Neoclassical Realism

Review: Neoclassical Realism and Theories of Foreign Policy Reviewed Work(s): The Perils of Anarchy: Contemporary Realism and International Security by Michael E. Brown; Useful Adversaries: Grand Strategy, Domestic Mobilization, and Sino-American Conflict, 1947-1958 by Thomas J. Christensen; Deadly Imbalances: Tripolarity and Hitler's Strategy of World Conquest by Randall L. Schweller; The Elusive Balance: Power and Perceptions during the Cold War by William Curti Wohlforth; From Wealth to Power: The Unusual Origins of America's World Role by Fareed Zakaria Review by: Gideon Rose Source: World Politics, Vol. 51, No. 1 (Oct., 1998), pp. 144-172 Published by: Cambridge University Press Stable URL: https://www.jstor.org/stable/25054068 Accessed: 12-08-2018 23:50 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide

range of content in a trusted digital archive. We use information technology and tools to increase productivity and

facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

https://about.jstor.org/terms

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to World Politics

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

Review Article

NEOCLASSICAL REALISM AND THEORIES OF FOREIGN POLICY

By GIDEON ROSE*

Michael E. Brown et al., eds. The Perils of Anarchy: Contemporary Realism and International Security. Cambridge: MIT Press, 1995, 519 pp.

Thomas J. Christensen. Useful Adversaries: Grand Strategy, Domestic Mobiliza tion, and Sino-American Conflict, 1947-1958. Princeton: Princeton Univer sity Press, 1996, 319 pp.

Randall L. Schweller. Deadly Imbalances: Tripolarity and Hitlers Strategy of World Conquest. New York: Columbia University Press, 1998,267 pp.

William Curti Wohlforth. The Elusive Balance: Power and Perceptions during the Cold War. Ithaca, N.Y.: Cornell University Press, 1993, 317 pp.

Fareed Zakaria. From Wealth to Power: The Unusual Origins of Americas World Role. Princeton: Princeton University Press, 1998,199 pp.

FOR two decades international relations theory has been dominated by the debate between neorealists and their various critics.1 Much of the skirmishing has occurred over questions about the nature of the international system and its effect on patterns of international out comes such as war and peace. Thus scholars have disputed whether a

multipolar system generates more conflict than a bipolar one, or

* For support, criticisms, and suggestions regarding earlier versions of this essay I am grateful to Richard Berts, Michael Desch, Michael Doyle, Aaron Friedberg, Philip Gordon, Ethan Kapstein, Jeff Legro, Sean Lynn-Jones, Andrew Moravcsik, Kenneth Pollack, Robert Powell, and especially Sheri Berman. I am also grateful for the comments of participants at discussions sponsored by the Research Program in International Security at Princeton University, the John M. Olin Institute for Strategic Studies at Harvard University, and the 1997 annual meeting of the American Political Science Asso ciation.

xThe seminal neorealist text is Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley, 1979). Debates over neorealism can be found in Robert O. Keohane, ed., Neorealism and Its Critics (New York: Columbia University Press, 1986); Barry Buzan et al., The Logic of Anarchy:

Neorealism to Structural Realism (New York: Columbia University Press, 1993); and David A. Baldwin, ed., Neorealism andNeoliberalism: The Contemporary Debate (New York: Columbia University Press, 1993). For the current state of the debate, see Robert Powell, "Anarchy in International Relations

Theory: The Neorealist-Neoliberal Debate," International Organization 48 (Spring 1994); and Brown et al., an invaluable collection of important recent articles on realism from the journal International Security.

World Politics 51 (October 1998), 144-72

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 145

whether international institutions can increase the incidence of inter

national cooperation. Because neorealism tries to explain the outcomes of state interactions, it is a theory of international politics; it includes some general assumptions about the motivations of individual states but does not purport to explain their behavior in great detail or in all cases. As Kenneth Waltz has written:

[A] theory of international politics ... can describe the range of likely outcomes of the actions and interactions of states within a given system and show how the range of expectations varies as systems change. It 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. . . .To the extent that dynamics of a system limit the freedom of its units, their behavior and the outcomes of their behavior

become predictable . . . [but in general] a theory of international politics bears on the foreign policies of nations while claiming to explain only certain aspects of them.2

From such a perspective, much of the daily stuff of international re lations is left to be accounted for by theories of foreign policy. These theories take as their dependent variable not the pattern of outcomes of state interactions, but rather the behavior of individual states. Theories

of foreign policy seek to explain what states try to achieve in the exter nal realm and when they try to achieve it. Theory development at this level, however, has received comparatively little attention.

