46. Kenneth A. Schultz, Democracy and Coercive Diplomacy (Cambridge: Cambridge University Press, 2001), pp. 175–96. Schultz’s insightful analysis of the Fashoda crisis emphasizes the role of incomplete information and signaling. 47. As part of the Morocco bargain with France, Britain gained more effective control over Egyptian finances. Until then, they had been supervised by the Caisse de la Dette, an awkward arrangement of several European creditor states, including France. The British, and especially Lord Cromer, its consul general in Egypt, wanted to dispense with the Caisse so they could control Egypt’s treasury as effectively as they controlled its politics. Zara S. Steiner, Britain and the Origins of the First World War (London: Macmillan, 1977), p. 29.

198

N O T E S T O PA G E S 3 2 – 3 7

48. Morocco’s position, opposite the British colony at Gibraltar, made it especially important to British military planners and foreign policy officials. It controlled entrance into Mediterranean, where Britain had extensive interests. It lay directly along the route from England to Suez to India. 49. Martin Gilbert, A History of the Twentieth Century, vol. 1; 1900–1933 (New York: William Morrow, 1997), p. 133. 50. Bismarck speaking to Eugen Wolf on 5 December 1888. 51. Britain had other good strategic reasons to be wary of a formal alliance with Germany. It would entail costly commitments to Germany and block chances for closer relations with France and Russia. Moreover, if either France or Russia did seriously challenge Britain, Germany’s own interest in the balance of power would induce it to back Britain, with or without an alliance. Britain also had ample evidence that Germany was secretly trying to stir up such challenges for Britain, primarily in colonial affairs, to force it into a pact with Germany. Francis Bertie of the U.K. Foreign Office makes that argument in a private memorandum (27 October 1901), published in Bourne, The Foreign Policy of Victorian England, pp. 464–69. Bourne agrees, p. 176. These arguments against closer ties with Germany were strengthened after Russia’s loss to Japan in 1904–5. Britain no longer needed to balance against Russia at Constantinople and so had less use for Germany as a counterweight to the Franco-Russian alliance. 52. Francis Bertie, UK Foreign Office, “Private and Secret” Memorandum of 27 October 1901, published in Bourne, The Foreign Policy of Victorian England, pp. 464, 465. 53. A decade earlier, London might well have seen the Berlin-to-Baghdad Railway as a useful counterweight to Russia, then its major rival in central Asia. But in 1904– 5, Russia was defeated by Japan. Thus weakened, Russia called off its challenge to Britain. They reached an agreement in 1907 ending the “Great Game” in central Asia. Russia’s obvious weakness had other crucial effects on European diplomacy. It allowed Germany to concentrate on France and England, rather than on any perceived Russian threat. With France already threatened by Germany and Russia now weaker and more vulnerable, the two had powerful incentives to strengthen their alliance. The rising German threat also induced closer cooperation between France and Britain, including detailed planning by their military staffs. 54. Stephen M. Walt, “Why Alliances Endure or Collapse,” Survival 39 (Spring 1997), p. 159. 55. Kenneth Bourne emphasizes this point in Britain and the Balance of Power in North America, 1815–1908 (London: Longmans, 1967). 56. Chaim D. Kaufmann and Robert A. Pape, Jr., “Explaining Costly International Moral Action: Britain’s Sixty-Year Campaign against the Atlantic Slave Trade,” International Organization 53 (Autumn 1999), pp. 631–68. 57. The Confederacy was so sensitive to Britain’s hostility to the slave trade that it actually banned the importation of slaves in its constitution. The goal was not to increase the scarcity value of existing slaves; it was to win European backing. According to the Confederate Constitution, art. I, sec. 9 (I): “The importation of negroes of the African race from any foreign country other than the slaveholding States or Territories of the United States of America, is hereby forbidden; and Congress is required to pass such laws as shall effectually prevent the same.” In a secession movement built on slave agriculture and fighting to preserve it, that is truly remarkable. It is eloquent

N O T E S T O PA G E S 3 8 – 4 4

199

testimony to the South’s need for British and other European support. Constitution of the Confederate States of America, 11 March 1861. 58. Quoted in Norman B. Ferris, The Trent Affair: A Diplomatic Crisis (Knoxville: University of Tennessee Press, 1977), p. 182. 59. America traditionally supported neutral shipping rights. Indeed, that was one reason the United States had fought Britain in 1812. The Royal Navy had been seizing sailors from American ships, and the United States found that intolerable. Britain, on the other hand, had long used its navy to interfere with enemy trade. As the world’s great sea power in the eighteenth and nineteenth centuries, it staunchly maintained its right to stop neutral shipping and confiscate contraband. Now, during the American Civil War, the United States turned the tables and asserted its own rights against neutral (British) shipping to maintain a blockade against the Confederacy. For the historic British position, see C. J. Bartlett, Defence and Diplomacy: Britain and the Great Powers, 1815–1914 (Manchester, UK: Manchester University Press, 1993), pp. 9–10. 60. Jasper Ridley, Lord Palmerston (London: Constable, 1970), pp. 742–43; the quote is from Howard Jones, The Course of American Diplomacy: From the Revolution to the Present (New York: Franklin Watts, 1985), p. 182. 61. Jones, The Course of American Diplomacy, p. 182. 62. Canada also encouraged European immigrant settlement in the prairies and British Columbia, partly for economic reasons, partly to head off American demographic expansion there. I am indebted to Duncan Snidal and Mario Seccareccia for their help on these points. 63. One speculation is that the United States hoped for Canada as compensation. 64. Bourne, Britain and the Balance of Power, p. 410. 65. Quoted in Ferris, The Trent Affair, p. 12. 66. “Letter of Instruction” to American Ambassador T. F. Bayard in London from Secretary of State Richard Olney, 20 July 1895. 67. This point tracks Michael Doyle and John Owen’s view that democracies recognize and respect the legitimacy of each other’s domestic governance. 68. Akira Iriye, After Imperialism: The Search for a New Order in the Far East, 1921–1931 (Cambridge: Harvard University Press, 1965); Michael A. Barnhart, Japan and the World since 1868 (London: Edward Arnold, 1995); Sydney Giffard, Japan among the Powers, 1890–1990 (New Haven: Yale University Press, 1994); W. G. Beasley, Japanese Imperialism, 1894–1945 (Oxford: Clarendon Press 1987). 69. England’s decision to retain these forts was made before the Treaty of Paris was formally ratified on 8 April 1784. According to Howard Jones, London’s secret decision was then communicated to Britain’s governor-general in Canada. The timing suggests that (a) the British acted in bad faith, and (b) their claim that the forts were retained because the United States refused to pay its debts is not the entire story. Howard Jones, The Course of American Diplomacy: From the Revolution to the Present (New York: Franklin Watts, 1985), p. 18. 70. Jonathan R. Dull, A Diplomatic History of the American Revolution (New Haven: Yale University Press, 1985), pp. 150–51. 71. Dull, A Diplomatic History of the American Revolution, p. 150. 72. The British had one more concern. For some time, the Foreign Office and Admiralty had forecast that the United States would become a major sea power in its own

200

N O T E S T O PA G E S 4 8 – 5 0

right, drawing on its vast industrial capacity to challenge Britain’s naval dominance. C. J. Bartlett cites that one of Britain’s “very pressing reasons [was] to try to remove causes of friction with their transatlantic ‘cousins.’ ” It is worth adding that Britain’s key decision was to remove the causes of friction, not exacerbate them by reasserting British naval power in the face of America’s potential threat. Bartlett, Defence and Diplomacy, p. 6. Chapter 3 A Contracting Theory of the Democratic Peace and Its Alternatives 1. James D. Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” American Political Science Review 88 (September 1994), pp. 577– 92. 2. Black’s Median Voter Theorem says that the median voter’s preference point cannot be defeated in a majority-rule election, given certain restrictions. The main restrictions are that there is a single-policy dimension and that preferences are single-peaked. In two-party competition, party platforms will also converge on that point, according to the Hotelling-Downs Theorem. Duncan Black, The Theory of Committees and Elections (Cambridge: Cambridge University Press, 1958); Anthony Downs, An Economic Theory of Democracy (New York: Harper and Row, 1957); Harold Hotelling, “Stability in Competition,” Economic Journal 39 (1929), pp. 41–57. 3. Actually, war can be rational ex post if the participants really enjoy fighting (and preparing to fight) and value it more highly than the other costs they suffer. 4. Stephen Van Evera, Causes of War: Power and the Roots of Conflict (Ithaca: Cornell University Press, 1999), chap. 2. 5. The great student of these perception and bias problems in international politics is Robert Jervis. The Logic of Images (Princeton: Princeton University Press, 1970); Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976); and Jervis, “War and Misperception,” in Robert I. Rotberg and Theodore K. Rabb, eds., The Origin and Prevention of Major Wars (Cambridge: Cambridge University Press, 1989), pp. 101–26. 6. James D. Fearon, “Rationalist Explanations for War,” International Organization 49 (Summer 1995), pp. 379–414. 7. It is conceivable, of course, that some participants actually relish fighting for its own sake. They would not accept bargains that prevent war unless they were fully compensated for the lost value of fighting. As for Pareto-improving solutions: they do not need to make everyone strictly better off. According to the standard definition, they must make no one worse off and at least one participant better off. 8. John Kennan provides a valuable summary of this literature in “The Economics of Strikes,” in Orley C. Ashenfelter and Richard Layard, eds., Handbook of Labor Economics, vol. 2 (Amsterdam: North-Holland, 1986), pp. 1091–1137. 9. George J. Borjas, Labor Economics (New York: McGraw-Hill, 1996), p. 379. 10. John R. Hicks, The Theory of Wages (London: Macmillan, 1932). 11. Ronald G. Ehrenberg and Robert S. Smith, Modern Labor Economics, 6th ed. (Reading, MA: Addison-Wesley Longman, 1997), p. 499. 12. The company’s optimistic forecast is surely honest if the union is its only partner. If there are other partners, however, then the company’s optimistic forecast could be

N O T E S T O PA G E S 5 0 – 5 1

201

designed to fool them. Corporations may put on a happy face for employees to fool investors who are watching closely. 13. In this simple example of wage bargaining, only the firm has private information. Therefore, only the firm needs to send costly signals of its resolve, based on its (hidden) financial position. Note that the strike occurs even though one side (the union) has no hidden information. Of course, in many real-world cases, both sides need to send costly signals because both have private information. 14. In pretrial negotiations, the scope of private information is obviously affected by statutes and court decisions governing disclosure. In union-management negotiations, the firm’s private information may be limited if it is a public company, subject to regulations that force disclosure of material information. In these ways and others, laws and regulations can restrict the scope of private information in domestic negotiations—opening a wider range of ex ante bargaining solutions. 15. Francesco Guicciardini, Storia d’Italia (ca. 1536–40). 16. Lawrence Freedman, US Intelligence and the Soviet Strategic Threat (Basingstoke, UK: Macmillan, 1977), pp. 65–67; Norman Friedman, The Fifty-Year War: Conflict and Strategy in the Cold War (Annapolis, MD: U.S. Naval Institute, 2000), pp. 200–201. Thanks to Bob Pape for discussing the bomber incident with me. 17. America’s exaggerated estimate of Soviet bomber production was not due entirely to Soviet deception. The CIA made a number of other high predictions, which were then confirmed by the deception. The mistakes had real consequences. Fearful that the Soviet Union was rapidly adding new bombers, America responded by building more B-52s for its Strategic Air Command. The U.S. response shows why exaggerating military capabilities can be a doubleedged sword. It can deter aggression (by masking weakness), but it can also stimulate an adversary to build more forces to counter the imaginary threat. 18. Fearon makes the important point that states could agree on their relative power, the cost of fighting, and even the respective likelihood of winning, but still go to war if they misjudged how highly their adversary valued victory, relative to its costs. That valuation could be private information, which is not easily discovered or credibly signaled. Fearon, “Rationalist Explanations for War,” pp. 394–95. 19. Macaulay’s sketch of William of Orange appears in The History of England, Vol. 2 chap. 7 (1848) “Long before he reached manhood he knew how to keep secrets, how to baffle curiosity by dry and guarded answers, how to conceal all passions under the same show of grave tranquility.” These traits are among the reasons why Macaulay thinks young William is mature enough to rule Holland. 20. “The single thing for which we can blame him [Cesare Borgia, Duke Valentino] is the election of Julius as pope. In this he made a bad choice because, as I have said, if he could not set up a pope to suit himself, he could exclude from the papacy whomever he wished. He should never have let the papacy go to any cardinal whom he had injured or who, on becoming pope, would need to fear him, because men do injury through either fear or hate. . . . In this choice, then, the Duke blundered, and it caused his final ruin.” Machiavelli repeatedly warns that resentment for old injuries is longlived and is not erased by more recent favors. Niccolo` Machiavelli, The Prince, in Machiavelli: The Chief Works and Others, vol. 1, trans. Allan Gilbert (Durham, NC: Duke University Press, 1989), chap. 7, p. 34.

202

N O T E S T O PA G E S 5 1 – 5 3

21. Machiavelli’s references to Pope Julius II in The Prince may be confusing because he often calls him San Piero ad Vincula. Like all high-ranking prelates, he had more than one name. Born Giuliano Della Rovere, he rose to the position of cardinal priest of San Piero in Vincoli before his election as pontiff. He was known by his elevated title rather than his family name. So he was “San Piero ad Vincula” during the election period, when Cesare Borgia made his fatal miscalculation. 22. Machiavelli was in Rome, representing Florence, in autumn 1503 when Julius was elected pope. He watched Julius win the election with Duke Valentino’s aid. Within days, he saw Julius betray Valentino, undermine his position in the Romagna region, and then imprison him. He records it all in letters to his employers, the rulers of Florence. On 4 November 1503, Machiavelli writes, “This Pontiff [Julius II] has been elected with very great good will. . . . To Duke Valentino, of whom he has made more use than of any other, it is said that he has promised to restore all the territory of Romagna, and has granted him Ostia for safety.” By 10 November, Machiavelli writes that the pope has put Valentino in a difficult position, “even though this Pontiff has always been considered a man of great fidelity.” The duke, now recognizing his peril, started to assemble a cavalry and infantry, and left Rome. On 23 November, Machiavelli writes that the pope had demanded Valentino give up key fortified cities. When Valentino refused, “the Pope has had him arrested. . . . This Duke now is in the Pope’s power.” He never escaped it. Machiavelli’s letters of October–November 1503 appear in “Legation 13,” in Machiavelli: The Chief Works and Others 1:142–60. 23. Machiavelli, The Prince 1:64. 24. Fearon, “Rationalist Explanations for War,” p. 391. 25. Cordelia speaking prophetically in King Lear, act 1, sc. 1. (Plaited means “hidden.”) This slow unfolding of cunning is central to the play. Lear freely (and impetuously) gives away his kingdom to two daughters, Goneril and Regan, who coax him with their obsequious professions of love. Once in possession of the kingdom, they abandon and betray their father, who is now powerless to stop them. Lear descends into madness as he reflects on his self-inflicted fate. The tragedy is compounded because, when Lear divided the kingdom, he denied a share to his third daughter, Cordelia. She had angered the King by refusing to become a sycophant like her sisters. Only later, when Lear is helpless, vulnerable, and crazed, does he comprehend Cordelia’s true devotion to him. The play is thus a game with incomplete information illustrating a time-inconsistent bargain (that is, Lear’s current gift of the kingdom in return for the prospect of a gentle retirement, surrounded by grateful children). Fortunately, Shakespeare chooses to describe it his own way, without diagramming the actors’ information sets. 26. Fiona McGillivray and Alastair Smith, “Trust and Cooperation through AgentSpecific Punishments,” International Organization 54 (Autumn 2000), pp. 808–24. McGillivray and Smith add an important point. When one country exploits another, it reaps immediate benefits but suffers long-term costs (a combination of direct punishments and foregone benefits of future cooperation). Democratic leaders are held accountable by their electorate for these consequences. Autocratic leaders are not. But does that actually make cooperation more likely among democracies, as McGillivray and Smith speculate? The answer depends on whether the net gains from defection are positive or negative. If there are real gains from defection, then democratic electorates will support it by definition. (If leaders or the electorate are ethically committed to

N O T E S T O PA G E S 5 4 – 5 7

203

fulfilling their bargains, that adds to the cost of deliberately breaking them and lowers the net gains.) 27. A rare counterexample is Charles de Gaulle’s cry of “Vive Que´ bec libre!” during a 1967 visit to Canada’s French-speaking province. His commitment to French-Canadian aspirations stimulated the movement for regional independence. Jean Lacouture, De Gaulle: The Ruler, 1945–1970 (New York: W. W. Norton, 1992), pp. 447–64. 28. Some of these features affect liberal democracies rather than constitutional democracies, which may not have all the features of liberalism. Some have monadic effects. But the most important features seem to be related to constitutionalism and have dyadic effects. 29. Kenneth N. Waltz, Theory of International Politics (Reading, MA: AddisonWesley, 1979), p. 126. 30. Waltz, Theory of International Politics. 31. Martin Wight is recognized as the founder of the “English School,” with its focus on international society and historical analysis of states-systems. The quote is from Martin Wight, Power Politics, ed. Hedley Bull and Carstein Holbraad (New York: Holmes and Meier, 1978), p. 139. 32. Robert Jervis, “Cooperation under the Security Dilemma,” World Politics 30 (January 1978), pp. 167–214, plus Jervis’s other extensive work on perception and signaling. 33. Andrew Kydd, “Sheep in Sheep’s Clothing: Why Security Seekers Do Not Fight Each Other,” Security Studies 7 (Autumn 1997), pp. 114, 114–55. 34. Some traditional Realists, such as Henry Kissinger, acknowledge that domestic institutions and values can shape foreign policies, almost always for the worse. A typical example is Kissinger’s thoughtful critique of the Habsburg emperors refusing to ally with non-Catholic powers during the Reformation era. The didactic point, however, is to show that this is self-indulgent foolishness. Even if the motives are worthy, the policy produces only quixotic ventures, overextension, and defeat. It leads to endless wars of principle rather than successful, amoral compromises—which usually have better moral consequences anyway. That was the consistent thrust of Realist criticism of American foreign policy during the early Cold War, led by such eminent figures as George Kennan and Hans Morgenthau. They were understandably worried that moralistic anticommunism would become the rationale for limitless involvement overseas and aggressive global conflict. The same issues arise now in the war against terrorism. The mandarins’ critique of democratic institutions and democratic foreign policy is often too general, depreciating the public’s contribution to policy debate. The democratic peace poses a problem for this perspective because, in this case at least, domestic institutions and values produce outcomes that are clearly desirable. See Henry Kissinger, Diplomacy (New York: Simon and Schuster, 1994), esp. chap. 3 (on Richelieu); George F. Kennan, American Diplomacy, 1900–1950 (New York: New American Library, 1952); George F. Kennan, Realities of American Foreign Policy (Princeton: Princeton University Press, 1954); Hans J. Morgenthau, Politics among Nations: The Struggle for Power and Peace (New York: Alfred A. Knopf, 1950). 35. This theme is developed in Bruce Russett and John R. Oneal, Triangulating Peace: Democracy, Interdependence, and International Organizations (New York: W. W. Norton, 2001).

204

N O T E S T O PA G E S 5 7 – 5 8

36. One prominent Realist account that does emphasize variations in fear is Stephen M. Walt, The Origin of Alliances (Ithaca: Cornell University Press, 1987). Also see Stephen M. Walt, Revolutions and War (Ithaca: Cornell University Press, 1996). 37. Alexander Wendt, “Constructing International Politics,” International Security 20 (Summer 1995), pp.71–81. 38. On this cycle of fear, action, and reaction, see Robert Jervis’s discussion of spiral models in Perception and Misperception in International Politics, chap. 3. 39. That is Deborah Welch Larson’s central point in Anatomy of Mistrust: U.S.Soviet Relations during the Cold War (Ithaca: Cornell University Press, 1997). 40. Stephen Walt does ask these questions in his analysis of alliances, which stresses balancing against threats rather than against capabilities. The Origin of Alliances. 41. The relative-gains debate was mainly between neo-Realists and liberal institutionalists. Some of the key works in the debate are Waltz, Theory of International Politics, esp. p. 126; Robert Gilpin, War and Change in World Politics (Cambridge: Cambridge University Press, 1981), pp. 87–88; Joseph M. Grieco, Cooperation among Nations: Europe, America, and Non-Tariff Barriers to Trade (Ithaca: Cornell University Press, 1990); Charles Lipson, “International Cooperation in Economic and Security Affairs,” World Politics 37 (October 1984), pp. 1–23; Michael Mastanduno, “Do Relative Gains Matter? America’s Response to Japanese Industrial Policy,” International Security 16 (Summer 1991), pp. 73–113; Duncan Snidal, “International Cooperation among Relative Gains Maximizers,” International Studies Quarterly 35 (December 1991), pp. 387–402; Duncan Snidal, “Relative Gains and the Pattern of International Cooperation,” American Political Science Review 85 (September 1991), pp. 701–26; and Robert Powell, “Absolute and Relative Gains in International Relations Theory,” American Political Science Review 85 (December 1991), pp. 1302–20. 42. Grieco, Cooperation among Nations, p. 28. 43. There is a much more straightforward explanation for U.S.-European divisions in the trade talks. Hard bargaining over agriculture can be understood as difficulty “dividing the pie.” The United States and Europe were engaged in a hard-fought struggle to win the bulk of the gains any new trade agreement would provide. These gains include political goals such as protecting vital domestic constituencies as well as improving macroeconomic performance and economic efficiency. So there was tough bargaining between the United States and the European Union (EU) over a variety of agricultural subsidies, mainly because the United States is an efficient agricultural producer and the EU wants to protect small, inefficient farmers within its lucrative market. The United States did the same thing in the mid-1950s, protecting Wisconsin cheese makers from the dangers of superior French products. Fears of enslavement had nothing to do with it. At the height of its Cold War power, the United States could have withstood the threat to national security posed by camembert cheese. The same kind of bargaining happens within the EU itself, where the protagonists are clearly unconcerned that the winner will enslave them. There, the hard bargaining is between the states that receive most of the subsidies (because their countryside has more inefficient producers) and the other states that have to pay them. For the past decade or more, the EU has also been very concerned about agricultural exports from states, such as Australia, that pose no conceivable military threat.

