international relations in the post cold war era

St r uct ur al Real i sm aft er t he Col d W ar Kenneth N. Waltz S ome students of in- ternational politics believe that realism is obsolete. 1 They argue that, although realism’ s concepts of anarchy , self-help, and power balancing may have been appropriate to a bygone era, they have been displaced by changed conditions and eclipsed by better ideas. New times call for new thinking. Changing conditions require revised theories or entirely different ones. T rue, if the conditions that a theory contemplated have changed, the theory no longer applies. But what sorts of changes would alter the international political system so profoundly that old ways of thinking would no longer be relevant? Changes of the system would do it; changes in the system would not. Within-system changes take place all the time, some important, some not. Big changes in the means of transportation, communication, and war Ž ghting, for example, strongly affect how states and other agents interact. Such changes occur at the unit level. In modern history , or perhaps in all of history , the introduction of nuclear weaponry was the greatest of such changes. Y et in the nuclear era, international politics remains a self-help arena. Nuclear weapons decisively change how some states provide for their own and possibly for others’ security; but nuclear weapons have not altered the anarchic structure of the international political system. Changes in the structure of the system are distinct from changes at the unit level. Thus, changes in polarity also affect how states provide for their security . SigniŽ cant changes take place when the number of great powers reduces to two or one. With more than two, states rely for their security both on their Kenneth N. Waltz, former Ford Professor of Political Science at the University of California, Berkeley, is a Research Associate of the Institute of War and Peace Studies and Adjunct Professor at Columbia University. I am indebted to Karen Adams and Robert Rauchhaus for help on this article from its conception to its completion. For insightful and constructive criticisms I wish to thank Robert Art, Richard Betts, Barbara Farnham, Anne Fox, Robert Jervis, W arner Schilling, and Mark Sheetz. 1. For example, Richard Ned Lebow , “ The Long Peace, the End of the Cold W ar, and the Failure of Realism,” International Organization, V ol. 48, No. 2 (Spring 1994), pp. 249– 277; Jeffrey W . Legro and Andrew Moravcsik, “ Is Anybody Still a Realist?” International Security, V ol. 24, No. 2 (Fall 1999), pp. 5– 55; Bruce Russett, Grasping the Democratic Peace: Principles for a Post– Cold War Peace (Princeton, N.J.: Princeton University Press, 1993); Paul Schroeder, “ Historical Reality vs. Neo- realist Theory ,” International Security, V ol. 19, No. 1 (Summer 1994), pp. 108– 148; and John A. V asquez, “ The Realist Paradigm and Degenerative vs. Progressive Research Programs: An Ap- praisal of Neotraditional Research on W altz’ s Balancing Proposition,” American Political Science Review, V ol. 91, No. 4 (December 1997), pp. 899– 912. International Security, V ol. 25, No. 1 (Summer 2000), pp. 5– 41 © 2000 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology . 5 own internal efforts and on alliances they may make with others. Competition in multipolar systems is more complicated than competition in bipolar ones because uncertainties about the comparative capabilities of states multiply as numbers grow , and because estimates of the cohesiveness and strength of coalitions are hard to make. Both changes of weaponry and changes of polarity were big ones with ramiŽ cations that spread through the system, yet they did not transform it. If the system were transformed, international politics would no longer be international politics, and the past would no longer serve as a guide to the future. W e would begin to call international politics by another name, as some do. The terms “ world politics” or “ global politics,” for example, suggest that politics among self-interested states concerned with their security has been replaced by some other kind of politics or perhaps by no politics at all. What changes, one may wonder , would turn international politics into some- thing distinctly different? The answer commonly given is that international politics is being transformed and realism is being rendered obsolete as democ- racy extends its sway , as interdependence tightens its grip, and as institutions smooth the way to peace. I consider these points in successive sections. A fourth section explains why realist theory retains its explanatory power after the Cold W ar . Democracy and Peace The end of the Cold W ar coincided with what many took to be a new demo- cratic wave. The trend toward democracy combined with Michael Doyle’ s rediscovery of the peaceful behavior of liberal democratic states inter se con- tributes strongly to the belief that war is obsolescent, if not obsolete, among the advanced industrial states of the world. 2 The democratic peace thesis holds that democracies do not Ž ght democra- cies. Notice that I say “ thesis,” not “ theory .” The belief that democracies constitute a zone of peace rests on a perceived high correlation between governmental form and international outcome. Francis Fukuyama thinks that the correlation is perfect: Never once has a democracy fought another democ- racy . Jack Levy says that it is “ the closest thing we have to an empirical law 2. Michael W . Doyle, “ Kant, Liberal Legacies, and Foreign Affairs, Parts 1 and 2,” Philosophy and Public Affairs, V ol. 12, Nos. 3 and 4 (Summer and Fall 1983); and Doyle, “ Kant: Liberalism and W orld Politics,” American Political Science Review, V ol. 80, No. 4 (December 1986), pp. 1151– 1169. International Security 25:1 6 in the study of international relations.” 3 But, if it is true that democracies rest reliably at peace among themselves, we have not a theory but a purported fact begging for an explanation, as facts do. The explanation given generally runs this way: Democracies of the right kind (i.e., liberal ones) are peaceful in relation to one another . This was Immanuel Kant’ s point. The term he used was Rechtsstaat or republic, and his deŽ nition of a republic was so restrictive that it was hard to believe that even one of them could come into existence, let alone two or more. 4 And if they did, who can say that they would continue to be of the right sort or continue to be democracies at all? The short and sad life of the W eimar Republic is a reminder . And how does one deŽ ne what the right sort of democracy is? Some American scholars thought that Wilhelmine Germany was the very model of a modern democratic state with a wide suffrage, honest elections, a legislature that controlled the purse, competitive parties, a free press, and a highly competent bureaucracy .

5 But in the French, British, and American view after August of 1914, Germany turned out not to be a democracy of the right kind. John Owen tried to Ž nesse the problem of deŽ nition by arguing that democracies that perceive one another to be liberal democracies will not Ž ght. 6 That rather gives the game away . Liberal democ- racies have at times prepared for wars against other liberal democracies and have sometimes come close to Ž ghting them. Christopher Layne shows that some wars between democracies were averted not because of the reluctance of democracies to Ž ght each other but for fear of a third party— a good realist reason. How , for example, could Britain and France Ž ght each other over Fashoda in 1898 when Germany lurked in the background? In emphasizing the international political reasons for democracies not Ž ghting each other , Layne gets to the heart of the matter .

7 Conformity of countries to a prescribed 3. Francis Fukuyama, “ Liberal Democracy as a Global Phenomenon,” Political Science and Politics, V ol. 24, No. 4 (1991), p. 662. Jack S. Levy , “ Domestic Politics and W ar ,” in Robert I. Rotberg and Theodore K. Rabb, eds., The Origin and Prevention of Major Wars (Cambridge: Cambridge University Press, 1989), p. 88. 4. Kenneth N. W altz, “ Kant, Liberalism, and W ar ,” American Political Science Review, V ol. 56, No. 2 (June 1962). Subsequent Kant references are found in this work. 5. Ido Oren, “ The Subjectivity of the ‘ Democratic’ Peace: Changing U.S. Perceptions of Imperial Germany ,” International Security, V ol. 20, No. 2 (Fall 1995), pp. 157ff.; Christopher Layne, in the second half of Layne and Sean M. Lynn-Jones, Should America Spread Democracy? A Debate (Cam- bridge, Mass.: MIT Press, forthcoming), argues convincingly that Germany’ s democratic control of foreign and military policy was no weaker than France’ s or Britain’ s. 6. John M. Owen, “ How Liberalism Produces Democratic Peace,” International Security, V ol. 19, No. 2 (Fall 1994), pp. 87– 125. Cf. his Liberal Peace, Liberal War: American Politics and International Security (Ithaca, N.Y .: Cornell University Press, 1997). 7. Christopher Layne, “ Kant or Cant: The Myth of the Democratic Peace,” International Security, V ol. 19, No. 2 (Fall 1994), pp. 5– 49. Structural Realism after the Cold War 7 political form may eliminate some of the causes of war; it cannot eliminate all of them. The democratic peace thesis will hold only if all of the causes of war lie inside of states. the causes of war T o explain war is easier than to understand the conditions of peace. If one asks what may cause war , the simple answer is ” anything.“ That is Kant’ s answer: The natural state is the state of war . Under the conditions of international politics, war recurs; the sure way to abolish war , then, is to abolish interna- tional politics. Over the centuries, liberals have shown a strong desire to get the politics out of politics. The ideal of nineteenth-century liberals was the police state, that is, the state that would conŽ ne its activities to catching criminals and enforcing contracts. The ideal of the laissez-faire state Ž nds many counterparts among students of international politics with their yen to get the power out of power politics, the national out of international politics, the dependence out of inter- dependence, the relative out of relative gains, the politics out of international politics, and the structure out of structural theory . Proponents of the democratic peace thesis write as though the spread of democracy will negate the effects of anarchy . No causes of con ict and war will any longer be found at the structural level. Francis Fukuyama Ž nds it ” perfectly possible to imagine anarchic state systems that are nonetheless peaceful.“ He sees no reason to associate anarchy with war . Bruce Russett believes that, with enough democracies in the world, it “ may be possible in part to supersede the ‘ realist’ principles (anarchy , the security dilemma of states) that have dominated practice . . . since at least the seventeenth cen- tury .” 8 Thus the structure is removed from structural theory . Democratic states would be so conŽ dent of the peace-preserving effects of democracy that they would no longer fear that another state, so long as it remained democratic, would do it wrong. The guarantee of the state’ s proper external behavior would derive from its admirable internal qualities. This is a conclusion that Kant would not sustain. German historians at the turn of the nineteenth century wondered whether peacefully inclined states could be planted and expected to grow where dangers from outside pressed daily upon them. 9 Kant a century earlier entertained the same worry . The 8. Francis Fukuyama, The End of History and the Last Man (New Y ork: Free Press, 1992), pp. 254– 256. Russett, Grasping the Democratic Peace, p. 24. 9. For example, Leopold von Ranke, Gerhard Ritter , and Otto Hintze. The American William Graham Sumner and many others shared their doubts. International Security 25:1 8 seventh proposition of his ” Principles of the Political Order“ avers that estab- lishment of the proper constitution internally requires the proper ordering of the external relations of states. The Ž rst duty of the state is to defend itself, and outside of a juridical order none but the state itself can deŽ ne the actions required. ” Lesion of a less powerful country ,“ Kant writes, ” may be involved merely in the condition of a more powerful neighbor prior to any action at all; and in the State of Nature an attack under such circumstances would be warrantable.“ 10 In the state of nature, there is no such thing as an unjust war . Every student of international politics is aware of the statistical data sup- porting the democratic peace thesis. Everyone has also known at least since David Hume that we have no reason to believe that the association of events provides a basis for inferring the presence of a causal relation. John Mueller properly speculates that it is not democracy that causes peace but that other conditions cause both democracy and peace. 11 Some of the major democra- cies— Britain in the nineteenth century and the United States in the twentieth century— have been among the most powerful states of their eras. Powerful states often gain their ends by peaceful means where weaker states either fail or have to resort to war . 12 Thus, the American government deemed the demo- cratically elected Juan Bosch of the Dominican Republic too weak to bring order to his country . The United States toppled his government by sending 23,000 troops within a week, troops whose mere presence made Ž ghting a war unnecessary . Salvador Allende, democratically elected ruler of Chile, was sys- tematically and effectively undermined by the United States, without the open use of force, because its leaders thought that his government was taking a wrong turn. As Henry Kissinger put it: ” I don’ t see why we need to stand by and watch a country go Communist due to the irresponsibility of its own people.“ 13 That is the way it is with democracies— their people may show bad judgment. ” W ayward“ democracies are especially tempting objects of interven- tion by other democracies that wish to save them. American policy may have been wise in both cases, but its actions surely cast doubt on the democratic peace thesis. So do the instances when a democracy did Ž ght another democ- 10. Immanuel Kant, The Philosophy of Law, trans. W . Hastie (Edinburgh: T . and T . Clark, 1887), p. 218. 11. John Mueller , “ Is W ar Still Becoming Obsolete?” paper presented at the annual meeting of the American Political Science Association, W ashington, D.C., August– September 1991, pp. 55ff; cf. his Quiet Cataclysm: Re ections on the Recent T ransformation of World Politics (New Y ork: HarperCollins, 1995).