Some, like Waltz himself, simply rule the subject out of bounds due to its complexity. Theories, he argues, must deal with the coherent logic of "autonomous realms." Because foreign policy is driven by both inter nal and external factors, it does not constitute such an autonomous realm, and therefore we should not strive for a truly theoretical explana tion of it. Instead, we must rest content with mere "analyses" or "ac counts," which include whatever factors appear relevant to a particular case.3 Others have rejected such diffidence, and their recent efforts to construct a general theory of foreign policy fall into several broad schools.

2 Waltz (fn. 1), 71-72. See also the discussion of this point in Thomas J. Christensen and Jack Sny der, "Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity," International Or ganization 44 (Spring 1990), 38 fn. 3; Fareed Zakaria, "Realism and Domestic Politics," in Brown et al.; and Schweller, Deadly Imbalances, 7-11. Robert Powell has questioned whether it is even useful or possible to speak of theories of international politics in isolation, since systemic theories must neces sarily include nontrivial assumptions about states' preferences and behavior to begin with; see Powell (fn.l).

3 "Much is included in an analysis," he writes; "little is included in a theory." Kenneth N. Waltz, "International Politics Is Not Foreign Policy," Security Studies 6 (Autumn 1996), 54-55. Waltz was re sponding to the suggestion that scholars should devise and test theories of foreign policy emerging from his neorealist framework; see Colin Elman, "Horses for Courses: Why Not Neorealist Theories of Foreign Policy?" Security Studies 6 (Autumn 1996).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

146 WORLD POLITICS

The first and most common school is composed o? Innenpolitik the ories, which stress the influence of domestic factors on foreign policy. The others are all variants of realism and highlight the influence of the international system on state behavior. "Offensive realism" (sometimes

called "aggressive realism") essentially reverses Innenpolitik logic and ar gues that systemic factors are always dominant. "Defensive realism" takes a softer line, arguing in practice that systemic factors drive some kinds of state behavior but not others.4

The works under review here collectively set out a fourth school, which I term "neoclassical realism." It explicitly incorporates both ex ternal and internal variables, updating and systematizing certain in sights drawn from classical realist thought. Its adherents argue that the scope and ambition of a country's foreign policy is driven first and fore

most by its place in the international system and specifically by its rel ative material power capabilities. This is why they are realist. 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.

Neoclassical realists argue that relative material power establishes the basic parameters of a country's foreign policy; they note, in Thucydides' formula, that "the strong do what they can and the weak suffer what they must."5 Yet they point out that there is no immediate or perfect

4 Offensive and defensive realism are not only theories of foreign policy, but both schools commonly address foreign policy behavior and it is this aspect of them that will be treated here. The distinction between offensive/aggressive and defensive realism was first made by Jack Snyder in Myths of Empire: Domestic Politics and International Ambition (Ithaca, N.Y.: Cornell University Press, 1991), 11-12, and has been widely adopted since then. See the following in Brown et al.: Sean M. Lynn-Jones and Steven E. Miller, "Preface"; Zakaria, "Realism and Domestic Politics"; and John Mearsheimer, "The False Promise of International Institutions." See also Benjamin Frankel, "The Reading List: Debating Re alism," Security Studies 5 (Autumn 1995), esp. 185-87; Fareed Zakaria, From Wealth to Power, Randall L. Schweller, "Neorealism's Status Quo Bias: What Security Dilemma?" Security Studies 5 (Spring 1996), esp. 114-15; Michael C. Desch, "Why Realists Disagree about the Third World," Security Stud ies 5 (Spring 1996), esp. 365; Eric J. Labs, "Beyond Victory: Offensive Realism and the Expansion of

War Aims," Security Studies 6 (Summer 1997); and Stephen M. Walt, "International Relations: One World, Many Theories," Foreign Policy 110 (Spring 1998), 37. Other authors make the same distinc

tion but use idiosyncratic terminology. Thus Robert G. Kaufman substitutes "pessimistic structural" for "offensive" and "optimistic structural" for "defensive"; Stephen G. Brooks substitutes "neorealist" for "offensive" and "postclassical" for "defensive"; and Charles Glaser calls his variant "contingent " in stead of "defensive" realism. See Kaufman, "A Two-Level Interaction: Structure, Stable Liberal De

mocracy, and U.S. Grand Strategy," Security Studies 3 (Summer 1994), 683ff; Brooks, "Dueling Realisms," International Organization 51 (Summer 1997); and Glaser, "Realists as Optimists: Cooper ation as Self-Help," in Brown et al. Finally, in an overview of recent realist theorizing, Joseph M. Grieco puts all neorealists into the defensive camp; see Grieco, "Realist International Theory and the Study of World Politics," in Michael W. Doyle and G. John Ikenberry, eds., New Thinking in Interna tional Relations Theory (Boulder, Colo.: Westview Press, 1997), esp. 166-67.

