CIAO DATE: 3/99

Security Under Anarchy: Defensive Realism Reconsidered _*

Jeffrey W. Taliaferro

Department of Political Science
Tufts University

International Studies Association
40th Annual Convention
Washington, D.C.
February 16–20, 1999

DRAFT: Do not cite or quote. Last revised: 1 February 1999

While neoliberal institutionalists and contructivists continue to debate the fundamental nature of world politics, new lines of research and vigorous debates have emerged within realism. ₁ Contemporary or neoclassical realism incorporates both systemic and unit-level variables to explain the foreign policies of particular states and international outcomes. ₂ While not abandoning Kenneth Waltz’s insights about the importance of anarchy on state behavior, neoclassical realism draws insights from classical realism as well. Neoclassical realists uses qualitative methodology and often detailed case-studies to test rival hypotheses, uncover new theoretical or historical puzzles, and generate policy prescriptions. Nonetheless, neoclassical realism is not monolithic. The most significant divide within the neoclassical realist literature is between offensive realism and defensive realism. The two branches share the same core assumptions, but proceed from different auxiliary assumptions about the implications of anarchy and the influence of unit-level variables. ₃

Offensive realists hold that anarchy compels states to maximize their relative power (or influence). ₄ The link between systemic imperatives and the actual foreign policies states pursue is relatively direct. Security and survival in the international system are never assured. States seek to maximize their security by maximizing their relative power and influence, since the stronger states in the international system stand the best chance of survival. Most states, however, do not engage in reckless, unlimited aim expansion (although clearly some do). Instead, states seek to maximize relative power at times and in places where the benefits of doing so outweigh the costs. Relative capabilities largely shape the intentions of states. As a state becomes more powerful, it will attempt to maximize its influence and control its international environment. In general, states pursue aggressive, influence-maximizing strategies when central decision-makers perceive an increase in relative capabilities. Defensive realism posits a more complex and problematic link between systemic imperatives and the behavior of states. Under the security dilemma even true security-seeking states may misperceive each other’s intentions and find themselves in spirals of mutual hostility or outright conflict. Defensive realists do not deny the existence of revisionist states. However, defensive realist theories also identify the conditions under which states will likely engage in security-driven expansion. National leaders’ perceptions of relative capabilities and the fine-grained structure of power are integral links between systemic imperatives and states’ foreign policies. Domestic politics constrain leaders’ ability to both mobilize national resources for national defense and later readjust foreign policies in response to the external environment. ₅

Recently defensive realism has come under attack from both within and outside the realist research program. Critics charge that defensive realism: (1) confuses what states ought to learn from the international system with what states actually learn; (2) fails to account for revisionist states; (3) and places too much emphasis on unit-level “pathologies” in explaining self-defeating foreign policies or negative international outcomes. Other critics, both realist and non-realist, claim that defensive realist theories are indeterminate in positing general explanations for foreign policy and international outcomes. Sometimes unit-level variables are causally important, but at other times, systemic variables are crucial. Sometimes leaders’ perceptions of material variables matter. At other times material variables alone explain state behavior or international outcomes. These criticisms are mistaken.

I argue that defensive realist assumptions generate more descriptively accurate theories of international relations and foreign policy, than do offensive realist assumptions. The paper addresses the major criticisms leveled against defensive realism by offensive realists, as well as some non-realist critics. In does this in two steps. First, the article clarifies the auxiliary assumptions of defensive realism. Offense-defense theory, balance-of-threat theory, window theory, and realist cooperation theory all share a set of implicit auxiliary assumptions regarding the link between systemic and unit-level variables. These assumptions concern: (1) the centrality of the security dilemma even in the (hypothetical) absence of “greedy” states, (2) the fine-grained structure of power, (3) national leaders’ perceptions, and (4) domestic politics. By making these auxiliary assumptions explicit, one can open a more fruitful dialogue between the two branches of neoclassical realism and between non-realists and realists in general.

Second, the paper discusses why the distinction between defensive realism and offensive realism remains valid. Recently several scholars have tried to transcend this dichotomy by positing alternative divisions or classifications. I address two of these newer classifications — post-classical realism and motivational realism. The first proceeds from a misinterpretation of the extant literature. The second refers to a philosophical difference that cuts across the existing divide between the offensive and defensive realism. ₆

The debate between offensive and defensive realism deserves attention for three reasons. First, realism is now (and will likely continue to be) the dominant research program in the study of international relations. It is important to determine which realist theory or theories are the most useful. ₇ Second, by explicitly integrating international relations theory and diplomatic history the two branches of neoclassical realism satisfy the growing demand for theoretical rigor and empirical richness. ₈ Finally, the debate between offensive realists and defensive realists has implications for both the development of international relations theory and the actual conduct of foreign policy. As Robert Jervis observes, “The study of international politics will be impoverished if it is totally divorced from contemporary events and hopes and fears for the future, but if it is to mature, it will have to develop some distance from them.” ₉

The remainder of the paper proceeds as follows. The next section briefly reviews the underlying assumptions of contemporary realism and shows how its offensive and defensive variants fit into the broader paradigm. The third section discusses the auxiliary assumptions that underlie defensive realist theories. The fourth section addresses the continued validity of the offensive-defensive realism dichotomy. Along the way, I respond to the major criticisms of defensive realism. I conclude that none of these criticisms poses a fatal blow for defensive realism and that several of them proceed from incomplete or flawed analysis. The conclusion discusses the broader implications of my argument and makes suggestions for future research.

 

Core Assumptions of Realism

Realism is both a political philosophy and an ever-evolving family of empirical theories. As a philosophical tradition, realism presents a fundamentally pessimistic view of human nature and political life. Realists see politics as a continual struggle among groups for power, prestige, influence and security in a world of scarce resources. This fundamentally pessimistic view of politics more than any particular theory, epistemology, or methodology unites political realists as varied as Thucydides, Jean Jacques Rousseau, Nicoló Machiavelli, Winston Churchill, Thomas Hobbes, Hans Morgenthau, Arnold Wolfers, Kenneth Waltz, Thomas Christensen and John Mearsheimer. ₁₀ There is no single empirical realist theory of international relations or foreign policy. Realism is not synonymous with Waltz’s systemic balance-of-power theory (neorealism) or any particular proposition derived from his theory. ₁₁ All variants of realism, however, share seven core assumptions about international politics.

First, the international system is anarchic. Thus, international politics is fundamentally different from domestic politics. In the international system there is no world government or universal sovereign to adjudicate disputes, enforce agreements, or prevent the use of force. Second, realism assumes that human beings do not face one another primarily as individuals but as members of groups, that commands their loyalty. Robert Gilpin observes, “[T] he name, size and organization of the competing groups... alter over time [but] the essential nature of inter-group conflict does not.” ₁₂ Regardless of their size, territoriality, or internal organizing principle, “states” are the primarily vehicle through which individuals relate to one another and the primary actors in international politics.

Third, the international system is mostly (although not exclusively) responsible for how states conduct themselves in the international arena. By this, I mean that the relative distribution of capabilities among states, as well as power trends, shape the broad parameters of states’ external behavior. Realism is an environment-based body of theories. As Jennifer Sterling-Folker notes, “The impact of structure on state behavior is more akin to an environment that surrounds states and acts as an extrinsic physical condition affecting and influencing the growth and development of states.” ₁₃ Contrary to the assertions of some critics, however, realism does not deny the importance of domestic variables in shaping how a state conducts its foreign policy. ₁₄ All realists, however, reject the notion that unit-level variables are the sole determinant of either individual states’ foreign policy or international outcomes.

Fourth, states are self-regarding, not other-regarding. ₁₅ The desire for survival in the absence of a universal sovereign requires states to look out for their own interests. The international system induces self-help behavior on the part of states. The policy implications of self-help constitute one of the principal divisions between offensive and defensive realists.

Fifth, states adopt purposeful policies in their pursuit of relative power or security. This assumption, often misidentified as realism’s “rationality” assumption, does not imply that states (or their leaders) behave according to the axioms of subjective expected utility theory or even the less stringent norms of bounded rationality. ₁₆ Unlike a neoclassical free market where the units have complete information and perfect foresight, states exist in an inherent uncertain environment. States attempt to respond to changes in the international distribution of power, but one cannot speak of optimizing or maximizing behavior in any meaningful sense. ₁₇ Sixth, states rely on the use of force (or the threat of the use of force) to protect their interests and enhance their security. War is always a possibility in the international system. Seventh, human reason cannot transcend the fundamentally conflictual nature of international politics. ₁₈ This assumption distinguishes realism from both the domestic and the international variants of liberalism. “What liberalism means by process is the interaction between humans as they attempt to reach collective agreements, and what they create in order to assist them in reaching those agreements.” ₁₉ It is not surprising, therefore, that most realists have little confidence in ability of international institutions to facilitate cooperation on the fundamental issues of peace and war. ₂₀

Neoclassical Realism versus Neorealism and Classical Realism

Neither neoclassical realism nor neorealism attempt to systemize the insights of classical realism. ₂₁ Although classical realism, neorealism and now neoclassical realism share a pessimistic worldview and an emphasis on power, the three bodies of literature vastly differ in epistemology, methodology and assumptions. Both neorealism and neoclassical realism finds the permissive cause of war in “third image” variables, whereas classical realism emphasized human nature or the animus dominandi as the permissive cause of conflict. Classical realists such as Morgenthau and Henry Kissinger focused on both individual statesmen and states with no generative mechanism linking the former to the latter. ₂₂ Neoclassical realism emphasizes the importance of power in international politics, which almost all realists now define in terms of material capabilities. Classical realism also emphasized the importance of power in international politics, but included factors such as national interests, statesmanship, and moral choice as well. ₂₃ Moreover, classical realism views power as an end in itself; states define their interests in terms of power and seek to maximize their power. ₂₄ Neorealism and now neoclassical realism see power as means to an end. Classical realism, neorealism and neoclassical realism draw upon different disciplines in constructing theories. Twentieth century classical realists drew heavily on history, sociology, international law, and in some cases, theology. Neorealism borrows heavy from microeconomics, to the exclusion of the other social sciences. ₂₅ Neoclassical realism not only draws upon microeconomics, but incorporates insights from social and cognitive psychology, organization theory, sociology and history, as well.