12. Edward Hallett Carr , Twenty Y ears’ Crisis: An Introduction to the Study of International Relations, 2d ed. (New Y ork: Harper and Row , 1946), pp. 129– 132. 13. Quoted in Anthony Lewis, “ The Kissinger Doctrine,” New Y ork Times, February 27, 1975, p. 35; and see Henry Kissinger , The White House Y ears (Boston: Little, Brown, 1979), chap. 17. Structural Realism after the Cold War 9 racy .

14 So do the instances in which democratically elected legislatures have clamored for war , as has happened for example in Pakistan and Jordan. One can of course say , yes, but the Dominican Republic and Chile were not liberal democracies nor perceived as such by the United States. Once one begins to go down that road, there is no place to stop. The problem is height- ened because liberal democracies, as they prepare for a war they may fear , begin to look less liberal and will look less liberal still if they begin to Ž ght one. I am tempted to say that the democratic peace thesis in the form in which its proponents cast it is irrefutable. A liberal democracy at war with another country is unlikely to call it a liberal democracy . Democracies may live at peace with democracies, but even if all states became democratic, the structure of international politics would remain anar- chic. The structure of international politics is not transformed by changes internal to states, however widespread the changes may be. In the absence of an external authority , a state cannot be sure that today’ s friend will not be tomorrow’ s enemy . Indeed, democracies have at times behaved as though today’ s democracy is today’ s enemy and a present threat to them. In Federalist Paper number six, Alexander Hamilton asked whether the thirteen states of the Confederacy might live peacefully with one another as freely constituted republics. He answered that there have been ” almost as many popular as royal wars.“ He cited the many wars fought by republican Sparta, Athens, Rome, Carthage, V enice, Holland, and Britain. John Quincy Adams, in response to James Monroe’ s contrary claim, averred ” that the government of a Republic was as capable of intriguing with the leaders of a free people as neighbor- ing monarchs.“ 15 In the latter half of the nineteenth century , as the United States and Britain became more democratic, bitterness grew between them, and the possibility of war was at times seriously entertained on both sides of the Atlantic. France and Britain were among the principal adversaries in the great power politics of the nineteenth century , as they were earlier . Their becoming democracies did not change their behavior toward each other . In 1914, democratic England and France fought democratic Germany , and doubts about the latter ’ s democratic standing merely illustrate the prob- lem of deŽ nition. Indeed, the democratic pluralism of Germany was an under- lying cause of the war . In response to domestic interests, Germany followed 14. See, for example, Kenneth N. W altz, “ America as Model for the W orld? A Foreign Policy Perspective,” PS: Political Science and Politics, V ol. 24, No. 4 (December 1991); and Mueller , “ Is W ar Still Becoming Obsolete?” p. 5. 15. Quoted in W alter A. McDougall, Promised Land, Crusader State (Boston: Houghton Mif in, 1997), p. 28 and n. 36. International Security 25:1 10 policies bound to frighten both Britain and Russia. And today if a war that a few have feared were fought by the United States and Japan, many Americans would say that Japan was not a democracy after all, but merely a one-party state. What can we conclude? Democracies rarely Ž ght democracies, we might say , and then add as a word of essential caution that the internal excellence of states is a brittle basis of peace. democratic wars Democracies coexist with undemocratic states. Although democracies seldom Ž ght democracies, they do, as Michael Doyle has noted, Ž ght at least their share of wars against others. 16 Citizens of democratic states tend to think of their countries as good, aside from what they do, simply because they are demo- cratic. Thus former Secretary of State W arren Christopher claimed that ” demo- cratic nations rarely start wars or threaten their neighbors.“ 17 One might suggest that he try his proposition out in Central or South America. Citizens of democratic states also tend to think of undemocratic states as bad, aside from what they do, simply because they are undemocratic. Democracies pro- mote war because they at times decide that the way to preserve peace is to defeat nondemocratic states and make them democratic. During W orld W ar I, W alter Hines Page, American ambassador to England, claimed that there ” is no security in any part of the world where people cannot think of a government without a king and never will be.“ During the V ietnam W ar , Secretary of State Dean Rusk claimed that the ” United States cannot be secure until the total international environment is ideologically safe.“ 18 Policies aside, the very existence of undemocratic states is a danger to others. American political and intellectual leaders have often taken this view . Liberal interven- tionism is again on the march. President Bill Clinton and his national security adviser , Anthony Lake, urged the United States to take measures to enhance democracy around the world. The task, one fears, will be taken up by the American military with some enthusiasm. Former Army Chief of Staff General Gordon Sullivan, for example, favored a new military ” model,“ replacing the negative aim of containment with a positive one: ” T o promote democracy , 16. Doyle, “ Kant, Liberal Legacies, and Foreign Affairs, Part 2,” p. 337. 17. W arren Christopher , “ The U.S.-Japan Relationship: The Responsibility to Change,” address to the Japan Association of Corporate Executives, T okyo, Japan, March 11, 1994 (U.S. Department of State, Bureau of Public Affairs, OfŽ ce of Public Communication), p. 3. 18. Page quoted in W altz, Man, the State, and War: A Theoretical Analysis (New Y ork: Columbia University Press, 1959), p. 121. Rusk quoted in Layne, “ Kant or Cant,” p. 46. Structural Realism after the Cold War 11 regional stability , and economic prosperity .“ 19 Other voices urge us to enter into a ” struggle to ensure that people are governed well.“ Having apparently solved the problem of justice at home, ” the struggle for liberal government becomes a struggle not simply for justice but for survival.“ 20 As R.H. T awney said: ” Either war is a crusade, or it is a crime.“ 21 Crusades are frightening because crusaders go to war for righteous causes, which they deŽ ne for themselves and try to impose on others. One might have hoped that Americans would have learned that they are not very good at causing democracy abroad. But, alas, if the world can be made safe for democracy only by making it democratic, then all means are permitted and to use them becomes a duty . The war fervor of people and their representatives is at times hard to contain. Thus Hans Morgenthau believed that ” the democratic selection and responsibility of government ofŽ cials destroyed international morality as an effective system of restraint.“ 22 Since, as Kant believed, war among self-directed states will occasionally break out, peace has to be contrived. For any government, doing so is a difŽ cult task, and all states are at times deŽ cient in accomplishing it, even if they wish to. Democratic leaders may respond to the fervor for war that their citizens sometimes display , or even try to arouse it, and governments are sometimes constrained by electoral calculations to defer preventive measures. Thus British Prime Minister Stanley Baldwin said that if he had called in 1935 for British rearmament against the German threat, his party would have lost the next election. 23 Democratic governments may respond to internal political impera- tives when they should be responding to external ones. All governments have their faults, democracies no doubt fewer than others, but that is not good enough to sustain the democratic peace thesis. That peace may prevail among democratic states is a comforting thought. The obverse of the proposition— that democracy may promote war against undemocratic states— is disturbing. If the latter holds, we cannot even say for sure that the spread of democracy will bring a net decrease in the amount of war in the world. 19. Quoted in Clemson G. T urregano and Ricky Lynn W addell, “ From Paradigm to Paradigm Shift: The Military and Operations Other than W ar ,” Journal of Political Science, V ol. 22 (1994), p. 15. 20. Peter Beinart, “ The Return of the Bomb,” New Republic, August 3, 1998, p. 27. 21. Quoted in Michael Straight, Make This the Last War (New Y ork: G.P . Putnam’ s Sons, 1945), p. 1. 22. Hans J. Morgenthau, Politics among Nations: The Struggle for Power and Peace, 5th ed. (New Y ork: Knopf, 1973), p. 248. 23. Gordon Craig and Alexander George, Force and Statecraft: Diplomatic Problems of Our Time, 2d ed. (New Y ork: Oxford University Press, 1990), p. 64. International Security 25:1 12 With a republic established in a strong state, Kant hoped the republican form would gradually take hold in the world. In 1795, America provided the hope. T wo hundred years later , remarkably , it still does. Ever since liberals Ž rst expressed their views, they have been divided. Some have urged liberal states to work to uplift benighted peoples and bring the beneŽ ts of liberty , justice, and prosperity to them. John Stuart Mill, Giuseppe Mazzini, W oodrow Wilson, and Bill Clinton are all interventionist liberals. Other liberals, Kant and Richard Cobden, for example, while agreeing on the beneŽ ts that democracy can bring to the world, have emphasized the difŽ culties and the dangers of actively seeking its propagation. If the world is now safe for democracy , one has to wonder whether democ- racy is safe for the world. When democracy is ascendant, a condition that in the twentieth century attended the winning of hot wars and cold ones, the interventionist spirit ourishes. The effect is heightened when one democratic state becomes dominant, as the United States is now . Peace is the noblest cause of war . If the conditions of peace are lacking, then the country with a capability of creating them may be tempted to do so, whether or not by force. The end is noble, but as a matter of right, Kant insists, no state can intervene in the internal arrangements of another . As a matter of fact, one may notice that intervention, even for worthy ends, often brings more harm than good. The vice to which great powers easily succumb in a multipolar world is inattention; in a bipolar world, overreaction; in a unipolar world, overextention. Peace is maintained by a delicate balance of internal and external restraints. States having a surplus of power are tempted to use it, and weaker states fear their doing so. The laws of voluntary federations, to use Kant’ s language, are disregarded at the whim of the stronger , as the United States demonstrated a decade ago by mining Nicaraguan waters and by invading Panama. In both cases, the United States blatantly violated international law . In the Ž rst, it denied the jurisdiction of the International Court of Justice, which it had previously accepted. In the second, it aunted the law embodied in the Charter of the Organization of American States, of which it was a principal sponsor . If the democratic peace thesis is right, structural realist theory is wrong. One may believe, with Kant, that republics are by and large good states and that unbalanced power is a danger no matter who wields it. Inside of, as well as outside of, the circle of democratic states, peace depends on a precarious balance of forces. The causes of war lie not simply in states or in the state system; they are found in both. Kant understood this. Devotees of the demo- cratic peace thesis overlook it. Structural Realism after the Cold War 13 The Weak Effects of Interdependence If not democracy alone, may not the spread of democracy combined with the tightening of national interdependence fulŽ ll the prescription for peace offered by nineteenth-century liberals and so often repeated today? 24 T o the suppos- edly peaceful inclination of democracies, interdependence adds the propulsive power of the proŽ t motive. Democratic states may increasingly devote them- selves to the pursuit of peace and proŽ ts. The trading state is replacing the political-military state, and the power of the market now rivals or surpasses the power of the state, or so some believe. 25 Before W orld W ar I, Norman Angell believed that wars would not be fought because they would not pay , yet Germany and Britain, each other ’ s second- best customers, fought a long and bloody war .