5 Robert B. Strassler, ed., The Landmark Thucydides:A Comprehensive Guide to the Peloponnesian War (New York: Free Press, 1996), 5.89.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 147

transmission belt linking material capabilities to foreign policy behav ior. Foreign policy choices are made by actual political leaders and elites, and so it is their perceptions of relative power that matter, not simply relative quantities of physical resources or forces in being. This

means that over the short to medium term countries' foreign policies may not necessarily track objective material power trends closely or continuously. Furthermore, those leaders and elites do not always have complete freedom to extract and direct national resources as they might

wish. Power analysis must therefore also examine the strength and structure of states relative to their societies, because these affect the proportion of national resources that can be allocated to foreign policy. This means that countries with comparable gross capabilities but dif ferent state structures are likely to act differently. And finally, systemic pressures and incentives may shape the broad contours and general di rection of foreign policy without being strong or precise enough to de termine the specific details of state behavior. This means that the influence of systemic factors may often be more apparent from a dis tance than from up close?for example, in significantly limiting the

menu of foreign policy choices considered by a state s leaders at a par ticular time, rather than in forcing the selection of one particular item on that menu over another.

For all these reasons, the neoclassical realists believe, understanding the links between power and policy requires close examination of the contexts within which foreign policies are formulated and imple mented.6 After briefly sketching out the schools theoretical competi tors, the remainder of this essay will discuss its major works and distinctive characteristics and assess its contribution to the field.7

Four Theories of Foreign Policy

Statesmen, historians, and political philosophers have long pondered what causes states to adopt certain kinds of foreign policies. Yet most

6 In their stress on intervening variables, constrained choice, and historical context, as in other ways, neoclassical realists have much in common with historical institutionalists in comparative politics, who study "intermediate-level institutions that mediate the effects of macro-level socioeconomic struc tures." Neoclassical realists would agree that "this focus on how macrostructures ... are magnified or mitigated by intermediate-level institutions allows us to explore the effects of such overarching struc tures on political outcomes, but avoid the structural determinism that often characterizes . .. [purely systemic] approaches." Kathleen Thelen and Sven Steinmo, "Historical Institutionalism in Compara tive Politics," in Sven Steinmo et al., eds., Structuring Politics: Historical Institutionalism in Comparative

Analysis (Cambridge: Cambridge University Press, 1992), 11. 7 For reasons of space and coherence this essay will focus on the general features of neoclassical re

alism as a theory of foreign policy rather than on the empirical contributions the various neoclassical realist authors have made to the literatures on their particular historical subjects.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

148 WORLD POLITICS

have sought answers in intricate combinations of case-specific factors, considering it hubris to think that one could construct parsimonious theories of foreign policy that would have much explanatory power.

Analysts interested in theory building, meanwhile, have tended to fol low one of three distinct paths.

The most common approach has been to assume that foreign policy has its sources in domestic politics. These Innenpolitik theories argue that internal factors such as political and economic ideology, national character, partisan politics, or socioeconomic structure determine how countries behave toward the world beyond their borders. A pure, monadic version of such theorizing in a liberal vein would be the no tion that the behavior of democracies is different from that of non

democracies. A modified, dyadic version would be the notion of the "democratic peace," which holds that the behavior of democracies is different when they deal with each other. There are many variants of the Innenpolitik approach, each favoring a different specific domestic independent variable, but they all share a common assumption?that foreign policy is best understood as the product of a country's internal dynamics. To understand why a particular country is behaving in a par ticular way, therefore, one should peer inside the black box and examine the preferences and configurations of key domestic actors.8

The chief problem with Innenpolitik theories is that pure unit-level explanations have difficulty accounting for why states with similar do mestic systems often act differently in the foreign policy sphere and why dissimilar states in similar situations often act alike. Some scholars grounded in the neorealist model of international politics have sought to avoid this problem by applying that model to individual state behav ior as well as to international outcomes. They have generated two the

8 For a brief history of Innenpolitik theorizing about foreign policy, see Zakaria, in Brown et al.; for a powerful restatement of the Innenpolitik tradition in modern social science terms, see Andrew