N O T E S T O PA G E S 5 8 – 6 2

205

Grieco does not rule out this more straightforward answer—namely, that the protagonists were simply trying to win a game of distributional bargaining that had nothing to do with long-term military concerns. In the case of U.S.-EU trade, bargaining is particularly difficult because the EU’s negotiating position is based on a prior deal among its member governments. That makes it especially difficult for EU negotiators to trade away the self-declared interests of major members such as France, where the central government is determined to protect small farmers, even if the trade-off would produce overall economic gains for the EU. Despite all these difficulties, it is worth remembering that the United States and Europe struck a series of far-reaching bargains on most aspects of trade during the Tokyo and Uruguay Rounds. 44. Adolf Berle diary entry, in B. B. Berle and T. B. Jacobs, eds., Navigating the Rapids, 1918–1971: From the Papers of Adolf A. Berle (New York: Harcourt Brace Jovanovich, 1973), pp. 374–5. 45. Geography also affects threats, but its impact depends on prevailing military technology. Take the Golan Heights, which Israel won from Syria during the 1967 war. Its strategic importance has declined with the proliferation of short-range rockets and cruise missiles, which can overfly this high ground. 46. Incomplete information is the centerpiece of Thomas Schelling’s classic work on deterrence. Thomas C. Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960). Also, Thomas C. Schelling, Arms and Influence (New Haven: Yale University Press, 1966). 47. Robert O. Keohane, “The Demand for International Regimes,” International Organization 36 (Spring 1982), pp. 325–55. 48. Alexander Wendt, Social Theory of International Politics (Cambridge: Cambridge University Press, 1999). 49. Thomas Risse-Kappen, “Collective Identity in a Democratic Community: The Case of NATO,” in Peter J. Katzenstein, ed., The Culture of National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996), p. 371. 50. Emanuel Adler, “Seizing the Middle Ground: Constructivism in World Politics,” European Journal of International Relations 3 (September 1997), p. 347. Also see Bruce Conin, Community under Anarchy: Transnational Identity and the Evolution of Cooperation (New York: Columbia University Press, 1999); Colin H. Kahl, “Constructing a Separate Peace: Constructivism, Collective Liberal Identity, and Democratic Peace,” Security Studies 8 (Winter 1998/99–Spring 1999), pp. 94–144; Michael C. Williams, “The Discipline of the Democratic Peace: Kant, Liberalism and the Social Construction of Security Communities,” European Journal of International Relations 7 (December 2001), pp. 525–53. 51. Desiderius Erasmus, Institutio Principis Christiani, trans. Percy Ellwood Corbett (London: Grotius Society/Sweet and Maxwell, 1921), p. 47. 52. Hedley Bull’s Anarchical Society is the most sustained and coherent statement of the English School. It is also the most penetrating, along with the work of Martin Wight. Wight preceded Bull and laid out the main themes of “international society,” but published little in his lifetime. Wight’s influence stemmed from his teaching, mainly at the London School of Economics and his central role on the British Committee on the Theory of International Politics, which included such major figures as Herbert Butterfield, Adam Watson, Michael Howard, Coral Bell, and Hedley Bull. After

206

N O T E S T O PA G E S 6 2 – 6 5

Wight’s death, his manuscripts were edited and published by Bull and others. Hedley Bull, The Anarchical Society: A Study of Order in World Politics (New York: Columbia University Press, 1977); Hedley Bull and Adam Watson, eds., The Expansion of International Society (Oxford: Clarendon Press, 1984); Martin Wight, “An Anatomy of International Thought,” Review of International Studies 13 (July 1987), pp. 221–27; Martin Wight, International Theory: The Three Traditions, ed. Gabriele Wight and Brian Porter (Leicester, UK: Leicester University Press for the Royal Institute of International Affairs, 1991); Wight, Power Politics; Martin Wight, Systems of States, ed. Hedley Bull (Leicester, UK: Leicester University Press for the London School of Economics, 1977); Martin Wight, “Western Values in International Relations,” in Herbert Butterfield and Martin Wight, eds., Diplomatic Investigations: Essays in the Theory of International Politics (Cambridge: Harvard University Press, 1966), pp. 89–131; Martin Wight, “Why Is There No International Relations Theory?” in Butterfield and Wight, Diplomatic Investigations, pp. 17–34. 53. Hedley Bull, The Anarchical Society: A Study of Order in World Politics, 2nd ed. (New York: Columbia University Press, 1995), p. 15. 54. Bull, The Anarchical Society, p. 13. 55. Bull, The Anarchical Society, chap. 8. 56. Bull, The Anarchical Society, p. 17. 57. Bull, The Anarchical Society, p. 191. 58. Wight, Power Politics, p. 138. 59. Wight, Power Politics, pp. 142–43. Bull makes the same point without the Hobbesian overtones. The Anarchical Society, pp. 184–86. 60. Martha Finnemore considers this lack of clarity, both about methods and about theoretical claims, one of the main defects of the English School. Iver Neumann speaks more gently about the School’s “methodological and epistemological pluralism,” but reaches a similar conclusion: it prevents scientific certainty. Martha Finnemore, “Exporting the English School?” Review of International Studies 27 (July 2001), pp. 509– 13; Iver B. Neumann, “The English School and the Practices of World Society,” Review of International Studies 27 (July 2001), pp. 503–7. 61. Bull, The Anarchical Society, p. 189. 62. Bull, The Anarchical Society, pp. 190–93. 63. The best analysis is by Andrew Moravcsik, The Choice for Europe: Social Purpose and State Power from Messina to Maastricht (Ithaca: Cornell University Press, 1998). 64. Arbitration was used in the Trent shipping damages case, involving the United States and Great Britain, and in the Venezuelan Boundary Dispute, involving British Guyana, Venezuela, and, behind the scenes, the United States. Both are discussed in chap. 2. 65. Celeste A. Wallander, Mortal Friends, Best Enemies: German-Russian Cooperation after the Cold War (Ithaca: Cornell University Press, 1999); Celeste A. Wallander and Robert O. Keohane, “Risk, Threat, and Security Institutions,” in Helga Haftendorn, Robert O. Keohane, and Celeste Wallander, eds., Imperfect Unions: Security Institutions over Time and Space (New York: Oxford University Press, 1999), pp. 21–47. 66. Other capacities of international institutions, such as peacekeeping in turbulent regions, have not been relevant to the democratic peace.

N O T E S T O PA G E S 6 5 – 6 7

207

67. International institutions might promote the democratic peace in the future by following NATO’s example. That is, they could provide a stable political environment, underwritten by outside powers, that eases security concerns for new democracies and gives them time to strengthen their domestic regimes. Indeed, NATO’s own expansion into central Europe is designed to do exactly that. 68. The connection between information, institutions, and international outcomes is central to the work of Robert Keohane. One of Keohane’s major contributions is to show how international institutions promote cooperation by facilitating better information. Robert O. Keohane, “The Demand for International Regimes” International Organization 36 (Spring 1982), pp. 325–55; Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton: Princeton University Press, 1984); Wallander and Keohane, “Risk, Threat, and Security Institutions,” pp. 21–47. 69. Michael Cox, G. John Ikenberry, and Takashi Inoguchi, eds., American Democracy Promotion: Impulses, Strategies, and Impacts (New York: Oxford University Press, 2000); Tony Smith, America’s Mission: The United States and the Worldwide Struggle for Democracy in the Twentieth Century (Princeton: Princeton University Press, 1994). 70. On the hierarchical character of the Soviet informal empire in Eastern Europe, see Alexander Wendt and Daniel Friedheim, “Hierarchy under Anarchy: Informal Empire and the East German State,” in Thomas J. Biersteker and Cynthia Weber, eds., State Sovereignty as a Social Construct (Cambridge: Cambridge University Press, 1996), pp. 240–72. 71. Prominent work on hegemonic stability theory includes Robert Gilpin, U.S. Power and the Multinational Corporation (New York: Basic Books, 1975); Charles P. Kindleberger, The World in Depression, 1929–1939 (Berkeley and Los Angeles: University of California Press, 1973); Stephen D. Krasner, “State Power and the Structure of International Trade,” World Politics 28 (April 1976), pp. 317–43; Keohane, After Hegemony; Duncan Snidal, “The Limits of Hegemonic Stability Theory,” International Organization 39 (Autumn 1985), pp. 579–614; Arthur A. Stein, “The Hegemon’s Dilemma: Great Britain, the United States, and the International Economic Order,” International Organization 38 (Spring 1984), pp. 355–86. The most recent and novel contribution, stressing the creation of international “constitutional orders,” is G. John Ikenberry, After Victory: Institutions, Strategic Restraint, and the Rebuilding of Order after Major Wars (Princeton: Princeton University Press, 2001). 72. Charles Lipson, Standing Guard: Protecting Foreign Capital in the Nineteenth and Twentieth Centuries (Berkeley and Los Angeles: University of California Press, 1985), chap. 2. 73. Karl Marx, “Civil War in France,” (1871), quoted in Margot Light, The Soviet Theory of International Relations (New York: St. Martin’s Press, 1988), p. 238. 74. Marx, “Civil War in France” (1871), in Robert C. Tucker, The Marx-Engels Reader, 2d ed. (New York: W. W. Norton, 1978), p. 651. 75. V. I. Lenin, Imperialism, The Highest Stage of Capitalism; A Popular Outline, rev. translation (New York: International Publishers, 1933). 76. Light, The Soviet Theory of International Relations, p. 212. 77. Peter Jones and Siaˆ n Kevill, comp., China and the Soviet Union 1949–1984 (Harlow, UK: Longman, 1985).

208

N O T E S T O PA G E S 6 7 – 6 8

78. Alfred D. Low, The Sino-Soviet Confrontation since Mao Zedong: Dispute, Detente, or Conflict? (Boulder, CO: Social Science Monographs; New York: Columbia University Press, 1987). 79. Charles McGregor, The Sino-Vietnamese Relationship and the Soviet Union, Adelphi Papers no. 232 (London: International Institute for Strategic Studies, 1988); Anne Gilks, The Breakdown of the Sino-Vietnamese Alliance, 1970–1979 (Berkeley: Institute of East Asian Studies, University of California, Center for Chinese Studies, 1992); William J. Duiker, China and Vietnam: The Roots of Conflict (Berkeley: Institute of East Asian Studies, University of California, 1986). 80. The Corn Laws, a tariff on grain imports, were repealed in 1846 and mark the beginning of Britain’s commitment to free-trade policies. Paul A. Pickering and Alex Tyrell, The People’s Bread: A History of the Anti–Corn Law League (London: Leicester University Press, 2000). 81. Wendy Hinde, Richard Cobden (New Haven: Yale University Press, 1987), p. 214. Cobden was also a strong supporter of nonintervention, disarmament, and compulsory arbitration (p. 202). 82. Japan’s colonial conquests and its drive for greater autonomy are obviously far different from the recommendations of classical liberals. The problem for Japan was that its trade ties with the United States, especially its import of almost all its refined petroleum, made it vulnerable to economic sanctions. By the summer of 1941, the Roosevelt administration forced the leaders of Imperial Japan to make a choice. If they wished to continue trading with the United States and avoid an oil embargo, they would have to give up their conquests in China and forgo others, which they were planning in Southeast Asia and the Dutch East Indies. Their alternative was to create a more self-sufficient empire by force and coercive diplomacy. That would require secure sealines for commerce and communication throughout the South Pacific. The British and American navies were obvious threats to those sea-lines. That is why, once the Japanese made the fateful decision to expand their empire despite strenuous Anglo-American opposition, they rapidly moved to secure their sea-lines. They simultaneously attacked the British in Malaya (on the way to Singapore) and the Americans in the Philippines and Hawaii. That prepared the way for a twofold conquest. They would take Southeast Asia and the Netherlands East Indies to provide Japan with oil, rubber, and other raw materials. They would take Singapore and islands in the South Pacific to provide a defense perimeter for the new empire. 83. John R. Oneal and Bruce M. Russett, “The Classical Liberals Were Right: Democracy, Interdependence, and Conflict, 1950–1985,” International Studies Quarterly 41 (June 1997), pp. 267–94; John R. Oneal and Bruce M. Russett, “Assessing the Liberal Peace with Alternative Specifications: Trade Still Reduces Conflict,” Journal of Peace Research 30 (July 1999), pp. 423–42; Solomon W. Polachek, John Robst, and Yuan-Ching Chang, “Liberalism and Interdependence: Extending the Trade-Conflict Model,” Journal of Peace Research 36 (July 1999), pp. 405–22; C. H. Anderton and J. R. Carter, “The Impact of War on Trade: An Interrupted Times-Series Study,” Journal of Peace Research 38 (July 2001), pp. 445–57; Eric Gartzke, Quan Li, and Charles Boehmer, “Investing in the Peace: Economic Interdependence and International Conflict,” International Organization 55 (Spring 2001), pp. 391–438.

N O T E S T O PA G E S 6 8 – 7 2

209

84. Susan M. McMillan, “Interdependence and Conflict,” International Studies Quarterly 41, suppl. 1 (May 1997), pp. 33–58. McMillan’s survey covers formal models, case studies, and quantitative research. 85. The classical liberal argument assumes that war interrupts trade, raising the costs of war for states with close commercial ties. That argument is undercut if war interrupts trade only briefly or if alternative markets can be readily found. Katherine Barbieri and Jack S. Levy have published findings that show trade is not severely hampered by war. Their findings have been challenged by other scholars. Katherine Barbieri and Jack S. Levy, “Sleeping with the Enemy: The Impact of War on Trade,” Journal of Peace Research 36 (July 1999), pp. 463–79. 86. James D. Morrow, “How Could Trade Affect Conflict?” Journal of Peace Research 36 (July 1999), pp. 481–89. 87. Andrew Moravcsik, “Taking Preferences Seriously: A Liberal Theory of International Politics,” International Organization 51 (Autumn 1997), pp. 513–53. 88. Moravcsik, “Taking Preferences Seriously,” p. 527. 89. Moravcsik recognizes that it is misleading to speak of purely domestic preferences leading to international outcomes. The reason is that rational domestic actors take into account the likely strategies and reactions of others and build that anticipation into their own strategies. So international constraints come in the back door. 90. Moravcsik, “Taking Preferences Seriously,” p. 524. 91. Joanne Gowa makes the same point from the Realist side. 92. That method, which informs most institutionalist work, is developed in Kenneth Oye, ed., Cooperation under Anarchy (Princeton: Princeton University Press, 1985). 93. Anne-Marie Burley and Walter Mattli, “Europe before the Court: A Political Theory of Legal Integration,” International Organization 47 (Winter 1993), pp. 41–76. 94. James Cameron and Karen Campbell, eds., Dispute Resolution in the World Trade Organisation (London: Cameron May, 1998). 95. John Mueller, Retreat from Doomsday: The Obsolescence of Major War (New York: Basic Books, 1989); Mueller, “The Essential Irrelevance of Nuclear Weapons: Stability in the Postwar World,” International Security 13 (Fall 1988), pp. 55–79. Also see Carl Kaysen, “Is War Obsolete? A Review Essay,” International Security 14 (Spring 1990), pp. 42–64. 96. Lyn Macdonald, Somme (London: M. Joseph, 1983); A. H. Farrar-Hockley, The Somme (1964; reprint, London: Pan Books, 1983); David French, “The Meaning of Attrition, 1914–1916,” English Historical Review 103 (April 1988), pp. 385–405. 97. Ian Kershaw, Hitler, 1936–1945: Nemesis (New York: W. W. Norton, 2000), pp. 119–22, on Mussolini’s intervention in the Munich negotiations, which Kershaw calls “decisive.” 98. The Somme and other trench warfare in World War I drove home the horrors of modern conventional war for the combatants. Deadly as it was, that war was fought mainly in rural areas. World War II, by contrast, involved airborne attacks on major cities. The bombing of London, Cologne, Dresden, Berlin, and Tokyo showed that mobilized civilian populations were vulnerable to punishing attacks. The use of nuclear weapons at Hiroshima and Nagasaki showed that future attacks on population centers could be more awful still. 99. Jervis, The Logic of Images; Jervis, Perception and Misperception in International Politics; Jervis, “War and Misperception,” pp. 101–26.

210

N O T E S T O PA G E S 7 2 – 7 8

100. Larson, Anatomy of Mistrust. 101. Jervis, “Cooperation under the Security Dilemma,” pp. 167–214. 102. Jervis, “War and Misperception,” p. 103. 103. Jervis, “War and Misperception,” p. 112. 104. The classic discussion of “reciprocal fear of surprise attack,” is by Thomas Schelling in Strategy of Conflict (Cambridge: Harvard University Press, 1960), pp. 207–29. Jervis explores these issues in “War and Misperception,” p. 120. Chapter 4 Why Democratic Bargains Are Reliable 1. The best argument for open, freely-contested discourse is still John Stuart Mill’s “Of the Liberty of Thought and Discussion,” which stresses its utilitarian value for truth-finding, John Stuart Mill, On Liberty (1859; reprint, London: Penguin, 1974), chap. 2. 2. A more detailed taxonomy of transparency would highlight differences in the kinds of information disclosed (such as tangible military hardware versus “orders of battle” and strategy) and the means of disclosure (ranging from self-disclosure to investigation by others, from voluntary cooperation to obstruction). James Marquardt has discussed these issues in unpublished work (University of Chicago, 1995). Kenneth Abbott has shown how such different kinds of information provision are reflected in different kinds of international agreements. Kenneth Abbott, “Trust but Verify: The Production of Information in Arms Control Treaties and Other International Agreements,” Cornell Journal of International Law 26 (Winter 1993), pp. 1–58. 3. Democratic institutions not only inform and constrain, they inform about constraints, letting citizens and outsiders know the limits of state power. Whether information or institutional constraints play a larger role in the democratic peace is analyzed in Kenneth A. Schultz, “Do Democratic Institutions Constrain or Inform? Contrasting Two Institutional Perspectives on Democracy and War,” International Organization 53 (Spring 1999), pp. 233–66. 4. For a simple, memorable demonstration of using reputation as a hostage, recall President George H. W. Bush’s pledge not to raise taxes. Since politicians are always making pledges not to raise taxes, the voting public is usually skeptical. To give his pledge added weight, Bush chose to frame it in pithy, clear language that staked his reputation on its fulfillment. Over and over, he repeated the swaggering phrase, “Read my lips. No new taxes.” Fortunately for Bush-the-candidate, the pledge was very convincing. Unfortunately for Bush-the-president, the pledge was well remembered. When President Bush faced strong pressures to raise taxes as part of a budget deal, he finally agreed to raise taxes. He might have sincerely thought that he had already carried out his tax pledge since he had not raised taxes for some time after the election. Candidate Bush, however, had set no time limit on his pledge, and its brevity clearly suggested that he would never agree to raise taxes since he was so dead set against them. When he did agree to hike taxes, he was severely punished by voters. Many other examples could be cited where politicians make very visible pledges to convince voters they are not lying. Newt Gingrich’s “Contract with America,” which Republican congressional candidates signed in 1994, is a good example.

N O T E S T O PA G E S 7 8 – 7 9

211

5. The basic law on treaties is the 1969 Vienna Convention, which clarified and codified existing legal practices. Naturally, it defined the term treaty. It states that “for the purposes of the present Convention: ‘treaty’ means an international agreement concluded between States in written form and governed by international law, whether embodied in a single instrument or in two or more related instruments and whatever its particular designation” (pt. I, art. 2a) American treaty law is considerably more complex. Many agreements that the Vienna Convention considers “treaties” are given lesser status in U.S. law. In the United States, treaties refer only to international agreements that have been ratified by twothirds of the Senate. Treaties, under the U.S. constitution, are the supreme law of the land; all other agreements have less far-reaching consequences and, to avoid confusion, are referred to by other names. Agreements made by the president alone are called executive agreements. Still other agreements are approved by both the Senate and the House, but without supermajorities. In international legal parlance, these differences do not matter. All are equally binding treaty obligations. Other states have their own treaty laws, of course. Most are closer to the usage of the Vienna Convention. But whether or not domestic laws distinguish among different kinds of agreements, diplomats definitely do. They use variations in the publicness and formality of agreements to signal each other about the depth of their commitments. Basic international treaty law is stated in the Vienna Convention on the Law of Treaties, signed 22 May 1969, entered into force 27 January 1980; reprinted in American Journal of International Law 63 (1969), pp. 875ff.; and in International Legal Materials 8 (1969), pp. 679ff. The classic statement of treaty law and interpretation is Lord McNair [Baron Arnold Duncan McNair], The Law of Treaties (Oxford: Clarendon Press, 1961). The relevant sections of the U.S. Constitution are art. II, sec. 2 [2] and art. VI [2]. The authorization for other agreements beside treaties comes from the president’s general powers to conduct foreign policy, also contained in art. II. 6. Charles Lipson, “Why Are Some International Agreements Informal?” International Organization 45 (Autumn 1991), pp. 495–538. 7. Charles Howard McIlwain, Constitutionalism, Ancient and Modern (Ithaca: Cornell University Press, 1947), p. 93. 8. Marc Bloch, Feudal Society, vol. 2, Social Classes and Political Organization, trans. L. A. Manyon (London: Routledge and Kegan Paul, 1961), pp. 451–52. 9. Strong states often provide aid and comfort to fragile partners, precisely because they are fragile and their partners want them to survive. That aid can include a number of asymmetric agreements. Though these agreements advance both sides’ interests (by definition), they are very different from the usual international bargains for mutual advantage. What is important for our purposes is that the larger, stronger partner does not need to rely on them. Rather, it is using the agreements to try to create a more reliable partner for the future. 10. Adam Przeworski, Democracy and the Market: Political and Economic Reforms in Eastern Europe and Latin America (Cambridge: Cambridge University Press, 1991), pp. 12–13. 11. Strategic surprise, though rare and difficult, is still possible within democracies because national security policymaking is typically centralized and subject to special provisions of secrecy. International monetary affairs are very similar. These special conditions pose special risks to partners. But even in these areas, large matters of policy

212

N O T E S T O PA G E S 7 9 – 8 1

and grand strategy are subject to extensive discussion and legislative oversight. Even in these centralized and secretive policy arenas, strategic surprise is much harder for democracies to achieve than for nondemocracies, where secrecy is far more pervasive. Michael I. Handel, The Diplomacy of Surprise: Hitler, Nixon, Sadat (Cambridge: Harvard Center for International Affairs, 1981). 12. David Collier and Stephen Levitsky, “Democracy with Adjectives: Conceptual Innovation in Comparative Research,” World Politics 49 (April 1997), pp. 430–51; Miriam Fendius Elman, “Unpacking Democracy: Presidentialism, Parliamentarism, and Theories of Democratic Peace,” Security Studies 9 (Summer 2000), pp. 91–126. Elman identifies two broad institutional categories. The “more majoritarian,” she says, are Westminster-style parliaments and semipresidential systems. “Less majoritarian” are coalitional parliaments and presidential systems. Westminster parliaments feature single-member districts, which are won by a plurality of the vote. They tend to eliminate small parties unless they have strong ties to a particular region, where they can win outright. A party that wins 10 percent or 20 percent of votes across the country might not win a plurality in any district and might wind up with no seats at all. The typical result is a parliament dominated by two (or sometimes three) national parties, with the winner usually controlling an outright majority of seats. Proportional voting, by contrast, allows smaller parties to win some seats (close to their percentage of the vote) and often prevents larger parties from winning a clear majority of parliamentary seats. That produces coalition governments and gives small parties disproportionate power, since they can credibly threaten to break the government by leaving. 13. The U.S. system also grants unusual powers to the federal courts, which can overturn laws if the courts deem them unconstitutional. Such judicial power is rare among democracies. 14. J. Mark Ramseyer, “The Puzzling (In)dependence of Courts: A Comparative Approach,” Journal of Legal Studies 23 (June 1994), p. 722. Ramseyer, an expert on Japan, argues that Japan lacks an independent judiciary because it lacks effective party rotation. Thomas E. Baker nicely captures Ramseyer’s important conclusion: The pragmatic explanation for the persistence of judicial independence in the federal government is found in the theory of repeated games designed into the nomination and confirmation process. Rational players in the executive and legislative branches have kept federal courts independent because under the Constitution regular elections in the political branches are assured, and there is little likelihood for either major party to dominate both branches permanently.

Thomas E. Baker, “A Catalogue of Judicial Federalism in the United States,” University of South Carolina Law Review 46 (Summer 1995), p. 844. I am indebted to Daniel Klerman for his comments on this point. 15. James D. Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” American Political Science Review 88 (September 1994), pp. 577– 92; James D. Fearon, “Signaling Foreign Policy Interests: Tying Hands versus Sinking Costs,” Journal of Conflict Resolution 41 (February 1997), pp. 68–90. 16. Joe Eyerman and Robert A. Hart, Jr., “An Empirical Test of the Audience Cost Proposition.” Journal of Conflict Resolution 40 (December 1996), pp. 597–616; Peter J. Partell and Glenn Palmer, “Audience Costs and Interstate Crises: An Empirical As-

N O T E S T O PA G E S 8 1 – 8 3

213

sessment of Fearon’s Model of Dispute Outcomes,” International Studies Quarterly 43 (June 1999), pp. 389–405; Kenneth A. Schultz, “Looking for Audience Costs,” Journal of Conflict Resolution 45 (February 2001), pp. 32–60. Alistair Smith extends the modeling process in “International Crises and Domestic Politics,” American Political Science Review 92 (September 1998), pp. 623–38. 17. Christopher F. Gelpi and Michael Griesdorf, “Winners or Losers? Democracies in International Crisis, 1918–1994,” American Political Science Review 95 (September 2001), pp. 646. 18. Jean-Se´ bastien Rioux, “A Crisis-Based Evaluation of the Democratic Peace Proposition,” Canadian Journal of Political Science 31 (June 1998), pp. 263–83. Rioux links the selection effect (democracies select crises they can win and avoid those they cannot) to domestic audience costs. His data, like Gelpi and Griesdorf’s, is drawn from the International Crisis Behavior Project. 19. Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” pp. 577–92. 20. Partell and Palmer, “Audience Costs and Interstate Crises,” p. 402. 21. Partell and Palmer’s measure of executive constraints is drawn from Polity III’s seven-point scale. 22. Kenneth A. Schultz, Democracy and Coercive Diplomacy (Cambridge: Cambridge University Press, 2001), p. 6. Also see chap. 8. 23. The work of Ian Macneil is among the most imaginative explorations of contracting as ways of projecting promises into the future. Macneil’s stress on relational agreements, which depend on the sustained connections between the individual parties, as opposed to arms-length formal contracts enforced by courts, is directly relevant to international agreements. Ian R. Macneil, The New Social Contract: An Inquiry into Modern Contractual Relations (New Haven: Yale University Press, 1980); Ian Macneil, “Reflections on Relational Contract,” Journal of Institutional and Theoretical Economics (Zeitschrift fu¨ r die gasamte Staatswissenschaft) 141 (December 1985), pp. 541–46. 24. William Pitt (the Elder), earl of Chatham, speech, 14 January 1766. 25. Deborah Larson’s work is especially good at spelling out these reputational and perception issues in modern diplomacy. 26. Robert Keohane calls them “inconvenient commitments.” 27. Kenneth Medhurst, “Spain’s Evolutionary Pathway from Dictatorship to Democracy,” in Geoffrey Pridham, ed., The New Mediterranean Democracies: Regime Transition in Spain, Greece, and Portugal (London: Frank Cass, 1984), p. 30. Also see, David Gilmour, The Transformation of Spain: From Franco to the Constitutional Monarchy (London: Quartet Books, 1985); and Kenneth Maxwell and Steven Spiegel, The New Spain: From Isolation to Influence (New York: Council on Foreign Relations, 1994), pp. 4–47. 28. Charles T. Powell, “Spain’s External Relations, 1898–1975,” in Richard Gillespie, Fernando Rodrigo, and Jonathan Story, eds., Democratic Spain: Reshaping External Relations in a Changing World (London: Routledge, 1995), pp. 11–29; Jonathan Story, “Spain’s External Relations Redefined, 1975–1989,” in Richard Gillespie, Fernando Rodrigo, and Jonathan Story, eds., Democratic Spain: Reshaping External Relations in a Changing World (London: Routledge, 1995), pp. 30–49; and Fernando Rodrigo, “Western Alignment: Spain’s Security Policy,” in Gillespie, Rodrigo, and Story, Democratic Spain, pp. 50–66.