Neoclassical realism (in both its offensive and defensive variants) is distinct from Waltzian neorealism. Neorealism is strictly a theory of international relations. ₂₆ It addresses two questions: (1) Why do balances of power recurrently form in international politics and (2) which of two gross distributions of power (bipolarity or multipolarity) is more prone to great power war? ₂₇ The theory has only one causal variable — polarity — and one dependent variable — the likelihood of great power war. Neorealism, however, has a limited empirical range. Waltz’s causal variable has arguably only changed twice in the past two hundred years: from multipolarity to bipolarity in 1945 and from bipolarity to unipolarity in 1989-1991. ₂₈ Neorealism makes no empirical claims with respect to the foreign policies of particular states. A strictly systemic theory, Waltz observes, “can tell us what pressures are exerted and what possibilities are posed by systems of different structure, but it cannot tell us just how, and how effectively, the units of a system will respond to those pressures and possibilities.” ₂₉

Neoclassical realists seek to explain both international outcomes and the foreign policies of individual states. Barry Posen writes: “Waltz stresses the influence of general systemic constrains and incentives on the behavior of all states, and on the behavior of the system as a whole. Students of ‘Realpolitik’ focus on how these general constraints and incentives combine with the unique situations of individual states to lead them to specific foreign or military polities.” ₃₀ Furthermore, while building on Waltz’s assumptions about anarchy, neoclassical realists explicitly reject the injunction that theories ought not to include explanatory variables at different levels of analysis. ₃₁ Fareed Zakaria writes, “a good account of a nation’s foreign policy should include systemic, domestic, and other influences, specifying what aspects of policy can be explained by what factors.” ₃₂ As Gideon Rose observes:

[Neoclassical realism] explicitly incorporates both external and internal variables.... Its adherents argue that the scope and ambition of a country’s foreign policy is driven first and foremost by its place in the international system and specifically by its relative material power capabilities. This is why they are realists. They argue further, however, that the impact of such power capabilities on foreign policy is indirect and complex, because systemic pressures must be translated through intervening variables at the unit level. This is why they are neoclassical. ₃₃

 

The Offensive-Defensive Realist Debate

The debate between offensive realism and defensive realism is not primarily an empirical one. Instead, the debate stems from four auxiliary assumptions about the relationship between systemic and unit-level variables. These assumptions are implicit in prominent defensive realist theories of international relations and foreign policy. Below I codify defensive realism’s auxiliary assumptions, discuss their broader theoretical implications, and address various criticisms. Table 1 below illustrates these auxiliary assumptions, the variables involved and the levels of analysis at which we would expect them to operate.

[Table 1 here, currently unavailable]

 

The Security Dilemma

The security dilemma is the core concept that unites all defensive realist theories of foreign policy and international relations. Robert Jervis defines the security dilemma as a situation “in which the means by which a state tries to increase its security decreases the security of others.” ₃₄ The anarchic nature of the international system produces uncertainty. States can never be certain of other states’ present or future intentions. As discussed above, anarchy induces self-help behavior on states. States undertake security-seeking policies vis-à-vis potential adversaries. Adversaries, however, react to having their own relative security reduced. They take steps to increase their own security, thus mitigating the security policies of the first state. Nevertheless, why should this be the case? Why should the efforts of one state to make itself secure make other states less so?

Charles Glaser posits three ways through which making one’s adversaries insecure can prove self-defeating, leaving the state worse off than before its initial action. ₃₅ First, even purely security seeking policies can set in motion a process that reduces the state’s own military capabilities — the ability to perform particular military missions. An adversary may buy or manufacture additional arms in order to restore its military capabilities. This action-reaction process, however, may not leave both states worse off, because it shifts the offense-defense balance. If each side deploys a new weapon system that favor offensive military operations in response to the other, both side will be less able to defend themselves and thus be less secure. Conversely, if each side deploys a new weapon system that favor defensive military operation in response to the other, both sides will increase their capacity for self-defense and thus become more secure. This action-reaction dynamic may also vary with the size and composition of each side’s armed forces and the particular missions those forces are equipped to undertake.

Second, self-help strategies may increase the value an adversary places on expansion as a means of self-defense, which in turn makes deterrence harder. Both defensive and offensive realists agree that expansion can increase a state’s security; they disagree over whether expansion always generates increased security. ₃₆ Under certain circumstances, defensive realists hold expansion can provide a state with more secure borders, strategic depth, and control of natural resources and industries valuable for building military capabilities and enhancing wealth. ₃₇ In general, security-driven expansion becomes more likely when leaders perceive offensive weapons and military doctrines to be more efficacious than defensive ones. In these situations, a state has a strong incentive to expand, if it believes that doing with redress the power imbalance (and security) between with its adversary and if it has the military capabilities to undertake such expansion. ₃₈

Military build ups and the development of new weapons technologies can increase an adversary’s insecurity by making it less able to defend itself. New alliances or the expansion of existing alliances to include neutrals or an adversary’s former allies can have the same effect. For example, fear of a possible Anglo-French alliance was the principal reason why Wilhelmine Germany initiated the 1905 Morocco Crisis. ₃₉ Moreover, both military build-ups and alliances can change the adversary’s beliefs about the state’s motives. An adversary may conclude that a state harbors “greedy” motives — that is a desire to expand for reasons other than security. Finally, self-help strategies, such as arms build-ups, may simply be a waste of finite resources.

Critics of defensive realism claim that the security dilemma ignores the problem of “greedy” states. Revisionist states, not the security dilemma, are the prime cause of conflict in the international system. Zakaria faults defensive realism for assuming states pursue “a minimal attempt to assure survival” and only expand when threatened. ₄₀ Schweller writes, “Predatory states motivated by expansion and absolute gains, not security and the fear of relative losses, are the prime movers of neorealist theory. Without some possibility of their existence, the security dilemma melts away, as do most concepts associate with contemporary realism.” ₄₁ Andrew Kydd presents the most rigorous deductive critique of the security dilemma to date. He constructs a formal model in which all states in the international system are security seekers. The model assumes that all states have complete information about one another’s preferences. The result would be no-war equilibrium: no state would have an incentive to arm against or attack another. This equilibrium would exist regardless of the relative distribution of power, the offense-defense balance, trends in power growth, or the cumulatively of resources. He then complicates the model by adding uncertainty about future intentions. Uncertainty may undermine the no-war equilibrium, since greedy states cannot achieve their maximum pay-off by not attacking. Security-seekers, on the other hand, may have incentive to launch preemptive or preventive attacks to forestall later conflicts on less favorable terms.

While states can never be completely certain of one another’s present or future intentions, states posses abundant means to reduce uncertainty to manageable levels. Two factors, according to Kydd, combine to give states adequate information about each other’s present motivations. First, for modern democratic states, the policy making process is so open that they cannot help but reveal their true preferences. “Thus if a democracy is a security-seeker, other states will know this, uncertainty about present motivations will be diminished, and cooperation will often result between the democracy and the security-seeking state. Of course, if the democracy is greedy, this will revealed as well, producing justified distrust on the part of other states and a lack of cooperation.” ₄₂ Second, all states, both democracies and non-democracies, can employ costly signals to reveal their benign intentions. “A costly signal is an act which one type of actor in a game can take that other types would find too costly. It thereby serves to differentiate the type from the other types and to identify it to other players.” ₄₃ He cites ideological moderation, policies toward domestic minorities, weaker neighbor states, and arms control as examples of such costly signals.

Schweller, Kydd and Zakaria raise interesting questions about the security dilemma. None of those critiques, however, invalidates the centrality of the concept. Zakaria and Schweller are incorrect in claiming that defensive realism assumes states merely pursue minimal security or survival. Instead, defensive realists assume (as do offensive realists) that at minimum all states seek to preserve their territorial integrity and political autonomy. This assumption does not preclude the existence of states seeking more than territorial integrity and continued autonomy. As Sean Lynn Jones points out, defensive realism can begin with the assumption that states seek to maximize security and “follow the same maximizing logic that Zakaria believes should underpin any structural theory of international politics.” ₄₄ Even if a defensive realist theory adopts the assumption that states seek to maximize security there were always be some systemic constraints that prevent states from achieving that goal. Defensive realists recognize that no state in the international system can be “absolutely” secure without becoming a global hegemon. Since the chances of establishing global hegemony are extremely low (and since the establishment of a world state would end international politics), for the purposes of theory construction, one can assume that states constantly seek security. ₄₅

There will always be uncertainty about states’ current and future capabilities and interests in an anarchic environment. No state has perfect information about another state’s intentions. Quite frequently, decision-makers face either a paucity of information about other states’ capabilities and intentions or the opposite problem of information overload. This is particularly true in crises or in areas of disputed great power influence. In the July 1914 Crisis, for example, uncertainty about Great Britain’s commitment to Belgium’s defense contributed to both French and German miscalculations. ₄₆ Kydd admits no amount of costly signaling between states can eliminate the problem of uncertainty.

Kydd relies on essentially Innenpolitik argument to challenge the existence of the security dilemma. ₄₇ He places considerable emphasis on costly signaling and the transparency of democratic political systems. He claims that his argument differs from the normative and institutional variants of the democratic peace thesis. Upon closer examination, however, this aspect of Kydd’s argument against the security dilemma is virtually identical to those of normative democratic peace theorists. Security-seeking states can employ costly signals of benign intentions — ideological moderation, unilateral arms control, and non-aggressive policies toward domestic minorities and weaker neighbor states. Greedy states would not undertake such costly signals. Does this differ from the normative democratic peace propositions? It does not. Democratic states supposedly project internal norms of peaceful conflict resolution outward. This signals their benign intentions toward other democracies. This mutual recognition of shared norms for domestic conflict resolution, supposedly produces a shared, but exclusive peace among democratic states. ₄₈

Some democratic peace theorists acknowledge that is possible (but highly improbable) that democratic states might wage war on each other, but claim that the historical record thus far shows democracies have almost never gone to war with other democracies. ₄₉ Of course, most of the empirical evidence in support of the democratic peace proposition involves very narrow definitions of democracy and war. ₅₀ The past two centuries have seen many cases in which (1) pairs of security-seeking democratic states (or rather, democratic states who leaders’ perceived themselves as security-seeking) stumbled into or came perilously close to stumbling into inadvertent conflicts, ₅₁ (2) the security-seeking intentions of democratic states were unclear, ₅₂ and (3) democratic and non-democratic states undertook security-seeking policies that other side found threatening. ₅₃

Fine Grained Structure of Power

The second auxiliary assumption closely relates to the security dilemma. Defensive realists assume that the fine-grained structure of power has more direct impact on the behavior of states than does the gross distribution of power in the international system. ₅₄ The ease or difficulty with which a state can enhance its relative security or power will vary with the fine-grained structure of power. How does the fine-grained structure of power differ from the gross structure of power?