Moravcsik, "Taking Preferences Seriously: A Liberal Theory of International Politics," International Organization 51 (Autumn 1997). On the concept of the democratic peace, see Michael E. Brown et al., eds., Debating the Democratic Peace (Cambridge: MIT Press, 1996). Other notable recent examinations of Innenpolitik variables include Jack S. Levy, "Domestic Politics and War," in Robert I. Rotberg and Theodore K. Rabb, eds., The Origin and Prevention of Major Wars (New York: Cambridge University Press, 1988); Richard Rosecrance and Arthur A. Stein, eds., The Domestic Bases of Grand Strategy (Ithaca, N.Y.: Cornell University Press, 1993); David Skidmore and Valerie M. Hudson, eds., The Limits of State Autonomy: Societal Groups and Foreign Policy Formulation (Boulder, Colo.: Westview, 1993); Joe D. Hagan, "Domestic Political Systems and War Proneness," Mershon International Studies Review 38, supplement 2 (October 1994); idem, "Domestic Political Explanations in the Analysis of Foreign Policy," in Laura Neack et al., eds., Foreign Policy Analysis: Continuity and Change in Its Second Generation (Englewood Cliffs, N.J.: Prentice Hall, 1995); and Matthew Evangelista, "Domestic Struc tures and International Change," in Doyle and Ikenberry (fn. 4).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 149

ories of foreign policy, offensive and defensive realism, which both start from the assumption that the international system is composed of uni tary, rational states motivated by a desire for security. The theories dif fer over what incentives they assume the international system offers such states and how they are likely to respond, as well as over the degree to which they assume the tension inherent in anarchy can be modulated by other factors such as the state of military technology.

Offensive realism assumes that international anarchy is generally Hobbesian?that apart from situations of bipolarity or nuclear deter rence, security is scarce and states try to achieve it by maximizing their relative advantage.9 In the offensive realist world rational states pursu ing security are prone to take actions that can lead to conflict with oth ers?and usually do: "States begin with a defensive motive, but are forced to think and sometimes act offensively because of the structure of the international system."10 Domestic differences between countries are considered to be relatively unimportant, because pressures from the international system are assumed to be strong and straightforward enough to make similarly situated states behave alike, regardless of their internal characteristics. According to this view, foreign policy ac tivity is the record of nervous states jockeying for position within the framework of a given systemic power configuration. To understand why a state is behaving in a particular way, offensive realists suggest, one should examine its relative capabilities and its external environment, because those factors will be translated relatively smoothly into foreign policy and shape how the state chooses to advance its interests.

Defensive realism, in contrast, assumes that international anarchy is often more benign?that is, that security is often plentiful rather than scarce?and that normal states can understand this or learn it over time

from experience.11 In the defensive realist world rational states pursuing security can often afford to be relaxed, bestirring themselves only to re spond to external threats, which are rare. Even then, such states gener ally respond to these threats in a timely manner by "balancing" against them, which deters the threatener and obviates the need for actual conflict. The chief exception to this rule is when certain situations

9 Examples of offensive realist analysis include John Mearsheimer, "Back to the Future: Instability in Europe after the Cold War," in Brown et al.; idem (fn. 4); and Labs (fn. 4).

10 Mearsheimer (fn. 4), 337 fn. 24. 11 Prominent defensive realist authors include Stephen Van Evera, Stephen M. Walt, Jack Snyder,

Barry Posen, and Charles L. Glaser; for citations to works in the defensive realist camp, see Zakaria (fn. 2), 476 fn. 34. For some of the reasons why defensive realists view systemic incentives as less

Hobbesian than offensive realists do, see Brooks (fn. 4).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

150 WORLD POLITICS

lead security-seeking states to fear each other, such as when prevailing modes of warfare favor the offensive.12 Foreign policy activity, in this view, is the record of rational states reacting properly to clear systemic incentives, coming into conflict only in those circumstances when the security dilemma is heightened to fever pitch. But this dance is repeat edly interrupted, according to defensive realists, by rogue states that misread or ignore the true security-related incentives offered by their environment.

Innenpolitik theories of foreign policy privilege domestic indepen dent variables, while offensive realism privileges systemic ones. Al though both schools are clear, bold, and predictive, the predictions of both are often oversimplified and inaccurate. (Pure systemic theories face the reverse anomaly from their Innenpolitik counterparts: states in similar structural positions do not always act alike.) The adherents of defensive realism also view it as a systemic theory, but in practice they rely on both systemic and domestic independent variables to account for different kinds of foreign policy behavior. Defensive realists view the international system as the cause of what might be called "natural" conduct, which includes a resort to aggression only if military technol ogy or certain other factors provide clear incentives to strike first. They consider the remainder of aggressive behavior to be "unnatural" and ac count for it by auxiliary hypotheses involving domestic variables.