26 Interdependence in some ways promotes peace by multiplying contacts among states and contributing to mutual understanding. It also multiplies the occasions for con icts that may promote resentment and even war .

27 Close interdependence is a condition in which one party can scarcely move without jostling others; a small push ripples through society . The closer the social bonds, the more extreme the effect becomes, and one cannot sensibly pursue an interest without taking others’ interests into account. One country is then inclined to treat another country’ s acts as events within its own polity and to attempt to control them. That interdependence promotes war as well as peace has been said often enough. What requires emphasis is that, either way , among the forces that shape international politics, interdependence is a weak one. Interdependence within modern states is much closer than it is across states. The Soviet economy was planned so that its far- ung parts would be not just interdependent but integrated. Huge factories depended for their output on products exchanged 24. Strongly afŽ rmative answers are given by John R. Oneal and Bruce Russett, “ Assessing the Liberal Peace with Alternative SpeciŽ cations: T rade Still Reduces Con ict,” Journal of Peace Research, V ol. 36, No. 4 (July 1999), pp. 423– 442; and Russett, Oneal, and David R. Davis, “ The Third Leg of the Kantian T ripod for Peace: International Organizations and Militarized Disputes, 1950– 85,” International Organization, V ol. 52, No. 3 (Summer 1998), pp. 441– 467. 25. Richard Rosecrance, The Rise of the T rading State: Commerce and Coalitions in the Modern World (New Y ork: Basic Books, 1986); and at times Susan Strange, The Retreat of the State: The Diffusion of Power in the World Economy (New Y ork: Cambridge University Press, 1996). 26. Norman Angell, The Great Illusion, 4th rev . and enlarged ed. (New Y ork: Putnam’ s, 1913). 27. Katherine Barbieri, “ Economic Interdependence: A Path to Peace or a Source of Interstate Con ict?” Journal of Peace Research, V ol. 33, No. 1 (February 1996). Lawrence Keely , War before Civilization: The Myth of the Peaceful Savage (New Y ork: Oxford University Press, 1996), p. 196, shows that with increases of trade and intermarriage among tribes, war became more frequent. International Security 25:1 14 with others. Despite the tight integration of the Soviet economy , the state fell apart. Y ugoslavia provides another stark illustration. Once external political pressure lessened, internal economic interests were too weak to hold the country together . One must wonder whether economic interdependence is more effect than cause. Internally , interdependence becomes so close that integration is the proper word to describe it. Interdependence becomes inte- gration because internally the expectation that peace will prevail and order will be preserved is high. Externally , goods and capital ow freely where peace among countries appears to be reliably established. Interdependence, like integration, depends on other conditions. It is more a dependent than an independent variable. States, if they can afford to, shy away from becoming excessively dependent on goods and resources that may be denied them in crises and wars. States take measures, such as Japan’ s managed trade, to avoid excessive dependence on others. 28 The impulse to protect one’ s identity— cultural and political as well as economic— from encroachment by others is strong. When it seems that ” we will sink or swim together ,“ swimming separately looks attractive to those able to do it. From Plato onward, utopias were set in isolation from neighbors so that people could construct their collective life uncontaminated by contact with others. With zero interdependence, neither con ict nor war is possible. With integration, international becomes national politics. 29 The zone in between is a gray one with the effects of interdependence sometimes good, providing the beneŽ ts of divided labor , mutual understanding, and cultural enrichment, and sometimes bad, leading to protectionism, mutual resentment, con ict, and war . The uneven effects of interdependence, with some parties to it gaining more, others gaining less, are obscured by the substitution of Robert Keohane’ s and Joseph Nye’ s term ” asymmetric interdependence“ for relations of dependence and independence among states. 30 Relatively independent states are in a stronger position than relatively dependent ones. If I depend more on you than you depend on me, you have more ways of in uencing me and affecting my 28. On states managing interdependence to avoid excessive dependence, see especially Robert Gilpin, The Political Economy of International Relations (Princeton, N.J.: Princeton University Press, 1987), chap. 10; and Suzanne Berger and Ronald Dore, eds., National Diversity and Global Capitalism (Ithaca, N.Y .: Cornell University Press, 1996). 29. Cf. Kenneth N. W altz, in Steven L. Spiegel and W altz, eds., Con ict in World Politics (Cambridge, Mass.: Winthrop, 1971), chap. 13. 30. Robert O. Keohane and Joseph S. Nye, Power and Interdependence, 2d ed. (New Y ork: Harper- Collins, 1989). Structural Realism after the Cold War 15 fate than I have of affecting yours. Interdependence suggests a condition of roughly equal dependence of parties on one another . Omitting the word “ dependence” blunts the inequalities that mark the relations of states and makes them all seem to be on the same footing. Much of international, as of national, politics is about inequalities. Separating one “ issue area” from others and emphasizing that weak states have advantages in some of them reduces the sense of inequality . Emphasizing the low fungibility of power furthers the effect. If power is not very fungible, weak states may have decisive advantages on some issues. Again, the effects of inequality are blunted. But power , not very fungible for weak states, is very fungible for strong ones. The history of American foreign policy since W orld W ar II is replete with examples of how the United States used its superior economic capability to promote its political and security interests. 31 In a 1970 essay , I described interdependence as an ideology used by Ameri- cans to camou age the great leverage the United States enjoys in international politics by making it seem that strong and weak, rich and poor nations are similarly entangled in a thick web of interdependence. 32 In her recent book, The Retreat of the State, Susan Strange reached the same conclusion, but by an odd route. Her argument is that ” the progressive integration of the world economy , through international production, has shifted the balance of power away from states and toward world markets.“ She advances three propositions in support of her argument: (1) power has ” shifted upward from weak states to stronger ones“ having global or regional reach; (2) power has ” shifted sideways from states to markets and thus to non-state authorities deriving power from their market shares“ ; and (3) some power has ” evaporated“ with no one exercising it. 33 In international politics, with no central authority , power does sometimes slip away and sometimes move sideways to markets. When serious slippage occurs, however , stronger states step in to reverse it, and Ž rms of the stronger states control the largest market shares anyway . One may doubt whether markets any more escape the control of major states now than they 31. Keohane and Nye are on both sides of the issue. See, for example, ibid. , p. 28. Keohane emphasized that power is not very fungible in Keohane, ed., “ Theory of W orld Politics,” Neorealism and Its Critics (New Y ork: Columbia University Press, 1986); and see Kenneth N. W altz, “ Re ection on Theory of International Politics: A Response to My Critics,” in ibid. Robert J. Art analyzes the fungibility of power in detail. See Art, “ American Foreign Policy and the Fungibility of Force,” Security Studies, V ol. 5, No. 4 (Summer 1996). 32. Kenneth N. W altz, “ The Myth of National Interdependence,” in Charles P . Kindleberger , ed., The International Corporation (Cambridge, Mass.: MIT Press, 1970). 33. Strange, Retreat of the State, pp. 46, 189. International Security 25:1 16 did in the nineteenth century or earlier— perhaps less so since the competence of states has increased at least in proportion to increases in the size and complications of markets. Anyone, realist or not, might think Strange’ s Ž rst proposition is the important one. Never since the Roman Empire has power been so concentrated in one state. Despite believing that power has moved from states to markets, Strange recognized reality . She observed near the beginning of her book that the ” authority— the ‘ power over ’ global outcomes enjoyed by American society , and therefore indirectly by the United States government— is still superior to that of any other society or any other govern- ment.“ And near the end, she remarked that the ” authority of governments tends to over-rule the caution of markets.“ If one wondered which government she had in mind, she answered immediately: ” The fate of Mexico is decided in W ashington more than W all Street. And the International Monetary Fund (IMF) is obliged to follow the American lead, despite the misgivings of Ger- many or Japan.“ 34 The history of the past two centuries has been one of central governments acquiring more and more power . Alexis de T ocqueville observed during his visit to the United States in 1831 that ” the Federal Government scarcely ever interferes in any but foreign affairs; and the governments of the states in reality direct society in America.“ 35 After W orld W ar II, governments in W estern Europe disposed of about a quarter of their peoples’ income. The proportion now is more than half. At a time when Americans, Britons, Russians, and Chinese were decrying the control of the state over their lives, it was puzzling to be told that states were losing control over their external affairs. Losing control, one wonders, as compared to when? W eak states have lost some of their in uence and control over external matters, but strong states have not lost theirs. The patterns are hardly new ones. In the eighteenth and nineteenth centuries, the strongest state with the longest reach intervened all over the globe and built history’ s most extensive empire. In the twentieth century , the strongest state with the longest reach repeated Britain’ s interventionist behav- ior and, since the end of the Cold W ar , on an ever widening scale, without building an empire. The absence of empire hardly means, however , that the extent of America’ s in uence and control over the actions of others is of lesser moment. The withering away of the power of the state, whether inter- 34. Ibid., pp. 25, 192. 35. Alexis de T ocqueville, Democracy in America, ed. J.P . Mayer , trans. George Lawrence (New Y ork: Harper Perennial, 1988), p. 446, n. 1. Structural Realism after the Cold War 17 nally or externally , is more of a wish and an illusion than a reality in most of the world. Under the Pax Britannica, the interdependence of states became unusually close, which to many portended a peaceful and prosperous future. Instead, a prolonged period of war , autarky , and more war followed. The international economic system, constructed under American auspices after W orld W ar II and later amended to suit its purposes, may last longer , but then again it may not. The character of international politics changes as national interdependence tightens or loosens. Y et even as relations vary , states have to take care of themselves as best they can in an anarchic environment. Internationally , the twentieth century for the most part was an unhappy one. In its last quarter , the clouds lifted a little, but twenty-Ž ve years is a slight base on which to ground optimistic conclusions. Not only are the effects of close interdepend- ence problematic, but so also is its durability . The Limited Role of International Institutions One of the charges hurled at realist theory is that it depreciates the importance of institutions. The charge is justiŽ ed, and the strange case of NA TO’ s (the North Atlantic T reaty Organization’ s) outliving its purpose shows why realists believe that international institutions are shaped and limited by the states that found and sustain them and have little independent effect. Liberal institution- alists paid scant attention to organizations designed to buttress the security of states until, contrary to expectations inferred from realist theories, NA TO not only survived the end of the Cold W ar but went on to add new members and to promise to embrace still more. Far from invalidating realist theory or casting doubt on it, however , the recent history of NA TO illustrates the subordination of international institutions to national purposes. explaining international institutions The nature and purposes of institutions change as structures vary . In the old multipolar world, the core of an alliance consisted of a small number of states of comparable capability . Their contributions to one another ’ s security were of crucial importance because they were of similar size. Because major allies were closely interdependent militarily , the defection of one would have made its partners vulnerable to a competing alliance. The members of opposing alli- ances before W orld W ar I were tightly knit because of their mutual dependence. In the new bipolar world, the word ” alliance“ took on a different meaning. One country , the United States or the Soviet Union, provided most of the International Security 25:1 18 security for its bloc. The withdrawal of France from NA TO’ s command struc- ture and the defection of China from the Soviet bloc failed even to tilt the central balance. Early in the Cold W ar , Americans spoke with alarm about the threat of monolithic communism arising from the combined strength of the Soviet Union and China, yet the bloc’ s disintegration caused scarcely a ripple. American ofŽ cials did not proclaim that with China’ s defection, America’ s defense budget could safely be reduced by 20 or 10 percent or even be reduced at all. Similarly , when France stopped playing its part in NA TO’ s military plans, American ofŽ cials did not proclaim that defense spending had to be increased for that reason. Properly speaking, NA TO and the WTO (W arsaw T reaty Organization) were treaties of guarantee rather than old-style military alliances. 36 Glenn Snyder has remarked that “ alliances have no meaning apart from the adversary threat to which they are a response.” 37 I expected NA TO to dwindle at the Cold W ar ’ s end and ultimately to disappear .

