Redsm and America's Rise

Redsm and Sean M.Lynn-Jones America's Rise

A Review Essay ~

Fareed Zakaria, From Wealth to Power: The Unusual Origins of America's World Role. Princeton, N.J.: Princeton University Press, 1998. Realism is usually regarded as a dominant-and monolithic-approach in the study of international relations. In the past few years, however, some of the most vigorous and interesting debates in international relations theory have emerged among different types of realism. It has become clear that realism is not a single theory, but a family of theories. One of the most significant divides within realism is between offensive realism and defensive realism.* Offensive realists generally argue that the international system fosters conflict and aggression. Security is scarce, making international competition and war likely. Rational states often are compelled to adopt offensive strategies in their search for security.2Defen- sive realists, on the other hand, argue that the international system does not

Sean M. Lynn-Jones is Co-Editor of International Security.

I thank Dale Copeland, Michael Desch, Colin Elman, Miriam Fendius Elman, Eugene Gholz, Charles Glaser, Eric Labs, Jack Levy, Dan Lindley, John Mearsheimer, Taylor Seybolt, Jack Snyder, Jeffrey Taliaferro, Brian Taylor, Bradley Thayer, and Stephen Van Evera for their extraordinarily helpful comments on an earlier draft of this essay. I alone am responsible for the opinions expressed here and any errors that remain.

1. For discussions of contemporary debates and differences among realists, especially between offensive and defensive realists, see Sean M. Lynn-Jones and Steven E. Miller, "Preface," in Michael E. Brown, Sean M. Lynn-Jones, and Steven E. Miller, eds., fie Perils of Anarchy: Contemporary Realism and International Security (Cambridge, Mass.: MJT Press, 1995), pp. ix-xiii; Benjamin Frankel, "Restating the Realist Case: An Introduction," Security Studies, Vol. 5, No. 3 (Spring 19961, pp. xiv-xx; and Gideon Rose, "Neoclassical Realism and Theories of Foreign Policy," WorZd Politics, Vol. 51, No. 1 (October 1998), which classifies several contemporary realists as "neoclassical" realists. For a different set of distinctions, see Stephen G. Brooks, "Dueling Realisms," International Organization, Vol. 51, No. 3 (Summer 1997), pp. 445477. 2. Offensive realism also has been called "aggressive realism." The clearest statements of offensive realism are John J. Mearsheimer, "Back to the Future: Instability in Europe after the Cold War," International Security, Vol. 15, No. 1 (Summer 1990), pp. 5-56; and Mearsheimer, "The False Promise of International Institutions," International Security, Vol. 19, No. 3 (Winter 1994/95), pp. 549. See also Eric J. Labs, "Beyond Vicfory: Offensive Realism and the Expansion of War Aims," Security Studies, Vol. 6, No. 4 (Summer 1997), pp. 149; and Robert Gilpin, War and Change in World Politics (Cambridge: Cambridge University Press, 1981).

International Security, Vol. 23, No. 2 (Fall 1998), pp. 157-182 0 1998 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

157

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 158

necessarily generate intense conflict and war, and that defensive strategies are often the best route to ~ecurity.~ Another debate pits classical realists against structural realists, who are also known as neorealists! Classical realists generally argue that power is the most important factor in international politics. States attempt to maximize their power, at least partly because the desire for increased power is rooted in human nature. Structural realists, on the other hand, do not build their theories on the assumption that human nature contains an innate drive for power. Instead, they posit that international politics is shaped by states' desire to survive in the anarchic international ~ystem.~Offensive realism shares classical realism's emphasis on power and pessimism, but follows neorealism's structural 10gic.~ These debates deserve attention for two reasons. First, the prominence of realist theories in the study of international politics makes it important to

3. Important examples of defensive realism include Stephen Van Evera, Causes of War, Vol. 1, The Structure of Power and the Roots of War (Ithaca, N.Y.: Cornell University Press, forthcoming 1999); Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987); Walt, Rmolution and War (Ithaca, N.Y.: Comell University Press, 1996); Barry R. Posen, The Sources of Military Doctrine: France, Britain, and Germany between the world Wars (Ithaca, N.Y.: Cornell Univer- sity Press, 1984); Jack Snyder, Myths of Empire; Domestic Politics and International Ambition (Ithaca, N.Y.: Cornell University Press, 1991); and Charles L. Glaser, "Realists as Optimists: Cooperation as Self-Help," International Security, Vol. 19, No. 3 (Winter 1994/95), pp. 50-90. The theoretical foundations of defensive realism were laid by Robert Jervis, "Cooperation under the Security Dilemma," World Politics, Vol. 30, No. 2 (January 1978), pp. 167-214, and Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979), although neither Jervis nor Waltz can readily be classified as a defensive realist. 4. Most writers use structural realism and neorealism interchangeably, as I do. Some scholars draw a distinction between Waltz's neorealism and structural realism more generally. See Barry Buzan, Charles Jones, and Richard Little, The Logic of Anarchy: Neorealism to Structural Realism (New York Columbia University Press, 1993). 5. The seminal example of classical realism is Hans J. Morgenthau, Politics among Nufions: The Struggle for Power and Peace (New York Knopf, 1948 and later editions). Structural realism, or neorealism, is developed in Waltz, Theory of International Politics. Waltz discusses the differences between classical realism and neorealism in Kenneth N. Waltz, "The Origins of War in Neorealist Theory," in Robert I. Rotberg and Theodore K. Rabb, eds., The Origin and Prevention of Major Wars (Cambridge: Cambridge University Press, 1989), pp. 39-44. For other discussions of the differences between classical realism and neorealism, see Robert 0. Keohane, "Realism, Neorealism, and the Study of World Politics"; Richard Ashley, "The Poverty of Neorealism"; and Kenneth N. Waltz, "Reflections on Theory of International Politics: A Response to My Critics," all in Keohane, ed., Neorealism and Its Critics (New York: Columbia University Press, 19861, pp. 7-16,255300,322-345. 6. Mearsheimer's explication of offensive realism, for example, rests on his argument that aggressive state behavior is a response to the anarchic structure of international politics, not a consequence of the flaws inherent in human nature. For a discussion of how both offensive and defensive realism fit into a neorealist framework, see Colin Elman, "Horses for Courses: Why Not Neorealist Theories of Foreign Policy?" Security Studies, Vol. 6, No. 1 (Autumn 19961, pp. 27-30 and the diagram on pp. 50-53.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America's Rise 1 159

determine which realist theory is the most powerful. Second, the debate between offensive and defensive realism has implications for foreign policy. If the offensive realists are correct, the prospects for constructing policies to avoid war are limited. On the other hand, defensive realism offers somewhat more optimistic conclusions about the ability of states to remain secure without threatening other^.^ Fareed Zakaria's From Wealth to Power' attempts to contribute to these debates in contemporary realist theory in three ways. First, it presents one of the most detailed critiques of defensive realism to date. Second, it argues that a modified version of offensive realism, which Zakaria labels "state-centered realism," can persuasively explain why states expand their territory and their interests. Third, it makes an interesting case for why realist theories should incorporate internal state strength-the capability of a government vis-a-vis society-as a component of national power. Zakaria develops these arguments by examining US. attempts to expand between 1865 and 1908 and offers a fascinating interpretation of how the United States emerged as a world power. Despite these contributions, several of the book's arguments are not convincing. Zakaria neither refutes defensive realism nor advances a compelling alter- native. In this essay, I argue that the overriding weakness of From Wealth to Power is its failure to offer a persuasive critique of defensive realism. There are three reasons why Zakaria does not credibly refute the theory. First, From Wealth to Power ultimately offers a weak critique of defensive realism because the book deduces the wrong hypothesis from the theory-one that virtually guarantees that defensive realism will appear to fail. Zakaria claims that defensive realism predicts that "nations try to expand their political interests abroad when central decision-makers perceive an increase in threats" (p. 42), but defensive realists would not recognize this hypothesis as their owngSecond, From Wealth to Power selects cases that offer weak tests of the relative power of the hypothe-

7. For a discussion of this point and of how contemporary theories of international relations are relevant to the making of foreign policy, see Stephen M. Walt, "International Relations: One World, Many Theories," Foreign Policy, No. 110 (Spring 1998), pp. 29-46. 8. Fareed Zakaria, From Wealth to Paver: The Unusual Origins of America's World Role (Princeton, N.J.: Princeton University Press, 1998). Subsequent references to pages in this book appear paren- thetically in the text. In the interests of full disclosure, I should pint out that I am thanked in the acknowledgments of From Wealth to Power. These thanks reflect the fact that I offered extensive comments on earlier drafts on the book's theory chapter. This review is a continuation of a debate that began with those comments. 9. As I argue below (p. 170), in its most general form the correct defensive-realist hypothesis is: "States attempt to expand when expansion increases their security."