Neoclassical realism challenges important elements of all three of these perspectives. Innenpolitik theories are misguided, the neoclassical realists say, because if there is any single, dominant factor shaping the broad pattern of nations' foreign policies over time, it is their relative material power vis-?-vis the rest of the international system?and so this is where analysis of foreign policy should begin. Defensive realism is misguided for a similar reason, because its emphasis on countries' re sponses to threats overlooks the fact that one s perceptions of threat are partly shaped by one's relative material power. The theory is further

12 Modern offense-defense theory is rooted in Robert Jervis s presentation of the security dilemma; see Jervis, "Cooperation under the Security Dilemma," World Politics 30 (January 1978). Recent de fensive realist works stressing the importance of offense-defense variables are Glaser (fn. 4); Ted Hopf, "Polarity, the Offense-Defense Balance, and War," American Political Science Review 85 (June 1991); Sean M. Lynn-Jones, "Offense-Defense Theory and Its Critics," Security Studies 4 (Summer 1995); Stephen Van Evera, "Offense, Defense, and the Causes of War," International Security 22 (Spring 1998); and Charles L. Glaser and Chaim Kaufmann, "What Is the Offense-Defense Balance and Can

We Measure It?" International Security 22 (Spring 1998); see also Jack S. Levy, "The Offensive/De fensive Balance of Military Technology: A Theoretical and Historical Analysis," International Studies Quarterly 28 (1984). In addition to military technology, the offense-defense balance is sometimes held to incorporate judgments about whether power resources are cumulative and therefore offer a tempt ing target for potential aggressors; for an analysis of this question, see Peter Liberman, Does Conquest

Pay? (Princeton: Princeton University Press, 1996).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 151

flawed because its first-order systemic argument does not account for much actual behavior, thus forcing its adherents to contract out the bulk of their explanatory work to domestic-level variables introduced on an ad hoc basis.13

The neoclassical realists believe that Innenpolitikers9 preferred inde pendent variables must be relegated to second place analytically because over the long run a state's foreign policy cannot transcend the limits and opportunities thrown up by the international environment. "A good theory of foreign policy," one of them writes, "should first ask what effect the international system has on national behavior, because the most powerful generalizable characteristic of a state in international relations is its relative position in the international system."14 Moreover, because the influence of structural factors such as relative power is not always obvious even to political actors themselves, neoclassical realists caution that analysts who do not begin by looking carefully for such in fluence may mistakenly attribute causal significance to other factors that are more visible but in reality are only epiphenomenal.

By making relative power their chief independent variable, the neo classical realists are forced to choose sides in the perennial debate about just how that concept should be defined and operationalized. They generally confront this issue directly, setting out their reasons for re serving the term "power" to refer to "the capabilities or resources . . .

with which states can influence each other" (Wohlforth, 4).15 They dis tinguish between these power resources and a country's foreign policy

13 Stephen Van Evera (fn. 12), for example, has recently argued that "a chief source of insecurity in Europe since medieval times has been [the] false belief that security was scarce." In general, he claims, "States are seldom as insecure as they think they are ... [the] exaggeration of insecurity, and the belli cose conduct it fosters, are prime causes of national insecurity and war" (pp. 42-43). Neoclassical real ists question the point of constructing an elaborate systemic theory around the assumption that states are driven by a quest for security only then to argue that on security-related questions states suffer from false consciousness most of the time. The original neoclassical realist critique of defensive realism along these lines is Zakaria (fn. 2); see also Schweller (fn. 4).

14 Zakaria (fn. 2), 482. 15 Neoclassical realists acknowledge that in contrast to this "material" definition, the "relational" def

inition of power?in Robert Dahl s formulation, "As ability to get B to do something it would not oth erwise do"?has certain strengths, but they find it so fraught with theoretical and empirical difficulties as to be practically unusable. In addition to stressing the problems of empirically operationalizing a re lational definition, they argue that employing such an approach makes it difficult to say much about the causal role of power factors relative to other potential independent variables. As Wohlforth writes: "If one defines power as control [over other actors, outcomes, or the international system as a whole], one must infer the relationship of power from outcomes_Inferring the balance of power from out comes and then using the balance of power to explain those outcomes appears to be a dubious analyt ical exercise." For a clear discussion of these issues, see Wohlforth, 1-17. For arguments against the use of broad material definitions of power, see Robert Dahl, "The Concept of Power," Behavioral Science 2 (July 1957); and David A. Baldwin, Paradoxes of Power (New York Basil Blackwell, 1989). See also

Waltz (fn. 1), 191-92; and Robert O. Keohane, "Realism, Neorealism and the Study of World Poli tics," in Keohane (fn. 1), 11.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

152 WORLD POLITICS

"interests," by which they mean the goals or preferences that guide the country's external behavior.