38 In a basic sense, the expectation has been borne out. NA TO is no longer even a treaty of guarantee because one cannot answer the question, guarantee against whom? Functions vary as structures change, as does the behavior of units. Thus the end of the Cold W ar quickly changed the behavior of allied countries. In early July of 1990, NA TO announced that the alliance would “ elaborate new force plans consistent with the revolutionary changes in Europe.” 39 By the end of July , without waiting for any such plans, the major European members of NA TO unilaterally announced large reductions in their force levels. Even the pretense of continuing to act as an alliance in setting military policy disappeared. With its old purpose dead, and the individual and collective behavior of its members altered accordingly , how does one explain NA TO’ s survival and expansion? Institutions are hard to create and set in motion, but once created, institutionalists claim, they may take on something of a life of their own; they may begin to act with a measure of autonomy , becoming less dependent on the wills of their sponsors and members. NA TO supposedly validates these thoughts. Organizations, especially big ones with strong traditions, have long lives. The March of Dimes is an example sometimes cited. Having won the war 36. See Kenneth N. W altz, “ International Structure, National Force, and the Balance of W orld Power ,” Journal of International Affairs, V ol. 21, No. 2 (1967), p. 219. 37. Glenn H. Snyder , Alliance Politics (Ithaca, N.Y .: Cornell University Press, 1997), p. 192. 38. Kenneth N. W altz, “ The Emerging Structure of International Politics,” International Security, V ol. 18, No. 2 (Fall 1993), pp. 75– 76. 39. John Roper , “ Shaping Strategy without the Threat,” Adephi Paper No. 257 (London: Interna- tional Institute for Strategic Studies, Winter 1990/ 91), pp. 80– 81. Structural Realism after the Cold War 19 against polio, its mission was accomplished. Nevertheless, it cast about for a new malady to cure or contain. Even though the most appealing ones— cancer , diseases of the heart and lungs, multiple sclerosis, and cystic Ž brosis— were already taken, it did Ž nd a worthy cause to pursue, the amelioration of birth defects. One can fairly claim that the March of Dimes enjoys continuity as an organization, pursuing an end consonant with its original purpose. How can one make such a claim for NA TO? The question of purpose may not be a very important one; create an organi- zation and it will Ž nd something to do. 40 Once created, and the more so once it has become well established, an organization becomes hard to get rid of. A big organization is managed by large numbers of bureaucrats who develop a strong interest in its perpetuation. According to Gunther Hellmann and Rein- hard W olf, in 1993 NA TO headquarters was manned by 2,640 ofŽ cials, most of whom presumably wanted to keep their jobs. 41 The durability of NA TO even as the structure of international politics has changed, and the old purpose of the organization has disappeared, is interpreted by institutionalists as evidence strongly arguing for the autonomy and vitality of institutions. The institutionalist interpretation misses the point. NA TO is Ž rst of all a treaty made by states. A deeply entrenched international bureaucracy can help to sustain the organization, but states determine its fate. Liberal institutional- ists take NA TO’ s seeming vigor as conŽ rmation of the importance of interna- tional institutions and as evidence of their resilience. Realists, noticing that as an alliance NA TO has lost its major function, see it mainly as a means of maintaining and lengthening America’ s grip on the foreign and military poli- cies of European states. John Kornblum, U.S. senior deputy to the undersecre- tary of state for European affairs, neatly described NA TO’ s new role. “ The Alliance,” he wrote, “ provides a vehicle for the application of American power and vision to the security order in Europe.” 42 The survival and expansion of NA TO tell us much about American power and in uence and little about institutions as multilateral entities. The ability of the United States to extend the life of a moribund institution nicely illustrates how international institu- tions are created and maintained by stronger states to serve their perceived or misperceived interests. 40. Joseph A. Schumpeter , writing of armies, put it this way: “ created by wars that required it, the machine now created the wars it required.” “ The Sociology of Imperialism,” in Schumpeter , Imperialism and Social Classes (New Y ork: Meridian Books, 1955), p. 25 (emphasis in original). 41. Gunther Hellmann and Reinhard W olf, “ Neorealism, Neoliberal Institutionalism, and the Future of NA TO,” Security Studies, V ol. 3, No. 1 (Autumn 1993), p. 20. 42. John Kornblum, “ NA TO’ s Second Half Century— T asks for an Alliance,” NATO on T rack for the 21st Century, Conference Report (The Hague: Netherlands Atlantic Commission, 1994), p. 14. International Security 25:1 20 The Bush administration saw , and the Clinton administration continued to see, NA TO as the instrument for maintaining America’ s domination of the foreign and military policies of European states. In 1991, U.S. Undersecretary of State Reginald Bartholomew’ s letter to the governments of European mem- bers of NA TO warned against Europe’ s formulating independent positions on defense. France and Germany had thought that a European security and defense identity might be developed within the EU and that the W estern European Union, formed in 1954, could be revived as the instrument for its realization. The Bush administration quickly squelched these ideas. The day after the signing of the Maastricht T reaty in December of 1991, President George Bush could say with satisfaction that “ we are pleased that our Allies in the W estern European Union . . . decided to strengthen that institution as both NA TO’ s European pillar and the defense component of the European Union.” 43 The European pillar was to be contained within NA TO, and its policies were to be made in W ashington. W eaker states have trouble fashioning institutions to serve their own ends in their own ways, especially in the security realm. Think of the defeat of the European Defense Community in 1954, despite America’ s support of it, and the inability of the W estern European Union in the more than four decades of its existence to Ž nd a signiŽ cant role inde- pendent of the United States. Realism reveals what liberal institutionalist “ theory” obscures: namely , that international institutions serve primarily na- tional rather than international interests. 44 Robert Keohane and Lisa Martin, replying to John Mearsheimer ’ s criticism of liberal institutionalism, ask: How are we “ to account for the willingness of major states to invest resources in expanding international institutions if such institutions are lacking in sig- niŽ cance?” 45 If the answer were not already obvious, the expansion of NA TO would make it so: to serve what powerful states believe to be their interests. With the administration’ s Bosnian policy in trouble, Clinton needed to show himself an effective foreign policy leader . With the national heroes Lech W alesa and V aclav Havel clamoring for their countries’ inclusion, foreclosing NA TO membership would have handed another issue to the Republican Party in the 43. Mark S. Sheetz, “ Correspondence: Debating the Unipolar Moment,” International Security, V ol. 22, No. 3 (Winter 1997/ 98), p. 170; and Mike Winnerstig, “ Rethinking Alliance Dynamics,” paper presented at the annual meeting of the International Studies Association, W ashington, D.C., March 18– 22, 1997, at p. 23. 44. Cf. Alan S. Milward, The European Rescue of the Nation-State (Berkeley: University of California Press, 1992). 45. Robert O. Keohane and Lisa L. Martin, “ The Promise of Institutionalist Theory ,” International Security, V ol. 20, No. 1 (Summer 1995), p. 40. Structural Realism after the Cold War 21 congressional elections of 1994. T o tout NA TO’ s eastward march, President Clinton gave major speeches in Milwaukee, Cleveland, and Detroit, cities with signiŽ cant numbers of East European voters. 46 V otes and dollars are the life- blood of American politics. New members of NA TO will be required to im- prove their military infrastructure and to buy modern weapons. The American arms industry , expecting to capture its usual large share of a new market, has lobbied heavily in favor of NA TO’ s expansion. 47 The reasons for expanding NA TO are weak. The reasons for opposing expansion are strong. 48 It draws new lines of division in Europe, alienates those left out, and can Ž nd no logical stopping place west of Russia. It weakens those Russians most inclined toward liberal democracy and a market economy . It strengthens Russians of the opposite inclination. It reduces hope for further large reductions of nuclear weaponry . It pushes Russia toward China instead of drawing Russia toward Europe and America. NA TO, led by America, scarcely considered the plight of its defeated adversary . Throughout modern history , Russia has been rebuffed by the W est, isolated and at times sur- rounded. Many Russians believe that, by expanding, NA TO brazenly broke promises it made in 1990 and 1991 that former WTO members would not be allowed to join NA TO. With good reason, Russians fear that NA TO will not only admit additional old members of the WTO but also former republics of the Soviet Union. In 1997, NA TO held naval exercises with Ukraine in the Black Sea, with more joint exercises to come, and announced plans to use a military testing ground in western Ukraine. In June of 1998, Zbigniew Brzezinski went to Kiev with the message that Ukraine should prepare itself to join NA TO by the year 2010. 49 The farther NA TO intrudes into the Soviet Union’ s old arena, the more Russia is forced to look to the east rather than to the west. The expansion of NA TO extends its military interests, enlarges its responsi- bilities, and increases its burdens. Not only do new members require NA TO’ s protection, they also heighten its concern over destabilizing events near their 46. James M. Goldgeier , “ NA TO Expansion: The Anatomy of a Decision,” Washington Quarterly, V ol. 21, No. 1 (Winter 1998), pp. 94– 95. And see his Not Whether but When: The U.S. Decision to Enlarge NATO (W ashington, D.C.: Brookings, 1999). 47. William D. Hartung, “ W elfare for W eapons Dealers 1998: The Hidden Costs of NA TO Expan- sion” (New Y ork: New School for Social Research, W orld Policy Institute, March 1998); and Jeff Gerth and Tim W einer , “ Arms Makers See Bonanza in Selling NA TO Expansion,” New Y ork Times, June 29, 1997, p. I, 8. 48. See Michael E. Brown, “ The Flawed Logic of Expansion,” Survival, V ol. 37, No. 1 (Spring 1995), pp. 34– 52. Michael Mandelbaum, The Dawn of Peace in Europe (New Y ork: T wentieth Century Fund Press, 1996). Philip Zelikow , “ The Masque of Institutions,” Survival, V ol. 38, No. 1 (Spring 1996). 49. J.L. Black, Russia Faces NATO Expansion: Bearing Gifts or Bearing Arms? (Lanham, Md.: Rowman and LittleŽ eld, 2000), pp. 5– 35, 175– 201. International Security 25:1 22 borders. Thus Balkan eruptions become a NA TO and not just a European concern. In the absence of European initiative, Americans believe they must lead the way because the credibility of NA TO is at stake. Balkan operations in the air and even more so on the ground exacerbate differences of interest among NA TO members and strain the alliance. European members marvel at the surveillance and communications capabilities of the United States and stand in awe of the modern military forces at its command. A ware of their weaknesses, Europeans express determination to modernize their forces and to develop their ability to deploy them independently . Europe’ s reaction to America’ s Balkan operations duplicates its determination to remedy deŽ cien- cies revealed in 1991 during the Gulf W ar , a determination that produced few results. Will it be different this time? Perhaps, yet if European states do achieve their goals of creating a 60,000 strong rapid reaction force and enlarging the role of the WEU, the tension between a NA TO controlled by the United States and a NA TO allowing for independent European action will again be bothersome. In any event, the prospect of militarily bogging down in the Balkans tests the alliance and may indeŽ nitely delay its further expansion. Expansion buys trouble, and mounting troubles may bring expansion to a halt. European conditions and Russian opposition work against the eastward extension of NA TO. Pressing in the opposite direction is the momentum of American expansion. The momentum of expansion has often been hard to break, a thought borne out by the empires of Republican Rome, of Czarist Russia, and of Liberal Britain. One is often reminded that the United States is not just the dominant power in the world but that it is a liberal dominant power . T rue, the motivations of the artiŽ cers of expansion— President Clinton, National Security Adviser Anthony Lake, and others— were to nurture democracy in young, fragile, long-suffering countries. One may wonder , however , why this should be an American rather than a European task and why a military rather than a political-economic organization should be seen as the appropriate means for carrying it out. The task of building democracy is not a military one. The military security of new NA TO members is not in jeopardy; their political development and economic well-being are. In 1997, U.S. Assistant Secretary of Defense Franklin D. Kramer told the Czech defense ministry that it was spending too little on defense. 