The gross structure of power refers to the relative share of the international system’s total resources each state controls. The fine-grained structure of power refers to the relative distribution of material capabilities that enable states to carry out particular diplomatic and military strategies. ₅₅ This in turn influences the severity of the security dilemma between particular states or in regional subsystems. Thus, one can think of the fine-grained structure of power as mediating the effects of systemic imperatives on the behavior of states.

Offense-defense theory addresses the best known components of the fine-grained structure of power — the offense-defense balance and offense-defense distinguishability. ₅₆ The former refers to the relative ease or difficulty of conquering territory. Offense-defense distinguishability, on the other hand, refers to potential adversaries’ ability to distinguish offensive weapons and military postures from purely defensive ones. ₅₇ The fine-grained structure of power, however, is broader than the objective offense-defense balance between states. It also includes, but is not limited to the material components of threat, the rate of technological innovation, diffusion and emulation among states, and states’ transportation and communications infrastructure. Each of these factors will influence a state’s ability to defend or conquer territory and extract resources from occupied territory.

The fine-grained structure of power provides incentives or disincentives for particular types of diplomatic and military strategies. If the objective balance in military technology favors the defense of territory, all else equal, states would have an incentive to pursue defensive military strategies. The nuclear revolution, specifically the development of secure second-strike capabilities by the declared nuclear states, provides strong disincentives for conquering territory. Mutual Assured Destruction (MAD) has effectively made intended war a non-option for dyads of declared nuclear states. ₅₈

The fine-grained structure of power and its various components are not systemic variables. ₅₉ Rather the different components of the fine-grained structure of power — threat the offense-defense balance, technological innovation and diffusion, states’ transportation and communications infrastructure — are intervening variables between systemic and unit-level variables. For example, it makes little sense to speak of a systemic offense-defense balance. ₆₀ “The offense-defense balance is well defined only for specific dyads of states, not for the entire international system.... Consequently, the offense-defense balance will frequently vary across dyads.” ₆₁ Similarly, balance-of-threat theory does not posit (or even imply) that states generally balance against the greatest threat in the international system. Rather, they generally balance against states that pose an immediate threat to their survival or interests. The scope of those interests, in turn, depends on a state’s relative material capabilities. Recall Walt defines threat as a composite of a state’s aggregate power, offensive military capabilities, geographic proximity and perceived aggressive intentions. ₆₂ All but the last of theses are components of the fine-grained structure of power, since they pertain to specific dyads or regional subsystems. A state’s perceived aggressive intentions, as I will argue below, involve the perceptions of national leaders.

The fine-grained structure of power assumption generally provokes the most outcries from critics of defensive realism. As with the criticisms of the security dilemma, none of these criticisms withstands scrutiny. Zakaria, among others, faults defensive realism for allegedly assuming security is plentiful in the international system. He writes: “By smuggling in normative assumptions about state behavior, defensive realism ends up regarding much foreign policy as abnormal and then explaining it by attributing abnormality to guilty parties.” ₆₃ Similarly, Mearsheimer writes, “Some defensive realists argue that the constraints of the international system are so powerful that conquest rarely pays, and therefore the world should be populated with only status quo powers. Revisionist powers are almost sure to lose if their act on their malevolent intentions, so it makes no sense for states to act aggressively toward each other.” ₆₄

As others note, these criticisms of defensive realism largely stems from (admittedly confusing) quotations by prominent defensive realists taken out of context. ₆₅ Stephen Walt, for example, states that security under anarchy would only be plentiful “ 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. ₆₆ Balancing is the predominant form of alliance behavior in international politics, but defensive realists do not believe balancing is ubiquitous or that it always happens in a timely fashion. ₆₇ Similarly, Van Evera writes, “states are seldom as insecure as they think they are. Moreover, if they are insecure, this insecurity often grows from their own efforts to escape imagined insecurity.” ₆₈ This does not imply, however, that objective security is plentiful in the international system. Rather, because of anarchy and uncertainty most states (including revisionist ones) will tend to over-estimate the degree of insecurity they face. Furthermore, the historical record abounds with cases of states that pursued security-driven expansion policies: Prussia during the Seven Years’ War, France under Louis XIV, Austria-Hungary (1914-1918), Wilhelmine Germany (1897-1918), Japan (1931-1945), and the United States intervention in Korea and Vietnam. “Modern great powers have been overrun by unprovoked aggressors only twice, but they have been overrun by provoked aggressors six times — usually by aggressors provoked by the victim’s fantasy-driven defensive bellicosity.” ₆₉

A second criticism of the fine-grained structure of power assumption revolves around issues of definition and measurement. Some offensive realists claim that defensive realist concepts — security seeking, the offense-defense balance, and threat — are too malleable, ambiguous and “difficult to operationalize” to be of any use in theory construction or testing. ₇₀ This criticism does not withstand scrutiny. One can make the same argument against any concept or variable of interest. The international relations literature abounds in competing (and contradictory) definitions of power, state strength, alliances, institutions, ideas, belief systems, and norms. ₇₁

Leaders’ Perceptions

Defensive realism’s third auxiliary assumption specifies the link between material capabilities and states’ behavior. The relative distribution of capabilities and the fine-grained structure of power influence state behavior through the medium of national leaders’ perceptions. Central decision-makers do not always update their estimates and policy preferences in response to new information about relative capabilities. They often inadvertently draw upon historical analogies and other cognitive short cuts to process incoming information and select policy options. ₇₂ Many defensive realists posit an explicit role for such pre-existing belief systems, images of adversaries, cognitive biases, loss aversion and risk propensity in the process of intelligence gathering, net assessment, military planning and foreign policy decision making. As Rose notes:

Neoclassical realists... argue that the notion of a smoothly functioning transmission belt [between systemic variables and unit-level behavior] is inaccurate and misleading. The international distribution of power can drive countries’ behavior only by influencing the decisions of flesh and blood officials, they point out, and would-be analysts of foreign policy thus have no alternative but to explore in detail how each country’s policymakers actually understand their situation. What this means in practice in that the translation of capabilities into national behavior is often rough and capricious over the short and medium term. ₇₃

Much of what neoclassical realists (both offensive and defensive realists) seek to explain would be simply inexplicable without reference to the perceptions of central decision-makers. ₇₄ “All policies are future-oriented... A decision to reform, retrench, or go to war reflect expectations about future trends and assessments of the likely effect of today’s policies on tomorrow’s distribution of power resources.” ₇₅

The role of such perceptional variables becomes particularly important during periods of power transition. ₇₆ Consider, for example, Great Britain’s response to relative decline between 1895 and 1905. Aaron Friedberg finds that senior officials, particularly in the Treasury and the Admiralty, relied on crude indicators of relative economic and military power. Reliance on such measures downplayed “the significance of major political, doctrinal, economic, and technological developments” in the last years of the nineteenth century. ₇₇ Christensen finds France failed to balance against the rising threat of Prussia on the eve of the Seven Weeks’ War (1866), largely because Napoleon III and his generals overestimated Austria’s military strength and perceived military technology as defense dominant. Jeffrey Taliaferro finds that loss aversion and perceptions of relative vulnerability vis-à-vis the West and the Soviet Union drove Japanese foreign policy in 1940-1941. This not only led the Japanese cabinet and military chiefs of staff to devote additional resources to a failing war against Nationalist China, but also to pursue high-risk strategies culminating in war with the vastly stronger United States in 1941. ₇₈

Perceptional variables play an important role in foreign policy even during non-crisis periods and periods when the “objective” gross distribution of power remains relatively stable. William Wohlforth finds that during the Cold War Soviet leaders often reacted to perceived shifts in the superpowers’ relative capabilities and prestige, not just the actual distribution of capabilities. He finds that “perceptions of power are more dynamic than measurements of material relationships. Rapid shifts in behavior may be related to perceived shifts in the distribution of power which are not captured by typical measures of capabilities.” ₇₉ Benjamin Miller finds that benign images of the opponent, balancing beliefs, and ideological similarity, along with multipolarity, the presence of common fears and a regional images of third-area conflicts, are necessary conditions for the emergence of great power concerts. This explains, in part, the creation and durability of the Concert Europe (1815-1854), as well as the failure of the United States and the Soviet Union to form a superpower condominium in the Middle East. ₈₀

Finally, leaders’ perceptions play a critical and at times, pernicious role in shaping how states respond to the fine-grained structure of power. Often time the “objective” offense-defense balance in military technology is sharply at odds with civilian and military leaders’ perceptions of the offense-defense balance. Officials often draw upon the “lessons of history” in formulating military doctrine or allow organizational priorities to override legitimate security requirements. ₈₁ The most oft cited instance in which this happened is the so-called “cult of the offensive” in Europe in the decade before the First World War. The prevailing military technology in Europe at the turn of the century made it easier to defend territory than to conquer territory. Nonetheless, drawing upon the “lessons” of the Franco-Prussian War (1871), the European great powers adopted offensive military doctrines culminating in the French general staff’s Plan XVII in 1913 and the Russian general staff’s Plan 20 in 1914. ₈₂

The difference between offensive realism and defensive realism on the role of perception is really a matter of degree. Defensive realist theories tend to treat leaders’ perceptions of material capabilities as an essential intervening variable. Indeed, most variants of offense-defense theory explicitly turn on leaders’ perceptions and misperceptions of the actual offense-defense balance. ₈₃ Offensive realists, on the other hand, either rely on perceptions to operationalize explanatory variables or address them in an ad-hoc manner.