214

N O T E S T O PA G E S 8 4 – 8 8

29. Mancur Olson, Jr., “Why the Transition from Communism Is so Difficult,” Eastern Economic Journal 21 (Fall 1995), pp. 437–61. 30. Martin Meredith, Our Votes, Our Guns: Robert Mugabe and the Tragedy of Zimbabwe (New York: Public Affairs, 2002). As for the 2002 election in Zimbabwe, the Economist called it “a blatantly rigged ballot,” a view shared by independent observers. “The Poll that Bob Stole,” Economist, “Special Report: Zimbabwe’s Future,” 16 March 2002, pp. 27–29. 31. It should be clear that I am referring here to the domestic regime. 32. Median voters probably work in the same direction as well, although that is theoretically unclear when multiple policy issues are considered. 33. Kenneth Schultz stresses the opposition party’s role in informing other states whether a particular policy will be sustained or abandoned. 34. New York Times (national edition), 13 September 1996. 35. The United Nations report was issued by Roberto Garreto´ n, special rapporteur on the situation of human rights in the Democratic Republic of the Congo. The report on ethnic massacres in eastern Congo is horrifying (UN Document E/CN.4/1998/64, on 23 January 1998).Garreto´ n and UN Human Rights commissions continued to make reports over the next several years. 36. Les Roberts, International Rescue Committee, “Mortality Study, Eastern Democratic Republic of Congo,” February–April 2001. The study shows 2.5 million “excess” deaths during the first thirty-two months of fighting. http://www.theirc.org/ mortality.cfm. 37. Brantly Womack, Introduction to Brantly Womack, ed., Contemporary Chinese Politics in Historical Perspective (Cambridge: Cambridge University Press, 1991), p. 1. 38. I thank Hein Goemans for raising this important question and helping me think about it. 39. Another answer might be that countries maintain secrecy because it gives them bargaining advantages. With more private information, they can reach solutions that are closer to their reservation price. They can do better than the more open countries sitting across the bargaining table. This bargaining edge may be a benefit from secrecy, but that is not the reason why dictatorships keep a tight lid on information. They do it to stay in power. Moreover, the immediate bargaining advantages it confers must be balanced against the long-run strategic advantages of mutual transparency, if it can be prudently achieved. 40. As Brantly Womack has written of the politics of post-Tiananmen China, “The reality . . . . is a polarization between a regime stigmatized by its use of repressive measures and a profoundly alienated urban society whose vocal spokesmen are in exile abroad.” Womack, introduction, p. 16. 41. As Alexander Solzhenitsyn observed in the early 1970s, “the lie has been incorporated into the state system as the vital link holding everything together, with billions of tiny fasteners, several dozen to each man.” Quoted in Timur Kuran, “Now Out of Never: The Element of Surprise in the East European Revolution of 1989,” World Politics 44 (October 1991), pp. 26–27. 42. Robert Dahl, among others, has set a lower hurdle for the news media, saying that the job of the news media is to help citizens develop an “adequate understanding” of political issues, including an understanding of their own preferences. For an informed and subtle discussion of the media’s proper role in modern democracy, see Michael

N O T E S T O PA G E 8 8

215

Schudson, The Power of News (Cambridge: Harvard University Press, 1995), chap. 10. Dahl’s views can be found in “Liberal Democracy in the United States,” in William S. Livingston, ed., A Prospect of Liberal Democracy (Austin: University of Texas Press, 1979), pp. 57–72, which Schudson cites. 43. The connections between adult literacy, popular newspapers, and electoral democracy are intriguing and underexplored by social historians. One goal should be to understand the direction of causation between the spread of the franchise and the growth of popular newspapers with political content (or, if causation runs in both directions, their relative weights). I expect that the lower costs of gathering and communicating information over long distances played an important role in fostering national newspapers during this period, and that may have had multiple effects on the character of local political contests. The fact that papers were closely associated with political parties probably spurred competition in the development of mass-market papers, lest the opposition party have the field to itself. The rise of mass literacy clearly had an important impact, which was reinforced because this new “middle-class” public was the target for mass distribution of all kinds of consumer goods. With more and more readers available, cheap newspapers began to be sold on the streets. Until then, newspapers were expensive and sold only by subscription, which was the sole source of revenues. Helping to lower the costs was the major new source of revenues: advertising. Hence, there was a close connection between the rise of mass literacy, the spread of a mass consumer market, and the development of newspapers to reach this new market for advertisers. Finally, especially within Europe, the “contagion” effects across countries are interesting. Rather than approaching each state separately, we can ask how the early development of a political press in some states affected its rise elsewhere. In the United States, these issues are explored in Michael Schudson, Discovering the News: A Social History of American Newspapers (New York: Basic Books, 1978); in Britain, Raymond Williams, The Long Revolution (New York: Columbia University Press, 1961). The rise of mass consumer markets, with a special focus on distribution systems, is the central focus of Alfred D. Chandler’s work, The Visible Hand: The Managerial Revolution in American Business (Cambridge: Harvard University Press, 1977). As Chandler points out in a later work, there were some first-mover advantages for consumer brands, which led them to spend heavily on national advertising campaigns to achieve economies of scale and to forestall entry by competitors. Alfred D. Chandler, Scale and Scope: The Dynamics of Industrial Capitalism (Cambridge: Harvard University Press, 1990), pp. 64–65. 44. The closest any polity came to press freedom was Norway, and it was not a fully independent state at the time. On the overall record of European press freedom, see Robert Justin Goldstein, Political Repression in Nineteenth-Century Europe (London: Croom Helm, 1983), chap. 2. 45. Special taxes on newspapers raised their prices and limited circulation, and sometimes made it impossible for newspapers to operate profitably at all. One measure of their effect was the rapid growth in circulation and numbers of papers during periods when taxes were temporarily reduced. 46. On the expansion of the franchise, see Adam Przeworski and John Sprague, Paper Stones: A History of Electoral Socialism (Chicago: University of Chicago Press, 1986), p. 36, table 2.3, col. 3, which gives the dates of universal male suffrage in European states.

216

N O T E S T O PA G E S 8 8 – 9 0

47. Goldstein, Political Repression in Nineteenth-Century Europe, p. 37. 48. Plebiscites in Uruguay and Chile in 1980 played an important role in breaking those authoritarian governments. Their rulers had expected clear victories, or else they would not have called the elections. When they did not get them, the regime’s opponents discovered that disaffection was far more widespread than previously thought—invigorating them and weakening the repressive regimes. Kathryn Sikkink offered very helpful ideas to me on this point. 49. Susanne Lohmann, “Dynamics of Informational Cascades: The Monday Demonstrations in Leipzig, East Germany, 1989–1991,” World Politics 47 (October 1994), pp. 42–101; Kuran, “Now Out of Never,” pp. 7–48; Timur Kuran, “Sparks and Prairie Fires: A Theory of Unanticipated Revolution,” Public Choice 61 (April 1989), pp. 41– 74. 50. Richard Baum, Burying Mao: Chinese Politics in the Age of Deng Xiaoping (Princeton: Princeton University Press, 1994), pp. 245–310; Richard Baum, “The Road to Tiananmen: Chinese Politics in the 1980s,” in Roderick MacFarquhar, ed., The Politics of China, 1949–1989 (Cambridge: Cambridge University Press, 1993), pp. 340– 471, esp. pp. 407ff; Ruan Ming, Deng Xiaoping: Chronicle of an Empire, trans. and ed. Nancy Liu, Peter Rand, and Lawrence Sullivan (Boulder, CO: Westview Press, 1994), pp. 211–26. I am indebted to Tang Tsou and Adam H. Arkel for their help on the events surrounding Tiananmen and the democracy movement. For background and documents, see Michael Oksenberg, Lawrence Sullivan, and Marc Lambert, eds., Beijing Spring, 1989 (Armonk, NY: M. E. Sharpe, 1990). 51. The goal of the Chinese regime, to quote Brantly Womack again, is “modernization, but it is unwilling to acknowledge the unfolding of a society that is autonomous in its economic and political decision making. Hence, there is a tension between the decentralizing reforms of the 1980s and the Leninist denial of privacy and citizenship. That tension has led to vacillation between periods of pragmatic suspension of orthodoxy (with everyone merely being encouraged to ‘get rich’) and periods of crackdown and retrenchment.” “Introduction,” p. 5. 52. These new technologies not only give dissidents new ways of communicating; they have another, more sinister side. They allow much more extensive government inspection of personal records, such as credit card purchases, phone calls, income, and the like. Two-way television may present even more intrusive possibilities. There is a complex and evolving interplay between increased state efforts to collect and analyze personal information, on the one hand, and the dispersion of sites to create information and better ways to encrypt it, on the other. 53. Lohmann, “Dynamics of Informational Cascades,” pp. 42–101. 54. The Mothers’ Marches at the Plaza de Mayo in Buenos Aires took place in front of Government House (the Casa Rosadao) in Buenos Aires. They are a compelling example of nonviolent demonstrations that catalyze real political change. The marches began spontaneously among the mothers and grandmothers of the “disappeared.” For months they had been rebuffed in their efforts to find out what had happened to their loved ones. They suspected what we now know, that the Argentine military junta had killed them. Death squads had been operating, officially and unofficially, since 1974, when a state of siege had been imposed. The mothers, exasperated by official silence, met informally to discuss their grievances and simply decided, on the spot, to walk to the Plaza de Maya on 30 April 1977. It was illegal for more than two people to gather, much less to demonstrate or picket, but they quietly defied the law and got away with

N O T E S T O PA G E S 9 0 – 9 1

217

it. They decided to meet again the next Thursday, and then every Thursday afternoon, for a vigil. They soon adopted a distinctive emblem, a baby’s diaper around their heads, and began to demonstrate in other official locations, such as police stations and local churches. These demonstrations spread word of their cause and showed the depth of their commitment. They served as the nucleus of a human rights movement that ultimately undermined the generals’ regime. Ian Guest, Behind the Disappearances: Argentina’s Dirty War against Human Rights and the United Nations (Philadelphia: University of Pennsylvania Press, 1990); Allison Brysk, The Politics of Human Rights in Argentina: Protest, Change, and Democratization (Stanford, CA: Stanford University Press, 1994); Marguerite Guzman Bouvard, Revolutionizing Motherhood: The Mothers of the Plaza de Mayo (Wilmington, DE: Scholarly Resources, 1994); the eventual trials of the junta members are reported in Amnesty International, Argentina: The Military Juntas and Human Rights. Report of the Trial of the Former Junta Members, 1985 (London: Amnesty International, 1987). I am indebted to Kathryn Sikkink for her valuable advice on this issue. 55. Tang Tsou, “The Tiananmen Tragedy: The State-Society Relationship, Choices, and Mechanisms in Historical Perspective,” in Womack, Contemporary Chinese Politics in Historical Perspective pp. 265–327. 56. This protection for speech rights is a fundamental difference between tolerant liberal democracies in the European Enlightenment tradition and illiberal democracies, which utilize majoritarian rule to advance a specific set of social objectives and suppress its alternatives. These illiberal democracies are a kind of midpoint between the sham elections of one-party states (where rulers would be rejected at the polls) and the free elections of multiparty democracies. There are important limits on the scope and freedom of electoral competition, but the winners in these illiberal races might well have gotten the most votes in a free contest. Hitler staged no votes after he became Fu¨ rher, but he would have easily won a fair election in 1939. The Islamic Republic of Iran holds quite extensive elections and has some electoral competition, but within strict limits, at least during the regime’s first two decades. Wilhelmine Germany had an elected parliament and active party competition, but there were also tight limits on the policies over which parliament could exercise any responsibility. Its role in national security matters and foreign affairs was especially restricted. Any explanation for “peace among democracies” needs to consider cases like these and reasonable counterfactuals like the Hitler example and ask if they are covered by the proposed explanation. For example, any explanation that factually asserts that most people will oppose war because it is too costly should consider the popular enthusiasm for military expansion in Nazi Germany, and, indeed, the expansive nationalism of many nineteenthcentury Western democracies and many new ones today. 57. For an important study of these open discourse and disagreements within democracies, see Amy Gutmann and Dennis Thompson, Democracy and Disagreement (Cambridge: Harvard University Press, 1996), esp. chap. 3. Gutmann and Thompson make the fundamental point that publicity within democracies has two distinct purposes. One is to connect officials and citizens, ensuring that their reasons for preferring policies are openly given. There is a kind of practical reciprocity here. The other rationale is, as Gutmann and Thompson put it, “the independent moral value of openness in government” (p. 8). To that, I would add the independent moral value of citizens’ expression of their views as aspects of their human autonomy.

218

N O T E S T O PA G E S 9 1 – 9 5

58. Human freedom, as Michael Doyle argues, is the central principle of the liberal polity, the generative force behind its rights and institutions. Michael W. Doyle, “Kant, Liberal Legacies, and Foreign Affairs,” pt. 1, Philosophy and Public Affairs 12 (Summer 1983), p. 207. 59. This point is raised in Frederick Schauer, Free Speech: A Philosophical Enquiry (Cambridge: Cambridge University Press, 1982), p. 36. 60. For an analysis of the scope and limits of democratic rights to free speech, see Cass R. Sunstein, Democracy and the Problem of Free Speech (New York: Free Press, 1993). 61. Fareed Zakaria, “The Rise of Illiberal Democracy,” Foreign Affairs 76 (November/December 1997), pp. 22–43; Daniel A. Bell and Kanishka Jayasuriya, “Understanding Illiberal Democracy: A Framework,” in Daniel A. Bell, David Brown, Kanishka Jayasuriya, and David Martin Jones, Towards Illiberal Democracy in Pacific Asia (New York: St. Martin’s Press, 1995). 62. For a comparison of veto points in different democratic systems, see George Tsebelis, “Decision Making in Political Systems: Veto Players in Presidentialism, Parliamentarism, Multicameralism, and Multipartyism,” British Journal of Political Science 25 (July 1995), pp. 289–325. I am indebted to Peter Lange for his comments on this point. 63. Louis Henkin, Constitutionalism, Democracy, and Foreign Affairs (New York: Columbia University Press, 1990); Michael J. Glendon, Constitutional Diplomacy (Princeton: Princeton University Press, 1990). 64. Mark Ramseyer, “The Puzzling (In)dependence of Courts,” pp. 721–47. 65. The involvement of courts in foreign policy issues is a decidedly mixed blessing. They can solidify commitments, but they also lack a background in foreign policy and necessarily give legal answers to political questions. The larger problem, however, is that they add yet another voice to the cacophony speaking on foreign policy issues. That is understandable in a democracy, but it can confuse foreign partners—even if it generally helps sustain promises to them. 66. Robert A. Pastor, Whirlpool: U.S. Foreign Policy toward Latin America and the Caribbean (Princeton: Princeton University Press, 1992), p. 46. 67. Timothy Stater, “Climax: Senate Ratification, 1977–1978,” in G. Harvey Summ and Tom Kelly, eds., The Good Neighbors: America, Panama, and the 1977 Canal Treaties (Athens: Ohio University Center for International Studies, 1988), pp. 75–96; J. Michael Hogan, The Panama Canal in American Politics: Domestic Advocacy and the Evolution of Policy (Carbondale: Southern Illinois University Press, 1986), chap. 4, “Selling New Canal Treaties,” and chap. 8, “Public Opinion and the Senate Debate.” 68. William L. Furlong and Margaret E. Scranton, The Dynamics of Foreign Policymaking: The President, the Congress, and the Panama Canal Treaties (Boulder, CO: Westview, 1984); George D. Moffett, The Limits of Victory: The Ratification of the Panama Canal Treaties (Ithaca: Cornell University Press, 1985). 69. The U.S. policy debate went beyond ocean resources. For conservative critics, the Law of the Sea Treaty was a compendium of fundamental policy mistakes, all made in the name of international cooperation. According to this view, the treaty set up a burdensome regulatory environment that did a lot to tax entrepreneurial companies and very little to protect and encourage deep-seabed mining, where U.S. firms were leaders. Rather than capturing joint gains, it was a bald effort to establish political control by

N O T E S T O PA G E S 9 5 – 9 6

219

rent-seeking Third World states. Why should the United States agree to that? Some new cooperative arrangements might well be desirable, but they could never be negotiated by softhearted U.S. multilateralists, who were more interested in placating America’s opponents than in using America’s wealth and power to advance its interests and extract maximum concessions. So went the conservative critique, led by President Reagan. Reagan often opposed multilateral solutions, but he was no isolationist. Quite the contrary; he envisioned America not just as an ideal exemplar, but as an active player in world affairs. He vigorously supported free trade and was ready to fight communist guerrillas wherever they popped up. His policies were a kind of right-wing variation on the idealistic internationalism of Woodrow Wilson, rather than a simple continuation of the realpolitik of Richard Nixon and Henry Kissinger. Reagan himself was imbued with Wilson’s self-confident vision of America’s unique mission in the world. What Reagan really wanted was to extend all forms of American power (military, diplomatic, and corporate); to contain communism in Europe and East Asia and roll it back on the cheap elsewhere, using local proxy force; and to spread the worldly gospel of “Americanism” and free enterprise. His opposition to the Law of the Sea was wholly consistent with this broader stance. For the official administration view on the Law of the Sea, see President Reagan’s statements on 9 July 1982 and 10 March 1983, plus an article by his chief negotiator, James L. Malone, “Who Needs the Sea Treaty?” Foreign Policy, no. 54 (Spring 1984), pp. 44–63, plus an exchange of letters between Malone and Elliot L. Richardson, who had been President Carter’s special representative, in Foreign Policy, no. 55 (Summer 1984), pp. 178–82. The Reagan statements are “Statement on United States Actions Concerning the Conference on the Law of the Sea,” Public Papers of the President of the United States: Ronald Reagan, 1982, bk. 2 (Washington, DC: Government Printing Office, 1983), pp. 911–12; and “Statement on United States Oceans Policy,” Public Papers of the President of the United States: Ronald Reagan, 1983, bk. 1 (Washington, DC: Government Printing Office, 1984), pp. 378–79. Also see, Kenneth A. Oye, “International Systems Structure and American Foreign Policy,” in Kenneth A. Oye, Robert J. Lieber, and Donald Rothchild, eds., Eagle Defiant: United States Foreign Policy in the 1980s (Boston: Little, Brown, 1983), pp. 25–26, for his discussion of Reagan administration views on multilateralism; and the articles in Lawrence Juda, ed., The United States without the Law of the Sea Treaty: Opportunities and Costs (Wakefield, RI: Times Press, 1983). For the comparisons of Wilson, Nixon, and Reagan, see Henry Kissinger, Diplomacy (New York: Simon and Schuster, 1994). 70. The best treatments are James K. Sebenius, Negotiating the Law of the Sea (Cambridge: Harvard University Press, 1984); and Robert L. Friedheim, Negotiating the New Ocean Regime (Columbia: University of South Carolina Press, 1993). 71. David Reynolds, The Creation of the Anglo-American Alliance, 1937–1941: A Study in Competitive Co-operation (Chapel Hill: University of North Carolina Press, 1982). Most studies focus on U.S. aid to Britain and Russia, but the Americans also made some effort to help the French. John McVickar Haight, Jr., American Aid to France, 1938–1940 (New York: Atheneum, 1970). 72. Robert A. Divine, The Reluctant Belligerent, 2d ed. (New York: Wiley, 1979); Robert A. Divine, Roosevelt and World War II (Harmondsworth, UK: Penguin Books,

220

N O T E S T O PA G E S 9 6 – 9 7

1970); Wayne S. Cole, Roosevelt and the Isolationists, 1932–1945 (Lincoln: University of Nebraska Press, 1983). 73. Waldo Heinrichs, “President Franklin D. Roosevelt’s Intervention in the Battle of the Atlantic,” Diplomatic History 10 (Fall 1986), p. 311–32. 74. Roosevelt’s effectiveness in educating the public and preparing it for war are superbly analyzed in David M. Kennedy, Freedom from Fear: The American People in Depression and War, 1929–1949 (New York: Oxford University Press, 1999), chaps. 13–15. 75. Roosevelt had to maneuver carefully to send the anti-Nazi aid he did. A quartercentury earlier, President Woodrow Wilson’s decision to send aid to belligerents had been an important step toward America’s involvement in World War I. Congressional leaders remembered. In the late 1930s, fearing that the U.S. would again be dragged into war, Congress passed new laws preventing some types of assistance. Roosevelt wanted to eliminate these restrictions but dared not do so directly. He feared it would not just fail but backfire. According to Robert Dallek, it could “agitate the animus toward Britain and increase fears of war, which had produced the legislation in the first place. The [1940] election and recent opinion polls suggested that the country remained strongly opposed to involvement in the fighting and painfully divided about risking war to save Britain. In short, Roosevelt understood that a stable consensus for large-scale aid financed by the United States depended on avoiding suggestions that the country was duplicating the experience of World War I.” That is why Roosevelt had to do much more than just lobby the Congress. He first had to persuade the public at large. Robert Dallek, Franklin D. Roosevelt and American Foreign Policy, 1932–1945 (Oxford: Oxford University Press, 1979), p. 253; William L. O’Neill, A Democracy at War: America’s Fight at Home and Abroad in World War II (Cambridge: Harvard University Press, 1993), pp. 22–25. 76. The Lend-Lease Bill, HR-1776, passed the Senate overwhelmingly on 8 March 1941 and the House of Representatives three days later. Roosevelt succeeded not so much because he lobbied Congress, but because he promoted the policy successfully to the public at large. When the policy battle began, public opinion was divided. But that was before Roosevelt’s Fireside Chat (29 December 1940) and subsequent publicity by the administration. By the time the vote finally took place, the public supported Lend-Lease by a two-to-one margin. O’Neill, A Democracy at War, pp. 24–25. 77. Wolfgang Leonhard, Betrayal: The Hitler-Stalin Pact of 1939 (New York: St. Martin’s, 1989), p. 28. 78. Hitler’s aims can be seen in the invasion of Holland and Belgium in 1940, which were incidental to a larger goal. “The purpose of the German invasion,” according to Gerhard Weinberg, “was to crush the French and British forces on the continent so that Germany would have quiet in the West while conquering living space from the Soviet Union in the East.” Gerhard L. Weinberg, A World at Arms: A Global History of World War II (Cambridge: Cambridge University Press, 1994), p. 122. 79. Hitler’s decision to hasten the onset of war before his rivals could prepare is the reason he was actually disappointed by the appeasement at Munich. He was seeking the pretext for a wider war, and he wanted it sooner rather than later. He was after a far bigger prize than the Sudetenland. Chamberlain’s readiness to make concessions, and Mussolini’s desire to act as intermediary, robbed Hitler of the war he wanted during the Munich crisis. Weinberg, A World at Arms, pp. 27–29, 142, 178.