50 Y et investing in defense slows economic growth. By common calculation, defense spending stimulates economic growth 50. Ibid., p. 72. Structural Realism after the Cold War 23 about half as much as direct investment in the economy . In Eastern Europe, economic not military security is the problem and entering a military alliance compounds it. Using the example of NA TO to re ect on the relevance of realism after the Cold W ar leads to some important conclusions. The winner of the Cold W ar and the sole remaining great power has behaved as unchecked powers have usually done. In the absence of counterweights, a country’ s internal impulses prevail, whether fueled by liberal or by other urges. The error of realist predictions that the end of the Cold W ar would mean the end of NA TO arose not from a failure of realist theory to comprehend international politics, but from an underestimation of America’ s folly . The survival and expansion of NA TO illustrate not the defects but the limitations of structural explanations. Structures shape and shove; they do not determine the actions of states. A state that is stronger than any other can decide for itself whether to conform its policies to structural pressures and whether to avail itself of the opportunities that structural change offers, with little fear of adverse affects in the short run. Do liberal institutionalists provide better leverage for explaining NA TO’ s survival and expansion? According to Keohane and Martin, realists insist “ that institutions have only marginal effects.” 51 On the contrary , realists have noticed that whether institutions have strong or weak effects depends on what states intend. Strong states use institutions, as they interpret laws, in ways that suit them. Thus Susan Strange, in pondering the state’ s retreat, observes that “ international organization is above all a tool of national government, an instrument for the pursuit of national interest by other means.” 52 Interestingly , Keohane and Martin, in their effort to refute Mearsheimer ’ s trenchant criticism of institutional theory , in effect agree with him. Having claimed that his realism is “ not well speciŽ ed,” they note that “ institutional theory conceptualizes institutions both as independent and dependent vari- ables.” 53 Dependent on what?— on “ the realities of power and interest.” Insti- tutions, it turns out, “ make a signiŽ cant difference in conjunction with power realities.” 54 Y es! Liberal institutionalism, as Mearsheimer says, “ is no longer a clear alternative to realism, but has, in fact, been swallowed up by it.” 55 Indeed, it never was an alternative to realism. Institutionalist theory , as Keohane has 51. Keohane and Martin, “ The Promise of Institutionalist Theory ,” pp. 42, 46. 52. Strange, Retreat of the State, p. xiv; and see pp. 192– 193. Cf. Carr , The T wenty Y ears’ Crisis, p. 107: “ international government is, in effect, government by that state which supplies the power necessary for the purpose of governing.” 53. Keohane and Martin, “ The Promise of Institutionalist Theory ,” p. 46. 54. Ibid., p. 42. 55. Mearsheimer, “ A Realist Reply ,” p. 85. International Security 25:1 24 stressed, has as its core structural realism, which Keohane and Nye sought “ to broaden.” 56 The institutional approach starts with structural theory , applies it to the origins and operations of institutions, and unsurprisingly ends with realist conclusions. Alliances illustrate the weaknesses of institutionalism with special clarity . Institutional theory attributes to institutions causal effects that mostly originate within states. The case of NA TO nicely illustrates this shortcoming. Keohane has remarked that “ alliances are institutions, and both their durability and strength . . . may depend in part on their institutional characteristics.” 57 In part, I suppose, but one must wonder in how large a part. The T riple Alliance and the T riple Entente were quite durable. They lasted not because of alliance institutions, there hardly being any , but because the core members of each alliance looked outward and saw a pressing threat to their security . Previous alliances did not lack institutions because states had failed to Ž gure out how to construct bureaucracies. Previous alliances lacked institutions because in the absence of a hegemonic leader , balancing continued within as well as across alliances. NA TO lasted as a military alliance as long as the Soviet Union appeared to be a direct threat to its members. It survives and expands now not because of its institutions but mainly because the United States wants it to.

NA TO’ s survival also exposes an interesting aspect of balance-of-power theory . Robert Art has argued forcefully that without NA TO and without American troops in Europe, European states will lapse into a “ security com- petition” among themselves. 58 As he emphasizes, this is a realist expectation. In his view , preserving NA TO, and maintaining America’ s leading role in it, are required in order to prevent a security competition that would promote con ict within, and impair the institutions of, the European Union. NA TO now is an anomaly; the dampening of intra-alliance tension is the main task left, and it is a task not for the alliance but for its leader . The secondary task of an alliance, intra-alliance management, continues to be performed by the United States even though the primary task, defense against an external enemy , has disappeared. The point is worth pondering, but I need to say here only that it 56. Keohane and Nye, Power and Interdependence, p. 251; cf. Keohane, “ Theory of W orld Politics,” in Keohane, Neorealism and Its Critics, p. 193, where he describes his approach as a “ modiŽ ed structural research program.” 57. Robert O. Keohane, International Institutions and State Power: Essays in International Relations Theory (Boulder , Colo.: W estview , 1989), p. 15. 58. Robert J. Art, “ Why W estern Europe Needs the United States and NA TO,” Political Science Quarterly, V ol. 111, No. 1 (Spring 1996). Structural Realism after the Cold War 25 further illustrates the dependence of international institutions on national decisions. Balancing among states is not inevitable. As in Europe, a hegemonic power may suppress it. As a high-level European diplomat put it, “ it is not acceptable that the lead nation be European. A European power broker is a hegemonic power . W e can agree on U.S. leadership, but not on one of our own.” 59 Accepting the leadership of a hegemonic power prevents a balance of power from emerging in Europe, and better the hegemonic power should be at a distance than next door . Keohane believes that “ avoiding military con ict in Europe after the Cold W ar depends greatly on whether the next decade is characterized by a con- tinuous pattern of institutionalized cooperation.” 60 If one accepts the conclu- sion, the question remains: What or who sustains the ” pattern of institutionalized cooperation“ ? Realists know the answer . international institutions and national aims What is true of NA TO holds for international institutions generally . The effects that international institutions may have on national decisions are but one step removed from the capabilities and intentions of the major state or states that gave them birth and sustain them. The Bretton W oods system strongly affected individual states and the conduct of international affairs. But when the United States found that the system no longer served its interests, the Nixon shocks of 1971 were administered. International institutions are created by the more powerful states, and the institutions survive in their original form as long as they serve the major interests of their creators, or are thought to do so. ” The nature of institutional arrangements,“ as Stephen Krasner put it, ” is better explained by the distribution of national power capabilities than by efforts to solve problems of market failure“ 61 — or , I would add, by anything else. Either international conventions, treaties, and institutions remain close to the underlying distribution of national capabilities or they court failure. 62 Citing examples from the past 350 years, Krasner found that in all of the instances ” it was the value of strong states that dictated rules that were applied in a 59. Quoted in ibid., p. 36. 60. Robert O. Keohane, “ The Diplomacy of Structural Change: Multilateral Institutions and State Strategies,” in Helga Haftendorn and Christian T uschhoff, eds., America and Europe in an Era of Change (Boulder , Colo.: W estview , 1993), p. 53. 61. Stephen D. Krasner , “ Global Communications and National Power: Life on the Pareto Frontier ,” World Politics, V ol. 43, No. 1 (April 1991), p. 234. 62. Stephen D. Krasner , Structural Con ict: The Third World against Global Liberalism (Berkeley: University of California, 1985), p. 263 and passim. International Security 25:1 26 discriminating fashion only to the weak.“ 63 The sovereignty of nations, a universally recognized international institution, hardly stands in the way of a strong nation that decides to intervene in a weak one. Thus, according to a senior ofŽ cial, the Reagan administration ” debated whether we had the right to dictate the form of another country’ s government. The bottom line was yes, that some rights are more fundamental than the right of nations to noninter- vention. . . . W e don’ t have the right to subvert a democracy but we do have the right against an undemocratic one.“ 64 Most international law is obeyed most of the time, but strong states bend or break laws when they choose to. Balancing Power: Not T oday but T omorrow With so many of the expectations that realist theory gives rise to conŽ rmed by what happened at and after the end of the Cold W ar , one may wonder why realism is in bad repute. 65 A key proposition derived from realist theory is that international politics re ects the distribution of national capabilities, a propo- sition daily borne out. Another key proposition is that the balancing of power by some states against others recurs. Realist theory predicts that balances disrupted will one day be restored. A limitation of the theory , a limitation common to social science theories, is that it cannot say when. William W ohlforth argues that though restoration will take place, it will be a long time coming. 66 Of necessity , realist theory is better at saying what will happen than in saying when it will happen. Theory cannot say when “ tomorrow” will come because international political theory deals with the pressures of structure on states and not with how states will respond to the pressures. The latter is a task for theories about how national governments respond to pressures on them and take advantage of opportunities that may be present. One does, however , observe balancing tendencies already taking place. Upon the demise of the Soviet Union, the international political system became unipolar . In the light of structural theory , unipolarity appears as the least durable of international conŽ gurations. This is so for two main reasons. 63. Stephen D. Krasner , “ International Political Economy: Abiding Discord,” Review of International Political Economy, V ol. 1, No. 1 (Spring 1994), p. 16. 64. Quoted in Robert T ucker , Intervention and the Reagan Doctrine (New Y ork: Council on Religious and International Affairs, 1985), p. 5. 65. Robert Gilpin explains the oddity . See Gilpin, “ No One Leaves a Political Realist,” Security Studies, V ol. 5, No. 3 (Spring 1996), pp. 3– 28. 66. William C. W ohlforth, “ The Stability of a Unipolar W orld,” International Security, V ol. 24, No. 1 (Summer 1999), pp. 5– 41. Structural Realism after the Cold War 27 One is that dominant powers take on too many tasks beyond their own borders, thus weakening themselves in the long run. T ed Robert Gurr , after examining 336 polities, reached the same conclusion that Robert W esson had reached earlier: ” Imperial decay is . . . primarily a result of the misuse of power which follows inevitably from its concentration.