Instead of assuming that states seek security, neoclassical realists as sume that states respond to the uncertainties of international anarchy by seeking to control and shape their external environment. Regardless of the myriad ways that states may define their interests, this school ar gues, they are likely to want more rather than less external influence, and pursue such influence to the extent that they are able to do so.16 The central empirical prediction of neoclassical realism is thus that over the long term the relative amount of material power resources countries possess will shape the magnitude and ambition?the envelope, as it were?of their foreign policies: as their relative power rises states will seek more influence abroad, and as it falls their actions and ambitions

will be scaled back accordingly. Yet a theory of foreign policy limited to systemic factors alone is

bound to be inaccurate much of the time, the neoclassical realists argue, which is why offensive realism is also misguided. To understand the way states interpret and respond to their external environment, they say, one must analyze how systemic pressures are translated through unit level intervening variables such as decision-makers' perceptions and do

mestic state structure. In the neoclassical realist world leaders can be

constrained by both international and domestic politics. International anarchy, moreover, is neither Hobbesian nor benign but rather murky and difficult to read. States existing within it have a hard time seeing clearly whether security is plentiful or scarce and must grope their way forward in twilight, interpreting partial and problematic evidence ac cording to subjective rules of thumb.

In this respect, therefore, neoclassical realists occupy a middle ground between pure structural theorists and constructivists. The former im plicitly accept a clear and direct link between systemic constraints and unit-level behavior; the latter deny that any objective systemic con straints exist at all, arguing instead that international reality is socially constructed and that "anarchy is what states make of it."17 Neoclassical

16 One member of the school writes that "classical realists have written carelessly about power-max imization/ leaving unclear whether states expand for material resources or as a consequence of mate rial resources. [Neoclassical realism] makes the latter assumption; increased resources give rise to greater ambitions. States are not resource-maximizers but influence-maximizers" (Zakaria, 19). Schweller considers this assumption too limiting and advocates incorporating a broader range of po tential state preferences into neoclassical realist theorizing; see Deadly Imbalances, 18-26, 217 fn. 37; and idem (fn. 4).

17 See Alexander Wendt, "Anarchy Is What States Make of It," International Organization 46 (Spring 1992); and idem, "Constructing International Politics," International Security 20 (Summer 1995).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 153

realists assume that there is indeed something like an objective reality of relative power, which will, for example, have dramatic effects on the outcomes of state interactions. They do not assume, however, that states necessarily apprehend that reality accurately on a day-to-day basis.

Hans Morgenthau famously argued that with his theory one could peer over the statesman's shoulder; neoclassical realists believe the same but feel that in doing so one sees through a glass, darkly. The world states end up inhabiting, therefore, is indeed partly of their own making.

It might be asked why, given their outlook, these authors are not best described simply as "classical" realists?why we must add yet another bit of jargon to an already burgeoning lexicon. The reason is that un fortunately there is no simple, straightforward classical realism. Rather, the term covers a host of authors who differ greatly from one another in assumptions, objectives, and methodologies, and thus is not helpful for current purposes.18 What sets the authors under discussion apart as a distinct school worthy of recognition is both the common nature of their quest?to develop an explicit and generalizable theory of foreign policy?and the common threads of their argumentation. Their central concern is to build on and advance the work of previous students of rel ative power by elaborating the role of domestic-level intervening vari ables, systematizing the approach, and testing it against contemporary competitors. The differences among the four general theories are sum marized in Table 1.

Because neoclassical realism stresses the role played by both inde pendent and intervening variables, it carries with it a distinct method ological preference?for theoretically informed narratives, ideally supplemented by explicit counterfactual analysis, that trace the ways different factors combine to yield particular foreign policies. The neo classical realist archetype is Thucydides' History of the Peloponnesian

War, which grounds its narrative in the theoretical proposition that the "real cause" of the war was "the growth of the power of Athens, and the alarm which this inspired in Sparta," and then describes how systemic

18 Michael Doyle has recently distinguished three separate theoretical strands within the classical realist tradition: Machiavelli's "fundamentalism," which emphasizes the importance of individual am bition; Hobbes s "structuralism," which emphasizes the importance of the international system; and Rousseau's "constitutionalism," which emphasizes the importance of unit-level factors such as the na ture and strength of state-society relations. All three strands, he argues, have their fons et origo in

Thucydides' "complex" realism, which incorporates variables from each level of analysis; see Michael W. Doyle, Ways of War and Peace (New York: W. W. Norton, 1997). For analysis of previous modern

"classical" realists, see Michael Joseph Smith, Realist Thought from Weber to Kissinger (Baton Rouge: Louisiana State University Press, 1986).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