N O T E S T O PA G E S 9 7 – 9 8

221

80. Weinberg, A World at Arms, p. 64. 81. Stalin, who is rarely accused of being too sanguine about world affairs, was actually too optimistic about the chances for longer-term cooperation with Hitler. In the summer of 1940, Gerhard Weinberg notes, “from the perspective of Moscow, Germany had every incentive to maintain its good relations with the Soviet Union—which had opened such marvelous opportunities for her—and might in fact be willing to make even more far-reaching new arrangements for the division of yet greater spoils.” As Russia saw it, there was a chance to collude beyond the Molotov-Ribbentrop Pact. What Stalin may not have fully understood was that any compromise would have been temporary because Nazi Germany’s thirst for more territory was unquenchable and was bound to threaten the Soviet Union. “War had been an intended and even a preferred part of National Socialist policy from the beginning,” Weinberg writes, “not so much out of a preference for fighting for its own sake, but from the entirely accurate conviction that the aim of German expansion could be secured only by war. . . . The vast reaches of additional land to be obtained would never be granted peacefully, and war was therefore both necessary and inevitable. The bulk of the land to be conquered was in Russia.” Weinberg, A World at Arms, pp. 164, 20–21, also pp. 200–201 for additional evidence that Stalin thought longer-term cooperation with Hitler was possible. 82. According to Gerhard Weinberg, the great historian of World War II and of German foreign policy, “It should not be surprising that already in mid and late May of 1940, as soon as it became clear that the German offensive in the West was going forward as quickly and successfully as Hitler could possibly hope, he began to turn his thoughts to the attack on the Soviet Union. He was beginning to discuss this project with his military associates in late May, and in June had them starting on the first preparations of plans for such an operation.” As Weinberg later notes, the invasion of Poland was less an end in itself than a dress rehearsal for the invasion of Russia. Weinberg, A World at Arms, pp. 179, 190. 83. R. Ernest Dupuy and Trevor N. Dupuy, The Encyclopedia of Military History, from 3500 B.C. to the Present, 2d rev. ed. (New York: Harper and Row, 1986), p. 1077; Weinberg, A World at Arms, pp. 196–99. 84. Weinberg, A World at Arms, p. 198. 85. Graham Ross, The Great Powers and the Decline of the European States System, 1914–1945 (London: Longman, 1983), p. 18. 86. The country’s need to speak with one coherent voice is strongest when foreign threats are strongest. That is why when countries go to war, they so often adopt “governments of national unity.” Likewise, the powers of heads of government usually increase during wartime, along with those of the central government generally. The same trends are evident during periods of extended threat. During the first two decades of the Cold War, when the dangers were most palpable, the U.S. presidency strengthened markedly. That rationale for strong executive power eroded in the latter stages of the Cold War and its aftermath. That is one reason why, in the 1990s, the power of Congress grew relative to that of the president, and why the power of federal states grew relative to that of Washington. The logic of a strong, centralized government and, within it, a strong executive, is simply not as compelling when there is no clear and present danger. By the same logic, a prolonged and difficult war on the terror would likely reverse these trends. For a discussion of how wartime pressures had a permanent impact on the size of the U.S. government, see Bartholomew H. Sparrow, From the Outside In: World War

222

N O T E S T O PA G E S 9 8 – 1 0 2

II and the American State (Princeton: Princeton University Press, 1996). For a theoretical analysis of the relationship between the chief executive’s power and foreign threats, see Daniel Verdier, Democracy and International Trade: Britain, France, and the United States, 1860–1990 (Princeton: Princeton University Press, 1994). 87. The exclusion of the wider public—and often the exclusion of informed elites— means that nondemocratic governments also have no way to improve public policy by discussion and criticism. That is another price of narrowing the policymaking domain to one voice. 88. “If the sovereign can change the law at will,” writes Stephen Holmes, “then law cannot possibly bind or restrict the sovereign.” That, I think, is why absolute monarchs find it so difficult to make stable international bargains and reassure their neighbors. If a sovereign can abandon a commitment at will, then commitments will not bind the sovereign—or reassure partners on the other side of the bargain. Stephen Holmes, Passions and Constraint: On the Theory of Liberal Democracy (Chicago: University of Chicago Press, 1995), p. 145. 89. James D. Fearon, “Rationalist Explanations for War,” International Organization 49 (Summer 1995), pp. 379–414. 90. M. S. Anderson, The Rise of Modern Diplomacy, 1450–1919 (London: Longman, 1993), pp. 13–14. 91. Anderson, The Rise of Modern Diplomacy, p. 14. 92. Charles Wilson, The Transformation of Europe, 1558–1648 (Berkeley and Los Angeles: University of California Press, 1976), p. ix. 93. Albert Speer, Inside the Third Reich (London: Cardinal, 1975), p. 258. 94. Lenin statement, 27 September 1905, quoted in Joseph D. Douglass, Why the Soviets Violate Arms Control Treaties (London: Pergamon-Brassey’s, 1988), p. 123. 95. Quoted in Arnold Beichman, The Long Pretense: Soviet Treaty Diplomacy from Lenin to Gorbachev (New Brunswick, NJ: Transaction Books, 1991), p. 7. 96. Ramseyer, “The Puzzling (In)dependence of Courts,” pp. 721–47. 97. Douglass C. North and Barry R. Weingast, “Constitutions and Commitment: The Evolution of Institutions Governing Public Choice in Seventeenth-Century England,” Journal of Economic History 49 (December 1989), pp. 803–32. 98. The problems of commercial dealings with the English Crown had been starkly revealed in 1640. At the time, merchants regularly deposited their surplus coin and bullion in the Royal Mint, which was housed in the Tower. At the outbreak of the Civil War, Charles I became so desperate that he seized these private funds and returned them only after the depositors provided him with a coerced loan. After England’s Civil War, and especially after the Glorious Revolution of 1688, Parliament was in a strong position to stop such predation. It regularized government finances and assisted London’s merchant community by chartering a major banking house like the ones in Amsterdam and Hamburg. The result was the new Bank of England, which combined both public and private functions. Parliament took care to ensure that the new institution came under the direct control of Parliament, not the Crown. The bank was initially proposed by William Paterson, who knew the government needed to raise capital. He combined his idea for a new bank with a novel way of raising money for the government. Instead of simply loaning the government money for a fixed period, as was normally done, the Bank would establish “a fund of Perpetual

N O T E S T O PA G E S 1 0 2 – 3

223

Interest.” In effect, the bank would establish a national debt. Both the bank and the national debt proposals were debated at length by the cabinet and then passed in 1694. They reflect not only England’s emerging commercial growth but also a new level of confidence in the state’s financial commitments. Creditors could hardly be expected to purchase government debt unless they figured it would be paid regularly and promptly. As Douglass North has argued, “A capital market entails security of property rights over time and will simply not evolve where political rulers can arbitrarily seize assets or radically alter their value. Establishing a credible commitment to secure property rights over time requires either a ruler who exercises forbearance and restraint in using coercive force, or the shackling of the ruler’s power to prevent arbitrary seizure of assets. The first alternative was seldom successful for very long in the face of the ubiquitous fiscal crises of rulers (largely as a consequence of repeated warfare). The latter entailed a fundamental restructuring of the polity such as occurred in England as a result of the Glorious Revolution of 1688, which resulted in parliamentary supremacy over the crown.” Douglass C. North, “Institutions,” Journal of Economic Perspectives 5 (Winter 1991), p. 101; John Giuseppi, The Bank of England: A History from Its Foundation in 1694 (London: Evans Brothers, 1966), pp. 8–11. 99. Niccolo` Machiavelli, Discourses on Livy, trans. Harvey C. Mansfield and Nathan Tarcov (ca. 1519; reprint, Chicago: University of Chicago Press, 1996), bk. I, chap. 59. I am indebted to Nathan Tarcov for his comments on this issue. 100. Machiavelli, Discourses on Livy, bk. I, chap. 59. 101. Machiavelli, Discourses on Livy, bk. I, chap. 59. 102. Revolutionary regimes pose some of the same risks. Like older autocratic regimes, they can usually act with stealth. Their newness itself adds to the uncertainty. In addition, there may be systematic bias in overestimating revolutionary dangers to surrounding states. The concerns of neighboring states are easy to understand. Revolutionary regimes typically have closed decisionmaking by small leadership cadres. They have often proclaimed universal goals that challenge the legitimacy of neighboring regimes and that purport to be universally applicable. Former patriots from the ancien re´ gime have often fled abroad and have nothing good to say about the new rulers, who have stolen their property, position, and dignity. Finally, as an entirely new government, they have no reputation for keeping agreements. It is not surprising that their presence amplifies the security dilemma. In the argument of this chapter, the main problem presented by revolutionary regimes is not so much their conflictual aims—although these undoubtedly impede agreements because they make it more difficult to locate joint gains—but rather the high expected variance that surrounds their most likely policies, and the aversion to risks that other states have. See Stephen M. Walt, “Revolution and War,” World Politics 44 (April 1992), pp. 321–68. 103. Although nondemocracies have the advantage of surprise, they have at least two systematic military weaknesses. First, they often need to use their armies for domestic repression and social control, rather than purely for defense against external threats. It is difficult to shape an army that can perform effectively both as a police force and as a fighting force. Second, it may be harder to extract resources voluntarily from society, especially high-quality recruits and experienced, educated leaders. Their foot soldiers and sailors are all too often lower-class conscripts who have been forced into service and whose morale reflects the miserable terms of their recruitment.

224

N O T E S T O PA G E S 1 0 3 – 6

104. The classic statement of the security dilemma is John H. Herz, “Idealist Internationalism and the Security Dilemma,” World Politics 2 (January 1950), pp. 157–80. Its modern restatement is Robert Jervis, “Cooperation under the Security Dilemma,” World Politics 30 (January 1978), pp. 167–214. 105. Oliver Williamson’s work emphasizes the vulnerability of transaction-specific assets to predation by trading partners. Beth and Robert Yarbrough utilize this framework to argue that asset specificity leads to the creation of stronger international institutions, which can deter opportunism and adjudicate disputes. Oliver R. Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting (New York: Free Press, 1985); Beth V. Yarbrough and Robert M. Yarbrough, Cooperation and Governance in International Trade: The Strategic Organizational Approach (Princeton: Princeton University Press, 1992). 106. Kenneth P. Thomas, “Structure and Mobility of the Automobile Industry, 1960– 1994,” in Capital beyond Borders: How Capital Mobility Strengthens Firms in Their Bargaining with States (New York: Macmillan, 1997), chap. 4. 107. Charles Lipson, “International Cooperation in Economic and Security Affairs,” World Politics 37 (October 1984), pp. 1–23. 108. The concept of security communities was developed by Karl Deutsch in the 1950s. Recently, it has been extended empirically and theoretically by Emanuel Adler, Michael Barnett, and their colleagues. Karl W. Deutsch, Sidney A. Burrell, et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957); Emanuel Adler and Michael Barnett, “Security Communities in Theoretical Perspective,” in Emanuel Adler and Michael Barnett, eds., Security Communities (Cambridge: Cambridge University Press), pp. 3–28; in the same volume, Adler and Barnett, “A Framework for the Study of Security Communities,” pp. 29–65. 109. Established democracies may well wish to provide financial aid and other assistance to emerging democracies, to strengthen them out of long-run self-interest. But helping young democracies is quite different from relying on them for future performance. 110. A simple, everyday example of “proxy exchange” is the payment of hourly wages for, say, lawyers or accountants. If we wanted to pay directly for an increment of “accounting service,” how would we possibly measure it? It is easy to measure an hour; it is hard to measure an increment of accounting service. So if we want to exchange money for accounting services, we need to use a proxy variable like hourly wages (or a unit payment for the entire job) to effect the exchange. Unfortunately, paying by the hour does not solve all problems of opportunism or eliminate the need for safeguards. Some accountants have been known to shirk while collecting the full rate. Some fritter away the afternoon working on minor details in order to keep the meter running. The same can be said of professors. Indeed, Adam Smith said it, and suggested a variant of piecework as the remedy. Whatever the specific arrangements, proxy variables present complicated trade-offs for the contracting parties. They permit exchanges that would otherwise be blocked, but they also create problems of their own, such as rents. They certainly do not eliminate the need for private governance mechanisms. Smith’s comments are in The Wealth of Nations, bk. V, chap. 1, pt. III, art. II: “Of the Expense of the Institutions for the Education of Youth.”

N O T E S T O PA G E S 1 0 6 – 9

225

111. Janice Gross Stein, “Deterrence and Reassurance,” in Philip E. Tetlock, Jo L. Husbands, Robert Jervis, Paul C. Stern, and Charles Tilly, eds., Behavior, Society, and Nuclear War, vol. 2 (New York: Oxford University Press, 1991), p. 32. 112. This incentive to deceive is a basic feature of international politics, as James Fearon has observed. Of course, what one country wants to hide, others want to discover. That fuels the point and counterpoint of espionage and counterintelligence—spy versus spy. 113. William Shakespeare, “Introduction to the Play,” The Second Part of Henry the Fourth. 114. Harold James, International Monetary Cooperation since Bretton Woods (Washington, DC: International Monetary Fund; New York: Oxford University Press, 1996), pp. 218–19; Joanne Gowa, Closing the Gold Window: Domestic Politics and the End of Bretton Woods (Ithaca: Cornell University Press, 1983); John S. Odell, U.S. International Monetary Policy (Princeton: Princeton University Press, 1982), chaps. 4–5. 115. An additional benefit of circumscribed agreements is that they save negotiating time. Negotiating time may be in short supply if the negotiators themselves are senior officials. They have only so much time and energy, and many demands upon them. The more common problem is that extensive negotiations sometimes bog down over details and miss the big picture. That is Thomas Schelling’s criticism of the latter years of superpower arms control. The agreements grew longer and longer, but that did not necessarily make them any better at stabilizing the nuclear balance. An equally important criticism is that negotiating all these time-consuming details delays the starting date for the agreement itself. So the opportunity cost is high. That, too, is an important point, and it is one reason why interim agreements are so useful. They allow the parties to implement what they have already agreed upon while they continue negotiating on other matters. SALT I was an interim agreement, and so was the 1947 GATT agreement. In the GATT’s case, the interim agreement stood for decades because the proposed comprehensive agreement, the International Trade Organization, was not ratified by the United States and was therefore abandoned. For Schelling’s views on the desirability of less-extensive, more-informal arms control agreements, see Thomas C. Schelling, “What Went Wrong with Arms Control?” Foreign Affairs 64 (Winter 1985–86), pp. 219–33. His views are endorsed by Kenneth Adelman, a senior arms control official in the Reagan administration. Kenneth L. Adelman, “Arms Control with and without Agreements,” Foreign Affairs 63 (Winter 1984–85), pp. 240–63. 116. Robert Lovett, memo, 28 April 1958, cited in Per Fredrik Ilsaas Pharo, “A Precondition for Peace: Transparency and the Test-Ban Negotiations, 1958–1963,” International History Review 22 (September 2000), p. 557, n. 3. 117. W. W. Rostow, Open Skies: Eisenhower’s Proposal of July 21, 1955 (Austin: University of Texas Press, 1982); Ann M. Florini, “The Open Skies Negotiations,” in Richard Dean Burns, ed., Encyclopedia of Arms Control and Disarmament, vol. 2 (New York: Charles Scribner’s Sons, 1993), pp. 1113–20; James J. Marquardt, “Open Skies— Not a Moment Too Soon,” Bulletin of the Atomic Scientists 58 (January/February 2002), pp. 18–20. 118. Pharo, “A Precondition for Peace,” p. 560. 119. In 1986, William F. Rowell wrote, “The abiding Soviet interest in reducing the flow of information provided through arms control verification has surfaced repeatedly in Soviet complaints that the underlying purpose of U.S. verification proposals is to

226

N O T E S T O PA G E S 1 0 9 – 1 3

gain intelligence information. In 1963 Nikita Khrushchev told Averell Harriman, negotiator of the Limited Test Ban Treaty, ‘The trouble with you is you want to spy. That’s your purpose.’ Two decades later, the Soviets contended that the on-site inspection provisions of the April 1984 draft chemical weapons treaty amounted to ‘asking nations voluntarily to sanction intelligence activities by the other side on their territory.’ ” William F. Rowell, Arms Control Verification: A Guide to Policy Issues for the 1980s (Cambridge, MA: Ballinger, 1986). 120. The first on-site inspection provisions in superpower arms control came in December 1987. They were included in the treaty on Intermediate-Range Nuclear Forces (INF), signed by President Reagan and General Secretary Gorbachev (8 December 1987). This change in Soviet verification policy opened the door to other agreements with on-site inspectors. 121. On-site inspections were traumatic for the Soviet military, according to William E. Odom. “Secrecy had long been a barrier not just against foreign intelligence services but also against the Soviet public and, increasingly, the party leadership as well.” William E. Odom, The Collapse of the Soviet Military (New Haven: Yale University Press, 1998), p. 135. 122. William Diebold, Jr., The End of the ITO, Princeton University Essays in International Finance, no. 16 (Princeton: October 1952). 123. Moffett, The Limits of Victory: The Ratification of the Panama Canal Treaties; Hogan, The Panama Canal in American Politics, pp. 190–91, and 200–205. 124. [John Jay], “Federalist Paper no. 64,” in [Alexander Hamilton, James Madison, and John Jay], The Federalist Papers, ed. Clinton Rossiter (1797; reprint, New York: Mentor, 1961), p. 394, italics in the original. 125. Stephen Holmes, Passions and Constraint: On the Theory of Liberal Democracy (Chicago, IL: University of Chicago Press, 1995), p. 162. Chapter 5 Leadership Succession as a Cause of War 1. This point was anticipated by the Federalist Papers, which urged acceptance of the new U.S. Constitution. In Federalist no. 5, John Jay wrote that “weakness and divisions at home would invite dangers from abroad; and that nothing would tend more to secure us from them than union, strength, and good government within ourselves.” 2. The United States designates its vice president as successor in article 2, section 1 of the Constitution. It was applied for the first time in 1841, when President William Henry Harrison died soon after taking office and was replaced by his vice president, John Tyler. There were no objections to this transfer of power, but there were some doubts whether Tyler had actually become president or was merely exercising the powers of office. The Constitution’s language was imprecise. In practice, Tyler was recognized as the new president, setting the pattern for future vice presidents. If any legal doubts remained, they were erased with new, more precise language in the TwentyFifth Amendment, ratified in 1967. Peter Calvert, “Political Succession and Political Change,” in Peter Calvert, ed., The Process of Political Succession (New York: St. Martin’s, 1987), p. 2. 3. From the beginning, the U.S. Constitution specified a detailed sequence of succession, beginning with the vice president, in case the president died in office. The Twenty-

N O T E S T O PA G E S 11 4 – 1 7

227

Fifth Amendment was added much later to deal with presidential illness or incapacity. Amendment 25, sections 3 and 4, outline the procedures for dealing with serious health problems, whether temporary or permanent. 4. Stephen M. Walt, Revolutions and War (Ithaca: Cornell University Press, 1996). 5. On the different goals of “greedy” and “nongreedy” states, see Charles L. Glaser, “Political Consequences of Military Strategy: Expanding and Refining the Spiral and Deterrence Models,” World Politics 44 (July 1992), pp. 497–538. 6. See chapter 3, present volume, the section “Rationalist Theories of War.” Also, James D. Fearon, “Rationalist Explanations for War,” International Organization 49 (Summer 1995), pp. 379–414. 7. There are obviously other reasons for war if states are not unified rational actors or if they do not consider war itself costly (independent of the goals achieved). 8. Charles Lipson, Standing Guard: Protecting Foreign Capital in the Nineteenth and Twentieth Centuries (Berkeley and Los Angeles: University of California Press, 1985). 9. Thomas Childers, introduction to The Formation of the Nazi Constituency, 1919– 1933 (Totowa, NJ: Barnes and Noble, 1986), p. 1; Andreas Dorpalen, Hindenburg and the Weimar Republic (Princeton: Princeton University Press, 1964). 10. Hitler had two strategic objectives when he refounded the Nazi Party in 1925. One was to establish his own absolute control over the party. The other, says Bullock, was to “build up the Party and make it a force in German politics within the framework of the constitution.” Hitler made no secret of these aims. After the failed putsch, he told fellow prisoners, “When I resume active work it will be necessary to pursue a new policy. Instead of working to achieve power by an armed coup, we shall have to hold our noses and enter the Reichstag against the Catholic and Marxist deputies. If outvoting them takes longer than out-shooting them, at least the result will be guaranteed by their own Constitution.” His judgment was shrewd. Once he managed to win the chancellorship by constitutional means, his final rise to dictatorial control was seamless. Hitler’s statement was made to Kurt Ludecke and is quoted in Alan Bullock, Hitler: A Study in Tyranny, rev. ed. (New York: Harper and Row, 1962), p. 130. On the putsch itself, see Harold J. Gordon, Jr., Hitler and the Beer Hall Putsch (Princeton: Princeton University Press, 1972), pp. 618–19. Gordon concludes that the putsch made Hitler into “the undisputed leader of a serious political movement.” 11. Alan Bullock, Hitler and Stalin: Parallel Lives (New York: Random House, 1992). 12. Hitler’s contempt for Weimar—and, indeed, for constitutionalism and the rule of law—was so complete that he never even bothered to repeal or revise the Weimar constitution. In a purely formal sense, then, it remained in force. This implies, if there were any doubts, that Hitler’s rise to power by constitutional means was nothing more than a cynical ploy. The pretense was cast aside after it had served its purpose. Elmar M. Hucko, ed., The Democratic Tradition: Four German Constitutions (Leamington Spa, UK: Berg, 1987), p. 61. Hitler’s relationship to Weimar is examined in Anthony J. Nicholls, Weimar and the Rise of Hitler, 2d ed. (New York: St. Martin’s, 1979). 13. Napole´ on III was cautious in using force and was always drawn to the process of concert diplomacy, but his taste for adventure and his antipathy to the Vienna settlement repeatedly undermined his own policies. His policies in the Near East, asserting France’s traditional right to protect Catholics, may have been legally sound, but they

228

N O T E S T O PA G E S 11 8 – 2 0

were also risky and provocative. They were a direct blow to Ottoman prestige, provoked Russia to assert its own rights to protect Orthodox Christians, and accidentally unleashed serious conflicts of interest over the twitching body of the Ottoman Empire. The conflicts in Napole´ on III’s policies are wonderfully told in A.J.P. Taylor’s The Struggle for Mastery in Europe, 1848–1918 (London: Oxford University Press, 1954), pp. 24ff. On Napole´ on III’s role in the Crimean War, which definitively ended the informal cooperation of the concert, see A.J.P. Taylor, “Crimea: The War That Would Not Boil,” in From Napoleon to the Second International: Essays on Nineteenth-Century Europe (London: Penguin Press, 1993). For a more sympathetic view of Napole´ on III and Second Empire foreign policy, see William E. Echard, Napoleon III and the Concert of Europe (Baton Rouge: Louisiana State University Press, 1983). 14. James F. McMillan, Napoleon III (London: Longman, 1991), chap. 4. 15. The proximate cause of France’s war with Prussia was their dispute over a vacant throne, a throwback to so many conflicts in the previous century. The Prussians wanted a relative of their own ruling Hohenzollern family to take the Spanish throne. The French, having watched Bismarck consolidate Prussian dominance in Germany, saw this as a threat to their national security. It would leave them surrounded by Hohenzollerns, much as they had once feared being surrounded by Habsburgs. France reacted clumsily. They expressed their fears by issuing public declarations and threatening to use force. The result was a pyrrhic victory. France won the Hohenzollern succession issue, but then escalated the crisis by demanding that the kaiser renounce all future efforts to install a Hohenzollern in Spain. This effort to humiliate Prussia backfired. Far from avoiding war, Bismarck seized the opportunity to complete German unification under Prussian leadership and, in a larger sense, to determine who should be Europe’s most powerful state. After Prussia defeated France at the Battle of Sedan (2 September 1870), an old and infirm Napole´ on III was deposed, and the Third (French) Republic was proclaimed two days later. William S. Halperin, “The Origins of the Franco-Prussian War Revisited: Bismarck and the Hohenzollern Candidature for the Spanish Throne,” Journal of Modern History 45 (March 1973), pp. 83–91; Lawrence D. Steefel, Bismarck, the Hohenzollern Candidacy, and the Origins of the Franco-German War of 1870 (Cambridge: Harvard University Press, 1962); William Carr, The Origins of the Wars of German Unification (London: Longman, 1991), chap. 4. 16. The proposed Zambian laws would have prevented candidacy by anyone whose parents were born outside the country. The law was directed specifically at former leader Kenneth Kaunda, who was born in Zambia but whose parents came from neighboring Malawi. According to the Economist, foreign aid donors pressured Chiluba to drop these antidemocratic plans. Nevertheless, both Chiluba and Kaunda’s rule show how even legitimately elected leaders can undermine nascent democratic forms. The pressure to eliminate anti-Kaunda laws also shows how international coalitions, composed of powerful democracies, can play at least a limited role fostering political competition in less-hospitable climes. Economist, 25 May 1996, p. 50. 17. Tony Smith argues persuasively that U.S. promotion of democracy is a fundamental foreign policy goal, one that has shaped many of its bilateral policies. America’s rhetorical support for democracies is not a sham, he concludes. It is a real motivation for policy. Nor is it a simple mask for strategic or economic interests, even though they are obviously important, and sometimes dominant, policy considerations in their own right. Henry Kissinger says much the same thing, arguing that Woodrow Wilson’s sup-