“ 67 The other reason for the short duration of unipolarity is that even if a dominant power behaves with moderation, restraint, and forbearance, weaker states will worry about its future behavior . America’ s founding fathers warned against the perils of power in the absence of checks and balances. Is unbalanced power less of a danger in international than in national politics? Throughout the Cold W ar , what the United States and the Soviet Union did, and how they interacted, were domi- nant factors in international politics. The two countries, however , constrained each other . Now the United States is alone in the world. As nature abhors a vacuum, so international politics abhors unbalanced power . Faced with unbal- anced power , some states try to increase their own strength or they ally with others to bring the international distribution of power into balance. The reac- tions of other states to the drive for dominance of Charles V , Hapsburg ruler of Spain, of Louis XIV and Napoleon I of France, of Wilhelm II and Adolph Hitler of Germany , illustrate the point. the behavior of dominant powers Will the preponderant power of the United States elicit similar reactions? Unbalanced power , whoever wields it, is a potential danger to others. The powerful state may , and the United States does, think of itself as acting for the sake of peace, justice, and well-being in the world. These terms, however , are deŽ ned to the liking of the powerful, which may con ict with the preferences and interests of others. In international politics, overwhelming power repels and leads others to try to balance against it. With benign intent, the United States has behaved and, until its power is brought into balance, will continue to behave in ways that sometimes frighten others. For almost half a century , the constancy of the Soviet threat produced a constancy of American policy . Other countries could rely on the United States for protection because protecting them seemed to serve American security interests. Even so, beginning in the 1950s, W estern European countries and, 67. Quoted in T ed Robert Gurr , “ Persistence and Change in Political Systems, 1800– 1971,” Ameri- can Political Science Review, V ol. 68, No. 4 (December 1974), p. 1504, from Robert G. W esson, The Imperial Order (Berkeley: University of California Press, 1967), unpaginated preface. Cf. Paul Kennedy , The Rise and Fall of Great Powers: Economic Change and Military Con ict from 1500 to 2000 (New Y ork: Random House, 1987). International Security 25:1 28 beginning in the 1970s, Japan had increasing doubts about the reliability of the American nuclear deterrent. As Soviet strength increased, W estern European countries began to wonder whether the United States could be counted on to use its deterrent on their behalf, thus risking its own cities. When President Jimmy Carter moved to reduce American troops in South Korea, and later when the Soviet Union invaded Afghanistan and strengthened its forces in the Far East, Japan developed similar worries. With the disappearance of the Soviet Union, the United States no longer faces a major threat to its security . As General Colin Powell said when he was chairman of the Joint Chiefs of Staff: ” I’ m running out of demons. I’ m running out of enemies. I’ m down to Castro and Kim Il Sung.“ 68 Constancy of threat produces constancy of policy; absence of threat permits policy to become capricious. When few if any vital interests are endangered, a country’ s policy becomes sporadic and self-willed. The absence of serious threats to American security gives the United States wide latitude in making foreign policy choices. A dominant power acts internationally only when the spirit moves it. One example is enough to show this. When Y ugoslavia’ s collapse was followed by genocidal war in successor states, the United States failed to respond until Senator Robert Dole moved to make Bosnia’ s peril an issue in the forthcoming presidential election; and it acted not for the sake of its own security but to maintain its leadership position in Europe. American policy was generated not by external security interests, but by internal political pressure and national ambition. Aside from speciŽ c threats it may pose, unbalanced power leaves weaker states feeling uneasy and gives them reason to strengthen their positions. The United States has a long history of intervening in weak states, often with the intention of bringing democracy to them. American behavior over the past century in Central America provides little evidence of self-restraint in the absence of countervailing power . Contemplating the history of the United States and measuring its capabilities, other countries may well wish for ways to fend off its benign ministrations. Concentrated power invites distrust be- cause it is so easily misused. T o understand why some states want to bring power into a semblance of balance is easy , but with power so sharply skewed, what country or group of countries has the material capability and the political will to bring the ” unipolar moment“ to an end? 68. ” Cover Story: Communism’ s Collapse Poses a Challenge to America’ s Military ,” U.S. News and World Report, October 14, 1991, p. 28. Structural Realism after the Cold War 29 balancing power in a unipolar world The expectation that following victory in a great war a new balance of power will form is Ž rmly grounded in both history and theory . The last four grand coalitions (two against Napoleon and one in each of the world wars of the twentieth century) collapsed once victory was achieved. V ictories in major wars leave the balance of power badly skewed. The winning side emerges as a dominant coalition. The international equilibrium is broken; theory leads one to expect its restoration. Clearly something has changed. Some believe that the United States is so nice that, despite the dangers of unbalanced power , others do not feel the fear that would spur them to action. Michael Mastanduno, among others, believes this to be so, although he ends his article with the thought that ” eventually , power will check power .“ 69 Others believe that the leaders of states have learned that playing the game of power politics is costly and unnecessary . In fact, the explanation for sluggish balancing is a simple one. In the aftermath of earlier great wars, the materials for constructing a new balance were readily at hand. Previous wars left a sufŽ cient number of great powers standing to permit a new balance to be rather easily constructed. Theory enables one to say that a new balance of power will form but not to say how long it will take. National and international conditions determine that. Those who refer to the unipolar moment are right. In our perspective, the new balance is emerging slowly; in historical perspectives, it will come in the blink of an eye. I ended a 1993 article this way: ” One may hope that America’ s internal preoccupations will produce not an isolationist policy , which has become impossible, but a forbearance that will give other countries at long last the chance to deal with their own problems and make their own mistakes. But I would not bet on it.“ 70 I should think that few would do so now . Charles Kegley has said, sensibly , that if the world becomes multipolar once again, realists will be vindicated. 71 Seldom do signs of vindication appear so promptly . The candidates for becoming the next great powers, and thus restoring a balance, are the European Union or Germany leading a coalition, China, Japan, and in a more distant future, Russia. The countries of the European Union have 69. Michael Mastanduno, “ Preserving the Unipolar Moment: Realist Theories and U.S. Grand Strategy after the Cold W ar ,” International Security , V ol. 21, No. 4 (Spring 1997), p. 88. See Josef Joffe’ s interesting analysis of America’ s role, ’ ” Bismarck’ or ’ Britain’ ? T oward an American Grand Strategy after Bipolarity ,” International Security, V ol. 19, No. 4 (Spring 1995). 70. W altz, “ The Emerging Structure of International Politics,” p. 79. 71. Charles W . Kegley , Jr ., “ The Neoidealist Moment in International Studies? Realist Myths and the New International Realities,” International Studies Quarterly, V ol. 37, No. 2 (June 1993), p. 149. International Security 25:1 30 been remarkably successful in integrating their national economies. The achievement of a large measure of economic integration without a correspond- ing political unity is an accomplishment without historical precedent. On questions of foreign and military policy , however , the European Union can act only with the consent of its members, making bold or risky action impossible. The European Union has all the tools— population, resources, technology , and military capabilities— but lacks the organizational ability and the collective will to use them. As Jacques Delors said when he was president of the European Commission: ” It will be for the European Council, consisting of heads of state and government . . . , to agree on the essential interests they share and which they will agree to defend and promote together .“ 72 Policies that must be arrived at by consensus can be carried out only when they are fairly inconsequential. Inaction as Y ugoslavia sank into chaos and war signaled that Europe will not act to stop wars even among near neighbors. W estern Europe was unable to make its own foreign and military policies when its was an organization of six or nine states living in fear of the Soviet Union. With less pressure and more members, it has even less hope of doing so now . Only when the United States decides on a policy have European countries been able to follow it. Europe may not remain in its supine position forever , yet signs of funda- mental change in matters of foreign and military policy are faint. Now as earlier , European leaders express discontent with Europe’ s secondary position, chafe at America’ s making most of the important decisions, and show a desire to direct their own destiny . French leaders often vent their frustration and pine for a world, as Foreign Minister Hubert V é drine recently put it, “ of several poles, not just a single one.” President Jacques Chirac and Prime Minister Lionel Jospin call for a strengthening of such multilateral institutions as the International Monetary Fund and the United Nations, although how this would diminish America’ s in uence is not explained. More to the point, V é drine complains that since President John Kennedy , Americans have talked of a European pillar for the alliance, a pillar that is never built. 73 German and British leaders now more often express similar discontent. Europe, however , will not be able to claim a louder voice in alliance affairs unless it builds a platform for giving it expression. If Europeans ever mean to write a tune to go with their libretto, they will have to develop the unity in foreign and military affairs that they are achieving in economic matters. If French and 72. Jacques Delors, “ European Integration and Security ,” Survival, V ol. 33, No. 1 (March/ April 1991), p. 106. 73. Craig R. Whitney , “ NA TO at 50: With Nations at Odds, Is It a Misalliance?” New Y ork Times, February 15, 1999, p. A1. Structural Realism after the Cold War 31 British leaders decided to merge their nuclear forces to form the nucleus of a European military organization, the United States and the world will begin to treat Europe as a major force. The European Economic Community was formed in 1957 and has grown incrementally to its present proportions. But where is the incremental route to a European foreign and military policy to be found? European leaders have not been able to Ž nd it or even have tried very hard to do so. In the absence of radical change, Europe will count for little in international politics for as far ahead as the eye can see, unless Germany , becoming impatient, decides to lead a coalition. international structure and national responses Throughout modern history , international politics centered on Europe. T wo world wars ended Europe’ s dominance. Whether Europe will somehow , some- day emerge as a great power is a matter for speculation. In the meantime, the all-but-inevitable movement from unipolarity to multipolarity is taking place not in Europe but in Asia. The internal development and the external reaction of China and Japan are steadily raising both countries to the great power level. 