Zakaria’s offensive realist theory of foreign policy posits increases in state and national power generate increased international ambition, but that national leaders only expand at times and in places than minimize costs and risks. He claims that defensive realism predicts states expand when faced with increases in threat. Instead of attempting to construct and then use objective measures of relative power and threat, Zakaria operationalizes these explanatory variables in terms of central decision-makers’ perceptions. Similarly, Eric Labs’ war aims theory turns in large part on civilian and military officials’ perceptions of relative power and strategic opportunities. Great powers expand their war aims in response to strategic opportunities that may arise as a result of events on the battlefield or belief that that costs of expansion are low. ₈₄

Schweller, on the other hand, does rely on objective measures of relative power. He uses the Correlates of War data set to measure the distribution of total resources (military and economic) among the great powers in the late 1930s and early 1940s. Based on those rankings, he further divides the great powers into two tiers: poles (Nazi Germany, the Soviet Union, and the United States) and lesser great powers (Great Britain, France, Italy and Japan). Such measures are crucial to both theoretical predictions of balance-of-interests theory and Schweller’s empirical claim that the international system immediately preceding World War II was tripolar, not multipolar. Alliance behavior in Europe and East Asia between 1935 and 1942, Schweller argues, stemmed from both objective tripolarity and the revisionist aspirations of two poles (Germany and the Soviet Union) and two lesser great powers (Italy and Japan).

Nonetheless, leaders’ perceptions and misperceptions of relative capabilities play an important, but largely unexplained, role in both balance-of-interests theory and Schweller’s historical analysis of World War II. Hitler correctly perceived the global international system as tripolar and the European subsystem as bipolar. ₈₅ Stalin, on the other hand, misperceived the actual distribution of power in Europe as tripolar (Britain, Germany, and the Soviet Union). The Soviet leader believed Britain and Germany would exhaust each other in case of general war, therefore, the Soviet Union had no need not balance against Hitler. ₈₆ Balance-of-interests theory is silent on why Hitler correctly perceived the “true” distribution of power while Stalin did not. Furthermore, it is less clear that the heads of state and government of the other great powers (Neville Chamberlain, Eduard Daladier, Benito Mussolini, Franklin Roosevelt, and Konoe Fumimaro) correctly perceived the “true” distribution of power in the international system or the European and Asian subsystems.

Domestic Politics

Neoclassical realism — in both its offensive and defensive variants — posits an important role for domestic politics in shaping states’ foreign policies. “It does not simply state that domestic politics matter in foreign policy, but specifies the conditions under which they matter.” ₈₇ State autonomy vis-à-vis civil society, coalition logrolling, organizational politics, and civil-military relations can all constrain national leaders’ ability to mobilize resources. During periods of immediate threat, such as international crises, the calculations of central decision-makers are paramount. Domestic politics, however, play a larger role in shaping states’ foreign policies over the medium to long term. The inclusion of domestic politics in neoclassical realist theories is deductively sound. As Sterling-Folker notes:

The proposition that domestic actors operation under a dual pressure from the anarchic environment and their own domestic processes goes a long way in explaining why the international choices that domestic actors make often appear objectively inefficient... While the anarchic environment encourages the goal of survival and comparative assessment of process, it is domestic process that is responsible for the ability of states to emulate the [successful] processes of others... Because domestic processes are not identical, no group addresses the pressures or environment in quite the same way or emulates process of others in quite the same manner. ₈₈

Offensive and defensive realist theories of foreign policy draw a distinction between the resources of civil society (population, industrial capabilities, raw materials, etc.) and the ability of the “state’ (that is, the central government) to mobilize those resources. Christensen, a leading defensive realist, develops the concept of national political power, defined as “the ability of state leaders to mobilize their nation’s human and material resources behind security policy initiatives.” ₈₉ Similarly, Zakaria, a leading offensive realist, writes: “Foreign policy is made not by the nation as whole, but by its government. Consequently, what matters is state power, not national power. State power is that portion of national power the government can extract for its purposes and reflects the ease with which central decision makers can achieve their ends.” ₉₀

The difference between offensive and defensive realists centers on the longer-term influence of domestic variables on state’s foreign policy. While some offensive realists acknowledge domestic constraints, they pay less attention to how leaders’ efforts to mobilize national resources at time “t” constrain their ability to change foreign policies in response to systemic imperatives at time “t + 1.” Zakaria’s state-centric theory, for example, posits a causal chain: increased state and national power leads to more expansionist foreign policies. In the case of the United States between 1865 and 1908, increases in both the power of the federal government and American economic capacity and population, were preconditions for a more assertive foreign policy. The theory is largely silent about the magnitude and character of state expansion. Does increased state power produce increasingly ambitious foreign policies, ad infinitum? At what point do states stop or at least modify expansionist foreign policies? ₉₁ How does the relative distribution of power between states either provide a motive for expansion or limit a state’s opportunities for expansion?

Christensen’s domestic mobilization theory models explicitly address the problem of how domestic politics constrain states’ ability to adjust their foreign policies. ₉₂ During the Cold War China and the United States faced a common threat to their security: the Soviet Union. A purely systemic theory, such as neorealism, would predict a Sino-American alliance against the Soviet Union. This did not happen and even after Nixon’s opening to China in 1972 there was never a Sino-American alliance. What explains this? In the late 1940s and late 1950s American and Chinese leaders sought to mobilize domestic resources to balance against the Soviet threat, but lacked sufficient “national political power” to do as they pleased. The Truman administration and later Mao Zedong had to use domestically popular but unnecessary foreign policies in secondary areas as a diversion for necessary, but unpopular, policies in primary areas. These policies in secondary areas set in motion a chain of events culminating in US-Chinese conflict during the Korean War and the 1958 Quemoy-Matsu Crisis. In short, leaders’ very efforts to mobilize resources in response to systemic pressures, latter contained their ability to re-adjust foreign policies in response to changes in the external environment. Christensen writes:

In both cases, the strategies adopted [by Truman and Mao] required significant public sacrifice in peacetime, so the leaders faced difficulties in selling those strategies to their respective publics. The manipulation or extension of short-term conflict with the other nation, while not desirable on straightforward or domestic grounds, became useful in gaining and maintaining public support for the core grand strategies. ₉₃

Offensive realists cannot criticize defensive realist theories for incorporating domestic variables, since they themselves do the same. However, does this attention to domestic variables remove particular offensive and defensive realist theories from the realist paradigm altogether? Are these not liberal theories of foreign policy parading under the “realist” banner? The answer to both question is no. Both offensive realism and defensive realism rest on a “top-down” conception of international politics. They treat the pressures of the external environment as analytically superior to the actions of states or the preferences of actors within the state. Christensen’s theory, for example, clearly states that leader’s respond to the constraints of the international system. Most variants of liberalism (with the notable exception of neoliberal institutionalism or functional regime theory) proceed from a “bottom-up” conception of politics in which the demands of individuals and groups within society are analytically before domestic or international politics. ₉₄

Does the inclusion of domestic-level variables in neoclassical realism herald a new synthesis between realism and liberalism? Again, the answer is no. Purely Innenpolitik theories fail to explain why states with identical domestic political systems and ideologies often pursue different external strategies, or why different types of states pursue similar strategies in the international arena. ₉₅ Because of realism’s top-down conception of international politics and gives analytical primacy to the international system, one cannot reconcile either branch of neoclassical realism with liberalism. This is not to say that Innenpolitik theories and research programs, such as the various democratic peace propositions or other variants of domestic liberalism, are completely without empirical power. Rather, I contend that one cannot have syntheses of theories that proceed from completely different and mutually incompatible core assumptions. ₉₆

 

Why Retain the Offensive-Defensive Realist Dichotomy?

Over the past decade, the distinction between offensive realism and defensive realism has become common in the international relations literature. However, is the distinction between offensive realism and defensive realism still relevant? Have not other variants or classifications largely subsumed or superseded the offensive realism-defensive realism divide? The answers to these questions are yes and no, respectively. Below, I consider two prominent attempts to re-categorize contemporary realist theories along lines different from the offensive-defensive dichotomy: postclassical realism and motivational realism. The existing dichotomy between offensive realism and defensive realism largely subsumes these new classifications.

Why Not Postclassical Realism?

Stephen Brooks presents the most ambitious attempt to reclassify the divisions within contemporary realism. Realists, he argues, diverge on three implicit assumptions about state behavior under anarchy. The first assumption concerns whether or not states respond to the probability of international conflict or the mere possibility of international conflict. The second involves the discount rate — how states weigh short-term versus long-term objectives. The third assumption involves states’ preferences for short-term military preparedness or longer-term economic prosperity. These three assumptions, particularly the possibility/probability distinction, Brooks contends, largely subsume the debate between offensive realism and defensive realism. ₉₇

Brooks posits two camps in contemporary realism: neorealism and postclassical realists. Both camps are state-centric and both have a systemic focus. Neorealism holds that the mere possibility of conflict shapes states’ behavior, causing leaders to adopt a worst-case worldview. States ultimate aim under anarchy is to seek security, not power (material capabilities). States, therefore, heavily discount the future by favoring short-term military preparedness over longer-term economic objectives, when and if the two conflict. Brooks claims that the micro-foundations of neorealism are inconsistent with subjective expected utility theory, which expects decision-makers to weigh the subjective probabilities of outcomes, not their mere possibility. Instead, neorealism implicitly posits a decision-making model characterized by loss aversion, an inattention to probabilities, and an inability to make inter-temporal trade-off. Brooks argues that the neorealist decision-making model derives from either prospect theory or Luce and Raiffa’s minimax model. ₉₈ He cites the writings of Waltz and Mearsheimer as prominent examples (indeed the only examples) of neorealism.