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 1 160

sis Zakaria attributes to defensive realism and the hypothesis that he derives from state-centered realism: “Nations try to expand their political interests abroad when central decision-makers perceive a relative increase in state power” (p. 42). Finally, Zakaria’s arguments against the concepts and logic of defensive realism do not add up to a case for rejecting the theory. In addition to failing to refute defensive realism, Zakaria does not persuasively make a case for state-centered realism, which is a variant of offensive realism even though Zakaria calls it a form of classical realism. To be sure, his argument that countries expand when they become more powerful is more convincing than his critique of defensive realism. U.S. foreign policy in the late nineteenth century reveals a pattern of opportunistic expansion; the United States became more likely to expand as its leaders became aware of its increased capabilities. It is not clear from Zakaria’s analysis, however, that the strength or weakness of the state was more important in shaping policy than other factors, such as aggregate national power or international incentives for expansion. Zakaria argues that the United States was slow to expand in the nineteenth century because it had a weak state, but some of his measures of state strength are actually indicators of an absence of consensus among decisionmakers. Zakaria also overstates the explanatory power of state-centered realism. By focusing entirely on how increased power leads a state to expand, state- centered realism only looks at the opportunity to expand and neglects international incentives that might give a state a motive for expansion. Although the theory seems to account for the fact of US. expansion between 1865 and 1908, it says little about the magnitude of that expansion. The first section of this essay offers a summary of the main arguments and evidence presented in From Wealth to Power. The second examines the book‘s strengths. The third assesses Zakaria’s empirical and logical critique of defensive realism. The fourth discusses the limits of Zakaria’s state-centered realism. The final section offers conclusions and implications for further research.

From Wealth to Power: A Summary

From Wealth to Power attempts to answer a basic question: Why do states expand? As Zakaria explains, “The central question of this study [is]: under what conditions do states expand their political interests abroad?” (p. 8). Zakaria does not limit his inquiry to territorial annexation and the acquisition of imperial possessions. Instead, he defines expansion broadly as ”an activist

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America's Rise I 261

foreign policy that ranges from attention to international events to increases in diplomatic legations to participation in great-power diplomacy" (p. 51." From Wealth to Power explores the general question of why states expand in the context of a particular historical puzzle: why didn't the United States expand more rapidly between 1865 and 1889? And why did it expand more rapidly between 1889 and 1908? The American economy grew dramatically during the first period, but the United States acquired few additional territories, main- tained military forces that were minuscule compared with those of other great powers, and generally eschewed a larger diplomatic role.

THE CONTENDING THEORIES Zakaria tests two realist theories against each another to see which best explains American expansion (or the lack thereof) between 1865 and 1908: state- centered realism, which he describes as a modified version of classical realism, and defensive realism. Zakaria uses both realist theories as theories of foreign policy, which attempt to explain a state's "intentions, goals, and preferences" (p. 141.'' STATE-CENTERED REALISM. Zakaria's state-centered realism rests on the basic argument that as a state becomes relatively more powerful, it expands in an attempt to maximize its influence and control its international environment. In sum, this theory argues that a state's capabilities shape its intentions: it will expand when it can. Although Zakaria calls state-centered realism a variant of classical realism, it actually follows the logic of the offensive variant of structural realism, in two ways. First, the theory identifies the anarchic nature of the international system as the basic cause of state behavior: "In the anarchic, nonhierarchical international environment, states are driven by the system's competitive imperative: if a state does not attempt to maximize its influence, then another will seize

10. Some critics might find this definition so broad as to be meaningless. For example, both balancing and bandwagoning might qualify as expansion as Zakaria defines it, because each is a type of alignment or alliance formation that represents an increase in international activity. Most of the nineteenth-century cases of potential U.S. expansion that Zakaria considers are, however, examples of actual or contemplated territorial expansion. 11. Theories of international politics, unlike theories of foreign policy, attempt to explain intema- tional patterns and outcomes, such as the formation of balances of power. For a debate over whether structural realist theories are useful as theories of foreign policy, see Elman, "Horses for Courses"; Kenneth N. Waltz, "International Politics Is Not Foreign Policy," Security Studies, Vol. 6, No. 1 (Autumn 1996), pp. 54-57; and Elman, "Cause, Effect, and Consistency: A Response to Kenneth Waltz," in ibid., pp. 58-61.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 1 162

the opportunity in its stead” (p. 29).12Second, the likelihood that a given state will attempt to expand depends on its position in the international system-its relative power-instead of its regime type or other domestic-level factors. The theory thus holds that states respond to systemic incentives, that expansion is often a rational response to international incentives, and that the position of a state in the international system (i.e., its relative capabilities) determines its behavior. These arguments differ from classical realist claims that state behavior is shaped not only by the structure of the international system, but also by a lust for power that is inherent in human nature, and that revisionist and status quo powers have different motives and 0bje~tives.l~Thus, contrary to Zakaria, I classlfy state-centered realism as an example of an offensive variant of structural realism, not classical realism.14 Zakaria adds two further elements to the offensive-realist core of this theory. First, he argues that states will expand “in a rational way, measuring risks, opportunities, costs, and benefits” (p. 20). Second, contending that state strength visd-vis society determines how much national power can be used for foreign policy purposes, Zakaria argues that ”state power”-the scope, autonomy, cohesion, and extractive capabilities of the state-must be taken into account in assessing the impact of changes in national power (pp. 35-41).‘5

12. It would be hard to find a purer structural argument than this claim. More generally, Zakaria’s most comprehensive explication of the logic of how systemic pressures cause states to expand (pp. 28-30) rests entirely on his interpretation of the implications of Waltz’s structural realism. 13. See Waltz, “The Origins of War in Neorealist Theory,” pp. 39-44. Some contemporary realists adopt classical realism’s distinctions among states on the basis of unit-level factors such as motivations and the difference between revisionist and status quo powers. See, for example, Randall L. Schweller, “Bandwagoning for Profit: Bringing the Revisionist State Back In,” Inferna- tiunul Security, Val. 19, No. 1 (Summer 19941, pp. 72-107; and Schweller, “Neorealism’s Status-Quo Bias: What Security Dilemma?” Security Studies, Vol. 5, No. 3 (Spring 1996), pp. 90-121. Zakaria does draw on classical realism’s emphasis on the importance of power in international politics, but as he notes on pp. 3233, he excludes factors such as the national interest, statesmanship, and moral choice, which are prominent in classical realism. 14. It appears that Zakaria has decided to give the label ”classical realism” to what other scholars call offensive realism. In Figure 2.1, “Structural Realism” (p. 29), he depicts defensive realism and what he calls classical realism as competing variants of structural realism. Others also conclude that Zakaria’s realism should be classified as “offensiverealism,” even though he does not use the term. See Benjamin Frankel, ”The Reading List,“ Securify Sfudies, Vol. 5, No. 1 (Autumn 1995), p. 187; and Labs, “Beyond Victory,” p. 11. Frankel and Labs reach this conclusion on the basis of their reading of Fareed Zakaria, “Realism and Domestic Politics: A Review Essay,” lnternutional Securify, Vol. 17, No. 1 (Summer 1992), pp. 177-198, which presents an earlier version of many of the arguments developed in From Wealth to Power. Rose, “Neoclassical Realism and Theories of Foreign Policy,” argues that the inclusion of perceptions of power and state strength in state-centered realism makes Zakaria a neoclassical realist. The addition of these intervening variables means that Zakaria’s theory is not purely structural, but the theory‘s logic of expansion nevertheless rests on the premises of offensive, structural realism. 15. Some scholars might argue that the inclusion of state strength adds a domestic variable to the offensive realist core of state-centered realism and therefore removes the theory from the realist