74 China will emerge as a great power even without trying very hard so long as it remains politically united and competent. Strategically , China can easily raise its nuclear forces to a level of parity with the United States if it has not already done so. 75 China has Ž ve to seven intercontinental missiles (DF-5s) able to hit almost any American target and a dozen or more missiles able to reach the west coast of the United States (DF-4s). 76 Liquid fueled, immobile missiles are vulnerable, but would the United States risk the destruction of, say , Seattle, San Francisco, and San Diego if China happens to have a few more DF-4s than the United States thinks or if it should fail to destroy all of them on the ground? Deterrence is much easier to contrive than most Americans have surmised. Economically , China’ s growth rate, given its present stage of economic development, can be sustained at 7 to 9 percent for another decade or more. Even during Asia’ s near economic collapse of the 1990s, China’ s growth rate remained approximately in that range. A growth rate of 7 to 9 percent doubles a country’ s economy every ten to eight years. 74. The following four pages are adapted from W altz, “ The Emerging Structure of International Politics.” 75. Nuclear parity is reached when countries have second-strike forces. It does not require quantitative or qualitative equality of forces. See W altz, “ Nuclear Myths and Political Realities,” American Political Science Review, V ol. 84, No. 3 (September 1990). 76. David E. Sanger and Erik Eckholm, “ Will Beijing’ s Nuclear Arsenal Stay Small or Will It Mushroom?” New Y ork Times, March 15, 1999, p. A1. International Security 25:1 32 Unlike China, Japan is obviously reluctant to assume the mantle of a great power . Its reluctance, however , is steadily though slowly waning. Economi- cally , Japan’ s power has grown and spread remarkably . The growth of a country’ s economic capability to the great power level places it at the center of regional and global affairs. It widens the range of a state’ s interests and increases their importance. The high volume of a country’ s external business thrusts it ever more deeply into world affairs. In a self-help system, the possession of most but not all of the capabilities of a great power leaves a state vulnerable to others that have the instruments that the lesser state lacks. Even though one may believe that fears of nuclear blackmail are misplaced, one must wonder whether Japan will remain immune to them. Countries have always competed for wealth and security , and the competi- tion has often led to con ict. Historically , states have been sensitive to changing relations of power among them. Japan is made uneasy now by the steady growth of China’ s military budget. Its nearly 3 million strong army , undergoing modernization, and the gradual growth of its sea- and air-power projection capabilities, produce apprehension in all of China’ s neighbors and add to the sense of instability in a region where issues of sovereignty and disputes over territory abound. The Korean peninsula has more military forces per square kilometer than any other portion of the globe. T aiwan is an unending source of tension. Disputes exist between Japan and Russia over the Kurile Islands, and between Japan and China over the Senkaku or Diaoyu Islands. Cambodia is a troublesome problem for both Vietnam and China. Half a dozen countries lay claim to all or some of the Spratly Islands, strategically located and sup- posedly rich in oil. The presence of China’ s ample nuclear forces, combined with the drawdown of American military forces, can hardly be ignored by Japan, the less so because economic con icts with the United States cast doubt on the reliability of American military guarantees. Reminders of Japan’ s de- pendence and vulnerability multiply in large and small ways. For example, as rumors about North Korea’ s developing nuclear capabilities gained credence, Japan became acutely aware of its lack of observation satellites. Uncomfortable dependencies and perceived vulnerabilities have led Japan to acquire greater military capabilities, even though many Japanese may prefer not to. Given the expectation of con ict, and the necessity of taking care of one’ s interests, one may wonder how any state with the economic capability of a great power can refrain from arming itself with the weapons that have served so well as the great deterrent. For a country to choose not to become a great power is a structural anomaly . For that reason, the choice is a difŽ cult one to sustain. Sooner or later , usually sooner , the international status of countries has Structural Realism after the Cold War 33 risen in step with their material resources. Countries with great power econo- mies have become great powers, whether or not reluctantly . Some countries may strive to become great powers; others may wish to avoid doing so. The choice, however , is a constrained one. Because of the extent of their interests, larger units existing in a contentious arena tend to take on systemwide tasks. Profound change in a country’ s international situation produces radical change in its external behavior . After W orld W ar II, the United States broke with its centuries-long tradition of acting unilaterally and refusing to make long-term commitments. Japan’ s behavior in the past half century re ects the abrupt change in its international standing suffered because of its defeat in war . In the previous half century , after victory over China in 1894– 95, Japan pressed for preeminence in Asia, if not beyond. Does Japan once again aspire to a larger role internationally? Its concerted regional activity , its seeking and gaining prominence in such bodies as the IMF and the W orld Bank, and its obvious pride in economic and technological achievements indicate that it does. The behavior of states responds more to external conditions than to internal habit if external change is profound. When external conditions press Ž rmly enough, they shape the behavior of states. Increasingly , Japan is being pressed to enlarge its conventional forces and to add nuclear ones to protect its interests. India, Pakistan, China, and perhaps North Korea have nuclear weapons capable of deterring others from threatening their vital interests. How long can Japan live alongside other nuclear states while denying itself similar capabilities? Con icts and crises are certain to make Japan aware of the disadvantages of being without the military instruments that other powers command. Japanese nuclear inhibitions arising from W orld W ar II will not last indeŽ nitely; one may expect them to expire as generational memories fade. Japanese ofŽ cials have indicated that when the protection of America’ s extended deterrent is no longer thought to be sufŽ ciently reliable, Japan will equip itself with a nuclear force, whether or not openly . Japan has put itself politically and technologically in a position to do so. Consistently since the mid-1950s, the government has deŽ ned all of its Self-Defense Forces as con- forming to constitutional requirements. Nuclear weapons purely for defense would be deemed constitutional should Japan decide to build some. 77 As a secret report of the Ministry of Foreign Affairs put it in 1969: ” For the time 77. Norman D. Levin, “ Japan’ s Defense Policy: The Internal Debate,” in Harry H. Kendall and Clara Joewono, eds., Japan, ASEAN, and the United States (Berkeley: Institute of East Asian Studies, University of California, 1990). International Security 25:1 34 being, we will maintain the policy of not possessing nuclear weapons. How- ever , regardless of joining the NPT [Non-Proliferation T reaty] or not, we will keep the economic and technical potential for the production of nuclear weap- ons, while seeing to it that Japan will not be interfered with in this regard.“ 78 In March of 1988, Prime Minister Noboru T akeshita called for a defensive capability matching Japan’ s economic power . 79 Only a balanced conventional- nuclear military capability would meet this requirement. In June of 1994, Prime Minister T sutumu Hata mentioned in parliament that Japan had the ability to make nuclear weapons. 80 Where some see Japan as a ” global civilian power“ and believe it likely to remain one, others see a country that has skillfully used the protection the United States has afforded and adroitly adopted the means of maintaining its security to its regional environment. 81 Prime Minister Shigeru Y oshida in the early 1950s suggested that Japan should rely on American protection until it had rebuilt its economy as it gradually prepared to stand on its own feet. 82 Japan has laid a Ž rm foundation for doing so by developing much of its own weaponry instead of relying on cheaper imports. Remaining months or mo- ments away from having a nuclear military capability is well designed to protect the country’ s security without unduly alarming its neighbors. The hostility of China, of both Koreas, and of Russia combines with inevi- table doubts about the extent to which Japan can rely on the United States to protect its security . 83 In the opinion of Masanori Nishi, a defense ofŽ cial, the main cause of Japan’ s greater “ interest in enhanced defense capabilities” is its belief that America’ s interest in “ maintaining regional stability is shaky .” 84 Whether reluctantly or not, Japan and China will follow each other on the route 78. ” The Capability to Develop Nuclear W eapons Should Be Kept: Ministry of Foreign Affairs Secret Document in 1969,” Mainichi, August 1, 1994, p. 41, quoted in Selig S. Harrison, “ Japan and Nuclear W eapons,” in Harrison, ed., Japan’ s Nuclear Future (W ashington, D.C.: Carnegie Endow- ment for International Peace, 1996), p. 9. 79. David Arase, “ US and ASEAN Perceptions of Japan’ s Role in the Asian-PaciŽ c Region,” in Kendall and Joewono, Japan, ASEAN, and the United States, p. 276. 80. David E. Sanger , “ In Face-Saving Reverse, Japan Disavows Any Nuclear-Arms Expertise,” New Y ork Times, June 22, 1994, p. 10. 81. Michael J. Green, “ State of the Field Report: Research on Japanese Security Policy ,” Access Asia Review, V ol. 2, No. 2 (September 1998), judiciously summarized different interpretations of Japan’ s security policy . 82. Kenneth B. Pyle, The Japanese Question: Power and Purpose in a New Era (W ashington, D.C.: AEI Press, 1992), p. 26. 83. Andrew H anami, for example, points out that Japan wonders whether the United States would help defend Hokkaido. Hanami, “ Japan and the Military Balance of Power in Northeast Asia,” Journal of East Asian Affairs, V ol. 7, No. 2 (Summer/ Fall 1994), p. 364. 84. Stephanie Strom, “ Japan Beginning to Flex Its Military Muscles,” New Y ork Times, April 8, 1999, p. A4. Structural Realism after the Cold War 35 to becoming great powers. China has the greater long-term potential. Japan, with the world’ s second or third largest defense budget and the ability to produce the most technologically advanced weaponry , is closer to great power status at the moment. When Americans speak of preserving the balance of power in East Asia through their military presence, 85 the Chinese understandably take this to mean that they intend to maintain the strategic hegemony they now enjoy in the absence of such a balance. When China makes steady but modest efforts to improve the quality of its inferior forces, Americans see a future threat to their and others’ interests. Whatever worries the United States has and whatever threats it feels, Japan has them earlier and feels them more intensely . Japan has gradually reacted to them. China then worries as Japan improves its airlift and sealift capabilities and as the United States raises its support level for forces in South Korea. 86 The actions and reactions of China, Japan, and South Korea, with or without American participation, are creating a new balance of power in East Asia, which is becoming part of the new balance of power in the world. Historically , encounters of East and W est have often ended in tragedy . Y et, as we know from happy experience, nuclear weapons moderate the behavior of their possessors and render them cautious whenever crises threaten to spin out of control. Fortunately , the changing relations of East to W est, and the changing relations of countries within the East and the W est, are taking place in a nuclear context. The tensions and con icts that intensify when profound changes in world politics take place will continue to mar the relations of nations, while nuclear weapons keep the peace among those who enjoy their protection. America’ s policy of containing China by keeping 100,000 troops in East Asia and by providing security guarantees to Japan and South Korea is intended to keep a new balance of power from forming in Asia. By continuing to keep 100,000 troops in W estern Europe, where no military threat is in sight, and by extending NA TO eastward, the United States pursues the same goal in Europe. The American aspiration to freeze historical development by working to keep the world unipolar is doomed. In the not very long run, the task will exceed America’ s economic, military , demographic, and political resources; and the very effort to maintain a hegemonic position is the surest way to undermine 85. Richard Bernstein and Ross H. Munro, The Coming Con ict with China (New Y ork: Alfred A. Knopf, 1997); and Andrew J. Nathan and Robert S. Ross, The Great Wall and the Empty Fortress: China’ s Search for Security (New Y ork: W .W . Norton, 1997). 