154 WORLD POLITICS

TABLE 1 Four Theories of Foreign Policy

View of Theory International System View of Units Causal Logic

Innenpolitik theories

unimportant highly internal factors ?? foreign policy differentiated

Defensive realism

Neoclassical realism

Offensive realism

occasionally important; anarchy's implications variable

important; anarchy is murky

very important; anarchy is Hobbesian

highly systemic or internal ?? foreign policy differentiated incentives factors

(two sets of independent variables in practice, driving "natural" and "unnatural" behavior respectively)

differentiated systemic ?> internal ?> foreign policy incentives factors (independent (intervening variable) variables)

undifferentiated systemic incentives ?> foreign policy

incentives were translated through unit-level variables into the foreign policies of the various Greek city-states.19

In keeping with this tradition, the major neoclassical realist works to date have been narratives or case studies of how great powers have re sponded to relative material rise or decline: Fareed Zakaria on the

United States; William Curti Wohlforth on the Soviet Union; Thomas J. Christensen on the United States and China; Randall L. Schweller on the belligerents of World War II. These same authors have also tackled issues ranging from the formation of alliances to the role of do

mestic politics in war initiation to the challenges facing contemporary American policymakers. Their collective output represents some of the most substantial and sophisticated work on foreign policy currently available.20

19 Strassler (fn. 5), 1.23. For an excellent discussion of Thucydides as an international relations the orist, see Doyle (fn. 18), 49-92; other interesting recent treatments include Mark V. Kauppi, "Thucyd ides: Character and Capabilities," Security Studies 5 (Winter 1995); and Ashley J. Tellis, "Political

Realism: The Long March to Scientific Theory," Security Studies 5 (Winter 1995), 12-25. 20 Recent statecentric writings, particularly on foreign economic policy, represent a comparably rig

orous and impressive literature; for a sampling of this work, see G.John Ikenberry et al., eds., The State and American Foreign Economic Policy (Ithaca, N.Y.: Cornell University Press, 1988); and G.John Iken berry, ed., American Foreign Policy: Theoretical Essays, 2d ed. (New York: HarperCollins, 1996). Another approach, known as comparative foreign policy or foreign policy analysis, has generally produced little cumulation of knowledge or lasting impact; its recent offerings can be sampled in Charles F. Hermann et al., eds., New Directions in the Study of Foreign Policy (Winchester, Mass.: Unwin Hyman, 1987); and Neacketal.(fn.8).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

neoclassical realism 155

The Rise and Fall of the Great Powers

The primary subject of all the major neoclassical realist works is the impact of relative power on foreign policy?which makes them the third wave of books on this hardy realist theme in the last two decades. The first wave came in the 1980s, as Robert Gilpin, Paul Kennedy, and Michael Mandelbaum all used relative power as the ordering principle for impressive and wide-ranging studies of international politics over several centuries. They argued that beneath the apparent chaos of events lay substantial regularities. As Mandelbaum put it, "Similar se curity policies recur throughout history and across the international system in states that, whatever their differences, occupy similar posi tions in the system. . . . The security policies of very strong states are different from those of very weak ones, and both differ from those of states that are neither very strong nor very weak."21 When individual states moved from one rank to the next, moreover, their foreign policies eventually followed suit: "The historical record suggests," Kennedy wrote, "that there is a very clear connection in the long run between an individual Great Power's economic rise and fall and its growth and de cline as an important military power (or world empire)."22 The reason for this pattern, Gilpin explained, was that states were continually "tempted to try to increase [their] control over the environment.... A more wealthy and more powerful state ... will select a larger bundle of security and welfare goals than a less wealthy and less powerful state."23 The second wave consisted of works by Aaron L. Friedberg and

Melvyn P. Leffler that traced precisely how a shift in relative power led to a shift in the foreign policy of a particular country.24 Friedberg began his analysis with the relative decline of Britain's economic and military strength around the turn of the twentieth century; his goal was to un derstand when and how this decline started to affect Britain's external

behavior. As he noted: "Structural considerations provide a useful point from which to begin analysis of international politics rather than a place at which to end it. Even if one acknowledges that structures exist

21 Michael Mandelbaum, The Fates of Nations: The Search for National Security in the Nineteenth and Twentieth Centuries (Cambridge: Cambridge University Press, 1988), 4,2.