N O T E S T O PA G E S 1 2 0 – 2 1

229

port for democracy abroad made a lasting imprint on American foreign policy, primarily because it created a lasting connection between the aims of Washington policymakers and the values of the electorate. Tony Smith, America’s Mission: The United States and the Worldwide Struggle for Democracy in the Twentieth Century (Princeton: Princeton University Press, 1994); Henry Kissinger, Diplomacy (New York: Simon and Schuster, 1994). 18. The presence of powerful, stable democracies—with real interests in fostering similar governments elsewhere—is a genuine source of support for new and aspiring democracies. They can develop representative institutions in a far more favorable context than, say, in the 1930s, when Europe was dominated by fascist and communist dictatorships. 19. France, in the last days of its ancien re´ gime, had no interest in encouraging American democracy. As far as Paris was concerned, that was simply an unpleasant by-product of France’s significant material help. France wanted to bedevil its traditional rival, England, for reasons solely related to the European balance of power. Its aid to the American colonies was meant to foster a colonial revolt that would harass the British and pin down their naval resources, giving France a freer hand on the continent. This is the old game of Great Power maneuver and countermaneuver in remote areas, which usually relies on local hands with their own private agendas. It may be an old game, but it continues. One recurrent feature is that indigenous revolutionaries are hard to control and their agendas hard to reshape, despite Great Power efforts. In Afghanistan, where the revolt against Soviet occupation was led by Islamic fundamentalists, the United States provided vital support to the rebels. It was a cheap way of punishing the Soviet Union and pinning down their military resources, much like France’s aid to the American colonies. In Afghanistan, the United States had little in common with the rebels beyond opposing the Soviet Union. It was actually repelled by their theocratic goals, but that was less important than the Great Game afoot. One nasty by-product of this whole affair was that the United States helped train and equip cadres of fundamentalist militia who hated the West and its values and planned to terrorize it once they had dislodged the Soviet Union from Afghanistan. That is exactly what they have done. 20. Franklin remained anxious about the stability of the new democratic government. Not long after the convention, he wrote, “Our Constitution is in actual operation. Everything appears to promise that it will last; but in this world nothing is certain but death and taxes.” Benjamin Franklin, letter to Jean-Baptiste Leroy, 13 November 1789, Works of Dr. Benjamin Franklin, ed. William Duane, vol. 6 (Philadelphia: William Duane, 1817), pp. 231–32. The source of the “republic, if you can keep it” quote is Franklin’s fellow delegate, James McHenry; the question was asked by Mrs. Powel of Philadelphia. It is reported in Max Farrand’s Records of the Federal Convention, rev. ed. (New Haven: Yale University Press, 1937), vol. 3, app. A, item CXVI, “McHenry Anecdotes,” p. 85. I am indebted to Nathan Tarcov for his assistance on this point. 21. The U.S. presidency was always an elected office, with some indirect elements because the popular vote was counted separately in each state, rather than as a single national total. The winner of each state’s popular vote won all of its “electors” (equal to the number of that state’s representatives in Congress plus its two senators). The president was (and still is) the candidate who won a majority of electoral votes. The House of Representatives was always elected directly by popular vote, with winnertake-all in each congressional district. The number of congressional representatives

230

N O T E S T O PA G E S 1 2 1 – 2 3

from each federal state is proportional to its population. Senators, on the other hand, were originally selected for office by a vote of the legislators in their home states. This was really an indirect election rather than a pure appointment, since selection was done by an elective body. The senate became a directly elective body in 1913, when the Seventeenth Amendment to the constitution was passed. Each state, regardless of size, has two senators. Unlike members of Parliament, all these elected officials have fixed terms: two years for the house, four years for the president, six years for the senate. These constitutional forms represented a series of compromises, some of which diluted popular sovereignty and were later eliminated. The most egregious was the outright exclusion of slaves, women, and men without property from any representation and, in the case of slaves, from freedom itself. The composition of the legislature was a compromise between large states (which wanted representation purely by population) and small states (which wanted equal representation for each state). The appointed Senate was a triumph for constitutional conservatives, just as the House of Representatives was a victory for populist democrats. 22. John Adams stayed up late, making last-minute judicial and administrative appointments. He left the White House and the capital at 4 A M., 4 March 1801, declining to meet with Jefferson or attend his inauguration. William Seale, The President’s House (Washington, DC: White House Historical Association, 1986), 1:88. 23. Israel is a liberal, constitutional democracy in most ways, but not in one other. It has a constitutionally privileged religion, and within that religion, legally privileged orthodox practices. On the other hand, it permits freedom of worship, access to holy places of other religions, rights of peaceful assembly, minority voting rights, and most of the other civil rights associated with modern liberal democracy. It has maintained these rights despite years of violent external and internal assault on its national existence. 24. Algerian newspapers that report fairly on the civil war are sent faxes calling them enemies of Islam, tools of Israel and world Zionism, and candidates for assassination. These are not idle threats. Editors of many important newspapers have been killed and others are now in hiding. Practicing journalists are regularly killed by various fundamentalist factions in Algeria. Report on ITN London television news, 27 August 1996. 25. On the possibilities of democratization in Islamic countries more generally, see the essays in Ghassan Salame´ , Democracy without Democrats? The Renewal of Politics in the Muslim World (London: I. B. Tauris, 1994). 26. Edward D. Mansfield and Jack Snyder, “Democratization and the Danger of War,” International Security 20 (Summer 1995), pp. 5–38. 27. Richard Bonney, European Dynastic States, 1494–1660 (New York: Oxford University Press, 1991). 28. The War of Spanish Succession was fought between France and the Habsburg powers from 1702 to 1713–14. The War of Polish Succession (1733–35) was more about French ascendancy than about Polish rule and was fought across Europe, drawing in Austria, Spain, and Russia. The War of Austrian Succession (1740–48) began when the last male Habsburg died unexpectedly. Once again, the major continental powers were entangled by conflicts over who would take the throne. There were other wars of succession during the same era, such as the Bavarian (1778–79). Even the Glorious Revolution in England (1689) arose from a confluence of succession problems and

N O T E S T O PA G E S 1 2 3 – 2 8

231

religion. It drew in the Dutch, who provided a new Protestant king for England, and left the French supporting Catholic pretenders for years to come. 29. The same familial ties were also the basis of many business partnerships before the advent of limited liability corporations. 30. I wish to thank David Laitin for discussing these issues with me. 31. Regency arrangements, though fairly standard, were obviously more problematic than direct inheritance and often led to political disputes over who should act as regent, or as the regent’s key advisers. 32. One intriguing question is why Western states did not solve the political problem of childless kings by letting them adopt heirs, as royalty could in South Asia (a practice the British moved to exterminate in the nineteenth century as their rule intensified). Simple functionalist explanations would say that Western states should have allowed adoption of heirs—so that prediction is obviously wrong. So, too, would functionalist explanations say that succession should have been well specified beyond immediate male descendants. That, too, would have avoided interstate conflicts if the rules had been widely accepted as legitimate. That did not happen either. Even in early-modern Europe, where monarchical rule had been entrenched for centuries, succession was poorly specified beyond the most immediate descendants. Interdynastic struggles arose during moments of ambiguity. I am indebted to Susanne Rudolph and Lloyd Rudolph for pointing out these social-cultural differences in royal inheritance practices. 33. Disputes over which Capet child would inherit could be serious in the dynasty’s early years. Hugh Capet’s oldest child was a daughter, who was never seriously considered for the crown. It passed instead to Hugh’s son, Robert II. Robert II’s eldest child was also a daughter, and his firstborn son died as a youth. That left two brothers. Which of them should inherit the crown? What about the family lands? Should they be kept together or divided among the heirs? Each boy was supported by a family faction, one led by the king, the other by the queen. The king won, and the crown was passed to his eldest surviving son, Henry I, along with the vast bulk of the family lands. That largely settled the pattern of inheritance by the eldest surviving son. Andrew W. Lewis, Royal Succession in Capetian France: Studies on Familial Order and the State (Cambridge: Harvard University Press, 1981), pp. 26–27. On the complex questions of land inheritance by younger royal children, see Lewis’s book; and Charles T. Wood, The French Apanages and the Capetian Monarchy, 1224–1328 (Cambridge: Harvard University Press, 1966). 34. By the early-fourteenth century, Capetian blood rights were so well established that an infant son could inherit the crown unchallenged. Lewis, Royal Succession in Capetian France, pp. 1, 149–50. 35. Philip of Valois became Philip VI of France in 1328. His grandfather was the Capetian king Philip III, and his father was the younger brother of Philip the Fair (Philip IV). 36. J. H. Shennan, International Relations in Europe, 1689–1789 (London: Routledge, 1995), chap. 3; M. S. Anderson, The War of the Austrian Succession, 1740– 1748 (London: Longman, 1995). 37. Paul W. Schroeder, The Transformation of European Politics, 1763–1848 (Oxford: Clarendon Press, 1994), p. 7. 38. Peter Rodman, More Precious than Peace: The Cold War and the Struggle for the Third World (New York: Scribner’s, 1994).

232

N O T E S T O PA G E S 1 2 8 – 3 2

39. Samuel P. Huntington, “The Clash of Civilizations?” Foreign Affairs 72 (Summer 1993), pp. 22–49. 40. Both Islamic governance and the secular application of Islamic law are central features of “Islamist” political movements. As is often the case with revolutionary political movements, the ideals are relatively modern (dating to the eighteenth century, at the earliest), but they are mythically presented as age-old traditions and reminders of a distant, golden time. On the modernity of these Islamic political traditions, see Fred Halliday, “The Politics of ‘Islam’—A Second Look,” British Journal of Political Science 25 (July 1995), p. 404. The evolution of Islamic law is analyzed in Noel Couson’s A History of Islamic Law (Edinburgh: Edinburgh University Press, 1964). 41. Heinz Halm, Shiism (Edinburgh: Edinburgh University Press, 1991). 42. Fred Halliday, “The Politics of Islamic Fundamentalism: Iran, Tunisia and the Challenge to the Secular State,” in Akbar Ahmad and Hastings Donnan, eds., Islam in the Age of Post-Modernity (London: Routledge, 1994), pp. 91–113. 43. Iran’s population, 45 million in the early 1980s, was three times the size of Iraq’s. 44. Immediately after the revolution, the Iranians actually rejected the desirability of military professionalism. They relied on spontaneity instead of training, and a headlong charge instead of sophisticated battle tactics, according to a study by the U.S. Army. Stephen C. Pelletiere and Douglas V. Johnson II, Lessons Learned: The Iran-Iraq War (Carlisle Barracks, PA: Strategic Studies Institute, U.S. Army War College, 1991), pp. 2–3. 45. Wars for limited objectives, with a “seize and hold” strategy, are discussed in John J. Mearsheimer, Conventional Deterrence (Ithaca: Cornell University Press, 1983). 46. My thanks to David Laitin for his useful suggestions on this point. 47. The Soviet Union’s selection may have been smooth after Khrushchev, but it had pathologies of its own. Brezhnev continued in office long after his mental capacity and physical vigor had gone. He was followed, in swift succession, by Andropov and Chernenko, who were aged and infirm themselves. Both died in office (of natural causes) after brief tenures. 48. In the 1980s, Syria’s sick, aging ruler, Hafez al-Assad, began to train his eldest son, Basil, to succeed him as Syria’s ruler. Assad’s own path to power went through the military and the Ba’ath Party, but he was apparently hoping for a monarchical succession. His plans were disrupted when Basil was killed in a (genuine) automobile accident, leaving no obvious heir and no process, beyond Assad’s whim, for selecting one. When Basil’s brother, Bashar, was chosen, it became clear that Assad was trying to convert Syria into a family dynasty. With a future struggle for power looming, it is not surprising that factions of the armed forces associated with different political leaders began to skirmish, initially over who would control the profits from smuggling. When Hafez al-Assad died, Bashar did inherit, but it took some months before outsiders knew if he was firmly in control. There are rough parallels between Syria and North Korea, where Kim Il-Sung hoped to pass power to his son: dynastic succession within a Leninist system. Kim died in July 1994, but it took several years before his son and designated successor, Kim Jong-il, emerged as the country’s clear leader—creating the the first communist monarchy. As in Syria, a closed political system made it unclear for some time whether the son was truly in control.

N O T E S T O PA G E 1 3 3

233

President Suharto of Indonesia tried to follow the same path of leadership transition: avoid any public disclosure, train no successors, postpone any choice, and hope to position family members for what was, in effect, a new dynasty. Suharto held an iron grip on Indonesia for over three decades. No one knew whom he wanted to succeed him. As the Economist observed, “Much of the political uncertainty in Indonesia stems from Mr. Suharto’s reluctance to make his preference clear, and his apparent desire to stand, unopposed, for a seventh five-year term in the 1998 presidential election” Economist, 3 August 1996, p. 22. The transition was ultimately bloody and prolonged. If strongman authoritarian rule is to develop stability and continuity, it must eventually be encapsulated in a political party, military junta, or family dynasty. As the Syrian, North Korean, and Indonesian examples show, rulers often hope to turn their personal rule into a family legacy. But if hope springs eternal, so do the associated struggles and doubts, including opposition from political cronies who are not family members. It should also be noted that the long tenure of authoritarian rulers, combined with their desire to avoid alternate centers of power and control, undermines the development of independent leadership, teamwork, and effective bureaucratic governance. 49. The same succession problem arises in corporations that are run in quasi-dictatorial fashion, such as Armand Hammer’s Occidental Petroleum. Headstrong leaders like Hammer (who ran Occidental until he died, well into his nineties) simply refuse to train, promote, or keep qualified leaders, since they would be regarded as potential successors and might actually depose the old leader. When such rising stars come up independently through the ranks, they are fired or shunted aside as they near the top. Most vulnerable of all is the company’s second in command (its president or chief operating officer). That position is apt to turn over frequently, or be handed over to an unthreatening functionary. What finally limits this disorderly succession are the economic interests of the shareholders, backed by their legal rights as owners. Either the firm’s board of directors defends their interests, or the owners will vote for takeovers by more efficient and more profitable outside firms. What is striking about major public firms in the United States and Europe is how they have been able to limit the tenure of chief executives and ensure relatively smooth succession. In public corporations, where the chief executive is not the controlling shareholder, it is rare to find leaders who have stayed too long and who cannot be dislodged. Armand Hammer and J. Peter Grace are rare figures. At virtually all major U.S. corporations, senior leaders remain at the top for five or six years and then retire, as planned, in their late fifties or early sixties. Mandatory retirement age for executives is another form of insurance against ingrained leadership. Directors of public corporations now regularly demand clear, workable plans of succession. They want to see several insiders groomed to take over, and they reserve the right to choose someone outside the company for top leadership. Of course, the company’s performance is subject to constant, and often intense, scrutiny by stockholders and the press, who have access to a wide range of performance data. There is a clear analogy to regular democratic succession in open political systems, and a sharp contrast to dictators who cling to power until they die or are violently deposed. 50. The prospect of Malenkov’s leadership caused some confusion among NATO countries. Malenkov had called for peaceful coexistence, a relaxation of Cold War tensions, and negotiations over numerous issues. This posed serious questions for Western states that had built a strong military-diplomatic coalition around quite different assump-

234

N O T E S T O PA G E S 1 3 3 – 3 5

tions about Soviet policies. For Soviet policy during this period, see David J. Dallin, Soviet Foreign Policy after Stalin (Philadelphia: Lippincott, 1961), pp. 117–217; for the Western response, see M. Steven Fish, “After Stalin’s Death: The Anglo-American Debate over a New Cold War,” Diplomatic History 10 (Fall 1986), pp. 333–55. 51. The succession process is described in William J. Tompson, Khrushchev: A Political Life (New York: St. Martin’s, 1995), chap. 5. Replacing Stalin was even more difficult because Politburo members feared that the struggle to succeed him and modify his terror system might precipitate a wider social conflict, engulfing them all. “We were scared—really scared,” Khrushchev himself later wrote. “We were afraid the thaw might unleash a flood, which we wouldn’t be able to control and which could drown us. . . . It could have overflowed the banks of the Soviet riverbed and formed a tidal wave which would have washed away all the barriers and retaining walls of our society.” (If Gorbachev read these memoirs, he must have skimmed this passage.) The most delicate moment came in June 1953, when Khrushchev and his allies arrested Beria, the dreaded head of Stalin’s secret police. At the time, they were prepared for a civil war to break out, and were without a well-organized secret police to oppose it (p. 121). The Khrushchev quote is from Khrushchev Remembers: The Last Testament, trans. Strobe Talbott (Boston: Little, Brown, 1974), p. 79. 52. Tompson, Khrushchev, p. 189. 53. The PRI’s loss, coupled with popular demands for more transparency in elections and more voice in government, will transform these selection procedures. For the first time, there will be more open selection based on party primaries in Mexico. 54. For a discussion of how the United States is drawn into succession crises in Latin America, see Robert A. Pastor, Whirlpool: U.S. Foreign Policy toward Latin America and the Caribbean (Princeton: Princeton University Press, 1992), chap. 7. 55. Rising leaders in specialized bureaucracies are rarely called upon to make choices about the full range of policy issues. That is true in both democratic and nondemocratic governments. There are three important differences, however, in democratic recruitment from government bureaucracies. One is that within democracies, the bureaucracies themselves have somewhat greater autonomy, partly because they have responsibilities to different branches of divided governments and partly because of constitutional protections. This gives their leaders additional room for disagreement with senior elected officials. Second, the open process of news gathering means that differences of opinion among government agencies and within them is often widely reported. The use of regular government hearings on public questions, with Pentagon or Commerce Department officials testifying, aids the process. Finally, when any rising leader in a government bureaucracy or appointive office actually decides to seek elective leadership, he or she is subject to the full range of scrutiny. The example of retired U.S. general Colin Powell—who merely considered running for public office—is instructive. The public was favorably disposed because of Powell’s leadership and integrity, but they still wanted to know his positions on abortion rights, gun control, affirmative action, and so on—questions that he had not needed to discuss as a military leader or foreign-policy adviser. Thus, while democracies sometimes recruit leaders from the military, bureaucracies, and appointive office, there are many-layered stages for these rising stars to disclose their general policy preferences before they take office. There is simply no equivalent in any closed governments, from monarchies to dictatorships.

N O T E S T O PA G E S 1 3 5 – 3 6

235

The result is that new leaders in closed governments pose a much greater risk of policy surprise, to insiders and to foreigners. 56. Mao may also have worried that the attack on Stalin’s cult of personality was not a ringing endorsement of his own. 57. The most turbulent cases of Soviet succession came when the regime was new (after Lenin’s death) and after years of one-man rule by Stalin. Later decisions about leadership, though opaque to the outside world, had a more corporate form. 58. The losers in social revolutions often flee for their lives, carrying with them lurid tales of death, destruction, and political repression. According to Stephen Walt, the presence of these embittered emigre´ s in neighboring countries contributes to interstate conflicts, not least because the political exiles overstate the threats posed by revolutionary states. If the revolutionary state is opaque and changing rapidly—as almost all are— then it may not be possible for neighboring governments to reject these claims as selfserving exaggerations. The same kinds of information problems affect revolutionary states. They will almost certainly lose control of the old espionage and diplomatic systems of the prerevolutionary government, making it very difficult for them to understand foreign developments. They are also likely to see an influx of would-be revolutionaries coming to the new Jerusalem with tales stressing the policy differences between their old homelands and the new revolutionary state. They are likely to exaggerate the prospects for positive, revolutionary change abroad, if only a war or foreign intervention could spark it. Again, revolutionary states may lack the capacity to rule out these tales as mere exaggerations. Stephen Walt, “Revolution and War,” World Politics 44 (April 1992), pp. 321–68, and personal communications. 59. Rival candidates really do fear that foreign powers may try to manipulate the vote by handing incumbents a public foreign-policy triumph or defeat on the eve of election. In the United States, this fear even has a name: “the October Surprise.” The term entered the lexicon after the 1980 campaign, when Ronald Reagan’s campaign chief (and later CIA head) William Casey feared that President Jimmy Carter might succeed in freeing the Iranian hostages just before election day, and unfairly reap the rewards. There is a hot debate about what, if anything, Casey and his aides did to prevent the early release of hostages. In retrospect, it is clear that the fear of an October surprise was groundless. The Iranian government did not want Carter to triumph and, if anything, held the hostages longer to humiliate Carter and cause his electoral defeat. In still other cases, the opacity and complexity of international negotiations means that incumbents can claim a foreign-policy victory immediately before the election. The opposition has no chance for effective rebuttal, whether the victory is real or not. On 26 October 1972, after interminable secret negotiations with North Vietnam and just before the U.S. presidential elections, Henry Kissinger publicly announced that “peace is at hand.” It was one of the most controversial pronouncements in a controversial career, and one that many thought was aimed squarely at United States voters. Naturally, Kissinger denies that charge in his memoirs, but he does acknowledge that his comments came to “haunt him.” He also reveals that President Nixon actually did ask him to attack rival candidate George McGovern in the same press conference, which Kissinger declined to do. Nixon’s request shows exactly how incumbents can try to increase their popularity, and diminish their opponents’, by claiming foreign policy achievements on the eve of elections. This impact gives foreign states some leverage to intervene in democratic elections, by handing foreign policy triumphs to incumbents

236

N O T E S T O PA G E S 1 3 6 – 3 9

or by denying them such victories. Henry Kissinger, White House Years (Boston: Little, Brown, 1979), chap. 33. 60. This way of squaring the circle between democratic choice and anticommunism is one reason why the Reagan administration was so attracted to the work of Jeanne Kirkpatrick. Jeanne J. Kirkpatrick, “Dictatorships and Double Standards,” Commentary 68 (November 1979), pp. 34–45. Kirkpatrick argued that once political regimes became communist, there was no chance they would permit democratic liberalization. Rightwing dictatorships, on the other hand, might liberalize because they did not monopolize all political and economic power. Kirkpatrick’s defenders later pointed to Chile’s democratization as demonstrating her point. Of course, Kirkpatrick was writing before the collapse of the Soviet Union and its East European allies. Gorbachev’s liberalization and the emergence of noncommunist regimes without violent suppression by the ruling communist parties directly contradicted Kirkpatrick’s central conclusion. 61. Democracies’ presumption against violent interference in others’ fair elections closely tracks the normative explanation of the democratic peace. 62. The best and most evenhanded study of the United States’ role in Chile is by Paul Sigmund. He concludes that the Central Intelligence Agency was not directly involved in the 1973 coup and did not foment it. Likewise, he says the United States played a relatively small role in the 1970 election, although it clearly opposed Allende and at least considered a variety of covert acts. He cites Henry Kissinger’s statement that the United States put only $135,000 into the 1970 Chile presidential campaign, but notes it might have done more if the CIA and State Department had agreed on tactics. Allende won a plurality of the 1970 popular vote. In the absence of a majority, the Chilean constitution required a runoff in Congress between the top two finishers. The U.S. ambassador suggested bribing congressmen to block Allende, but apparently the United States did nothing. Once Allende was in office, the U.S. gradually escalated its opposition. Many have charged that the CIA played an important role in the coup that killed Allende and deposed his elected regime. Sigmund’s conclusion is that it did not, although the CIA did have close relations with the Chilean military and with U.S.-based multinationals, which organized an “invisible blockade” of Allende’s Chile. Sigmund concludes that “despite the articles, books, and films that have claimed the contrary, there is no evidence of U.S. participation in, or direct encouragement of, the coup.” Paul Sigmund, The United States and Democracy in Chile (Baltimore: Johns Hopkins University Press, 1993), p. 83. These are, of course, highly contentious topics, and there are a number of Latin American specialists and leftist scholars who argue that the U.S. role was larger and more instrumental in the coup. The case is not closed. More documents still might surface, particularly in connection with lawsuits filed in the United States. Chapter 6 Extending the Argument 1. The evidence presented here is limited to the extensions of my contracting theory of the democratic peace. It is not intended to summarize the larger literature covering all ways that democracies conduct foreign policy or, working together, produce different outcomes.