Postclassical realists, on the other hand, assume that states do not always engage in worst-case reasoning. The subjective probability of international conflict, not its mere possibility, shapes state behavior. Other material variables, in addition to the relative distribution of military capabilities, increase or decrease the likelihood of conflict. These include the offense-defense balance in military technology, geographic proximity, access to raw materials, international economic pressure, and the ease with which states can extract resources from conquered territory. Consequently, states do not necessarily subordinate long-term objectives to short-term security requirements. He also claims that postclassical realism assumes states seek to maximize relative power. State leaders do not maximize power because of an insatiable desire to dominate others, but rather to maintain maximum flexibility in deal with international contingencies. Brooks cites the work of Gilpin, Jervis, Walt, Van Evera, Barry Buzan, and Stephen Krasner, as prominent examples of postclassical realism.

Brooks does raise some interesting theoretical problems in contemporary realist theory. Rather than subsuming the extant intra-realist debate, however, Brooks substitutes idiosyncratic terms for offensive realism (which he calls neorealism) and defensive realism (which he calls postclassical realism). Moreover, there are several problems with the assumptions he attributes to defensive and offensive realist theories. First, it seems rather odd that Mearsheimer and Waltz are the only scholars who fall into the so-called neorealist (offensive realist) camp. More importantly, Brooks glosses over important theoretical differences between Waltz and Mearsheimer. Waltz, for example, assumes states in an anarchic environment seek security above all else. ₉₉ Mearsheimer, on the other hand, writes: “States seek to survive under anarchy by maximizing their power relative to other states, in order to main the means for self-defense. Relative power, not absolute levels of power matter most to states. Thus, states seek opportunities to weaken potential adversaries and improve their relative power positions.” ₁₀₀

Second, Brooks puts all other scholars into the defensive realist camp, despite vast theoretical differences among them. Gilpin’s hegemonic theory, which Brooks cites as an example of postclassical realism (defensive realism), has more in common with Mearsheimer’s systemic variant of offensive realism than either Stephen Walt’s balance-of-threat theory or Van Evera’s window theory. ₁₀₁ A more troubling problem concerns Brooks’ statement that neoclassical or defensive realists assume that anarchy leads states to maximize relative power. In fact, defensive realists do not assume anarchy induces power-maximizing behavior.

Third, Brooks misstates the implicit decision-making models of various realist theories. He is correct in noting the behavioral micro-foundations of Waltz’s balance-of-power theory are unclear. Waltz argues that structure shapes state behavior through both socialization and competition. This would seem to privilege structural constraints over subjective expected utility calculations in determining states’ behavior. ₁₀₂ But Brooks then claims that neorealist micro-foundations may be consistent with Daniel Kahneman and Amos Tversky’s prospect theory. As evidence, he cites the alleged assumption that leaders do not make intra-temporal trade-offs and respond to the mere possibility (not subjective probability) of international conflict. What he terms neoclassical realism, on the other hand, assumes that states adopt a probabilistic view of international conflict. When war becomes more probable, states rationally prefer shorter-term military preparations to longer-term objections.

Brooks’ distinction between probability and possibility is an interesting observation. In showing how that distinction produces different defensive and offensive realist predictions, however, he runs into several problems. First, the behavioral assumptions that Brooks attributes to offensive realism do not, in fact, proceed from prospect theory. He is correct in noting that some offensive realists assume the mere possibility of conflict shapes states’ behavior. However, prospect theory predicts individuals overweigh outcomes they perceive as certain relative to those they perceive as merely probable. They treat extremely probable (but uncertain) outcomes as if they were certain. Only when facing highly probable or near certain losses would prospect theory predict individuals to heavily discount the future. ₁₀₃ In many ways, the implicit decision model of various defensive realist theories is quite consistent with the predictions of prospect theory. ₁₀₄

Second, there are several circumstances in which defensive realists would expect states (or more properly state leaders) to heavily discount the future and subordinate long-term economic capacity for short-term military preparedness. For example, offense-defense theory predicts that the perceived efficacy of offensive or defensive military technologies influence national leaders’ discount rate. When leaders perceive military technology, doctrine, force posture, and deployments favor the offense, the efficacy of short-term military preparedness, rapid mobilization, fierce resistance to others’ expansion, truncated diplomacy and preventive military strikes increase dramatically. ₁₀₅

Why Not Motivational Realism?

Kydd presents another attempt to reclassify the divisions within contemporary realism. He draws a distinction between offensive and defensive realism, on the one hand, and what he calls motivational realism. The extant literature, Kydd argues, posits that anarchy generates uncertainty, which in turn generates insecurity. He claims that motivational realism challenges both offensive and defensive realism by pointing to the existence of many possible motivation states might have. In short, not all states are interested in security and greedy states are necessary to produce much of the action in international politics. ₁₀₆ Motivational realism is neither new nor necessarily antithetical to offensive and defensive realism. Instead, Kydd uses (perhaps inadvertently) game theory to lay bare an old philosophical division within political realism.

Some classical realists, such as Rousseau and Hobbes, emphasize the anarchic environment and the unintended consequences of states’ actions in explaining the recurrence of international conflict: the so-called “tragedy” school. ₁₀₇ Other classical realists, such as Morgenthau, Machiavelli and Reinhold Neibuhr, attribute the causes of international conflict to the power-lust inherent in political actors: the so-called “evil” school. ₁₀₈ Still other classical realists, such as Thucydides and Arnold Wolfers, saw the permissive cause of conflict as arising from a combination of human nature, the internal composition of states and the international environment. ₁₀₉

Motivational realism is not a new branch of international relations theory, but rather a new name for the older evil tradition in political realism. Two important points about the tragedy versus evil distinction are worth noting. First, the debate stems from normative assumptions about the relative importance of environment and actors’ motivations. There is no way to test empirically the relative importance of international environment and state’s innate characteristics as a general cause for war or peace. Second, as Table 2 below illustrates, the philosophical debate — tragedy versus evil — cuts across the theoretical divide between the offensive realism and defensive realism.

[Table 2 here, currently unavailable]

 

Kydd is correct in noting that some defensive realists, such as Jervis, Posen and Kaufmann, downplay actors’ motivations in favor of the security dilemma. ₁₁₀ However, other defensive realists, such as Miller, put equal (if not greater) emphasis on states’ motivations. ₁₁₁ Some offensive realists, such as Mearsheimer, argue that anarchy and the relative distribution of power largely determine states’ behavior and international outcomes. Zakaria contends: “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 the opportunity in its stead.” ₁₁₂ Still other defensive and offensive realists take a middle position in the tragedy versus evil debate. Schweller, for example, argues that both relative power and states’ interests are necessary to produce international conflict or cooperation. Van Evera’s window theory looks at both the structural incentives for aggressive military action (namely anticipated shifts in the relative distribution of power) and the motivations of the states involved. ₁₁₃

 

Conclusions: Directions for Future Research

Neoclassical realism is and will continue to be a vibrant research program in international relations. Theories in this school combine parsimony and theoretical rigor with attention to historical contingency. Both offensive and defensive variants of neoclassical realism address interesting and important questions. They generate a rich set of hypotheses on international relations and the foreign policies of major states.

Neoclassical realism is not monolithic. Offensive realists hold that anarchy induces relative power maximization. States expand their interests abroad when central decision-makers perceive an increase in relative power. Relative capabilities largely shape the intentions of states and the link between systemic imperatives and the actual foreign policies states pursue is relatively direct. Defensive realists hold that anarchy does not always provide incentives for power maximization. Often times, moderate foreign policies are the best route to security. The link between systemic imperatives and states’ foreign policy choices is often indirect, complex and problematic.

This paper sought to advance the debate within neoclassical realism by clarifying the auxiliary assumptions underpinning its defensive realist variant. Defensive realist theories of foreign policy and international relations share four auxiliary assumptions that distinguish them from offensive realism. First, the security dilemma is an inescapable feature of life in an anarchic environment. Second, the fine-grained structure of power influences the ease or difficulty with which a state can enhance its relative security or power. The likelihood of international conflict and cooperation varies with the offense-defense balance and level of objective threat between particular dyads or in particular regions. Third, the gross and the fine-grained structure of power shape the behavior of states through the perceptions (and misperceptions) of central decision-makers. Fourth, domestic politics both constrain national leaders’ ability to mobilize societal resources and limit their ability to readjust foreign policy in response to external changes. By making these auxiliary assumptions explicit, one can have a more fruitful dialogue both within neoclassical realism and between it and other research programs.

I addressed three common critiques of defensive realism and found them wanting. Defensive realism does not confuse what at states ought to learn from the international system with what states actually learn. The security dilemma neither implies all states pursue minimal security nor precludes the existence of revisionist states. Even in the hypothetical absence of revisionist states, however, there would still be a security dilemma. Defensive realism does not place too much emphasis on unit-level “pathologies” in explaining self-defeating foreign policies or negative international outcomes. Both offensive realists and defensive realist theories posit a role for unit-level variables. Defensive realist theories tend to be more explicit in their reliance of central decision-makers’ perceptions of relative capabilities.

The current debate between offensive and defensive realism suggests several avenues for future theorizing and empirical research. First, both offensive and defensive realists should be more explicit about the definition of essential terms and the auxiliary assumptions underlying their theories. Within the defensive realist camp, more attention should be devoted to the definition and measurement of threat and the offense-defense balance. ₁₁₄ Likewise, offensive realists should more explicit in the their definitions of relative power maximization and “greedy” states.

Second, both variants of neoclassical realism need to devote more attention to the sources of revisionist state behavior. ₁₁₅ Defensive realists note that states will sometimes engage in security-driven expansion and that, over the long run, self-aggrandizement may be self-defeating. Offensive realists pay considerable attention to revisionist states and see them as the prime movers in international politics. The sources of revisionism — whether driven by “greed” or “security” — remains underdeveloped in both camps. ₁₁₆

Third, both defensive and offensive realist theories argue that perceptions and misperceptions among top leaders may inhibit a state’s ability to respond to changes in relative capabilities. The link between objective changes in relative power and decision maker’s perceptions of relative power remains underdeveloped. Future research in both camps should be more explicit in incorporating insights from cognitive and social psychology. ₁₁₇ Finally, the current debate between the two branches of neoclassical realism need not be a permanent bifurcation. By combining defensive realism’s assumptions about the fine-grained structure of power with offensive realism’s assumptions about revisionist states, one might develop more powerful international relations and foreign policy theories.