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America's Rise I 163

DEFENSIVE REALISM. In Zakaria's view, defensive realism argues that insecurity causes states to attempt to expand. Defensive realism thus differs from Zakaria's variant of offensive realism, which attributes the drive for expansion to increased capabilities. According to Zakaria, defensive realism contends that states expand their political interests in an effort to make themselves more secure (p. 21). He claims that the theory predicts that a state is more likely to expand when international threats to its security increase.16In short, defensive realism argues that the necessity for security shapes a state's intentions: it will expand when it must. Zakaria selects defensive realism and state-centered realism as candidate explanations of why great powers expand because he believes that "the most important general characteristic of a state in international relations is its relative standing in the international system'' (p. 16). In his view, theories that begin by examining the effect of the international system make the best "first- cut" theories because they can generalize "across regimes, cultures, and peo- ples" and can therefore be applied to a larger number of cases than can a "country-specific explanation" (p. 16).17 Additional variables can be layered onto first-cut theories to provide more discriminating explanations of particular cases. Zakaria makes two modifications to defensive realism and state-centered realism to make the theories more testable (pp. 24, 42). First, instead of attempting to construct and then use objective measures of relative power and threat, Zakaria operationalizes these variables as perceptions." Second, recog-

camp altogether. Although Zakaria's theory stretches the bounds of realism, I accept his classifica- tion of state-centered realism as a realist theory. Other realists agree that domestic variables can play a role in realist theories. Stephen Walt, for example, notes that "Realists argue that systemic factors (e.g., relative power) exert a greater influence on state behavior than unit-level variables do, but no Realist maintains that unit-level factors exert no influence at all." Walt, "Alliances, Threats, and U.S. Grand Strategy: A Reply to Kaufman and Labs," Security Studies, Vol. 1, No. 3 (Spring 1992), p. 473, n. 1. 16. As I argue below (pp. 170-172), defensive realists would not agree that defensive realism leads to this hypothesis. 17. This preference resembles Kenneth Waltz's rejection of "reductionist" theories that attempt to explain international politics on the basis of state-level variables. See Waltz, Theory of International PoIitics, chaps. 24. Zakaria also appears to have eliminated some other candidate theories by conducting some simple tests that reveal them to be weak (pp. 48-55). For example, he suggests that claims that war-weariness prevented US. expansion after 1865 are undermined by evidence that states expand their international influence after wars, as the United States and the Soviet Union did after World War II (p. 52). And arguments that domestic economic problems prevented US. expansion in the 1860s and 1870s are difficult to reconcile with claims that similar economic difficulties prompted it to expand in the 1890s (p. 54). 18. Zakaria does not address the question whether relying on perceptions instead of reality vitiates realist theory by removing it from the objective, material realm or whether this step requires an explanation of perceptions and misperceptions-an explanation that might employ theories of

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Znternational Security 23:2 I 164

nizing the difficulty of measuring the beliefs of states, which are abstract entities, he contends that "statesmen, not states, are the primary actors in international affairs" (p. 42), and therefore focuses on their perceptions of power and threats. State-centered realism thus leads to the hypothesis: "Na- tions try to expand their political interests abroad when central decisionmakers perceive a relative increase in state power" (p. 42). The corresponding hypothesis for defensive realism is: "Nations try to expand their political interests abroad when central decision-makers perceive an increase in threats'' (p. 42). Before testing these hypotheses, Zakaria offers a critique of the concepts and logic of defensive realism (pp. 25-31). He argues that defensive realism rests on the ambiguous and malleable concept of security, that it fails to explain much great power expansion and must rely on domestic, state-level variables, that it misunderstands how the imperatives of the international system compel states to maximize their influence, and that it assumes that states can learn the right lessons from history."

EXPLAINING AMERICA'S RISE TO POWER American foreign policy from 1865 to 1908 provides a set of cases for testing the hypotheses derived from state-centered realism and defensive realism. From Wealth to Power identifies fifty-four individual cases that were "distinct opportunities to expand," and defines "opportunity" as "a serious discussion within the executive branch of the US. government of a specific possibility to extend American influence abroad" (pp. 54-55). Instead of constituting a single

psychology or domestic politics. He apparently concludes that relying on perceptual variables still makes it possible to construct a parsimonious theory, even if the sources of those perceptions are not explained. Other contemporary realists of all stripes also have employed perceptions of power and threat as explanatory variables. See, for example, Thomas J. Christensen, "Perceptions and Alliances in Europe, 1864-1940," International Organization, Vol. 51, No. 1 (Winter 1997), pp. 65-97; Thomas J. Christensen and Jack Snyder, "Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity," International Organization, Vol. 44, No. 2 (Spring 19901, pp. 137-168; Jack Snyder, "Perceptions of the Security Dilemma in 1914," in Robert Jervis, Richard Ned Lebow, and Janice Gross Stein, Psycho/ogy and Deferrence (Baltimore, Md.: Johns Hopkins University Press, 1985); Stephen Van Evera, "Offense, Defense, and the Causes of War," Internationul Security, Vol. 22, No. 4 (Spring 19981, pp. 543; and William Curti Wohlforth, The Elusive Balance: Power and Perceptions during the Cold War (Ithaca, N.Y.: Cornell University Press, 1993). For arguments that uphold the validity and necessity of relying on perceptual variables, see Wohlforth, "Realism and the End of the Cold War," pp. 97-98, 107-109, 127; and Sean M. Lynn-Jones, "Offense-Defense Theory and Its Critics," Security Studies, Vol. 4, No. 4 (Summer 1995), pp. 681482. 19. An earlier version of this critique appears in Zakaria, "Realism and Domestic Politics." I explicate and respond in detail to these criticisms below, pp. 173-178.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America‘s Rise 1 165

case, US. foreign policy in the period selected thus becomes many cases in which the relevant variables can be measured.” In each case, Zakaria assesses the historical record to determine whether U.S. decisionmakers acted on the basis of perceptions of power (as state-centered realism would predict) or on the basis of threats (as Zakaria’s interpretation of defensive realism would predict). Zakaria argues that US. foreign policy between 1865 and 1908 “provides an excellent set of critical cases for both state-centered realism and defensive realism” (p. 43). In his view, U.S. foreign policy during this period is a difficult or “least-likely” case for state-centered realism. Many scholars see the United States as an atypical great power with a national character that favors isola- tionism. Zakaria claims that this makes U.S. behavior a challenging case for state-centered realism; if the theory can explain this case, then it ”would surely apply to the other great powers” (p. 43). U.S. foreign policy in the period under question, according to Zakaria, also is a ”most-likely” or easy case for defensive realism. The United States enjoyed an ”extremely secure geographic position” (p. 431, and thus defensive realism would appear to explain American nonexpansion as the result of abundant security. Zakaria divides the years from 1865 to 1908 into two periods: 1865-1889, during which expansion was rare, and 1889-1908, when expansion was much more frequent. In the first period, the United States confronted twenty-two opportunities to expand, but did so in only six of them (pp. 55, 88). Zakaria argues that his state-centered realism explains the outcomes of fifteen of twenty-two cases of potential expansion, whether the result was expansion or not?’ U.S. policymakers generally did not attempt to expand U.S. influence during this period, because they ”presided over a weak, divided, and decen- tralized government that presented them with little usable power” (p. 55).