N O T E S T O PA G E S 1 3 9 – 4 4

237

2. Imre Lakatos, “Falsification and the Methodology of Scientific Research Programmes,” in Imre Lakatos and Alan Musgrave, eds. Criticism and the Growth of Knowledge (Cambridge: Cambridge University Press, 1976), pp. 91–196. 3. Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities (New Haven: Yale University Press, 1982), p. 13. The topic has been developed recently by E. O. Wilson, Consilience (New York: Knopf, 1998). 4. Dina A. Zinnes, “The Problem of Cumulation,” in James N. Rosenau, ed., In Search of Global Patterns (New York: Free Press, 1976), cited in Stuart A. Bremer, “Advancing the Scientific Study of War,” in Stuart A. Bremer and Thomas R. Cusack, eds., The Process of War: Advancing the Scientific Study of War (Luxembourg: Gordon and Breach, 1995), p. 2. 5. Thomas Hobbes, “Of the first and second Natural Laws, and of Contracts,” in Leviathan (1651; reprint, New York: E.P. Dutton, 1950), pt. I, chap. XIV, p. 113 (spelling modernized). 6. Hostages were actually used with some frequency to secure bargains in ancien re´ gime Europe. 7. There is another reason why reputation is an imperfect bond: It could be too high. If a financial bond could be set accurately in advance, then even states with good reputations could decide if it was profitable to breach the agreement and forfeit the bond. Assuming the bond was freely negotiated between the parties, then the innocent party would be fully compensated and both parties would be better off. This is called efficient breach. By comparison, reputation is a blunt instrument. With only this clumsy tool available, states cannot bargain in advance to set the precise level of a bond. As a result, states with good reputations may stick to an agreement because they cannot devise a well-calibrated bond that would encourage efficient breach. Losing their reputation would be too high a cost. In some cases, then, reputation deters too well. Unfortunately, those are not likely to be cases that involve very high stakes, such as war. 8. Robert Jervis makes this point in an early discussion of signaling. Reputation can serve as a bond, but only for actors with decent reputations to begin with. “An actor whose reputation is so tarnished that it can hardly be damaged any further (i.e. a chronic liar) cannot employ this valuable method of commitment. One cannot pledge one’s word, or one’s honor, to a course of action if one does not have any word or honor to place in jeopardy.” Robert Jervis, The Logic of Images (Princeton: Princeton University Press, 1970), p. 80. 9. Cathleen S. Fisher, preface to John R. Redick, Nuclear Illusions: Argentina and Brazil, Henry L. Stimson Center Occasional Paper, no. 25, (Washington, DC: December 1995), p. iv. 10. The Argentine-Brazilian Agreement for the Exclusive Peaceful Use of Nuclear Energy was signed in July 1991 and ratified in December of that year. Also in December, Argentina, Brazil, the ABACC, and the International Atomic Energy Agency signed a Quadripartite Agreement giving the IAEA full rights to conduct independent inspections, equivalent to verification under the Nuclear Non-Proliferation Treaty. Tom Zamora Collina and Fernando de Souza Barros, “Bilateral Nuclear Inspections for the Korean Peninsula: Can the Latin American Experience Help Reduce Tensions between North and South?” Korean Journal of Defense Analysis 8 (Summer 1996), pp. 116–17. For a description of the ABACC’s work, visit its home page; .

238

N O T E S T O PA G E S 1 4 4 – 4 7

11. Henry L. Stimson Center, “ Nuclear Confidence-Building between Argentina and Brazil,” at their Web site on Confidence-Building Measures: http://www. stimson.org/cbm/la/lachron.htm [accessed September 2001]. The Web site includes a chronology of the negotiations and measures adopted. 12. Collina and Barros, “Bilateral Nuclear Inspections for the Korean Peninsula,” pp. 114–15. Barros is the former president of the Brazilian Physics Society and was intimately involved in his country’s nuclear program. 13. Michael Barletta, “Democratic Security and Diversionary Peace: Nuclear Confidence-Building in Argentina and Brazil,” National Security Studies Quarterly (Summer 1999), p. 30 n. 4. 14. According to the Stimson Center, “Major progress, however, was not made until the mid-1980s when democratic elections were held in both countries.” Stimson Center, “Nuclear Confidence-Building between Argentina and Brazil.” Mitchell Reiss concurs, saying that elections of civilian presidents in both Argentina and Brazil gave the entire process essential momentum. Mitchell Reiss, Bridled Ambition: Why Countries Constrain Their Nuclear Capabilities (Washington, DC: Woodrow Wilson Center Press; distributed by the Johns Hopkins University Press, 1995), pp. 54–55. 15. Collina and Barros, “Bilateral Nuclear Inspections for the Korean Peninsula,” pp. 120–21. 16. Collina and Barros, “Bilateral Nuclear Inspections for the Korean Peninsula,” p. 116. In 1991, for instance, Germany began requiring full safeguards before it would export products useful to the nuclear industry. 17. The demands for IAEA inspections in Argentina and Brazil came before the United States defeated Iraq. That is an important date because, after the victory, it was discovered that Saddam Hussein had conducted a major effort to build a nuclear bomb, which had gone undiscovered by IAEA inspections. After that, the agency’s seal of approval was considerably less reassuring. On the international demands for IAEA inspections in Latin America, see Collina and Barros, “Bilateral Nuclear Inspections for the Korean Peninsula,” pp. 121–22. 18. Robert Manning, “The North Korean Nuclear Deal: Mistrust—and Verify,” Los Angeles Times, 23 October 1994, cited in Reiss, Bridled Ambition, p. 318 n. 194. 19. My discussion of the North Korean case is heavily indebted to Reiss, Bridled Ambition, chap. 6; as well as to Leon V. Sigal, Disarming Strangers: Nuclear Diplomacy with North Korea (Princeton: Princeton University Press, 1998). 20. Reiss, Bridled Ambition, p. 244. 21. Reiss, Bridled Ambition, p. 242. 22. Discussed by Spurgeon M. Kenny, Jr., in a press conference of the Arms Control Association, Arms Control Today 31 (April 2001), p. 13. 23. Rust Deming, deputy assistant secretary for East Asian and Pacific Affairs, U.S. Department of State, Testimony before the Senate Foreign Relations Committee, Subcommittee on East Asian and Pacific Affairs, Washington, DC, 14 July 1998. http:// www.state.gov/www/policy_remarks/1998/980714_deming_north_korea.html [accessed September 2001]. 24. The tit-for-tat moves in negotiations (involving the United States, the IAEA, North Korea, and South Korea) are explained in Sigal, Disarming Strangers, app. 1, pp. 257–59.

N O T E S T O PA G E S 1 4 8 – 5 1

239

25. The agreement is known as the Agreed Framework of October 21, 1994. There is also a side-letter from President Clinton to North Korean leader Kim Jong Il. Both are reprinted in Sigal, Disarming Strangers, app. 2, pp. 262–64. 26. Conservative U.S. lawmakers (generally Republicans concerned about President Clinton’s foreign policy) argued that the Agreed Framework was still not safe enough and rewarded North Korea for its long history of violations and bad behavior. Reiss, Bridled Ambition, pp. 277–78. 27. Robert Gallucci, “U.S. Policy toward North Korea: Where Do We Go From Here?” Testimony before the Senate Committee on Foreign Relations, 23 May 2001. http://www.georgetown.edu/admin/publicaffairs/pressreleases/ testimony_05252001.html [accessed September 2001]. 28. Robert Gallucci, response at 2001 press conference, reported in Arms Control Association, Arms Control Today 31 (April 2001), p. 19. 29. Secretary of State Warren Christopher on NBC News’ Meet the Press, 15 January 1995 30. “U.S. defends nuclear pact with N. Korea,” Reuters News Service, 24 January 1995. 31. Kurt T. Gaubatz, “Democratic States and Commitment in International Relations,” International Organization 50 (Winter 1996), pp. 109–39. 32. Gaubatz, “Democratic States and Commitment in International Relations,” p. 121. 33. Gaubatz, “Democratic States and Commitment in International Relations,” p. 134. 34. Gaubatz, “Democratic States and Commitment in International Relations,” pp. 134–35; Randolph M. Siverson and Juliann Emmons, “Birds of a Feather: Democratic Political Systems and Alliance Choices in the Twentieth Century,” Journal of Conflict Resolution 35 (June 1991), pp. 285–306. 35. William Reed, “Alliance Duration and Democracy: An Extension and CrossValidation of ‘Democratic States and Commitment in International Relations,’ ” American Journal of Political Science 41 (July 1997), p. 1072. His use of discrete-time event history analysis contrasts with Gaubatz’s use of continuous-time EHA. He also uses a slightly different definition of democracy, but finds that the results are robust across both operational definitions (p. 1076). 36. D. Scott Bennett, “Testing Alternative Models of Alliance Duration, 1816– 1984,” American Journal of Political Science 41 (July 1997), pp. 873. 37. Bennett, “Testing Alternative Models of Alliance Duration, 1816–1984,” p. 875. He adds a cautionary note. No single model of alliances explains their duration, and some plausible variables, like the level of alliance institutionalization, seem to have little effect. On the other hand, he explicitly rejects the neo-Realist claim that domestic variables have no effect. 38. Brian Lai and Dan Reiter, “Democracy, Political Similarity, and International Alliances, 1816–1992,” Journal of Conflict Resolution 44 (April 2000), pp. 203–27. 39. Cederman excludes world wars (to concentrate on dyads) and eliminates dyads where the countries are far apart, have little contact, and no reasons to face conflicts with each other. Instead of partitioning the data by time periods, he introduces dummy variables for the interwar years and the Cold War. Lars-Erik Cederman, “Back to Kant:

240

N O T E S T O PA G E S 1 5 1 – 5 3

Reinterpreting the Democratic Peace as a Macrohistorical Learning Process,” American Political Science Review 95 (March 2001), pp. 15–31. 40. Cederman, “Back to Kant,” p. 25. 41. Cederman, “Back to Kant,” p. 27. 42. Sara McLaughlin Mitchell, Scott Gates, and Ha˚ vard Hegre, “Evolution in Democracy-War Dynamics,” Journal of Conflict Resolution 43 (December 1999), p. 771. 43. Emanuel Adler, “Seasons of Peace: Progress in Postwar International Security,” in Emanuel Adler and Beverly Crawford, Progress in Postwar International Relations (New York: Columbia University Press, 1991), pp. 128–73; Arie M. Kacowicz, Zones of Peace in the Third World: South America and West Africa in Comparative Perspective (Albany: State University of New York Press, 1998), pp. 202–3. Like all work on pluralistic security communities, it draws on Karl W. Deutsch, Sidney A. Burrell, et al., Political Community and the North Atlantic Area (Princeton: Princeton University Press, 1957). 44. Paul Seabury and Angelo Codevilla, War: Ends and Means, rev. ed. (New York: Basic Books, 1990), p. viii. 45. Edward D. Mansfield and Jack Snyder, “Democratization and the Danger of War,” International Security 20 (Summer 1995), pp. 5–38; Edward Mansfield and Jack Snyder, “Democratization and War,” Foreign Affairs 74 (May/June 1995), pp. 79–97. More recently, Jack Snyder revisited these issues in From Voting to Violence: Democratization and Nationalist Conflict (New York: W. W. Norton, 2000). 46. Because Mansfield and Snyder used data (as it then existed) from the Correlates of War project, their numbers ended in 1980. Ironically, that omits the last wave of democratization, which prompted so much interest in their findings. The omission is critical, according to Reinhard Wolf, because none of the newly democratic states in Eastern Europe has become involved in war. Mansfield and Snyder disagree, arguing that their analysis deals with the “democratization” process, not the achievement of fullscale democracy, and that several Balkan states were at least moving toward democracy before they began fighting. Reinhard Wolf, “Correspondence on ‘Democratization and the Danger of War,’ ” International Security 20 (Spring 1996), pp. 176–80; Edward D. Mansfield and Jack Snyder, “Correspondence on ‘Democratization and the Danger of War,’ ” International Security 20 (Spring 1996), pp. 197–98. 47. Mansfield and Snyder, “Democratization and War,” p. 83. 48. Mansfield and Snyder, “Democratization and War,” p. 94. 49. Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968). 50. Michael D. Ward and Kristian S. Gleditsch, “Democratizing for Peace,” American Political Science Review 92 (March 1998), pp. 51–61. One other relevant finding is by John Oneal and Bruce Russett. They find, contrary to Mansfield and Snyder, no evidence that states are more conflict-prone if they have recently undergone regime change, either toward democracy or autocracy. Their data are for the Cold War era. John R. Oneal and Bruce M. Russett, “The Classical Liberals Were Right: Democracy, Interdependence, and Conflict, 1950–1985,” International Studies Quarterly 41 (June 1997), pp. 267–94. 51. In their response to critics, Mansfield and Snyder are emphatic that (a) they made no attempt to challenge the majority view on the democratic peace, and (b) they think that, in the long run, enlarging the zone of democracy increases the chances of

N O T E S T O PA G E S 1 5 4 – 5 5

241

peace. Mansfield and Snyder, “Correspondence on ‘Democratization and the Danger of War,’ ” pp. 197–98. 52. Regime maturity is examined in the democratic peace literature as early as Maoz and Russett’s 1992 work. They find immaturity is associated with higher levels of conflict. Their study, which is one of the first to establish the statistical basis for the democratic peace, does not explore contingent relationships between maturity and regime type. They do not compare, say, “immature autocracies” with “immature democracies,” or compare either with their more mature variants. Rather, they look at the independent effects of maturity, regardless of regime type. Zeev Maoz and Bruce Russett, “Alliance, Contiguity, Wealth, and Political Stability: Is the Lack of Conflict among Democracies a Statistical Artifact?” International Interactions 17 (1992), pp. 245–68. 53. Paul D. Senese, “Democracy and Maturity: Deciphering Conditional Effects on Levels of Dispute Intensity,” International Studies Quarterly 43 (September 1999), p. 494. 54. Cederman, “Back to Kant,” p. 28. 55. Paul D. Senese, “Between Dispute and War: The Effect of Joint Democracy on Interstate Conflict Escalation,” Journal of Politics 59 (February 1997), p. 16. 56. William R. Thompson, ed., Great Power Rivalries (Columbia: University of South Carolina Press, 1999). Goertz and Diehl have produced extensive empirical studies of these rivalries. Paul F. Diehl and Gary Goertz, War and Peace in International Rivalry (Ann Arbor: University of Michigan Press, 2000); Paul F. Diehl, ed., The Dynamics of Enduring Rivalries (Urbana: University of Illinois Press, 1998); Gary Goertz and Paul F. Diehl, “The Empirical Importance of Enduring Rivalries,” International Interactions 18 (December 1992), pp. 151–63; Gary Goertz and Paul F. Diehl, “Enduring Rivalries: Theoretical Constructs and Empirical Patterns,” International Studies Quarterly 37 (June 1993), pp. 147–71; Gary Goertz and Paul F. Diehl, “Taking Enduring Out of Enduring Rivalry: The Rivalry Approach to War and Peace,” International Interactions 21 (3) (1995), pp. 291–308; Gary Goertz and Paul F. Diehl, “The Initiation and Termination of Enduring Rivalries: The Impact of Political Shocks,” American Journal of Political Science 39 (February 1995), pp. 30–52; Gary Goertz and Paul F. Diehl, “Rivalries: The Conflict Process,” in John A. Vasquez, ed., What Do We Know about War? (Lanham, MD: Rowman and Littlefield, 2000), pp. 197–217; Gary Goertz and Patrick Regan, “Conflict Management in Enduring Rivalries,” International Interactions 22 (1997), pp. 321–40; Gary Goertz and Paul F. Diehl, “(Enduring) Rivalries,” in Manus I. Midlarsky, Handbook of War Studies, 2d ed. (Ann Arbor: University of Michigan Press, 2000). 57. Jonathan Wilkenfeld and Michael Brecher have conducted extensive work on “protracted conflicts.” Michael Brecher and Jonathan Wilkenfeld, Crises in the Twentieth Century, vol. 1, Handbook of International Crises (Oxford: Pergamon Press, 1988); Jonathan Wilkenfeld and Michael Brecher, Crises in the Twentieth Century, vol. 2, Handbook of Foreign Policy Crises (Oxford: Pergamon Press, 1988); Michael Brecher and Jonathan Wilkenfeld, Crisis, Conflict, and Instability (Oxford: Pergamon Press, 1989). 58. Goertz and Diehl, “Rivalries: The Conflict Process,” p. 216 n. 2. 59. Goertz and Diehl, “Rivalries: The Conflict Process,” p. 200 table 9.3. 60. Another group of rivalries fell somewhere between isolated disputes and enduring rivalries. These “proto” rivalries involved 3–5 disputes between two countries.

242

N O T E S T O PA G E S 1 5 6 – 5 7

There were 223 cases of proto-rivalry between 1816 and 1992, or about one-fifth of all rivalries. They produced about one-third of all wars. For comparison: isolated rivalries were three-quarters of all the cases, and slightly more than one-sixth of all wars. Enduring rivalries were one-twentieth of all cases and half of all wars. Goertz and Diehl, “Rivalries: The Conflict Process,” p. 199 table 9.2. 61. The Weimar regime was founded in the midst of revolutionary uprisings at the conclusion of World War I. Some returning troops were reconstituted as right-wing paramilitary, known as Freikorps. They turned major German cities into battlegrounds, fighting street battles against police and communists. In 1919, the left attempted its own coups in Munich, Bremen, and Berlin. In 1920, the right staged another major effort to seize power (the Kapp Putsch), which sparked an opposing series of left-wing uprisings in the industrial Ruhr Valley. Those leftist uprisings spread and took on a momentum of their own. There were also a number of communist uprisings at the local level, notably in Hamburg and Prussian Saxony. The last of these early coups came from the right, the notorious 1923 Beer Hall Putsch in Munich, led by Hitler and his new Nazi Party. It was backed by General Erich Ludendorff, one of Germany’s leading figures from the Great War. (I wish to thank Michael Geyer for his comments on this period in German history.) 62. Jon Jacobson, Locarno Diplomacy: Germany and the West, 1925–1929 (Princeton: Princeton University Press, 1972). 63. Paul R. Hensel, Gary Goertz, and Paul F. Diehl, “The Democratic Peace and Rivalries,” Journal of Politics 62 (November 2000), pp. 1173–88. 64. Goertz and Diehl, “Rivalries: The Conflict Process,” p. 215; also Goertz and Diehl, “(Enduring) Rivalries,” pp. 242–43. 65. D. Scott Bennett, “Democracy, Regime Change, and Rivalry Termination.” International Interactions 22 (1997), pp. 369–97. 66. Stuart A. Bremer, “Democracy and Militarized Interstate Conflict, 1816–1965,” International Interactions (Spring 1993), pp. 231–49; Maoz and Russett, “Alliance, Contiguity, Wealth, and Political Stability,” pp. 245–68; Zeev Maoz and Bruce Russett, “Normative and Structural Causes of Democratic Peace, 1946–1986,” American Political Science Review 87 (September 1993), pp. 624–38. 67. Suzanne Werner, “The Effects of Political Similarity on the Onset of Militarized Disputes, 1816–1985,” Political Research Quarterly 53 (June 2000), pp. 343–74. 68. Paul D. Senese, “Between Dispute and War,” pp. 1–27. Peter Partell reaches similar conclusions. Peter J. Partell, “Escalation at the Outset: An Analysis of Target Responses in Militarized Interstate Disputes,” International Interactions 23 (1997), pp. 1–35. 69. Senese’s dependent variable for this midrange of escalation includes several levels of conventional MIDs plus low levels of actual fighting that result in fewer than one-thousand battle deaths. About three-quarters of all fighting ends with under onethousand battle deaths, the threshold defined as “war” by the Correlates of War project (and so defined before any work began on the idea of a democratic peace). 70. William Reed, “A Unified Statistical Model of Conflict Onset and Escalation,” American Journal of Political Science 44 (January 2000), pp. 84–93. He reaches much the same conclusion in Robert A. Hart and William Reed, “Selection Effects and Dispute Escalation: Democracy and Status Quo Evaluations,” International Interactions 25 (1999), pp. 243–63. The data is on MIDs between 1816 and 1985.

N O T E S T O PA G E S 1 5 7 – 6 0

243

71. Reed, “A Unified Statistical Model of Conflict Onset and Escalation,” p. 90. 72. Hart and Reed, “Selection Effects and Dispute Escalation,” pp. 243–63. 73. William J. Dixon, “Democracy and the Peaceful Settlement of International Conflict,” American Political Science Review 88 (March 1994), p. 17. 74. According to William Dixon, “Nearly 80% of observations having a minimum democracy score of at least 6 occur prior to phase 3, and in only one instance does this level of minimum democracy appear in a conflict extending beyond a fourth phase.” Dixon, “Democracy and the Peaceful Settlement of International Conflict,” p. 25; also William J. Dixon, “Dyads, Disputes, and the Democratic Peace,” in Murray Wolfson, ed., The Political Economy of War and Peace (Boston: Kluwer, 1998), p 110. 75. Dixon, “Democracy and the Peaceful Settlement of International Conflict,” p. 28. 76. Dixon, “Dyads, Disputes and the Democratic Peace,” pp. 119–20. 77. Jonathan Wilkenfeld and Michael Brecher, “Interstate Crises and Violence: Twentieth-Century Findings,” in Midlarsky, Handbook of War Studies p. 291 (emphasis omitted in quotation). 78. “Specifically, when the proportion of democracies among the crisis actors is low, the probability of violence is .52. This probability decreases to .39 for a moderate proportion of democracies, and, finally, to a probability of .27 when the proportion of democracies is high.” Wilkenfeld and Brecher, “Interstate Crises and Violence,” p. 291. 79. These tests for violence in crises include standard control variables. Wilkenfeld and Brecher, “Interstate Crises and Violence,” p. 291 (emphasis omitted in quotation). 80. Joe Eyerman and Robert A. Hart, Jr., “An Empirical Test of the Audience Cost Proposition.” Journal of Conflict Resolution 40 (December 1996), pp. 597–616. Eyerman and Hart use data from 1945 to 1984. 81. Midlarsky, Handbook of War Studies, p. xvii. 82. I am not arguing, tautologically, that an issue is vital because a state shows enough resolve to win a specific dispute. There must be some independent confirmation of the issue’s importance. I infer which issues are crucial from a conventional understanding of national security priorities (for example, it is more important to protect national territory than colonies) plus the historical record of the states involved. In Britain’s case, for example, naval superiority directly involved the country’s territorial security and was a policy pursued diligently for generations. The case of empire in Victorian Britain is more problematic. It did not involve the security of the home islands, but most policymakers considered the defense of major colonial holdings, particularly India, a crucial dimension of policy. Even so, the conflict at Fashoda was several stages removed from defending the route to India, much less defending India itself. 83. David Lake, “Powerful Pacifists: Democratic States and War,” American Political Science Review 86 (March 1992), pp. 24–37; Allan C. Stam III, Win, Lose, or Draw: Domestic Politics and the Crucible of War (Ann Arbor: University of Michigan Press, 1996); Dan Reiter and Allan C. Stam III, “Democracy, War Initiation, and Victory,” American Political Science Review 92 (1998), pp. 377–89; Bruce Bueno de Mesquita, James Morrow, Randolph Siverson, and Alastair Smith, “An Institutional Explanation of the Democratic Peace,” American Political Science Review 93 (1999), pp. 791–808. 84. On democracies’ success in crises, see J. S. Rioux, “A Crisis-Based Evaluation of the Democratic Peace Proposition,” Canadian Journal of Political Science 31 (June 1998), pp. 263–83.