---

Endnotes

*: Prepared for Panel DF-02: “Intra-Realist Debates: Theoretical Progress or Degeneration?” at the Annual Meeting of the International Studies Association, Washington, D.C., February 16–20, 1999.  Back.

Note 1: Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979). Some draw a distinction between 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). Back.

Note 2: Gideon Rose coined the term “neoclassical realism” specifically in reference to the work of Fareed Zakaria, Randall L. Schweller, William C. Wohlforth and Thomas J. Christensen. See Rose, “Neoclassical Realism and Theories of Foreign Policy,” World Politics, Vol. 51, No. 1 (October 1998), 144-172. As will be readily apparent in this paper, I use the terms “neoclassical realism” or “contemporary realism” to denote a wider body of empirical theories sharing seven core assumptions about world politics, but distinct from Waltzian neorealism and classical realism. Unlike Rose, I do not see offensive realism and defensive realism as distinct from neoclassical realism. Back.

Note 3: The terms offensive/aggressive realism and defensive realism originated in Jack Snyder, Myths of Empire: Domestic Politics and International Ambition (Ithaca: Cornell University Press, 1991), 11-12. Overviews of the offensive-defensive realist debate include: Sean M. Lynn-Jones and Steven E. Miller, “Preface,” in The Perils of Anarchy: Neorealism and International Security, ed. Michael E. Brown, Sean M. Lynn-Jones, and Steven E. Miller (Cambridge: MIT Press, 1995), ix-xii; Benjamin Frankel, “Restating the Realist Case: An Introduction,” Security Studies, Vol. 5, No. 3 (spring 1996), xiv-xx. Back.

Note 4: Examples of offensive realism include John J. Mearsheimer, “Back to the Future: Instability in Europe after the Cold War,” International Security, 15, No. 1 (summer 1990), 5-56; Randall L. Schweller, “Neorealism’s Status Quo Bias: What Security Dilemma,” Security Studies, Vol. 5, No. 3 (spring 1996), 90-121; Fareed Zakaria, From Wealth to Power: The Unusual Origins of America’s World Role (Princeton: Princeton University Press, 1998) and idem, “Realism and Domestic Politics: A Review Essay,” International Security, Vol. 17, No. 1 (summer 1992), 177-188 and Eric J. Labs, “Beyond Victory: Offensive Realism and the Expansion of War Aims,” Security Studies, Vol. 6, No. 4 (summer 1997), 1-49. Back.

Note 5: Robert Jervis, “Cooperation under the Security Dilemma,” World Politics, Vol. 30, No. 2 (January 1978), 167-214 provides the theoretical foundations for defensive realism. See Stephen M. Walt, The Origins of Alliances (Ithaca: Cornell University Press, 1987) and Revolution and War (Ithaca: Cornell University Press, 1996); Snyder, Myths of Empire; Charles L. Glaser, “Realists as Optimists: Cooperation as Self-Help,” Security Studies, Vol. 5, No. 3 (spring 1996), 122-166; Stephen Van Evera, “Offense, Defense, and the Causes of War,” International Security, Vol. 22, No. 4 (spring 1998), 5-43; Thomas J. Christensen and Jack Snyder, “Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity,” International Organization, Vol . 44, No. 3 (spring 1990), 137-168 and William C. Wohlforth, The Elusive Balance: Power and Perceptions during the Cold War (Ithaca: Cornell University Press, 1990). Back.

Note 6: Andrew Kydd uses the term “motivational realism” to describe both Randall Schweller’s work and his own. See Kydd, “Sheep in Sheep’s Clothing: Why Security Seekers Do Not Fight Each Other,” Security Studies, Vol. 7, No. 1 (autumn 1997), 114-155. Stephen G. Brooks coined the term “post-classical realism.” See Brooks, “Dueling Realisms,” International Organization, Vol. 51, No. 3 (summer 1997), 177-198. Back.

Note 7: Sean M. Lynn-Jones, “Realism and America’s Rise: A Review Essay,” International Security, Vol. 23, No. 2 (fall 1998), 157-182. Back.

Note 8: Colin Elman and Miriam Fendius Elman, “Diplomatic History and International Relations Theory: Respecting Difference and Crossing Boundaries,” International Security, Vol. 22, No. 1 (Summer 1997), 5-21. Back.

Note 9: Robert Jervis, “Realism and the Study of World Politics,” International Organization, Vol. 52, No. 4 (autumn 1998), 971-991, at 991. Back.

Note 10: Randall L. Schweller, “New Realist Research on Alliances: Refining, Not Refuting Waltz’s Balancing Proposition,” American Political Science Review, Vol. 91, No. 4 (December 1997), 927-930. Also see Michael W. Doyle, Ways of War and Peace: Realism, Liberalism and Socialism (New York: WW Norton, 1997), 41-49. Back.

Note 11: Colin Elman and Miriam Fendius Elman, “Correspondence: History vs. Neorealism: A Second Look,” International Security, Vol. 20, No. 1 (summer 1995), 182-193. Richard Ashley coined the term neorealism in reference to Waltz’s theory in “The Poverty of Neo-realism,” International Organization, Vol. 38, No. 2 (spring 1984), 225-261. Back.

Note 12: Robert Gilpin, “The Richness of the Tradition of Political Realism,” in Keohane, ed. Neorealism and Its Critics, 309. Also see Gilpin, “No One Loves a Political Realist,” Security Studies, Vol. 5, No. 3 (spring 1996), 3-28 and Marcus Fischer, “Feudal Europe, 800-1300: Communal Discourses and Conflictual Practices,” International Organization, Vol. 46, No. 2 (spring 1992), 427-466. Back.

Note 13: Jennifer Sterling Folker, “Realist Environment, Liberal Profess and Domestic-Level Variables,” International Studies Quarterly, Vol. 41, No. 1 (March 1997), 1-25, at 5. Back.

Note 14: See, for example, Paul F. Diehl and Frank W. Wayman, “Realpolitik: Dead End, Detour, or Road Map?” in Reconstructing Realpolitik, ed. Diehl and Wayman (Ann Arbor: University of Michigan Press, 1994), 252. Also see Bruce Bueno de Mesquita and David Lalman, War and Reason: Domestic and International Imperatives (New Haven: Yale University Press, 1992), 10-19. Back.

Note 15: Jonathan Mercer, “Anarchy and Identity,” International Organization, Vol. 49, No. 2 (spring 1995), 229-252. Back.

Note 16: Keohane asserts that rationality is one of three hard core assumptions for both classical realism and neorealism. Keohane, “Theory of World Politics,” in Keohane, ed. Neorealism and Its Critics (New York: Columbia University Press, 1986), 165. Back.

Note 17: Waltz, “Reflections on Theory of International Politics : A Response to My Critics ,” in Keohane, ed. Neorealism and its Critics, 118. Also, João Resende-Santos, “Anarchy and the Emulation of Military Systems: Military Organization and Technology in South America, 1870-1914,” Security Studies, Vol. 5, No. 3 (spring 1996), 193-260, at 209, n. 56 and Miles Kahler, “Rationality in International Relations,” International Organization, Vol. 52, No. 4 (autumn 1998), 919-942, at 924-925. Back.

Note 18: Hans J. Morgenthau, Scientific Man versus Power Politics (Chicago: University of Chicago Press, 1946), 90-95. Back.

Note 19: Sterling-Folker, “Realist Environment,” 5. Back.

Note 20: For the definitive realist statement on international institutions see John J. Mearsheimer, “The False Promise of International Institutions,” International Security, Vol. 19, No. 3 (winter 1994/95), 5-49. For an impassioned but ultimately unconvincing defense of neoliberal institutionalism see Robert O. Keohane and Lisa L. Martin, “The Promise of Institutionalist Theory,” International Security, Vol. 20, No. 1 (summer 1995), 39-51. For a more sanguine realist view on international institutions see Randall L. Schweller and David Priess, “A Tale of Two Realisms: Expanding the Institutions Debate,” Mershon International Studies Review, Vol. 41, Supp. 1 (May 1997), 1-33. Back.

Note 21: Colin Elman, ”Horses for Courses: Why Not Neorealist Theories of Foreign Policy,“ Security Studies, Vol. 6, No. 1 (autumn 1996), 7-53, at 20-21; Keith L. Shimko, ”Realism, Neo-realism, and American Liberalism,“ Review of Politics, Vol. 54, No. 2 (spring 1992), 281-309. For a contrasting view see James E. Dougherty and Robert L. Pfaltzgraff, Jr. Contending Theories of International Relations, 3^rd ed. (New York: Harper & Row, 1990), 119-120. Back.

Note 22: See Ashley J. Tellis, “Reconstructing Political Realism: The Long March to Scientific Theory,” Security Studies, Vol. 5, No. 2 (winter 1995), 3-104. Back.

Note 23: Prominent classical realist works include Hans J. Morgenthau, Politics among Nations: The Struggle for Power and Peace (New York: Knopf, 1948 and later editions); Henry A. Kissinger, A World Restored: Castlereagh, Metternich and the Restoration of Peace, 1812-1822 (Boston: Houghton, Mifflin, 1957); Frederick L. Schuman, International Politics: The Destiny of the Western State System, 4^th ed. (New York: MacGraw Hill, 1948); Edward Hallett Carr, The Twenty Years’ Crisis (New York: Harper & Row, 1946); Arnold Wolfers, Discord and Collaboration: Essays on International Politics (Baltimore: Johns Hopkins University Press, 1962), and Raymond Aron, Peace and War: A Theory of International Relations, trans. Richard Howard and Annette Baker Fox (New York: Praeger, 1966). Back.

Note 24: Morgenthau, Politics Among Nations, 6. Back.

Note 25: See Schweller and Priess, “Tale of Two Realisms,” 7-8. Also, see John Gerard Ruggie, “What Makes the World Hang Together? Neo-Utilitarianism and the Social Constructivist Challenge,” International Organization, Vol. 52, No. 4 (autumn 1998), 855-917, at 889-890. Back.