20. Zakaria thus appears to be following the methodological advice of Gary King, Robert 0. Keohane, and Sidney Verba, Designing Social Inquiry: Scientific Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994), pp. 217-228. He multiplies the number of observations and measurements to make possible a broader test of a theory than a single case would permit. See also Arend Lijphart, ”Comparative Politics and the Comparative Method,” American Political Science Review, Vol. 65, No. 3 (September 1971), pp. 682493. 21. Zakaria presents a complete list of these cases and codings of outcomes on p. 88. Why some cases are on the list and others are excluded is sometimes unclear. For example, Canada and Mexico are listed as examples of opportunities to expand between 1865 and 1869, but they receive virtually no attention in the text. On the other hand, U.S. attempts in the same period to purchase the islands of Culebra and Culebrita off Puerto Rico, Fuerto Rico itself, Cuba, St. Bartholomew (now St. Barthklemy), St. Pierre and Martinique, and Tiger Island are not included in the list but are discussed on p. 59.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 166

Cases of potential U.S. expansion during this period include Secretary of State William Seward‘s attempts between 1865 and 1869 to purchase, annex, or control the Danish West Indies (the Virgin Islands), Haiti, Santo Domingo (now the Dominican Republic), Greenland and Iceland, the Hawaiian Islands, and Alaska. In most cases, congressional opposition and a belief that the United States was fiscally overextended thwarted Seward’s expansionist ambitions. From Wealth to Power describes a continuing pattern of unsuccessful attempts at U.S. expansion between 1869 and 1889. Efforts to purchase or otherwise acquire Santo Domingo and Cuba were blocked by congressional opposition. The United States negotiated a treaty to annex Santo Domingo, but the Senate rejected it in 1870. This vote revealed the insurmountable obstacles to Ameri- can expansion and discouraged subsequent U.S. administrations from further attempts to expand (pp. 68-69, 74).22Zakaria claims that this opposition generally was not based on substantive disagreements with the executive branch, but instead reflected a desire to assert the power of the legislative branch: ”Congress was hardly isolationist during the 1870s. It simply opposed the exercise of executive leadership” (p. 70). State structure, not ”domestic or party politics,” prevented U.S. central decisionmakers from extracting national power and using it to increase U.S. influence in the world (pp. 87-88). Zakaria argues that the United States reached a key turning point in the 1880s and 1890s: changes in U.S. state structure increased U.S. state power and made international expansion possible. The federal government gained power at the expense of the states, because a nationwide regulatory system was necessary in an era of national business and commerce (pp. 103-106). The federal bureaucracy became more autonomous after civil service reform pro- duced a merit-based system that reduced, but did not eliminate, patronage pressures (pp. 113-116). And the presidency finally began to gain the upper hand in the struggle for power with Congress: presidents reasserted their control over appointments and dismissals of cabinet members and other officials, used the veto much more often to reestablish control over policy, and began to initiate major national policies more frequently (pp. 107-113). In each case, the American public sided with the executive branch and forced Congress to acquiesce.= During the same period, the United States strengthened the

22. The impact of the Santo Domingo vote on subsequent U.S. actions suggests that many of Zakaria’s cases are not independent. On why this implies a need for caution in drawing conclusions, see King, Keohane, and Verba, Designing Social Inquiry, pp. 222-223. 23. In addition, the Supreme Court, in its 1890 case In re Neugle, provided a legal justification for presidential actions to support the national interest without Congress’s orders (p. 137).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise I 167

organizations it used to conduct its foreign relations. The Foreign Service and the armed forces became larger, more professional, and less corrupt (pp. 120- 126). These organizational changes and reforms were not as important as the overall growth in state capacity, but they strengthened U.S. capabilities to conduct a foreign policy (p. 120). In Zakaria’s view, the increases in U.S. state capacity in the 1880s and 1890s were not the result of international pressures and therefore do not conform to Charles Tilly’s well-known aphorism, “War made the state, and the state made war.’124Instead, “the American state came to the fore during a period that lacked rising threats” and gained strength largely in response to domestic pressures attributable to industrialization (p. 95). This argument is extremely important to Zakaria’s theory, because if international threats-which he regards as the central variable of defensive realism-aused the rise of the American state, then defensive realism (as he interprets it) would explain the change in state power-the central variable in Zakaria’s state-centered realism-and would therefore explain the pattern of U.S. expansion and win the theoretical contest hands down. The increase in US. state strength in the 1880s and 1890s enabled the United States to take advantage of opportunities to expand in the 1890s and early 1900s. Between 1889 and 1908, Zakaria identifies thirty-two opportunities to expand and concludes that the United States expanded in twenty-five of them (pp. 130, 175). Prominent examples of successful U.S. expansion include U.S. control over the Philippines, Cuba, and Puerto Rico after the victory in the Spanish-American War; the acquisition of the Isthmus of Panama and the construction of the Panama Canal; the U.S. intervention to help suppress the Boxer Rebellion in China; and President Theodore Roosevelt’s assistance in negotiating an end to the Russo-Japanese War. Zakaria argues that state- centered realism explains the outcomes (whether expansion or nonexpansion) in twenty-two of the thirty-two cases, whereas defensive realism explains only six cases, and four cannot readily be classified (p. 175). In his view, the weakness of defensive realism is revealed by the tendency of the United States to back down in the face of serious threats-particularly threats from Britain and Japan-and to expand primarily at the expense of weaker, nonthreatening states (pp. 174-180).

24. For Rlly’s quotation, see Charles Tilly, “Reflections on the History of European State-Making,” in Tilly, ed., The Formation of Nufional States in Western Europe (Princeton, N.J.: Princeton University Press, 1975), p. 42.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 168

In the book‘s conclusion, Zakaria reaffirms that defensive realism is a poor explanation of expansion and ”deficient as a general theory,” but recognizes that it might explain state behavior in crises or other situations where threats are high and unambiguous (p. 186). He suggests that defensive realists have allowed their preference that the United States expand only in a limited way to achieve clear security objectives to shape their theory of foreign policy.25 Instead of constructing a theory that predicts that states will seek the limited security objectives that defensive realists believe they should seek, Zakaria calls for facing the fact that states inexorably attempt to expand their influence as their relative power increases (p. 183). This leads to a less sentimental but more accurate interpretation of U.S. diplomatic history-an interpretation that recognizes that the United States has not been an exception to the rule that great powers seek influence beyond limited security aims because they can. Zakaria also suggests that the concept of state strength is important in understanding the past and future of international politics. He argues that changes in state power help to explain the rise of British imperialism in the nineteenth century and Germany’s more assertive policy under Wilhelm 11. Looking toward the future, he ends From Wealth to Power on an optimistic note: “States today are simply not as autonomous or as powerful as they were 150 years ago” (p. 191). They have lost power to international institutions, subna- tional units, and nongovernmental actors. These changes could tame the aggressive ambitions of great powers.26

The Strengths of From Wealth to Power

Although I have many criticisms of From Wealth to Power, it is important to recognize the book‘s strengths. Zakaria persuasively demonstrates that the increase in U.S. international activism and expansion at the end of the nineteenth century reflected U.S. decisionmakers’ increasing awareness of U.S. power. The evidence from the U.S. case supports his general argument that

25. Defensive realists called for limited containment and argued against ”rollback of the Soviet Union during the Cold War. See, for example, Barry R. Posen and Stephen Van Evera, ”Defense Policy and the Reagan Administration: Departure from Containment,” International Security, Vol. 8, No. 1 (Summer 1983), pp. 3-45; and Stephen M. Walt, ”The Case for Finite Containment: Analyzing US.Grand Strategy,” International Security, Vol. 14, No. 1 (Summer 19891, pp. 549. 26. Zakaria’s optimism about the weakening of states and its impact seems to be at odds with his argument that a rising tide of illiberal democracy in many states is increasing the arbitrary exercise of executive power. See Zakaria, ”The Rise of Illiberal Democracy,” Foreign Affairs, Vol. 76, No. 6 (November/December 1997), pp. 22-43. Moreover, if industrialization tends to increase state strength-as Zakaria argues it did in the US. case-currently weak states in developing countries will be strengthened as those countries industrialize.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise I 169

increased capabilities give a state more opportunities to expand.27The persua- siveness of this argument confirms that structural realism can offer a useful first-cut explanation of expansion. Zakaria’s analysis of American expansion is also a reminder and clarification of the central importance of power in international politics. Although it would seem obvious that power matters-a point on which most offensive and defensive realists agree-the concept periodically goes out of fashion in the study of international politics.’’ Zakaria defines power carefully (as capabilities) and distinguishes it from influence, thereby avoiding classical realism’s ambiguity over whether power is an end or a means in international politics (p. 19). In his realist theory, power is a means, whereas influence is the end. Zakaria’s addition of state strength to the list of factors that determine a country’s aggregate capabilities is a useful modification of realist theory. This is a modest innovation, because other scholars have analyzed the relationships between state strength, foreign policy, and international politics, but it reminds us of the need for care in measuring power.29 State strength needs to be integrated into any list of the components of power, which include: ”size of