244

N O T E S T O PA G E S 1 6 0 – 6 3

85. “If discussion reflects all social experience, and everyone can speak and criticize freely,” says Iris Marion Young, “then discussion participants will be able to develop a collective account of the sources of the problems they are trying to solve, and will develop the social knowledge necessary to predict likely consequences of alternative courses of action meant to address them. Their collective critical wisdom thus enables them to reach a judgement that is not only normatively right in principle, but also empirically and theoretically sound.” Iris Marion Young, Inclusion and Democracy (New York: Oxford University Press, 2000), pp. 30–31. 86. Lake, “Powerful Pacifists,” pp. 24–37. 87. To show how risk acceptance reduces the scope for agreement, consider a simple example. Let’s say two identical states wish to divide a territory that each values at $100 million. Each thinks it has a 50 percent chance of winning a war to claim it all. Fighting the war would cost each state $10 million, whether it wins or loses. Therefore, each state has an expected utility of $40 million if it fights (an expected value of $50 million for the results of the war minus $10 million to fight). If both states are riskneutral, each would prefer a negotiated outcome of, say, $41 million to fighting. In this case, there is a negotiating range between $40,000,000.01 and $59,999,999.99 that both states prefer to fighting. To avoid the war, they have to settle on a specific amount within that range and execute the bargain. On the other hand, if both states are riskacceptant, each might prefer an outcome where it either wins $90 million (total victory minus the $10 million cost) or loses $10 million (total loss minus the cost). If both states prefer a risky gamble like this to a certain outcome of, say, $59.99 million, then there is simply no scope for a mutually profitable bargain. In general, the more riskacceptant states are, the smaller the scope for profitable bargains. The more risk-averse states are, the wider the scope for agreements that are profitable to both sides. 88. Kenneth A. Schultz, “Do Democratic Institutions Constrain or Inform? Contrasting Two Institutional Perspectives on Democracy and War,” International Organization 53 (Spring 1999), pp. 233–66. 89. Stephen M. Walt, Revolutions and War (Ithaca: Cornell University Press, 1996). 90. Walt, Revolutions and War, p. 11. During these early phases of the French Revolution, however, Robespierre was not the dominant voice. The Girondins actively campaigned for war (pp. 64–65). 91. On these issues of debt renunciation and expropriation, see Charles Lipson, Standing Guard: Protecting Foreign Capital in the Nineteenth and Twentieth Centuries (Berkeley and Los Angeles: University of California Press, 1985). 92. Robert Service, A History of Twentieth-Century Russia (Cambridge: Harvard University Press, 1997), chap. 6. 93. There was also a Czech legion fighting alongside the Allies and the White Russians. American troops, by contrast, never actually fought the communists during the Civil War. J.F.N. Bradley, Allied Intervention in Russia, 1917–1920 (London: Weidenfeld and Nicolson, 1968); Evan Mawdsley, The Russian Civil War (Boston: Allen and Unwin, 1987); Michael Jabara Carley, Revolution and Intervention: The French Government and the Russian Civil War, 1917–1919 (Montreal: McGill-Queen’s University Press, 1983); Peter Kenez, Civil War in South Russia, 1918: The First Year of the Volunteer Army (Berkeley and Los Angeles: University of California Press, 1971); Peter Kenez, Civil War in South Russia, 1919–1920: The Defeat of the Whites (Berkeley and Los Angeles: University of California Press, 1977).

N O T E S T O PA G E S 1 6 3 – 6 6

245

94. Richard H. Ullman, Anglo-Soviet Relations, 1917–1921, vols. 2 and 3 (Princeton: Princeton University Press, 1968, 1972); David Carlton, Churchill and the Soviet Union (Manchester: Manchester University Press, 2000), chap. 1; Michael Kettle, Russia and the Allies, 1917–1920, vol. 3, Churchill and the Archangel Fiasco, November 1918–July 1919 (London: Routledge, 1992). 95. James W. Morley, The Japanese Thrust into Siberia, 1918–1920 (New York: Columbia University Press, 1957). 96. On autocrats’ willingness to take long-shot gambles to resurrect their political careers, see Hein Goemans, “Fighting for Survival: The Fate of Leaders and the Duration of War,” Journal of Conflict Resolution 44 (2000), pp. 555–79. 97. Stanley Karnow, Vietnam: A History (New York: Viking, 1983), p. 651. 98. Nguyen Co Thach, a North Vietnamese delegate to the Paris talks, said of Kissinger’s statement, “I think the speech . . . ‘Peace is at hand,’ is first of all for the voters in the United States, to say to them that the Vietnam question is no more, there is no more Vietnam question, so they can elect Nixon as a hero of peace.” Nguyen Co Thach, quoted in Vietnam: A Television History on the Public Broadcasting System, transcript at . 99. Suzanne Werner, “The Precarious Nature of Peace: Resolving the Issues, Enforcing the Settlement, and Renegotiating the Terms.” American Journal of Political Science 43 (July 1999), p. 929. 100. Werner, “The Precarious Nature of Peace,” p. 929. 101. Werner, “The Precarious Nature of Peace,” p. 932. 102. The classic theoretical article on collective action problems within alliances is Mancur Olson and Richard Zeckhauser, “An Economic Theory of Alliances,” Review of Economics and Statistics 48 (August 1966), pp. 266–79. Todd Sandler reviews the large literature that ensued in “The Economic Theory of Alliances: A Survey,” Journal of Conflict Resolution 37 (1993), pp. 446–83. The standard neo-Realist treatment of buck-passing is Kenneth N. Waltz, Theory of International Politics (Reading, MA: Addison-Wesley, 1979), esp. pp. 164–67. Also see Thomas J. Christensen and Jack Snyder, “Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity,” International Organization 44 (Spring 1990), pp. 137–68. There is now an extensive empirical literature (and a prolonged policy debate) on the specific issue of collective goods provision within NATO. John R. Oneal, “The Theory of Collective Action and Burden Sharing in NATO,” International Organization 44 (Summer 1990), pp. 379–402; John R. Oneal, “Testing the Theory of Collective Action: NATO Defense Burdens, 1950–1984,” Journal of Conflict Resolution 34 (September 1990), pp. 426–48; Christopher Coker, ed., Shifting into Neutral: The Burden Sharing Debate in the Western Alliance (London and Washington: Brassey’s, 1990); Simon Duke, The Burdensharing Debate: A Reassessment (New York: St. Martin’s Press, 1993); Alan Tonelson, “NATO Burden-Sharing: Promises, Promises,” Journal of Strategic Studies 23 (September 2000), pp. 29–58; Keith Hartley and Todd Sandler, “NATO Burden-Sharing: Past and Future,” Journal of Peace Research 36 (November 1999), pp. 665–80; Todd Sandler and Keith Hartley, The Political Economy of NATO: Past, Present, and into the Twenty-First Century (Cambridge: Cambridge University Press, 1999).

246

N O T E S T O PA G E S 1 6 6 – 7 5

103. On contributions and institutional control, see the introduction and conclusion in Barbara Koremenos, Charles Lipson, and Duncan Snidal, eds., Rational International Institutions, special issue of International Organization 55 (Autumn 2001). 104. The frustrations of waging war by committee shine through in discussions of the Kosovo War by its NATO commander, General Wesley Clark. Wesley K. Clark, Waging Modern War: Bosnia, Kosovo, and the Future of Combat (New York: Public Affairs, 2001). 105. Lisa L. Martin, Coercive Cooperation: Explaining Multilateral Economic Sanctions (Princeton: Princeton University Press, 1992). The problem of second-order sanctioning figures prominently in Robert Axelrod, “An Evolutionary Approach to Norms,” American Political Science Review 80 (December 1986), pp. 1095–1111. 106. Russell Hardin, Collective Action (Baltimore: Johns Hopkins University Press, for Resources for the Future, 1982). 107. For a more sanguine view about the future of collective security, see Charles A. Kupchan and Clifford A. Kupchan, “The Promise of Collective Security” International Security 20 (Summer 1995), pp. 57–59. My own view has fewer bright colors and more chiaroscuro. Charles Lipson, “Is the Future of Collective Security Like the Past?” in George Downs, ed., Collective Security beyond the Cold War (Ann Arbor: University of Michigan Press, 1994), pp. 105–31. Chapter 7 Reliable Partners and Reliable Peace 1. Commitment problems have long been considered important in international relations. They figure prominently in the pathbreaking work of Thomas Schelling and throughout the literature on deterrence. What is new is (1) greater attention to the information problems, which always surround credible commitment; (2) the development of incomplete-information game theory, which gives a formal grounding to this work; (3) greater attention to promises as well as threats; and (4) sharper focus on the connection between specific governance arrangements and credible commitments, including the development of governance arrangements as a substitute for incredible contracts (issues that have been developed most fully in studies of the U.S. Congress). In the international relations literature, several scholars, including Peter F. Cowhey, Kurt Taylor Gaubatz, Lisa Martin, and Kenneth Schultz are now working in this area, particularly on the connections between forms of government and their capacity to make commitments. Gaubatz, for instance, shows that military alliances among democracies are longer lasting than those between democracies and nondemocracies or among democracies. That is partial evidence, he says, for his conclusion that democracies can make more credible commitments. Kurt T. Gaubatz, “Democratic States and Commitment in International Relations,” International Organization 50 (Winter 1996), pp. 109–39. Peter Cowhey offers an imaginative and penetrating analysis of U.S. and Japanese capacities to make credible promises. Peter F. Cowhey, “Domestic Institutions and the Credibility of International Commitments: Japan and the United States,” International Organization 47 (Spring 1993), pp. 299–326. 2. Marc Bush, “Democracy, Consultation, and the Paneling of Disputes under GATT,” Journal of Conflict Resolution (2000), pp. 425–46; W. J. Davey, “The WTO Dispute Settlement System,” Journal of International Economic Law 3 (March 2000),

N O T E S T O PA G E S 1 7 5 – 8 7

247

pp. 15–18; J. P. Trachtman, “ The Domain of WTO Dispute Resolution,” Harvard International Law Journal 40 (1999), pp. 333–77; Frederick M. Abbott, “NAFTA and the Legalization of World Politics: A Case Study,” International Organization 54 (Summer 2000), pp. 519–47. For the increasing use of private arbitration forums, see Walter Mattli, “Private Justice in a Global Economy,” International Organization 55 (Autumn 2001); Alessandra Casella, “On Market Integration and the Development of Institutions: The Case of International Commercial Arbitration,” European Economic Review 40 (1996), pp. 155–86; and William Park, International Forum Selection (Deventer, Netherlands, and Boston: Kluwer Law International, 1995). My thanks to Walter Mattli for his comments on this issue of legalization. 3. The adaptation of multinational firms to such risky political environments is analyzed in Charles Lipson, Standing Guard: Protecting Foreign Capital in the Nineteenth and Twentieth Centuries (Berkeley and Los Angeles: University of California Press, 1985). 4. Hans-Ulrich Wehler, The German Empire, 1871–1918, trans. Kim Traynor (Leamington Spa, UK: Berg, 1985); Hans-Ulrich Wehler, “Bismarck’s Imperialism, 1862– 1890,” Past and Present 48 (1970), pp. 119–55. 5. David M. Pletcher, The Diplomacy of Annexation; Texas, Oregon, and the Mexican War (Columbia: University of Missouri Press, 1973); Frederick B. Hosen, comp., Unfolding Westward in Treaty and Law: Land Documents in United States History from the Appalachians to the Pacific, 1783–1934 (Jefferson, NC: McFarland, 1988); Frederick Merk, Manifest Destiny and Mission in American History: A Reinterpretation (1963; reprint, Cambridge: Harvard University Press, 1995); Anders Stephanson, Manifest Destiny: American Expansionism and the Empire of Right (New York: Hill and Wang, 1995). 6. Francis Fukuyama, The End of History and the Last Man (New York: Free Press, 1992). 7. Richard F. Hamilton, Who Voted for Hitler? (Princeton: Princeton University Press, 1982); Anthony J. Nicholls and Erich Matthias, eds., German Democracy and the Triumph of Hitler: Essays in Recent German History (New York: St. Martin’s Press, 1972). 8. Mansfield and Snyder attribute the war-proness of new democracies to the nationalism unleashed by statehood rather than to the contracting problems I highlight. Edward Mansfield and Jack Snyder, “Democratization and War,” Foreign Affairs 74 (May/ June 1995), p. 81. Also see Edward D. Mansfield and Jack Snyder, “Democratization and the Danger of War,” International Security 20 (Summer 1995), pp. 5–38. 9. Cowhey, “Domestic Institutions and the Credibility of International Commitments,” pp. 299–326. 10. To reiterate an earlier point: the presence of nuclear weapons may have encouraged peaceful relations among some states but it cannot explain the “democratic peace,” which predates it and extends to many pairs of states that lack nuclear weapons and are not under the nuclear “umbrella” of a major power. 11. Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge: Cambridge University Press, 1990); Robert D. Putnam with Robert Leonardi and Raffaella Y. Nanetti, Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press, 1993).

248

N O T E S T O PA G E S 1 8 8 – 8 9

12. This is the heart of Paul Schroeder’s capstone work, The Transformation of European Politics, 1763–1848 (Oxford: Clarendon Press, 1994). Drawing on primary documents, Schroeder specifically addresses the worldviews of European diplomats and their attitudes toward a balance of power based solely on force. He also deals specifically with these issues in “The Nineteenth-Century International System: Balance of Power or Political Equilibrium,” Review of International Studies 15 (April 1989), pp. 135–53. An interesting debate on the character of the Concert, with contributions from Schroeder and Robert Jervis, can be found in the American Historical Review 97 (June 1992). 13. Abstention from mutual injury is the precondition for cooperative social life, including nondestructive competition. That is the practical wisdom behind the moral command: “Do not do unto others what you would not have them do unto you.” (Phrased in the negative by the Jewish sage Hillel.) 14. Diego Gambetta, “Can We Trust Trust?” in Diego Gambetta, ed., Trust: Making and Breaking Cooperative Relations (Oxford: Basil Blackwell, 1988), p. 215.

Index “If you don’t find it in the index, look very carefully through the entire catalogue.” —Sears, Roebuck and Co., Consumer’s Guide, 1897

Page numbers in italics refer to figures and tables. ABACC (Brazilian-Argentine Agency for Accounting and Control of Nuclear Materials), 144, 146, 237n.10 Adams, John, 121, 230n.22 Adler, Emanuel, 60, 152 Afghanistan, 110, 229n.19 Africa, 30, 66, 181 democratic transitions in, 84 ivory trade in, 108–9 leadership succession in, 118, 119 new democracies of, 84, 181 Algeria, 121, 134, 136, 230n.24 Allende, Salvador, 84, 136, 137, 236n.62 Alsace–Lorraine, 25 Anderton, Charles, 68 Anglo-Boer War, 24, 43 Anglo-U.S. relations, 36–44, 61, 63, 68, 155, 159, 180 increased strength of, 10, 43–44, 150 in U.S. Civil War, 36–40, 54, 63, 175, 198n.57, 199nn.59, 63, 69, and 72 Venezuela–British Guiana border dispute and, 40–42, 45, 175 anticommunism, 66, 113, 128, 136, 236n.60 appeasement, lesson of, 71 Arafat, Yasser, 132 Argentina, 143–45, 146, 180, 186, 216n.54, 237n.10 Aristide, Jean-Bertrand, 118–19 aristocracies, press freedom and, 88 Articles of Confederation, U.S., 120 Asia, 66, 109, 118 new democracies in, 84, 181 Assad, Bashar al-, 232n.48 Assad, Hafez al-, 232n.48 assassination, political, 90 audience costs, 7, 14, 48, 81–82, 85, 158, 171, 173

Austria-Hungary, 88, 98, 123, 126–27, 188 autocracies: civil wars and, 28 definition of, 194n.13 Ayub Khan, Mohammad, 154 Babst, Dean, 17–18 “Back to the Future” (Mearsheimer), 19 Ballots and Bullets (Gowa), 22 Barbarossa, Operation, 71, 96–97, 221n.82 Barletta, Michael, 144 Barros, Fernando de Souza, 144–45, 237n.10 Bates, Edward, 38 Beer Hall Putsch, 117, 242n.61 Belgium, 142, 162 Bennett, D. Scott, 149–50, 156 Beria, Lavrentii, 133 Berle, Adolf A., Jr., 58–59 Bertie, Francis, 34 bilateral diplomacy, 64, 165–68 Bismarck, Otto von, 24, 33, 228n.15 Bloch, Marc, 78 Bolsheviks, 162–63 Borgia, Cesare, 51–52, 201n.20, 202n.21 Borjas, George, 49 “bounded competition,” 157 Brazil, 143–45, 146, 180, 186 Brazilian-Argentine Agency for Accounting and Control of Nuclear Materials (ABACC), 144, 146, 237n.10 Brecher, Michael, 155, 158 Bremer, Stuart A., 193n.1 Bretton Woods monetary agreement, 107 Bright, John, 68–69 British Guiana, see Venezuela–British Guiana border dispute British-U.S. relations, see Anglo-U.S. relations

250 Bulganin, Nikolai, 133 Bull, Hedley, 62–63, 73, 205n.52, 206n.59 Bullock, Alan, 117 Burundi, 86, 128 Bush, George H. W., 146, 210n.4 Bush, George W., 95 Canadian-U.S. relations: democratic peace and, 42 political ties in, 176 trade in, 103 Trent affair in, 36–38 U.S. territorial expansion and boundary issues in, 37, 39, 43, 159 Capet, Hugh, 124–25, 231n.33 Capetian dynasty, 124–25, 231nn.33 and 34 capitalism, war and, 67 Caribbean, U.S. involvement in, 40, 44, 155 Carr, Raymond, 26 Carrero Blanco, Luis, 83 Carter, Jimmy, 94–95, 110, 147, 235n.59 Carter, John, 68 Carter administration, 95, 110 Casey, William, 235n.59 Castro, Fidel, 132 Catholics, 123, 128 Cederman, Lars-Erik, 151, 154 central banks, 107 Central Europe, see Eastern and Central Europe Central Intelligence Agency (CIA), 133, 137, 235n.59, 236n.62 Charles II, King of Spain, 125–27 Charles IV (the Fair), King of France, 124–25 Chavez, Hugo, 84 Chile, 84, 136, 137, 138, 216n.48, 236nn.60 and 62 Chiluba, Frederick, 119, 228n.16 China, People’s Republic of, 59, 86, 89–90, 146, 175, 183, 216 leadership change in, 116, 130, 131 Soviet Union and, 67, 135 Christopher, Warren, 148 Churchill, Sir Winston, 27, 163 civil rights, 193n.26 Civil War, U.S.: Anglo-U.S. relations in, 36–40, 54, 63, 175, 198n.57, 199nn.59, 63, 69, and 72 as war of democracies, 24 civil wars, 28 Clausewitz, Karl von, 20 Cleveland, Grover, 41 Clinton, Bill, 95

INDEX

Clinton administration, 137 close calls (near wars), 1, 15, 56–57, 159–60, 169 see also specific incidences Cobden, Richard, 68–69 “cold peace,” 43, 184 Cold War, 68, 233n.50 containment in, 71 Korean peninsula in, 145–46 lessons of appeasement in, 71–72 limits of U.S. hegemonic control in, 66 nuclear standoff in, 183–84 secrecy in, 77, 107 succession issues in, 113, 128 see also Soviet-U.S. relations Collina, Tom Zamora, 144–45, 237n.10 commitments, comparative seriousness of, 77–78 communication technology, 89–90, 216n.52 communism, 66, 113, 128, 236n.60 collapse of, 18 Communist Party, Soviet, 133, 134, 135, 136 Concert of Europe, 17, 188 Congo, Democratic Republic of, 86 Congress, U.S., 94–95, 110, 229n.21 consilience, 193n.25 Constitution, U.S., 111, 120–21, 226nn.1, 2, and 3 constitutions, constitutionalism, 15, 18, 75, 92–93, 94, 104, 170, 182 in communist-era Eastern Europe, 78 contracting advantage and, 12, 14, 48, 171 limiting effects of, 2, 8, 14, 80, 135, 164 medieval, 78 reliability provided by, 77–82, 173 self-binding and, 93–100 see also leadership changes Constructivism, 55, 60–63 containment policy, 71 contracting advantage, 193n.26 description of, 4–7, 15, 170–71, 170 four sources of, 11–14 international relations theory and, 55–73 policy implications of, 8–11, 171–77, 184– 87, 187 rationalist theory of, 53–55 self-protecting mechanisms and, 140–48 theoretical implications of, 139–68, 140, 172, 236n.1, 237nn.7 and 8, 239n.37, 240n.50, 241nn.52 and 60, 243n.82, 244n.85, 245n.102, 246nn.104 and 107 theories of war and, 47–53, 55, 169, 171

INDEX

cooperation, 2, 3–4 co-religionists, war among, 63 Corn Laws, 68, 208n.80 corporate executives, succession and, 233n.49 Correlates of War Project, 20, 149, 151, 191, 194n.9, 240n.46 cost explanation, of democratic peace, 2–3, 18 coups d’etat, 3, 84, 118, 152, 236n.62 courts, independence of, 80, 93–94 courts, international, 185 effective lack of, 4 use of, 42, 63, 70, 174–75 credibility, long-term, 100–104, 173 Cuba, 86, 133, 137 Cuban Missile Crisis, 78 Cyprus, 66 Cyprus War, 154 Czechoslovakia, 71 Czech Republic, 119, 185 dauphin, focal role of, 132 defense pacts, 149 Delcasse´ , The´ ophile, 32 democracies: alliances among, 8, 14, 58–59, 67–68, 149– 50 bargaining among, 8, 9–10, 14 common interests of, 195n.24 continuity of governance in, 12, 14 coups supported by, 3 durability of agreements among, 108–11 economic interdependence of, 54, 68 freedom of press and speech in, 88–93 harmonious preferences and, 69–70 illiberal, 182, 217n.56 “joint democracy” and, 157–58 leadership changes in, 8–9, 135–38 learning effects and early conflicts between, 150–52 liberal, 54 limits on promises made by, 104–6 long-term credibility and, 100–104, 173 long-term rivalry among, 155–56 mutual assessment between, 160–61 mutual fear and confidence as advantages of, 180 national reputation and regime continuity and, 82–87 near outbreaks of war between, 1, 15, 56–57, 159–60, 169 opaqueness in, 105, 106–7 other advantages of, 193n.26

251 self-binding nature of promises made by, 93–100 slow promise-breaking process in, 7, 80–81, 101, 177 threats carried out by, 54–55 threats made against other democracies by, 1, 3, 10, 151 types of, 79–80 war-proneness of, 3 see also elections; new democracies; transparency democracy: definition of, 20, 194nn.12 and 13 “joint,” 157–58, 169 as “organized uncertainty,” 79 transition to, 83–84 U.S. promotion of, 228n.17 “Democratic,” “Democracy,” duplicitous use of terms, 78, 86, 119 democratic peace: alternative theories of, 74 “cold,” 43 common data and definitions on, 20 economics and, 67–68 effect of external shocks on, 161–62 empirical and statistical studies of, 1, 17–22, 56, 154–58, 191n.1, 193n.1, 194n.9, 240nn.46, 50, and 51 Gowa’s challenge to consensus on, 22–24, 195n.22 international relations theory and, 55–73 key questions about, 17–20 lack of theoretical explanation of, 1–2, 59 Layne’s challenge to consensus on, 24–26, 56–57 limits of, 177–79 near breaches of, 1, 15, 56–57, 159–60, 169 19th-century Anglo-U.S. relations and, 36– 44, 45 policy implications of, 1–2 rationalist theories of, 53–55 Russo-Finnish War and, 22–23, 26–28 Spanish-American War and, 3, 23, 24, 26 three incomplete explanations of, 2–4 valid criticisms of consensus on, 28–30 “warm,” 43 as weakest for new democracies, 9 see also contracting advantage demonstrations, 90–91, 216n.54 Depression, Great, 168 dictatorships, 2, 3, 47, 152 communist, 18

252 dictatorships (cont.) dangers of loosening controls in, 196n.37 deception and, 99–100 expansionism of, 71–72, 96–98 instability of, 85–86 succession disputes in, 128–30 transparency opposed by, 109 Diehl, Paul F., 155–56 diplomats, 50–51, 99, 177 “dispute propensities,” 151 Disraeli, Benjamin, 31 Dixon, William J., 29, 157–58, 243n.74 domestic issues, state policy and, 69 Doyle, Michael, 18, 20 Dreyfus, Alfred, 31–32 due process, 193n.26 Dutch East Indies, 68 Dutch Republic, 123, 126, 127 dyadic proposition, 21–22, 56, 151, 157–58, 169, 195n.16 dynastic succession, 112–13, 123–28, 134, 135, 138 inter- vs. intradynastic, 124 East Asia, 176, 180, 186 Eastern and Central Europe: communist-era constitutions and national names in, 78 democratization of, 18, 65, 78–79, 84, 89, 119–20, 153–54, 183 NATO expansion and, 184–85, 207n.67 Soviet domination of, 18, 66 Ecclesiastes, 152 economy, U.S., 58 Ecuador, 66, 118 Edward III, King of England, 125 Egypt, Anglo-French conflict in, 25, 30–33, 35, 45, 54, 159 EHA (event history analysis), 149 Ehrenberg, Ronald, 49 Eisenhower, Dwight D., 109, 133 elections, 2, 158, 173, 185 contracting advantage and, 12, 48 fraudulent, 26, 113 incumbents’ surprises and, 164, 235n.59 in new democracies, 116–22 peaceful changes facilitated by, 98, 135–37 political promises and, 6–7, 79–82, 93 public scrutiny and, 234–35 elections, U.S.: of 1972, 164, 235n.59 of 1980, 94, 95