Note 26: As is well known, Waltz disagrees with the contention that neorealism can explain foreign policy. He nonetheless uses his theory to make foreign policy predictions. See Waltz, “The Emerging Structure of International Politics,” International Security, Vol. 18, No. 2 (fall 1993), 44-79; “Correspondence,” International Security, Vol. 19, No. 1 (summer 1994), 198-199; “On the Nature of States and their Recourses to Violence,” U.S. Institute of Peace Journal, Vol. 3, No. 2 (April 1990), 6-7. Back.

Note 27: On the deductive reasoning for the stability of bipolar systems see Waltz, Theory of International Politics, chap. 8; Waltz, “Reflections on Theory of International Politics : A Response to My Critics,” in Neorealism and Its Critics ; Glenn Snyder and Paul Diesing, Conflict among Nations (Princeton: Princeton University Press, 1977), chap. 6. Neoclassical realists, on the other hand, do not universally accept the inherent stability of bipolar systems. See Ted Hopf, “Polarity, the Offense-Defense Balance and War,” American Political Science Review, Vol. 85, No. 3 (June 1991), 476-493 and Dale Copeland, “The Myth of Bipolar Stability,” Security Studies, Vol. 5, No. 3 (spring 1996), 29-89. Back.

Note 28: Alternatively, polarity has changed four times in the past century: (1) from multipolarity to bipolarity in 1933-35; (2) from bipolarity to tripolarity in 1934-35; (3) from tripolarity to bipolarity in 1943-45, and (4) from bipolarity to unipolarity in 1989-91. Schweller makes this assessment, using the capabilities index from the Correlates of War (COW) data set and his own separate power index. See Schweller, Deadly Imbalances, 27-31. Back.

Note 29: Waltz, Theory of International Politics, 73. Back.

Note 30: Barry R. Posen, The Sources of Military Doctrine (Ithaca: Cornell University Press, 1984), 34-35. Unlike Rose and Schweller, I date the rise of neoclassical realism to the mid-1980s. Other early examples of the shift away from neorealism’s purely systemic focus include Stephen Van Evera, “The Cult of the Offensive and the Origins of the First World War,” International Security, Vol. 9, No. 1 (Summer 1984), 58-107; Jack Snyder, The Ideology of the Offensive: Military Decisionmaking and the Disasters of 1914 (Ithaca: Cornell University Press, 1984). Back.

Note 31: Waltz, Theory of International Politics, 75. Back.

Note 32: Zakaria, “Realism and Domestic Politics,” 128. Back.

Note 33: Rose, “Neoclassical Realism and Theories of Foreign Policy,” 146. Back.

Note 34: Jervis, “Cooperation under the Security Dilemma.” For earlier discussions of the security dilemma see John Herz, “Idealist Internationalism and the Security Dilemma,” World Politics, Vol. 2, No. 2 (January 1950), 157-180 and Herbert Butterfield, History and Human Relations (London: Collins, 1951). Back.

Note 35: Charles L. Glaser, “The Security Dilemma Revisited,” World Politics, Vol. 50, No. 1 (October 1997), 175. Back.

Note 36: See Glaser, “Realists as Optimists,” 145. Back.

Note 37: See Peter Liberman, Does Conquest Pay? The Exploitation of Occupied Industrial Societies (Princeton: Princeton University, 1996); and idem “The Spoils of Conquest,” International Security, Vol. 18, No. 1 (fall 1993), 125-153; Lynn-Jones, “Realism and America’s Rise;” Van Evera, Causes of War, Vol. 1, 119-127. Like Glaser and Van Evera, I see this phenomena as an integral part of the security dilemma. Jervis, on the other hand, identified the incentives for security-driven expansion as separate from the security dilemma. See “Cooperation under the Security Dilemma,” 168-169, fn. 1. Back.

Note 38: Van Evera, Causes of War, Vol. 1, 135. Back.

Note 39: See Memorandum by Senior Foreign Office Counselor Friedrich von Holstein, 3 June 1904 in German Diplomatic Documents, 1871-1914, Vol. 3, trans. E.T.S. Dugdale (London: Harper & Brothers, 1930), 220-221 and German Emperor Wilhelm II to Reich Chancellor Bernhard von Bülow, 19 April 1904 in Bernhard von Bülow, Letters of Prince von Bülow, trans. Frederic Whyte (London: Hutchinson & Co., 1930), 54. Also see Glenn H. Snyder, “The Security Dilemma in Alliance Politics,” World Politics, Vol. 36, No. 3 (July 1984), 468-470. Back.

Note 40: Zakaria, From Wealth to Power, 30. Back.

Note 41: Schweller, “Neorealism’s Status Quo Bias,” 19. While Schweller speaks explicitly about Waltzian neorealism, his argument constitutes a general criticism of the security dilemma. Back.

Note 42: Kydd, “Sheep in Sheep’s Clothing,” 138. Back.

Note 43: Ibid., 140. Back.

Note 44: Lynn-Jones, “Realism and America’s Rise,” 176-177. Back.

Note 45: Lynn-Jones, “Offense-Defense Theory and its Critics,” 664-665, n. 10. Back.

Note 46: Jack S. Levy, “Preferences, Constraints and Choices in July 1914,” International Security, Vol. 15, No. 3 (winter 1990/91), 151-186. Back.

Note 47: See Glaser, “Realists as Optimists,” 143. Glaser also claims that costly signals can mitigate (but not eliminate) the security dilemma between states. The costly signals he identifies all pertain to military policy: arms control, unilateral defense, and unilateral restraint. The potential effectiveness of these signals depends on the “objective” offense-defense balance in military technology. Among potential adversaries, each sides’ military policies, not domestic policies and ideology, are more likely to exacerbate or mitigate the security dilemma. Also, see Walt, Origins of Alliances, chap. 6. Back.

Note 48: John M. Owen, “How Liberalism Produces Democratic Peace,” International Security, Vol. 19, No. 2 (fall 1994), 87-125. Back.

Note 49: On this point see Bruce Russett, “Correspondence: The Democratic Peace — And Yet it Moves,” International Security, Vol. 19, No. 4 (spring 1995), 164-175. Also see Christopher Layne, “Lord Palmerston and the Triumph of Realism: Anglo-French Relations, 1830-48,” in Miriam Fendius Elman, ed. Paths to Peace: Is Democracy the Answer? (Cambridge: MIT Press, 1997) 61-100. Back.

Note 50: Thomas Schwartz and Kiron Skinner, “The Myth of Democratic Pacifism,” Wall Street Journal, 7 January 1999. Back.

Note 51: These cases include the Trent Affair (1861), the Venezuela Crisis (1895-96), the Fashoda Crisis (1898) and the Ruhr Crisis (1923). See Christopher Layne, “Kant or Cant: The Myth of the Democratic Peace,” International Security, Vol. 19, No. 2 (fall 1994), 5-49. Also see Layne, “Lord Palmerston and the Triumph of Realism: Anglo-French Relations, 1830-48,” in Miriam Fendius Elman, ed. Paths to Peace: Is Democracy the Answer? (Cambridge: MIT Press, 1997) 61-100. Back.

Note 52: Jack S. Levy, “Preferences, Constraints and Choices in July 1914,” in Steven E. Miller, et. al, eds. Military Strategy and the Origins of the First World War, rev. and expanded ed. (Princeton: Princeton University Press, 1991), 226-262, at 233-234 and 243-245. Back.

Note 53: See Thomas J. Christensen, “Threats, Assurances and Last Chances for Peace: The Lesson’s of Mao’s Korean War Telegrams,” International Security, Vol. 17, No. 1 (summer 1992), 122-154. Back.

Note 54: Van Evera coined the term “fine-grained” structure of power. Van Evera, Causes of War, Vol. 1, 8-10. Back.

Note 55: Variables such as the offense-defense balance and threat have both objective components and perceptions. See the exchange between James W. Davis, Jr. and Stephen Van Evera in “Correspondence: Taking Offense at Offense-Defense Theory,” International Security, Vol. 23, No. 3 (winter 1998/99), 179-182 and 195-200. When I speak of the fine-grained structure power, I am referring to the objective components. Back.

Note 56: I use the term “offense-defense theory” as a convenient short hand. There are actually several different offense-defense theories of foreign policy and international relations. On this point see Charles L. Glaser and Chaim Kaufmann, “What is the Offense-Defense Balance and How Can We Measure It? International Security, Vol. 22, No. 4 (spring 1998), 44-82; Sean M. Lynn-Jones, “Realism, Security, and Offense-Defense Theories” (paper presented at the annual meeting of the American Political Science Association, Boston, 3-6 September 1998). Back.

Note 57: Jervis, “Cooperation under the Security Dilemma.” There are, of course, numerous definitions and ways of operationalizing the offense-defense balance between states. For a review of various such efforts see Jack S. Levy, “The Offense/Defense Balance of Military Technology: A Theoretical and Historical Analysis,” International Studies Quarterly, Vol. 38, No. 2 (June 1984), 219-238 and Lynn-Jones, “Offense-Defense Theory and its Critics.” Back.

Note 58: Van Evera, “Taking Offense at Offense-Defense Theory,” 195 and Robert Jervis, The Meaning of the Nuclear Revolution (Ithaca: Cornell University Press, 1989), 4-5, 19-21 and 29-35. Back.

Note 59: See Glenn H. Snyder, “Process Variables in Neorealist Theory,” Security Studies, Vol. 5, No. 3 (spring 1996), 167-182 and Deborah L. Spar and David A. Welch, “Asset Specificity and Structural Theories of International Politics,” (paper presented at the Annual Meeting of the American Political Science Association, Washington, DC, 1-4 September 1994). Back.

Note 60: Early variants of offense-defense theory tended to imply a global offense-defense balance in military technology. See, for example, Jervis, “Cooperation under the Security Dilemma,” and George H. Quester, Offense and Defense in International Politics (New York: John Wiley and Sons, 1977), 7-11. Back.

Note 61: Glaser and Kaufmann, “What is the Offense-Defense Balance?” 57. Back.

Note 62: See Walt, Origins of Alliances, chap. 1. Back.

Note 63: Zakaria, “Realism and Domestic Politics,” 126. Back.

Note 64: Mearsheimer, “Offensive Realism,” 2. Back.

Note 65: Lynn-Jones makes this point in “Realism and America’s Rise,” 175. Back.