27. This general proposition is not, however, unique to Zakaria’s state-centered realism or offensive realism more generally. Jack Snyder, for example, argues that “much imperial expansion is un- problematic: the strong conquer the weak because it pays.” Snyder, Myths of Empire, p. 10. 28. Critiques of the clarity of the concept of power and its explanatory utility include Inis L. Claude, Power and International Relations (New York Random House, 1962); Robert 0. Keohane and Joseph S. Nye, Jr,, Power and Interdependence: World Politics in Transition (Boston: Little, Brown, 1977); and David A. Baldwin, Paradoxes of Power (New York: Basil Blackwell, 1989). Critiques of hegemonic stability theory and arguments that norms play an important role in international politics also tend to play down the explanatory power of power. See Robert 0. Keohane, After Hegemony: Cooperation and Discord in the World Political Economy (Princeton, N.J.: Princeton Univer- sity Press, 19%); and Peter J. Katzenstein, ed., Culture and National Security: Norms and Identity in World Politics (New York Columbia University Press, 1996). 29. Thomas J. Christensen, Useful Adversaries: Grand Strategy, Domestic Mobilization, and Sino-Ameri- can Conflict, 2947-1958 (Princeton, N.J.: Princeton University Press, 1996), argues that grand strategies are influenced by ”national political power,” which resembles Zakaria’s concept of ”state power.” Prominent works that use variations in state strength to explain variations in foreign policies include Peter J. Katzenstein, ed., Between Power and Plenty: Foreign Economic Policies of Advanced Industrial States (Madison: University of Wisconsin Press, 1978); and Stephen D. Krasner, Defending the National Interest: Raw Materinls Investments and U.S.Foreign Policy (Princeton, N.J.: Princeton University Press, 1978). The concept also has been applied to problems of political development. See Joel Migdal, Strong Societies and Weak States: State-Society Relations and Capabilities in the Third World (Princeton, NJ.: Princeton University Press, 1988). The impact of the international security environment on state strength also has received much attention. Classic analyses include Otto Hintze, ”Military Organization and the Organization of the State,” in Felix Gilbert, ed., The Historical Essays of Otto Hinfze (Oxford: Oxford University Press, 1975). More recent discussions include Michael C. Desch, “War and Strong States, Peace and Weak States?” International Organi- zation, Vol. 50, No. 2 (Spring 1996), pp. 237-268; and Brian M. Downing, The Military Revolution and Political Change: The Origins of Democracy and Autocracy in Early Modern Europe (Princeton, N.J.: Princeton University Press, 1992).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 270

population and territory, resource endowment, economic capability, military strength, political stability, and c~mpetence.”~’ Finally, From Wealth to Power has the virtue of provoking further debate between offensive and defensive realism. Although Zakaria shuns the label, his theory should and will be classified as a variant of offensive realism. As I argue below, his critique of defensive realism is ultimately more stimulating than convincing, but it raises issues that deserve attention.

The Flawed Critique of Defensive Realism

Zakaria vigorously argues against defensive realism and contends that his state-centered realism better explains why states expand. These arguments, however, are seriously overstated. Defensive realism may have its weaknesses, but it is not guilty of the sins that Zakaria attributes to it. Neither the empirical nor the conceptual and logical critique of defensive realism in From Wealth to Power is convincing.

THE WEAKNESSES OF THE EMPIRICAL TESTS OF DEFENSIVE REALISM Zakaria’s empirical tests of defensive realism do not persuasively refute defensive realism. The most important weakness of Zakaria’s test of defensive realism is that he derives the wrong hypothesis from the theory. He argues that defensive realism leads to the prediction that ”nations expand their political interests when they become increasingly insecure” (p. 21), and then refines the hypothesis to predict that “nations try to expand their political interests abroad when central decision-makers perceive an increase in threats” (p. 42). The correct defensive realist hypothesis, however, is that states attempt to expand when expansion increases their security. In general terms, threats might cause a state to expand when three conditions are satisfied: (1) a threat exists; (2) expansion is an effective strategy for reducing the threat; (3) the state has the capability to adopt an expansionist strategy. Defensive realism specifies the conditions under which expansion is likely to be an effective means of reducing a threat and increasing a state‘s security. They include an offense-defense balance that favors the offense and makes conquest easy (or a shift in the offense-defense balance toward the offense); situations in which resources of conquered territories can easily be exploited; expectations that a state’s relative power will decline; and multipolarity, which

30. Waltz, Tkoy of International Politics, p. 131.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise ~ 171

makes it easier to defeat opponents pie~emeal.~’Defensive realists may argue that these conditions exist infrequently, but that does not change the logic of the theory: ”When these conditions are present, Realists would expect the state to adopt a strategy of security through expansion; when they are absent, Realists would expect the state to adopt a nonaggressive ~trategy.”~’ Defensive realists place considerable emphasis on the concepts of threat and insecurity that Zakaria incorporates into his ”defensive-realist” hypo these^.^^ In general, however, they have used these concepts to explain alliance formation and the outbreak of war, not the more general attempts of states to expand their interests. It is striking that Zakaria quotes only historians and statesmen as offering the hypothesis he attributes to defensive realism, whereas several political scientists are quoted making what he calls the classical (i.e., offensive) realist argument (compare pp. 21-22 and 19-20). The closest approximation to a comprehensive defensive realist theory of expansion (as Zakaria defines it) is offered in Jack Snyder‘s Myths of Empire, but that theory differs from Zakaria’s explication of defensive realism, and Snyder’s book is primarily an attempt to explain overexpansion-expansion that becomes irrational and self- defeating-not expansion per se.% It is not surprising that the hypothesis that states expand when they become more threatened and insecure does not withstand empirical scrutiny. If the conditions for its operation are not specified, this hypothesis generates absurd predictions. It would imply that weak states would always respond to threats from their powerful neighbors by attempting to expand. It would have predicted, for example, that Lithuania would have built an empire or invaded the Soviet Union as the Soviet threat to it increased during the 1920s and 1930s. The claim that increasing threats cause increasing expansion also implies that deterrence always fails: threats would provoke expansion, and every conflict would become a spiral of increasing hostility and aggre~sion.~~Given the