INDEX

Emmons, Juliann, 149 End of History and the Last Man, The (Fukayama), 19 “enduring rivalries,” 155–56 English School, 55, 62–63, 205n.52 Entente Cordiale (1904), 32 Erasmus, 62, 64 espionage, 99, 225n.119 ethnicity, 112 Europe: dynastic succession in, 112–13, 123–28 medieval, 78, 130–31 postwar elections in, 136 simultaneous development of democracy and press freedom in, 88 see also Eastern and Central Europe European Coal and Steel Community, 64 European Community, 84 European Court of Justice, 70 European Economic Community, 64 European Union (EU), 65, 70, 84, 185, 187 expansion of, 120 European-U.S. relations, 18, 19, 64–65, 186 trade and, 58, 204n.43 event history analysis (EHA), 149 Eyerman, Joe, 158 Farber, Henry, 22, 23 Fashoda, Anglo-French conflict in, 25, 30–33, 35, 45, 54, 159, 243n.82 Fearon, James, 49, 52, 81 Federalist Papers, 111, 226n.1 Finland, 22–23, 26–28, 97 Ford, Gerald, 94 Fox, Vicente, 134 France, 27, 38, 44, 57, 61, 64, 96, 113, 121, 127, 136, 142, 162–63, 167–68, 180, 188, 197nn.44, 45, and 47, 228n.15, 229n.19 conflict in Fashoda between Great Britain and, 25, 30–33, 35, 45, 54, 159, 243n.82 dynastic succession in, 123, 124–25, 132 historical changes in, 173 leadership change in, 117–18 power sharing system of, 80 Weimar Germany and, 155–56, 159–60 Franco, Francisco, 83 Franco-Prussian War, 228n.15 Franklin, Benjamin, 121, 229n.20 Frederick II (the Great), King of Prussia, 163–64 free trade, 185 French Revolution, 162, 173

INDEX

Fujimori, Alberto, 118 Fukayama, Francis, 19 Gallucci, Robert, 148 Gartzke, Erik, 68 Gaubatz, Kurt, 148–50 Gelpi, Christopher F., 81 German Democratic Republic (East Germany), 78 Germany, 3, 19, 64, 123, 176, 180 military rebuilding avoided by, 59 reassuring signals sent by, 65 Germany, Imperial, 25, 33–35, 43, 44, 68, 88, 97, 98, 142, 162, 178–79 Germany, Nazi, 26, 71, 96–97, 100, 117, 155– 56, 167–68, 181, 227n.10 Germany, Weimar Republic of, 79, 84, 155– 56, 159–60, 181, 227n.12, 242n.61 Ghent, Treaty of (1814), 43 Gilbert, Martin, 32 glasnost, 109 Gleditsch, Kristian S., 153 Gleditsch, Nils Petter, 21 globalization, 13, 181 Glorious Revolution, 102, 230n.28 Goertz, Gary, 155–56 Gorbachev, Mikhail, 18, 89, 131, 135 Gowa, Joanne, 22–24, 29, 151, 195n.22 Grace, J. Peter, 233n.49 Grand Alliance, 127 Great Britain, 27, 35, 57, 72, 88, 94, 96, 123, 127, 132, 142–43, 163, 167–68, 188 appeasement policy of, 71 conflict in Fashoda between France and, 25, 30–33, 35, 45, 54, 159, 243n.82 period of economic dominance of, 66–67 rise of parliamentary authority in, 102, 222n.98 war-proneness of, 3 see also Anglo-U.S. relations Great Depression, 168 Greece, 66, 166 Grieco, Joseph, 58 Griesdorf, Michael, 81 Gromyko, Andrei, 100 Guicciardini, Francesco, 50–51 Gurr, Ted Robert, 20 Habsburg dynasty, 112, 123, 125–27, 163–64, 203n.34, 228n.15 Haiti, 118–19 Halperin, William S., 228n.15

253 Hammer, Armand, 233n.49 Handbook of War Studies II (Midlarsky), 158 Hardy, Thomas, 188 Hart, Robert A., 157–58 Hayek, Friedrich von, 7 Hegel, G.W.F., 181 Hegre, Ha˚ vard, 28 Henry I, King of France, 231n.33 Hensel, Paul, 156 Hicks, John, 49 Hindenburg, Paul von, 117 history, “final stage,” 181–84 Hitler, Adolf, 26, 71, 96–98, 100, 117, 156, 168, 181, 220nn.78 and 79, 221nn.81 and 82, 227nn.10 and 12, 242n.61 Hobbes, Thomas, 30, 141, 179–80, 197n.42 Hohenzollern dynasty, 113, 123, 228n.15 Holland, 176 Holmes, Stephen, 111 House of Representatives, U.S., 229n.21 Hundred Years’ War, 125 Hungary, 119, 185 Huntington, Samuel P., 153 Imperialism, The Highest Stage of Capitalism (Lenin), 67 incomplete contracts, 5 incomplete information, 82–83, 99 India, 66, 243n.82 Indonesia, 68, 119, 232n.48 information: incomplete, 82–83, 99 private, 47, 49–52, 201n.14, 214n.39 information technology, 89–90, 216n.52 inheritance rules, 130–31 institutional explanation, of democratic peace, 2, 4, 64–65, 75 integrative cumulation, 139 intelligence agencies, 107 International Atomic Energy Agency (IAEA), 108, 144–45, 147, 238n.17 International Court of Justice, 78, 185 International Criminal Court, 185 international institutions, democratic peace and, 64–65 International Monetary Fund (IMF), 120 international order, stable contracts as source of, 179–81 international relations (IR) theory, democratic peace and, 55–73 international society, 62, 73 International Trade Organization, 110

254 Internet, 89 “intersubjective liberal identity,” 60 Iran, 57, 128–30, 163, 182, 232nn.43 and 44 Iran-Iraq War, 129–30 Iraq, 57, 65, 87, 129–30, 140, 147, 152, 163 Islamic fundamentalists, 121, 128, 136, 229n.19, 230n.24 Islamic Salvation Front, 121 Israel, 66, 121, 137, 230nn.23 and 24 Italy, 168, 230n.28 ivory trade, 108–9 Jaggers, Keith, 20 Japan, 19, 58, 163, 180, 185, 186 imperial expansion of, 42–43, 68, 208n.82 military rebuilding avoided by, 59 one-party rule in, 80, 94, 101, 104 as “opaque democracy,” 105, 177, 182–83 Pearl Harbor attack and, 68, 71, 96 Jay, John, 111 Jefferson, Thomas, 121 Jervis, Robert, 55, 72–73, 237n.8 “joint democracy,” 157–58, 169 Joint Nuclear Control Commission, 146 Jordan, 66 Juan Carlos, King of Spain, 83 judiciary, independence of, 80, 93–94 Julius II, Pope, 51–52, 201n.20, 202nn.21 and 22 Kabila, Laurent, 86 Kant, Immanuel, 2, 18, 24, 25, 62, 72, 166, 178 Kantor, Arnold, 147 Kashmir War, Second, 154 Kaunda, Kenneth, 119, 228n.16 Keohane, Robert O., 193, 206n.65, 207n.68, 213n.26 Khadaffi, Mohammar, 132 Khomeini, Ayatollah Ruholla, 182 Khrushchev, Nikita S., 109, 133, 135, 225n.119, 234n.51 Kim Jong-il, 232n.48 Kim Il-Sung, 232n.48 Kirkpatrick, Jeanne, 236n.60 Kissinger, Henry A., 164, 203n.34, 228n.17, 235n.59, 245n.98 Kitchener, Herbert, 30, 31 Korea, People’s Democratic Republic of (North Korea), 57, 65, 71, 86, 143, 232n.48 Korea, Republic of (South Korea), 71, 143, 180, 186

INDEX

Kosovo, 166 Kydd, Andrew, 55–56 Kyoto Protocol on Global Warming (1997), 95 Labor Party (Israel), 121 Lai, Brian, 22, 150 Lakatos, Imre, 8, 139 Lane, Christel, 9 Larson, Deborah Welch, 72 Latin America, 18, 66 new democracies of, 84, 118, 153–54, 181 Law of the Sea, 95, 218n.69 Layne, Christopher, 24–26, 28, 29, 36, 56, 70 leadership changes (succession), 171 as cause of war, 86, 112–38, 226n.1, 228n.15, 229n.19, 230n.28, 232n.48, 234n.51, 235n.59 in democracies, 8–9, 135–38 in new democracies, 113, 116–22, 136 in nondemocracies, 114–16, 123–24, 135, 138, 162–65 reputation and, 82–87 transparency and, 116, 132, 133, 135 turbulent vs. peaceful, 112–16 see also elections League of Nations, 27, 96, 167, 178 learning, two types of, 70 learning effects, 150–52 Lee, Robert E., 39 Lend-Lease program, 96, 220n.76 Lenin, V. I., 67, 162 “lesson of the Somme,” 71 Leviathan (Hobbes), 179 Levy, Jack, 1, 21–22 liberal democracies, 54 Liberal Democratic Party (Japan), 80, 104 liberalism, 68–69, 75 liberal values, 18 Libya, 65, 86, 133 Likud Party (Israel), 121 Lincoln, Abraham, 37–38 literacy, 215n.43 Lithuania, 97 Lithuanian War of Independence, 154 Locarno Agreements (1926), 156, 168 Louis XIV, King of France, 126–27 Low Countries, 88, 97 Macaulay, Thomas Babington, Lord, 51 McGovern, George, 164, 235n.59 Machiavelli, Niccolo`, 51–52, 54, 102, 201n.20, 202nn.21 and 22

255

INDEX

McIlwain, Charles Howard, 78 McMillan, Susan, 68 Mahuad, Jamil, 118 Maine, U.S.S., 3 Malenkov, Georgii, 133, 233n.50 Mansfield, Edward D., 153, 240nn.46, 50, and 51 Mao Zedong, 235n.56 Marchand, Jean-Baptiste, 25, 31, 38 Marx, Karl, 67 Marxists, 67–68 Mearsheimer, John, 19 median voter, 6, 48, 200n.2 medieval constitutionalism, 78 medieval Europe, 78, 130–31 Mexico, 134, 234n.53 Middle East, 66 Midlarsky, Manus I., 158 Militarized Interstate Disputes (MID) data, 20, 151, 154, 194nn.9 and 10 Mill, John Stuart, 160 Millett, Allan R., 27 minorities, rights of, 193n.26 Mobutu Sese Seko, 85–86 Molotov-Ribbentrop Pact (1939), 26, 96–97, 221n.81 monadic proposition, 21, 195n.16 monarchies, 26, 47, 48, 152, 222n.88 cost explanation and, 2, 3, 17 lack of press freedom in, 88 succession in, 112–13, 122–28, 132, 135, 138, 231nn.32 and 33 Moravcsik, Andrew, 69 Morocco, 32, 33, 34, 35, 36 Morrow, James, 69 Movement for Multi-Party Democracy, 119 Mueller, John, 70–71 Mugabe, Robert, 84, 119 multilateral relationships, weakness of, 178 Munich conference (1938), 71, 220n.79 Murray, Williamson, 27 Mussolini, Benito, 71, 96, 168 Myanmar, 134 NAFTA (North American Free Trade Agreement), 70 Napoleon III, Emperor of France, 117–18, 227n.13, 228n.15 NATO (North Atlantic Treaty Organization), 51, 64–65, 71–72, 83–84, 120, 166–67, 168, 184–85, 187, 207n.67 Nazi-Soviet Pact, 26, 96–97, 221nn.81 and 82

neo-Realism, 2, 29, 55 new democracies, 24, 54 in Africa, 84, 181 in Asia, 84, 181 commitment problems and, 153–54 in Eastern Europe, 18, 65, 84, 153–54, 183 instability of, 9, 181 international institutions and, 65 in Latin America, 84, 118, 153–54, 181 leadership change in, 113, 116–22, 136 limits on promises made by, 104 national reputation and, 83–84 secrecy in, 86–93 unreliability of promises made by, 78, 82, 85 war-proneness of, 1, 181–82 Nigeria, 134, 175 Nixon, Richard M., 164, 235n.59 Nixon administration, 137 nondemocratic governments, 3, 21, 66, 152, 188 leadership change in, 114–16, 123–34, 135, 138, 162–65 military weaknesses of, 223n.103 press restrictions in, 7 self-protective mechanisms for dealing with, 140–48 succession differences among, 130–34 see also dictatorships; monarchies; one-party states nongovernmental organizations (NGOs), 176 normative explanation, of democratic peace, 2–3 North, Douglass, 102 North American Free Trade Agreement (NAFTA), 70 North Atlantic Treaty Organization (NATO), 51, 64–65, 71–72, 83–84, 120, 166–67, 168, 184–85, 187, 207n.67 Nuclear Non-proliferation Treaty (NPT), 108, 145, 147, 237n.10 nuclear weapons, 57 Argentina and Brazil’s agreement on, 143– 45, 146, 180, 186, 237n.10, 238nn.11, 14, and17 as deterrent, 19, 134, 152, 183–84 North and South Korea and, 143, 145–48 treaties on, 108, 109–10, 145, 225nn. 115 and 119, 226nn.120 and 121 Nye, Joseph S., 193n.23 Occidental Petroleum, 233n.49 October Surprise, 164, 235n.59 oil, 68

256 Olney, Richard, 41 Olson, Mancur, 11, 139, 193n.25 Oneal, John, 68, 193n.1 one-party states, 47, 152 Japan as, 80, 94, 101, 104 judiciary in, 80 leadership succession in, 134 opposition parties, 7 Ostrom, Elinor, 187 Ottoman Empire, 30, 34, 117, 132 Owen, John, 18, 25, 43 Pakistan, 66, 84, 119 Palmer, Glenn, 82 Palmerston, Lord, 19, 36, 38 Panama Canal, 41, 44, 94–95, 110 Pareto-superior solutions, 49, 200n.7 Paris, Treaty of (1783), 43 Paris Commune, 67 parliamentary systems, 2, 80, 101–2, 212n.12 Partell, Peter J., 82 Pax Americana, 23, 24 peace: “cold,” 43, 184 limited goal of, 62 path to, 185 promise of, 187–89 “seasons” of, 152 “warm,” 43 see also cooperation; democratic peace peacekeeping missions, 185, 186–87 Pearl Harbor, 68, 71, 96 Peres, Shimon, 137 perestroika, 89, 109 Perpetual Peace (Kant), 18, 72 Persian Gulf regimes, 109 Peru, 66, 118 Pharo, Per Fredrik Ilsaas, 109 Philip IV (the Fair), King of France, 124 Philippines, 119 Pitt, William, the Elder, 82 Polachek, Solomon, 68 Poland, 97, 119, 185 Politburo, 133 Polity data sets, 20, 194n.9 polls, 91 Portugal, 83 Powell, Colin, 234n.55 presidency, U.S., 80, 113, 117, 121, 221n.86, 226nn.2 and 3 presidential systems, 2, 80, 192n.13 press freedom, 7, 87–93, 185, 215n.44

INDEX

Preval, Rene´ , 119 PRI, 134, 234n.53 Prisoners’ Dilemma, 3–4, 63 privacy, 90 private information, 47, 49–52, 201, 214n.39 promises, reciprocal exchange of, 6–7 property rights, 193n.26 proportional voting systems, 80 Protestants, 123, 128 “protracted conflicts,” 155 proxy agreements, 105–6, 108, 183, 224n.110 proxy variables, 141 Prussia, 113, 118, 127, 163–64, 188, 228n.15 Przeworski, Adam, 79 Putnam, Robert, 187 Quayle, Dan, 9 Ramseyer, J. Mark, 101 rationalist theories, 47–55 of democratic peace, 53–55 of war, 47–52, 55 Ray, James Lee, 21, 25 Reagan, Ronald, 94–95, 235n.59 Realism, 27–28, 29, 47, 55, 56–57, 59, 75, 184–85, 203n.34 Realpolitik, 19, 24–26, 36 Red Army, Soviet, 27, 97, 134 Reed, William, 149–50, 157–58 regime maturity, 241n.52 Reiss, Mitchell, 146 Reiter, Dan, 22, 150 relational contracts, 53 religion, 63, 112, 123, 128 Renaissance, 51–52, 99 “republican” regimes, 102 reputation, national, 53, 82–87, 143, 237nn.7 and 8 revolutionary regimes, 223n.102 Revolutionary War, 43 Rhodes, Cecil, 30 Ribbentrop, Joachim von, 100 Ribbentrop-Molotov Pact (1939), 26, 96–97, 221n.81 Rioux, Jean-Se´ bastien, 81 Risse, Thomas, 60 Robert II, King of France, 231n.33 Robespierre, Maximilien, 162 Romanov dynasty, 123 Rome, Treaty of (1957), 64 Roosevelt, Franklin D., 96, 220nn.74, 75, and 76

INDEX

Royal Navy, British, 33, 35, 38 rule of law, 2 Rummel, R. J., 21 Rush-Bagot Agreement (1817–18), 39 Russell, Lord, 40 Russett, Bruce, 21, 25, 54, 68, 193n.1 Russia, 119, 137 Russia, Imperial, 27, 35, 44, 88, 123, 188 Russian Civil War, 163 Russian Revolution, 162–63 Russo-Finnish War, 22–23, 26–28 Rwanda, 86, 128 Saddam Hussein, 130, 133, 163 “salami slicing,” 141 Salic Law, 125 Salisbury, Robert Gascoyne-Cecil, Lord, 30, 31–32 “sameness” of regime types, war and, 63 sanctions, 165–66 Saudi Arabia, 86 Scandinavia, 88 Schelling, Thomas, 246n.1 Schlieffen Plan, 97 Schroeder, Paul W., 127–28 Schultz, Kenneth A., 32, 82, 161 Scowcroft, Brent, 147 “seasons of peace,” 152 Second Kashmir War, 154 security communities, 60–61, 64, 224n.108 self-binding promises, 6 self-protection and self-improvement, 13–14 Senate, U.S., 94–95, 110, 229n.21 Senese, Paul, 154, 157–58 Seward, William, 38 Shakespeare, William, 52, 106–7, 202n.25 shari’a law, 128, 232n.40 Sharif, Nawaz, 119 Shaw, George Bernard, 90 Shi’a Muslims, 128, 129, 163 Sigal, Leon, 147 Sigmund, Paul, 236n.62 Singer, J. David, 18 Siverson, Randolph, 149 Small, Melvin, 18 Smith, Robert, 49 Smith, Tony, 228n.17 Snyder, Jack, 153, 240nn.46 and 51 “society of states,” 73 Southeast Asia, 67, 68 Soviet Union, 100, 109, 137, 146 Afghanistan and, 110, 229n.19

257 China and, 67, 135 collapse of, 18 Eastern Europe and, 18, 66 fraudulent weapons parades staged by, 51, 201n.17 glasnost and perestroika in, 89, 109 increased transparency in, 109–10 leadership succession in, 116, 131, 133, 135, 232n.47 Nazi invasion of, 71, 96–97, 221nn.81 and 82 in Russo-Finnish War, 22–23, 26–28 Soviet-U.S. relations: arms control agreements in, 108, 109–10 Cuban Missile Crisis resolved in, 77 deterrence in, 19, 134, 152, 183–84 wartime alliance in, 58 see also Cold War Spain, 83–84, 113, 123, 125–27 Spanish-American War, 3, 23, 24, 26, 41, 154 speech, freedom of, 87–93 spot transactions, 100–101 Stalin, Josef, 26–27, 71, 96–97, 133, 135, 221n.81, 235nn.56 and 57 strikes, 49–50, 201n.13 strong-party states, see one-party states succession, see leadership changes Suez Canal, 30–31 Suharto, 232n.48 Sumner, Charles, 38 Sunni Muslims, 128 Syria, 86, 133, 134, 143, 232n.48 Taiwan, 180, 186 television, 89 Texas, 39 Thatcher, Margaret, 80 theories: consilience and, 193n.25 progressive research programs for, 8 retrograde vs. progressive, 139 Theory of Wages, The (Hicks), 49 Thirty Years’ War, 127 Thompson, William, 155 threats: assessing, 57–58, 59, 100 following through on, 54–55 made between democracies, 1, 3, 10, 151 Thucydides, 54 time consistency, 141 trade, 18 free, 185 U.S.-Canada, 103

258 trade (cont.) U.S.-Europe, 58, 204n.43 war and, 68–69, 209n.85 transnational ties, 176–77 transparency, 6, 9, 36, 141 avoidance of war through, 53, 54–55, 59, 72–73, 75, 186 contracting advantage and, 7, 12, 14, 45, 47, 61, 171–73 effects in absence of, 49–52, 104–7, 177, 182–83, 186 importance of, 85, 93 leadership change and, 116, 132, 133, 135 limits of, 186 opposition to, 109 taxonomy of, 211n.2 treaties, 43 breaking of, 7, 96–100, 127, 142–43 comparative seriousness of, 77 definition of, 211n.5 as “incomplete contracts,” 5 Jay on, 111 Trent affair, 36, 37–39, 54, 159 Triple Entente, 35 Trotsky, Leon, 162 Truman administration, 110, 186 Turkey, 66, 136, 166 turno pacı´fico, 26 Tutsi people, 86 United Nations (UN), 86, 185 United States: Congress in, 229n.21 democracy promoted by, 228n.17 economy of, 58 as global power, 22, 23, 24, 65–66, 168, 178 Haiti and, 119 Japan and, 182–83, 186 judiciary in, 94 as new democracy, 120–21, 229nn.19 and 20 North Korea and, 145–48 power sharing system of, 80, 121, 212, 229nn.19 and 21 presidency in, 229n.21 regime change and, 136, 236nn.60 and 62 in Spanish-American War, 3, 23, 24, 26 too-close relationships between semidemocratic states and, 186 war-proneness of, 3 see also Anglo-U.S. relations; EuropeanU.S. relations Valois dynasty, 125 Venezuela, 84

INDEX

Venezuela–British Guiana border dispute, 40– 42, 45, 63, 155, 175 verification issues, 109–10, 141–42, 225n.119 Versailles, Treaty of (1918), 110, 156, 159 Vietnam War, 164, 235n.59, 245n.98 voting, proportional, 80 voting rights, 88 Wales, Prince of, focal role of, 132 Walt, Stephen, 35, 114, 235n.58 Waltz, Kenneth, 29, 55 war: Constructivist theory of, 62–63 cost explanation and, 2–3, 18 definition of, 20, 194n.9 democracies as winners of, 1 dictatorship succession and, 128–30 dynastic succession and, 123–28 incomplete information and, 99 mutual fear and confidence in avoidance of, 180 near outbreaks of, 1, 159–60, 169 neighboring states and, 191n.2 obsolescence of, 70–71 rationalist theories of, 47–52, 55, 169 risk acceptance and, 244n.87 “sameness” and, 63 succession issues and, 86, 112–38, 226n.1, 228n.15, 229n.19, 230n.28, 232n.48, 234n.51, 235n.59, 236n.62 trade and, 68–69, 209n.85 wars, specific: Anglo-Boer War, 24, 43 Civil War, U.S., see Civil War, U.S. Cyprus War, 154 Franco-Prussian War, 228n.15 Hundred Years’ War, 125 Iran-Iraq War, 129–30 Lithuanian War of Independence, 154 Revolutionary War, 43 Russian Civil War, 163 Russo-Finnish War, 22–23, 26–28 Second Kashmir War, 154 Spanish-American War, 3, 23, 24, 26, 41, 154 Thirty Years’ War, 127 Vietnam War, 164, 235n.59, 245n.98 War of Austrian Succession, 230n.28 War of 1812, 39, 43, 44, 68 War of Spanish Succession, 125–27, 230n.28 World War I, see World War I World War II, see World War II Ward, Michael D., 153 War of Austrian Succession, 230n.28

259

INDEX

War of 1812, 39, 43, 44, 68 War of Spanish Succession, 125–27, 230n.28 Warsaw Pact, 66, 89 Washington, George, 117, 121 Washington Treaty (1871), 40 Wehler, Hans-Ulrich, 179 Weinberg, Gerhard, 27 Weingast, Barry, 102 Wendt, Alexander, 60 Werner, Suzanne, 164–65 Whewell, William, 193n.25 Wight, Martin, 62–63, 73, 205n.52 Wilhelm II, Kaiser, 142, 179 Weltpolitik of, 33–35 Wilkenfeld, Jonathan, 155, 158 Wilson, Charles, 99 Wilson, Woodrow, 19, 96, 110, 167–68, 178, 228n.17

Womack, Brantly, 86 World Bank, 120 World Trade Organization (WTO), 70, 108, 120 World War I, 23, 68, 71, 97, 98, 167, 209n.98, 220n.75, 242n.61 Russia’s withdrawal from, 162–63 World War II, 58 excluded from Gowa’s analysis, 22–23, 195n.22 outbreak of, 96–98, 220n.75 Yeltsin, Boris, 137 Youngman, Henny, 7 Zaire, 85–86 Zambia, 119, 228n.16 Zimbabwe, 84, 119 Zinnes, Dina, 139