86. Michael J. Green and Benjamin L. Self, “ Japan’ s Changing China Policy: From Commercial Liberalism to Reluctant Realism,” Survival, V ol. 38, No. 2 (Summer 1996), p. 43. International Security 25:1 36 it. The effort to maintain dominance stimulates some countries to work to overcome it. As theory shows and history conŽ rms, that is how balances of power are made. Multipolarity is developing before our eyes. Moreover , it is emerging in accordance with the balancing imperative. American leaders seem to believe that America’ s preeminent position will last indeŽ nitely . The United States would then remain the dominant power without rivals rising to challenge it— a position without precedent in modern history . Balancing, of course, is not universal and omnipresent. A dominant power may suppress balancing as the United States has done in Europe. Whether or not balancing takes place also depends on the decisions of govern- ments. Stephanie Neuman’ s book, International Relations Theory and the Third World, abounds in examples of states that failed to mind their own security interests through internal efforts or external arrangements, and as one would expect, suffered invasion, loss of autonomy , and dismemberment. 87 States are free to disregard the imperatives of power , but they must expect to pay a price for doing so. Moreover , relatively weak and divided states may Ž nd it impos- sible to concert their efforts to counter a hegemonic state despite ample provo- cation. This has long been the condition of the W estern Hemisphere. In the Cold W ar , the United States won a telling victory . Victory in war , however , often brings lasting enmities. Magnanimity in victory is rare. Winners of wars, facing few impediments to the exercise of their wills, often act in ways that create future enemies. Thus Germany , by taking Alsace and most of Lorraine from France in 1871, earned its lasting enmity; and the Allies’ harsh treatment of Germany after W orld W ar I produced a similar effect. In contrast, Bismarck persuaded the kaiser not to march his armies along the road to V ienna after the great victory at Kö niggrä tz in 1866. In the T reaty of Prague, Prussia took no Austrian territory . Thus Austria, having become Austria- Hungary , was available as an alliance partner for Germany in 1879. Rather than learning from history , the United States is repeating past errors by extending its in uence over what used to be the province of the vanquished. 88 This alienates Russia and nudges it toward China instead of drawing it toward Europe and the United States. Despite much talk about the “ globalization” of international politics, American political leaders to a dismaying extent think of East or W est rather than of their interaction. With a history of con ict 87. Stephanie Neuman, ed., International Relations Theory and the Third World (New Y ork: St. Martin’ s, 1998). 88. T ellingly , John Lewis Gaddis comments that he has never known a time when there was less support among historians for an announced policy . Gaddis, “ History , Grand Strategy , and NA TO Enlargement,” Survival, V ol. 40, No. 1 (Spring 1998), p. 147. Structural Realism after the Cold War 37 along a 2,600 mile border , with ethnic minorities sprawling across it, with a mineral-rich and sparsely populated Siberia facing China’ s teeming millions, Russia and China will Ž nd it difŽ cult to cooperate effectively , but the United States is doing its best to help them do so. Indeed, the United States has provided the key to Russian-Chinese relations over the past half century . Feeling American antagonism and fearing American power , China drew close to Russia after W orld W ar II and remained so until the United States seemed less, and the Soviet Union more, of a threat to China. The relatively harmoni- ous relations the United States and China enjoyed during the 1970s began to sour in the late 1980s when Russian power visibly declined and American hegemony became imminent. T o alienate Russia by expanding NA TO, and to alienate China by lecturing its leaders on how to rule their country , are policies that only an overwhelmingly powerful country could afford, and only a foolish one be tempted, to follow . The United States cannot prevent a new balance of power from forming. It can hasten its coming as it has been earnestly doing. In this section, the discussion of balancing has been more empirical and speculative than theoretical. I therefore end with some re ections on balancing theory . Structural theory , and the theory of balance of power that follows from it, do not lead one to expect that states will always or even usually engage in balancing behavior . Balancing is a strategy for survival, a way of attempting to maintain a state’ s autonomous way of life. T o argue that bandwagoning represents a behavior more common to states than balancing has become a bit of a fad. Whether states bandwagon more often than they balance is an interesting question. T o believe that an afŽ rmative answer would refute bal- ance-of-power theory is, however , to misinterpret the theory and to commit what one might call “ the numerical fallacy” — to draw a qualitative conclusion from a quantitative result. States try various strategies for survival. Balancing is one of them; bandwagoning is another . The latter may sometimes seem a less demanding and a more rewarding strategy than balancing, requiring less effort and extracting lower costs while promising concrete rewards. Amid the uncertainties of international politics and the shifting pressures of domestic politics, states have to make perilous choices. They may hope to avoid war by appeasing adversaries, a weak form of bandwagoning, rather than by rearming and realigning to thwart them. Moreover , many states have insufŽ cient re- sources for balancing and little room for maneuver . They have to jump on the wagon only later to wish they could fall off. Balancing theory does not predict uniformity of behavior but rather the strong tendency of major states in the system, or in regional subsystems, to resort to balancing when they have to. That states try different strategies of International Security 25:1 38 survival is hardly surprising. The recurrent emergence of balancing behavior , and the appearance of the patterns the behavior produces, should all the more be seen as impressive evidence supporting the theory . Conclusion Every time peace breaks out, people pop up to proclaim that realism is dead. That is another way of saying that international politics has been transformed. The world, however , has not been transformed; the structure of international politics has simply been remade by the disappearance of the Soviet Union, and for a time we will live with unipolarity . Moreover , international politics was not remade by the forces and factors that some believe are creating a new world order . Those who set the Soviet Union on the path of reform were old Soviet apparatchiks trying to right the Soviet economy in order to preserve its position in the world. The revolution in Soviet affairs and the end of the Cold W ar were not brought by democracy , interdependence, or international insti- tutions. Instead the Cold W ar ended exactly as structural realism led one to expect. As I wrote some years ago, the Cold W ar “ is Ž rmly rooted in the structure of postwar international politics and will last as long as that structure endures.” 89 So it did, and the Cold W ar ended only when the bipolar structure of the world disappeared. Structural change affects the behavior of states and the outcomes their interactions produce. It does not break the essential continuity of international politics. The transformation of international politics alone could do that. T rans- formation, however , awaits the day when the international system is no longer populated by states that have to help themselves. If the day were here, one would be able to say who could be relied on to help the disadvantaged or endangered. Instead, the ominous shadow of the future continues to cast its pall over interacting states. States’ perennial uncertainty about their fates presses governments to prefer relative over absolute gains. Without the shadow , the leaders of states would no longer have to ask themselves how they will get along tomorrow as well as today . States could combine their efforts cheerfully and work to maximize collective gain without worrying about how each might fare in comparison to others. Occasionally , one Ž nds the statement that governments in their natural, anarchic condition act myopically— that is, on calculations of immediate inter- 89. Kenneth N. W altz, “ The Origins of W ar in Neorealist Theory ,” Journal of Interdisciplinary History, V ol. 18, No. 4 (Spring 1988), p. 628. Structural Realism after the Cold War 39 est— while hoping that the future will take care of itself. Realists are said to suffer from this optical defect. 90 Political leaders may be astigmatic, but responsible ones who behave realistically do not suffer from myopia. Robert Axelrod and Robert Keohane believe that W orld W ar I might have been averted if certain states had been able to see how long the future’ s shadow was. 91 Y et, as their own discussion shows, the future was what the major states were obsessively worried about. The war was prompted less by consid- erations of present security and more by worries about how the balance might change later . The problems of governments do not arise from their short time horizons. They see the long shadow of the future, but they have trouble reading its contours, perhaps because they try to look too far ahead and see imaginary dangers. In 1914, Germany feared Russia’ s rapid industrial and population growth. France and Britain suffered from the same fear about Germany , and in addition Britain worried about the rapid growth of Germany’ s navy . In an important sense, W orld W ar I was a preventive war all around. Future fears dominated hopes for short-term gains. States do not live in the happiest of conditions that Horace in one of his odes imagined for man: Happy the man, and happy he alone, who can say , T omorrow do thy worst, for I have lived today .

92 Robert Axelrod has shown that the ” tit-for-tat“ tactic, and no other , maxi- mizes collective gain over time. The one condition for success is that the game be played under the shadow of the future. 93 Because states coexist in a self-help system, they may , however , have to concern themselves not with maximizing collective gain but with lessening, preserving, or widening the gap in welfare and strength between themselves and others. The contours of the future’ s shadow look different in hierarchic and anarchic systems. The shadow may facilitate cooperation in the former; it works against it in the latter . W orries 90. The point is made by Robert O. Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N.J.: Princeton University Press, 1984), pp. 99, 103, 108. 91. Robert Axelrod and Robert O. Keohane, “ Achieving Cooperation under Anarchy: Strategies and Institutions,” in David Baldwin, ed., Neorealism and Neoliberalism: The Contemporary Debate (New Y ork: Columbia University Press, 1993). For German leaders, they say , “ the shadow of the future seemed so small” (p. 92). Robert Powell shows that “ a longer shadow . . . leads to greater military allocations.” See Powell, “ Guns, Butter , and Anarchy ,” American Political Science Review, V ol. 87, No. 1 (March 1993), p. 116; see also p. 117 on the question of the compatibility of liberal institutionalism and structural realism. 92. My revision. 93. Robert Axelrod, The Evolution of Cooperation (New Y ork: Basic Books, 1984). International Security 25:1 40 about the future do not make cooperation and institution building among nations impossible; they do strongly condition their operation and limit their accomplishment. Liberal institutionalists were right to start their investigations with structural realism. Until and unless a transformation occurs, it remains the basic theory of international politics. Structural Realism after the Cold War 41