31. This list is from Snyder, Myths of Empire, pp. 21-26. As I note above (n. 27), Snyder agrees that powerful states tend to expand because they can. Other defensive realists might offer slightly different lists. See, for example, Lynn-Jones, “Offense-Defense Theory and Its Critics,” pp. 668469. -I _- 32. bid., p. 22. 33. See, in particular, Walt, The Origins of Alliances; Walt, Revolution and War; and Van Evera, Causes of War. 34. Myths of Empire presents a defensive-realist theory to explain expansion, but offers a domestic-level theory to explain overexpansion. Other works of defensive realism explain aspects of expansion. See, for example, Van Evera, Causes of War; and Walt, Revolution and War. They tend to focus on explaining attempts to expand by war, however, instead of the broader form of expansion considered by Zakaria. 35. The deterrence and spiral models are discussed in Robert Jervis, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976), pp. 58-113.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 272

absurdity of such predictions, it is unremarkable that Zakaria’s state-centered realism fares better empirically than what he misconstrues as defensive realism. Zakaria’s case selection also makes his study a poor test of his ersatz version of defensive realism. Although he argues that the United States between 1895 and 1908 is a ”most-likely” case for defensive realism, because the apparent security of the United States should have explained its nonexpansion (p. 431, the relative paucity of significant threats to the United States actually makes it highly unlikely that defensive realism (as defined and explicated by Zakaria) will explain much of U.S. foreign policy during this period. Theories that explain events on the basis of international threats (as Zakaria’s interpretation of defensive realism does) fare poorly when threats are minor or nonexistent. This is the central point of Arnold Wolfers’s well-known story of the behavior of people in a house that is on fire: the imminent threat compels them all to adopt the uniform behavior of fleeing to the exits. In a house that is not on fire, behavior is far less uniform.36In the concluding chapter of From WeaItk to Power, Zakaria recognizes that his interpretation of defensive realism tends to apply ”in periods of international turmoil. When a state must act to ensure its survival and territorial integrity” (p. 186). This conclusion undermines his earlier claim that the cases he examines are a “most-likely” test for defensive realism. Zakaria also contradicts his earlier claim that these cases are “least- likely” (p. 43) cases for state-centered realism: “When power is high and threats low, a state has greater freedom to determine its interests and pursue them, and under such circumstances state-centered realism is a good guide to behavior” (p. 186). There is a second reason why U.S. expansion between 1865 and 1908 is a weak test of Zakaria’s version of defensive realism: the value of the independent variable-the level of external threat to the United States-remains low and changes modestly or not at all during this period. On the other hand, the values of the independent variables of state-centered realism, national power and state power, change dramatically. Whether measured by economic output, military capability, population, or state strength, U.S. national and state power increase greatly between 1865 and 1908. A better test would consider at least one additional case in which power declined and threats increased, or

36. Arnold Wolfers, Discord and Collaboration: Essays on International Politics (Baltimore, Md.: Johns Hopkins University Press, 1962), pp. 13-14. For a similar argument, see Posen, The Sources of Military Doctrine, p. 59.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise 1 173

cases in which power was constant and threats varied dramatically. Such cases would offer a better evaluation of how much of the variance in outcomes can be explained by each theory. If Zakaria had correctly specified defensive realism’s hypothesis on the causes of expansion, US. foreign policy in the second half of the nineteenth century would generally support defensive realism. (Recall that defensive realism, properly formulated, predicts that a state will expand when expansion makes that state more secure.) Between 1865 and 1889, the United States enjoyed a high degree of security. During the first half of the nineteenth century, it had expanded vigorously and achieved a preponderant position on the North American continent. Two oceans and the British Navy separated it from other great powers that might have posed a threat. As Zakaria correctly points out (pp. 177-178), U.S. decisionmakers did not see the Royal Navy as a benign shield. Nevertheless, the British fleet-and British policies to prevent the emergence of a hegemonic power in Europe-ensured that no other great power could pose a threat to the United States, and the possibility of a U.S. invasion of Canada deterred Britain from threatening the United States (and vice versa). Under these conditions, U.S. attempts to expand could not add much to U.S. security and might even reduce it by provoking fear and opposition in other countries. The overall record of U.S. foreign policy during this period is therefore remarkably consistent with defensive realism, particularly when it is compared to the behavior of other great powers. Finally, regardless of how the competing hypotheses are framed, it is difficult to evaluate the empirical evidence in From Wealth to Power. In many cases Zakaria incorporates most of the empirical material into a chronological and analytical narrative instead of a systematic measurement of variables in each opportunity to expand. The result is a lucid description of the period-often in elegant and entertaining prose-but how decisionmakers perceived power and threats, what each theory would have predicted, and how the outcome compares to these predictions is sometimes obscured. Explicitly measuring variables in each case would have yielded a book that was less pleasant to read, but more useful to political scientists who want to examine and possibly replicate Zakaria’s analysis of the evidence.

THE FLAWS OF THE CRITIQUE OF THE LOGIC OF DEFENSIVE REALISM From Wealth to Power also advances four arguments that claim that defensive realism suffers from logical and conceptual errors. These criticisms do not rest on the mistaken assertion that defensive realism predicts expansion in the face

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 174

of threats, but they are ultimately no more persuasive than Zakaria's empirical test of defensive realism. Zakaria's first argument against the concepts and logic of defensive realism holds that the concept of security, which is central to defensive realism, is too ambiguous, "malleable," and "difficult to operationalize" to be useful in constructing a theory (pp. 26-27, 185-186). He claims that virtually any grand strategy can be justified as a quest for security. Napoleonic France, Wilhelmine Germany, and nineteenth-century Britain all adopted expansive territorial goals and justified them in the name of security. Zakaria contends that attrib- uting policies to a quest for security fails to explain why some states attempt limitless expansion while others do not (p. 27). This critique does not withstand scrutiny. The fact that a term is misused by some does not render it useless. Criminal defendants routinely proclaim their innocence, but such proclamations do not mean that the distinction between guilt and innocence is invalid. All that is necessary is an independent means of assessing the validity of a claim. Zakaria's skepticism over whether the ambitious international goals of leaders like Napoleon and Hitler were actually driven by security requirements reveals that he believes that it is possible to assess security more ~bjectively?~In fact, throughout From Wealth to Power Zakaria offers assessments of how secure the United States was. He writes that "after the war [with Spain], the United States was probably more secure than at any point in its history until then" (p. 176), and concludes that the United States "was relatively secure" (p. 87) during the period under study. Zakaria's own assessments of the security of the United States and of the validity of leaders' claims to be acting according to security needs belie his argument that security is ambiguous and impossible to measure. Security may be a difficult concept to define, but so are Zakaria's central variables: expansion, power, influence, and state strength. The second argument that Zakaria advances against the logic of defensive realism is the claim that defensive realism cannot explain much great power behavior, because defensive realists generally assume that "'security [is] plentiful''' (p. 27) and the theory thus regards most cases of expansion as anoma- lies. According to Zakaria, defensive realists are therefore forced to argue that

37. Ironically, Zakaria's argument that the goals of Napoleon and Hitler vastly exceeded the actual security requirements of France and Germany suggests that his own estimate of security requirements is closer to the "minimal security" that he claims defensive realists advocate. Note that Snyder, Myths of Empire, also does not take leaders' statements of their states' security requirements at face value.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America's Rise I 175

domestic factors explain most expansionist policies. He claims that this reliance on domestic, state-level variables turns defensive realism into a "degenerative" research program that invokes auxiliary theories to explain what defensive realists see as the anomaly of frequent great power expansion (pp. 27-28).38 This critique is unpersuasive, for two reasons. First, defensive realists do not believe that security is always plentiful for all states. The quotation about plentiful security that Zakaria cites is from Stephen Walt and describes a condition that exists only "if balancing is the norm, if ideology exerts little effect or is often divisive, and if foreign aid and penetration are rather weak causes" of alliance formation.39 This condition is not a constant feature of international life. Although some defensive realists have explained expansion as a result of domestic factors, they also have identified numerous cases of security-driven expansion. Stephen Van Evera lists the expansion of Athens, Rome, France under Louis XIV, Prussia, Japan, Russia, the United States and the Soviet Union after 1945, Israel, and many more as example^.^' Second, even if some or all defensive realists believed that "security is plentiful," it would