Note 66: Walt, Origins of Alliances, 49. Emphasis added. Back.

Note 67: See Stephen M. Walt, “Alliances, Threats and U.S. Grand Strategy: A Reply to Kaufman and Labs,” Security Studies, Vol. 1, No. 3 (spring 1992), 448-482, at 450-451. Back.

Note 68: Van Evera, “Offense, Defense and the Causes of War,” 42-43. Back.

Note 69: Ibid., 43. Back.

Note 70: Zakaria, From Wealth to Power, 26-27 and 185-186; Mearsheimer, “Offensive Realism,” 14. Back.

Note 71: Lynn-Jones, “Realism and America’s Rise,” 174. Back.

Note 72: Jervis, Perception and Misperception, chap. 6. For a detailed analysis on the role of historical analogies in foreign policy decision making see Yuen Foong Khong, Analogies at War (Princeton: Princeton University, 1992) and Deborah Welch Larson, The Origins of Containment: A Psychological Explanation (Princeton: Princeton University, 1985). Back.

Note 73: Rose, “Neoclassical Realism and Theories of Foreign Policy,” 158. Back.

Note 74: Most works draw upon Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976) and idem, The Logic of Images in International Relations (Princeton: Princeton University Press, 1970). Back.

Note 75: Wohlforth, “Realism and the End of the Cold War,” 98. Back.

Note 76: Christensen, “Perceptions and Alliances in Europe,” 92. Also see William C. Wohlforth, “The Perception of Power: Russia in the Pre-1914 Balance,” World Politics, Vol. 39, No. 3 (April 1987); Walt, Revolution and War, chap. 2; Back.

Note 77: Aaron L. Friedberg, The Weary Titan: Britain and the Experience of Relative Decline, 1895-1905 (Princeton: Princeton University Press, 1988) , 284. Back.

Note 78: Christensen, “Perceptions and Alliances in Europe,” and Jeffrey W. Taliaferro, “Quagmires in the Periphery: Foreign Wars and Escalating Commitment in International Conflict,” Security Studies, Vol. 7, No. 3 (spring 1998), 94-144. Back.

Note 79: Wohlforth, Elusive Balance, 294. Back.

Note 80: Benjamin Miller, When Opponents Cooperate: Great Power Conflict and Collaboration in World Politics (Ann Arbor: University of Michigan Press , 1996), 110-119. Back.

Note 81: See Posen, Sources of Military Doctrine, 67-69. Back.

Note 82: Van Evera, “Offense, Defense and the Causes of War,” and idem, “The Cult of the Offensive and the Origins of the First World War,” International Security, Vol. 9, No. 1 (summer 1984), 58-107; Snyder, Ideology of the Offensive, Back.

Note 83: Not all offense-defense theorists, however, posit an central role for perceptions. See for example Glaser and Kaufmann, “What is the Offense-Defense Balance?” Back.

Note 84: Labs, “Beyond Victory,” 16. Back.

Note 85: Schweller, Deadly Imbalances, 94-100. Back.

Note 86: Ibid., 168. Back.

Note 87: Thomas J. Christensen, Useful Adversaries: Grand Strategy, Domestic Mobilization and Sino-American Conflict, 1947-1958 (Princeton: Princeton University Press, 1996), 256. Back.

Note 88: Sterling-Folker, “Realist Environment,” 21. Back.

Note 89: Christensen, Useful Adversaries, 26. Rose writes, “Christensen does not take a position in the theoretical debate between defensive realists and their critics over whether active foreign policies are usually driven by increase power or increase thereat; his model is general enough to accommodate both.” Rose, “Neoclassical Realism,” 164. I disagree. Christensen domestic mobilization model assumes that central decision-makers respond to international challenges. See Christensen, Useful Adversaries, 13-14. Back.

Note 90: Zakaria, From Wealth to Power, 9. Back.

Note 91: Lynn-Jones, “Realism and America’s Rise,” 181. Lynn-Jones notes that Zakaria’s broad definition of “expansion” subsumes just about everything from the acquisition of the Philippines, to the dispatch of ambassadors abroad, to President Theodore Roosevelt’s role in negotiating an end to the Russo-Japanese War. Back.

Note 92: Arguably, the theory Snyder presents in Myths of Empire addresses the same question. However, while his explanation for initial expansion is perfectly consistent with defensive realism, his explanation of “over-expansion” (the pursuit of apparently self-defeating expansionist policies) relies almost entirely on domestic variables. Back.

Note 93: Christensen, Useful Adversaries, 6. Back.

Note 94: Andrew Moravcsik, “Taking Preferences Seriously: A Liberal Theory of International Politics,” International Organization, Vol. 51, No. 4 (autumn 1997), 513-553. For overviews of the various democratic peace propositions see Bruce Russett, Grasping the Democratic Peace (Princeton: Princeton University Press, 1993), 24-43; Miriam Fendius Elman, “The Need for Qualitative Tests of the Democratic Peace Theory,” in Elman, ed. Paths to Peace, 1-57. Back.

Note 95: See Fendius Elman, “Conclusion: Testing the Democratic Peace Theory,” in Paths to Peace; Stephen M. Walt, “Alliances, Threats, and U.S. Grand Strategy: A Reply to Kaufman and Labs,” Security Studies, Vol. 1, No. 3 (spring 1992), 473, n. 1. Back.

Note 96: Jon M. Owen calls for a synthesis of realism and the democratic peace proposition. See Owen, “How Liberalism Produces the Democratic Peace,” in Debating the Democratic Peace, ed. Michael E. Brown, Sean M. Lynn-Jones and Steven E. Miller (Cambridge: MIT Press, 1997), 151-154. Back.

Note 97: Brooks, “Dueling Realisms,” 458. Back.

Note 98: Ibid., 460. On the minimax model see R. Duncan Luce and Howard Raiffa, Games and Decisions: Introduction and Critical Survey (New York: John Wiley & Sons, 1957), 278-289. For an overview of prospect theory and its application in international relations see Jack S. Levy, “Prospect Theory, Rational Choice, and International Relations,” International Studies Quarterly, Vol. 41, No. 1 (March 1997), 87-112. Back.

Note 99: Waltz, Theory of International Politics, 126. Back.

Note 100: Mearsheimer, “Back to the Future,” 148. Back.

Note 101: Both Mearsheimer and Gilpin emphasize relative power maximization, power transitions, hegemonic rivalry, and the hierarchy of prestige as an ordering principle in the international system. See Robert Gilpin, War & Change in World Politics (Cambridge: Cambridge University Press, 1980) and Mearsheimer, “Back to the Future,” 148. Back.

Note 102: Kahler, “Rationality in International Politics,” 924. Also see Jon Elster, Ulysses and the Sirens: Studies in Rationality and Irrationality, rev. ed. (Cambridge: Cambridge University Press, 1984), 113-114. Back.

Note 103: Robert Jervis, “The Political Implications of Loss Aversion,” Political Psychology, Vol. 13, No. 1 (spring 1992), 187-204 and Jack S. Levy, “Prospect Theory and International Relations: Theoretical Applications and Analytical Problems,” Political Psychology, Vol. 13, No. 1 (spring 1992), 283-310. Back.

Note 104: Taliaferro, “Quagmires in the Periphery,” 99-100. Back.

Note 105: Van Evera, “Offense, Defense, and the Causes of War,” 7-9. Back.

Note 106: Kydd, “Sheep in Sheep’s Clothing,” 153-155. Back.

Note 107: Kenneth N. Waltz, Man, The State and War (New York: Columbia University Press, 1954/59), esp. chaps. 6-8; Jean Jacques Rousseau, State of War (1756), in A Lasting Peace, trans. C. E. Vaughn (London: Constable, 1917); Thomas Hobbes, Leviathan (Oxford: The Clarendon Press, 1909). Back.

Note 108: Nicolò Machiavelli, The Prince, trans. Harvey C. Mansfield, Jr. (Chicago and London: University of Chicago Press, 1985), chaps. 15-19; Reinhold Neibuhr, The Children of Light and the Children of Darkness (New York: Scribner’s, 1944) and Morgenthau, Scientific Man vs. Power Politics Back.

Note 109: Thucydides, History of the Pelponnesian War, trans. Rex Warner (New York: Penguin Books, 1954, 1988); Wolfers, Discord and Collaboration, 83-84. For a concise restatement of the “tragedy” and “evil” schools in realist thought see Michael Spirtas, “A House Divided: Tragedy and Evil in Realist Theory,” Security Studies, Vol. 5, No. 3 (spring 1996), 72-107. For a concise analysis of how Thucydides’ account of the Peloponnesian War used multiple levels-of-analysis see Doyle, Ways of War and Peace, esp. 68-69, 76-78. Back.

Note 110: Jervis, “Cooperation under the Security Dilemma” and more recently Systems Effects: Complexity in Political and Social Life (Princeton: Princeton University Press, 1997); Barry R. Posen, “The Security Dilemma in Ethnic Conflict,” in Michael E. Brown, ed. Ethnic Conflict and International Security (Princeton: Princeton University Press, 1993), 103-124; Chaim D. Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,” International Security, Vol. 20, No. 4 (spring 1996), 136-175. Back.

Note 111: Miller, When Opponents Cooperate, chap. 3; Van Evera, Causes of War, Vol. 1. Back.

Note 112: Zakaria, From Wealth to Power, 29. Back.

Note 113: Schweller, “Tripolarity and the Second World War,” International Studies Quarterly, Vol. 37, No. 1 (March 1993), 73-103, at 76. See also Van Evera, Causes of War, Vol. 1. Back.

Note 114: Glaser and Kaufmann, “What is the Offense-Defense Balance.” Also see Stephen Biddle, “Recasting the Foundations of Offense-Defense Theory” (paper presented at the annual meeting of the American Political Science Association, Boston, 3-6 September 1998). Back.

Note 115: Rose, “Neoclassical Realism,” 165. Back.

Note 116: Glaser, “Security Dilemma Revisited,” 200-201. Back.

Note 117: See James M. Goldgeier, “Psychology and Security,” Security Studies, Vol. 6, No. 4 (summer 1997) and Mercer, “Anarchy and Identity.” Back.