~~ ~~ ~~~ ~~~~~~ ~~~~ 38. For a discussion of what makes a research program "degenerative," see Imre Lakatos, "Fal- sification and the Methodology of Scientific Research Programmes," in Lakatos and Alan Mus- grave, eds., Criticism and the Growth of Knowledge (Cambridge: Cambridge University Press, 1970), pp. 91-197 at 117-118. For a more general discussion of whether neorealism is a degenerating research program, see the articles by John A. Vasquez, Kenneth N. Waltz, Thomas J. Christensen and Jack Snyder, Colin Elman and Miriam Fendius Elman, Randall Schweller, and Stephen M. Walt in American Political Science Review, Vol. 91, No. 4 (December 1997), pp. 899-935. Zakaria, of course, relies on the domestic variable of state strength to enhance offensive realism's explanatory power, but he argues that this modification does not entail "sacrificing much parsimony" and produces "a spare but accurate realist theory of foreign policy" (p. 41). A discussion of whether theories and research programs should be judged by Lakatosian criteria is beyond the scope of this essay. Defensive realists and Zakaria apparently agree that it is better to enhance the explanatory power of a theory than to insist on locating explanations only at the level of the international system. It would be easy to accuse Zakaria of making the same error he attributes to defensive realism. It might be wiser, however, to fault him for criticizing defensive realism for incorporating domestic-level variables. 39. Walt, The Origins of Alliances, p. 49. Walt does argue that these conditions usually exist, but he does not claim that they always exist, and when they do exist security is plentiful only for "most states." See The Origins of Alliances, pp. 49,263,266,268. Moreover, Walt seems to equate "plentiful security" with a world in which "hegemony over the international system [is] extremely difficult to achieve." By this measure, security is often plentiful, because global hegemony by a single state is extremely rare or nonexistent. This standard is not, however, the one that Zakaria uses to determine how secure a state is. 40. Van Evera, Causes of War. To be sure, some-but not d-of these examples of expansion and aggressive war were caused by perceptions of insecurity, not the reality of insecurity, but Zakaria operationalizes defensive realism in precisely the same way. Van Evera's book was, of course, not available to Zakaria when he wrote From Wealth to Power, but he cites the dissertation on which it was based and several articles drawn from it at the head of his list of "important works" of defensive realism (p. 22, n. 37).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 I 176

not undermine the validity of defensive realism as a theory of expansion. The theory would continue to identify the conditions under which expansion is likely, even if theorists disagreed over how frequently those conditions exist.41 For example, in its simplest form offense-defense theory-which is embraced by many defensive realists-predicts that aggressive policies and expansion are more likely when offense has the There is room for debate over how to measure the offense-defense balance and whether it favors the offense or defense at a particular time, but the theory itself provides an international-level explanation of expansion. Zakaria’s third criticism of the logic of defensive realism holds that defensive realists have misunderstood how the microeconomic insights that underpin Kenneth Waltz’s structural realism should be used in constructing a theory of foreign policy (pp. 28-30). He argues that the proper analog to the economic assumption that firms engage in profit-maximizing behavior is the assumption that states seek to maximize their influence.43According to Zakaria, the competitive pressures of the international system and the desire of each state to survive drive all states to pursue this goal and prevent them from having ”limited conceptions of their security.” Defensive realists have erred in assuming that each state will engage in “a minimal attempt to ensure survival” and that it will moderate its foreign policy behavior to avoid international catas- trophes.44Instead, “the tyranny of small decisions” will often lead to adverse and unintended outcomes.45 This critique of the logic of defensive realism makes the mistake of assuming that defensive realism necessarily starts from the premise that states seek minimal security or survival. Defensive realism can, however, start with the

41. For example, Charles Glaser, a prominent defensive realist, explicitly labels his theory “con- tingent realism” because it predicts competitive behavior under some conditions and cooperation under others. See Glaser, ”Realists as Optimists.” 42. On offense-defense theory, see Jervis, ”Cooperation under the Security Dilemma”; Van Evera, “Offense, Defense, and the Causes of War”; Charles L. Glaser and Chairn Kaufmann, ”What Is the Offense-Defense Balance and Can We Measure It?” International Security, Vol. 22, No. 4 (Spring 1998), pp. 44-82; and Lynn-Jones, ”Offense-Defense Theory and Its Critics.” 43. Zakaria traces this error to what he sees as Kenneth Waltz’s “confused and contradictory” (p. 30) writings in Theory of International Politics and Man, the State, and War (New York Columbia University Press, 1959). Zakaria argues that Waltz sees the assumption of profit-maximization ”as a parallel to state behavior” (p. 30) but then assumes that states have survival, not influence- maximization, as their goal. He contrasts Theory of International Politics, pp. 74-76, 117-119, and 136-137, with Man, the State, and War, pp. 37-38, 227. 44. For other discussions of whether defensive realists assume that states have minimal security requirements that lead them to attempt to maintain the status quo, see Labs, “Beyond Victory,” pp. 7-11; and Schweller, “Neorealism‘s Status-Quo Bias.” 45. The term ”tyranny of small decisions” is credited to Alfred E. Kahn.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America's Rise 1 177

assumption that states seek to maximize their security and follow the same "maximizing" logic that Zakaria believes should underpin any structural theory of international politics. If security is defined as "one minus the probability that a state will be conquered or destroyed" security becomes a continuous variable.46A state cannot achieve absolute security in international politics without becoming a universal hegemon. Because global hegemony is unlikely-and a true world-state would mean that international politics no longer existed-states can, at least for the purposes of constructing a theory, be assumed to constantly seek security.47Defensive realism thus need not rest on the assumption that states seek finite, or minimal, security. This critique also misses the mark because Zakaria does not acknowledge that there is a difference between a state's aspirations and what it actually tries to achieve in a given situation.48A theory can begin with an assumption that actors attempt to maximize a certain objective (security, profits, or whatever) while simultaneously predicting that structural or systemic constraints will force those actors to settle for less. Defensive realism does not argue that states should have limited conceptions of security; it suggests that in some cases limited means (i.e,, less expansionist policies) are more likely to maximize security.49Instead of "relentlessly pursuing great influence" (p. 301, states are

46. I suggest this definition in "Offense-Defense Theory and Its Critics," pp. 664-665, n. 10. Note that this is a definition of the term "security," not a statement of the perceived security requirements of a given state or of how much security it can actually obtain at a given time. For another definition that also casts security in probablistic terms, see Andrew Kydd, "Sheep in Sheep's Clothing: Why Security Seekers Do Not Fight Each Other," Security Studies, Vol. 7, No. 1 (Autumn 1997), p. 121. Not all defensive realists may share my definition of security; my point in advancing it is to suggest that a defensive-realist theory can be based on an assumption of security-maximizing. 47. Assuming that states maximize security is preferable to assuming, as Zakaria does, that states maximize influence, because security is the ultimate end of states and influence is only a means to that end. There may be situations in which the quest for influence reduces security by provoking opposition and balancing alliances from other states. In an earlier essay, Zakaria suggests that there is only a semantic difference between assuming that states seek to maximize their influence and the assumption that they "constantly seek survival," which is essentially equivalent to assuming that they maximize their security as I have defined it. See "Realism and Domestic Politics," p. 194. This would only be true if influence-maximizing is always the best strategy for maximizing security. If one assumes that states assess the costs and benefits of any strategy (including the pursuit of influence) in terms of whether it increases or reduces their security, there may be cases in which the difference is more than semantic. 48. Zakaria's argument that states expand in an essentially rational manner, weighing the costs and benefits of expansion, seems to recognize this difference. This argument is an important qualification to the influence-maximizing assumption, because it recognizes that aspirations are not the same as behavior. Zakaria does not, however, discuss the argument in his critique of the logic of defensive realism. 49. I thank Charles Glaser for his advice on this point.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 1 178

compelled by systemic pressures to settle for as much security as they can ”produce” or ”buy” in a given situation. Thus security-maximizing states that engage in the kind of cost-benefit calculations that Zakaria attributes to them may recognize that international conditions dictate that they cautiously pursue a limited degree of security. The level of security that a given state can attain may change as international conditions change, because international politics is a dynamic process. For example, shifts in the offense-defense balance or the distribution of power affect how much security states can obtain. Zakaria’s fourth criticism of the logic of defensive realism suggests that the theory depends on the incorrect assumption that states can learn the correct lessons from history and avoid self-defeating expansionist beha~ior.~’For example, defensive realists conclude that alliances form to balance against potential hegemons and therefore states will realize that an excessively threatening policy-such as a bid for hegemony-will backfire. Zakaria argues that this assumption confuses what states actually learn with what defensive realists think “states should learn” (p. 31, emphasis in original). He suggests that states often learn ambiguous or even contradictory lessons from history, and that the lessons that states actually learn may not be the ones that defensive realists believe they should learn. Zakaria overlooks the fact that the mechanisms of defensive realism can work even if states do not base their behavior on the “lessons of history.” States that rationally calculate the costs and benefits of expansion will recognize when expansion does not pay and adjust their actions accordingly. From a defensive-realist perspective, a state that expands or otherwise acts in a way that reduces its security is likely to change its policies.51 For example, if an expansionist state provokes a large coalition to balance against it and to threaten it, it may retrench and increase its security by limiting its expansion. States can learn from their own behavior, even if they do not rely on lessons derived from history. In addition, states often will rely on lessons from pre- vious experiences when they assess the costs and benefits of their international actions, including e~pansion.~’

50. For a clear argument that states are influenced by the lessons of history, see Posen, The Sources of Military Doctrine, pp. 67-69. See Jervis, Perception and Misperception in International Politics, chap. 6, for a discussion of why leaders often learn the wrong lessons from history. 51. Offensive realism may rely on similar mechanisms. See Labs, ”Beyond Victory,” p. 13. For a discussion of the relationship between learning and neorealist theory, see Jack S. Levy, “Learning and Foreign Policy: Sweeping a Conceptual Minefield,” International Organization, Vol. 48, No. 2 (Spring 1994), pp. 296-298. 52. Robert Gilpin argues that ”foremost among the determinants of these perceptions [of costs and benefits] is the historical experience of the society. What, in particular, have been the consequences

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise I 179

Weaknesses of State-Centered Realism

Zakaria’s state-centered realism is a plausible first-cut theory that rests on an almost incontestable insight: states that have more capabilities demand more from the international system. The application of the theory to the United States in the late nineteenth century reveals that it is not a conceptual abstrac- tion. Actual decisionmakers based their perceptions of US. interests on their assessment of U.S. power. It is not clear from Zakaria’s evidence, however, that assessments of state power were the most important determinants of U.S. policy. In addition, it is important to recognize the limits of state-centered realism. It is not so much wrong as incomplete. The theory does not answer some important questions.

STATE POWER AND U.S. EXPANSION From Wealth to Power does not present a convincing case that the United States became increasingly activist and expansionist when its state became stronger internally. Some of the evidence offered in From Wealth to Power appears to support Zakaria’s argument that the United States expanded when its leaders saw the internal power of the state increase. But one important interpretation deserves particular scrutiny. Was the absence of U.S. expansion between 1865 and 1889 the result of a weak state or of the lack of political consensus? Did clashes between the executive and legislative branches during this period reflect the weakness of the American state or did they reveal divergent assessment of national power and interests? If congressional opposition to plans for expansion reflected political and policy disagreements more than state structure, then Zakaria’s attempt to explain U.S. policy as a consequence of state strength or weakness becomes less persuasive. Zakaria attempts to answer this objection (pp. 60, 63, 70, 87-88), but does not demonstrate that state structure was the root cause of executive-congressional disagreements.

THE LIMITS OF STATE-CENTERED REALISM The most significant limitation of state-centered realism is that it does not appear to offer predictions about the magnitude and character of a state’s expansion. Increased power leads to expansion, but how great is this expansion and does it lead to war? What factors cause states to engage in acute and

for the country from past attempts of its own and others to change the international system, and what lessons has the nation learned about war, aggression, appeasement, etc?” See Gilpin, War and Change in World Politics, p. 51. Zakaria cites this book approvingly (pp. 20,180). See also Posen, The Sources of Military Doctrine, p. 68.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Infernational Security 23:2 1 280

self-defeating overexpansion? No theory can explain everything, and state- centered realism should not be expected to do so. Nevertheless, for political scientists and decisionmakers, the most interesting issue is when expansion leads to war and when it does not, and state-centered realism is silent on this issue. This problem is a consequence of three features of state-centered realism. First, Zakaria’s broad definition of expansion lumps together dispatching am- bassadors and launching invasions. Thus the acquisition of the Philippines, for example, counts as one example of expansion, as does Theodore Roosevelt’s role in negotiating an end to the Russo-Japanese War. Second, state strength, Zakaria’s key variable in the US. case, explains ”underexpansion” much better than it explains overexpansion. A country with a weak state may expand less than the imperatives of influence-maximizing would suggest, but state-centered realism does not predict how much a state would expand in the absence of domestic constraint^.^^ Factors identified by defensive realism may offer better explanations of why some expansions are rapid and large, and lead to war, while others do not.54 Third, state-centered realism has difficulty explaining the magnitude and character of expansion because it relies entirely on changes in levels of relative national (and state) power to explain expansion. In practice, however, states are most likely to expand when opportunity combines with motive to make expansion an attractive route to security. Zakaria assumes that all states share a basic desire to maximize their influence, but how vigorously they pursue this goal depends not only on their relative power, as Zakaria claims, but also on the other international incentives for expansion that they confront (e.g., the magnitude of the threats they confront, and whether expansion is an effective response to those threats). Elements of defensive realism can enhance realist explanations that begin with measurements of power.55

Conclusions and ImpZications

From Wealth to Power makes three major arguments, with varying degrees of success. First, it argues that defensive realism is a weak theory that fails to

53. I thank Dale Copeland for bringing this point to my attention. 54. Zakaria might argue that a theory of international politics, not foreign policy, is necessary to explain events like the outbreak of war, but a theory of foreign policy can suffice because it only takes one state to start a war. 55. One could, in fact, argue that adding variables such as threat to analyses based on relative power alone is the essence of the entire defensive-realist research agenda. See Walt, The Origins of Alliances, pp. 263-266; Walt, Revolution and War, pp. 18-19; and Van Evera, Causes of War.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 Realism and America’s Rise I 282

explain why countries expand, and that the theory suffers from conceptual and logical errors. These arguments are not persuasive because Zakaria deduces the wrong hypothesis from defensive realism and tests this hypothesis in cases where it is likely to fail. Zakaria’s criticisms of the concepts and logic of defensive realism raise important issues, but do not undermine the theory. Second, Zakaria argues that states tend to expand when their relative capabilities increase. This argument is the most successful element in From Wealth to Power. Zakaria presents a strong deductive case for this proposition and amasses considerable evidence to show that increased power caused the United States to expand at the end of the nineteenth century. When U.S. decisionmakers became aware of US. capabilities, they sought to expand U.S. interests and territory. Third, Zakaria suggests that state power-the ability of a state to extract and harness its capabilities for international purposes-is more important than overall national power in shaping foreign policy. This argument implies that measurements of power in international relations must incorporate assessments of state strength. From Wealth to Power offers mixed support for the argument, however, because some of the measures Zakaria uses to assess state power are actually measures of the degree of elite consensus. Executive-congressional division, for example, is an indicator of elite dissensus, not necessarily of a weak state. Zakaria’s call for more attention to state power needs to be supported by more empirical evidence from the U.S. and other cases. From Wealth to Power has the following implications for further research and theorizing. First, defensive realists should respond to Zakaria’s critique by clarifying the extent to which international-level factors can explain expansion, aggressive behavior, and war. They should frame and test the hypotheses that defensive realism offers on international expansion and attempt to refute Zakaria’s claim that the search for security does not cause expansion. Defensive realists also need to be more explicit about the extent to which power, as opposed to threat, is an important explanatory variable. And Zakaria’s critique of the logic and concepts of defensive realism, although ultimately unpersuasive, suggests that defensive realists should be more explicit about their defini- tions of key terms such as security. Second, it may be possible to integrate elements of offensive realism and defensive realism to produce better realist explanations of expansion and other international phenomena. Zakaria’s state-centered realism focuses on how increased power gives states an opportunity to expand, but it neglects how other international factors give states a motive to expand. Instead of setting up

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021 International Security 23:2 1 182

offensive and defensive realism as polar opposites and dividing important explanatory variables between them, it might make more sense to combine those variables to produce a more powerful realist theory. Third, additional tests are necessary before scholars can accept Zakaria’s suggestive argument that state strength is an important determinant of foreign policy and grand strategy. If this potential innovation is to bear fruit, state strength needs to be measured objectively, and predictions derived from hypotheses about its impact need to be tested further against contending theories.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec.23.2.157 by guest on 02 October 2021