22 Paul Kennedy, The Rise and Fall of the Great Powers (New York: Random House, 1987), xxii, em phasis in original.

23 Robert Gilpin, War and Change in World Politics (Cambridge: Cambridge University Press, 1981), 94-95,22-23.

24 Aaron L. Friedberg, The Weary Titan: Britain and the Experience of Relative Decline, 1895-1905 (Princeton: Princeton University Press, 1988); and Melvyn P. Leffler, A Preponderance of Power: Na tional Security, the Truman Administration, and the Cold War (Stanford, Calif.: Stanford University Press, 1992).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

156 WORLD POLITICS

and are important, there is still the question of how statesmen grasp their contours from the inside, so to speak," and of what they will do about them.25

Friedberg found that in practice British officials reacted to decline haphazardly, pursuing policies that "simply ignored or papered over se rious underlying weaknesses in Britain's position or, in solving certain problems, created new and perhaps more dangerous ones." This is not the response one would expect from a unitary actor responding ratio nally to incentives from the international system, and he argued that to explain it properly one had to consider not merely changes in relative capabilities but also organizational, intellectual, and domestic political factors. Assessments of relative power by policy-making elites, Fried berg concluded, "are related to but not directly determined by reality" and are, "in turn, related to but not fully determinative of policy."26

Leffler's study of American foreign policy during the early cold war examined the opposite situation?a case in which relative power was increasing rather than decreasing. Instead of following the lead of most traditional or revisionist historians in highlighting the objective nature of either a postwar Soviet threat or an American ideological quest for global dominance, he took his stand with the postrevisionists and fo cused on the dynamic interaction between the two countries, their goals, and their relative strength. Most importantly, he demonstrated how changing capabilities helped to drive policymakers' perceptions of external threats, interests, and opportunities. Worries about the Soviet Union underlay the policies of the Truman administration, Leffler ar gued, but those worries were themselves partly the product of increased American strength: American policymakers were concerned not about an immediate or primarily military threat but rather about some poten tial future challenge to America's broader environment. Only the great est of powers, one might point out, have the luxury of viewing their national interests so expansively; certainly the United States did not do so earlier in its history, when its ideals and institutions were the same but its geopolitical position was different.

The neoclassical realists pick up where these earlier waves left off and demonstrate the applicability of this line of analysis to a wide vari ety of times and places. Thus, in his compelling study of U.S. foreign policy in the late nineteenth century, From Wealth to Power, Fareed Za karia asks: "Why, as states grow increasingly wealthy, do they build

25 Friedberg (fn. 24), 8. 26 Ibid., 295,290-91.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 157

large armies, entangle themselves in politics beyond their borders, and seek international influence?" (p. 3). Echoing the basic answer of the first wave, he argues that this behavior stems from the tendency of states to use the tools at their disposal to gain control over their envi ronment. William Curti Wohlforth, meanwhile, grounds his analysis of Soviet foreign policy during the cold war in the notion that "state be havior [is an] adaptation to external constraints conditioned by changes in relative power."27 And Thomas J. Christensen, in Useful Adversaries, argues that U.S. and Chinese foreign policies during the early cold war

were driven in the first place by shifting distributions of power in the international system.

The influence of relative power on national policies is not obliterated even by world-historical leaders?or at least so Randall L. Schweller contends in Deadly Imbalances, his neoclassical realist study of foreign policy dynamics before and during World War II. Conventional wis dom that explains the onset and course ofthat war largely by reference to the character and views of Adolph Hitler is misguided, Schweller ar gues, because the structure of the international system?that is, the dis tribution of material power capabilities across the units?had a critical impact on alliance patterns and foreign policies during the 1930s and 1940s. He documents the existence of a comprehensive international pecking order dominated by three poles (the United States, the Soviet Union, and Germany) and traces its influences on the behavior of pow ers of various different sizes. His analysis makes it clear that the con ventional neorealist division between bipolar and multipolar systems is inadequate for many purposes and that a much closer look at the dis tribution of power may be necessary in order to uncover the foreign policy effects that system structure should be expected to produce.

Perception and Misperception in International Politics

In stressing the primacy of relative power, the neoclassical realists part company with the Innenpolitikers. They separate themselves from many other structural theorists, however, through a further contention that the impact of such power on policy is indirect and problematic. The first intervening variable they introduce is decision-makers' percep tions, through which systemic pressures must be filtered.

27 William Curti Wohlforth, "Realism and the End of the Cold War," in Brown et al., 8. This arti cle follows through on the argument of Wohlforth's book The Elusive Balance and should be read as its final chapter.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

158 WORLD POLITICS

Purely systemic explanations of foreign policy presume a reasonably accurate apprehension by officials of the distribution of power and a reasonably direct translation of such apprehensions into national pol icy. "In most structural realist formulations," Friedberg noted, "assess

ment [of relative power] through rational calculation plays the part of a reliable but invisible transmission belt connecting objective [material] change to adaptive behavior."28 Robert O. Keohane has made the same point, arguing that for most systemic theorists "the link between sys tem structure and actor behavior is forged by the rationality assump tion, which enables the theorist to predict that leaders will respond to the incentives and constraints imposed by their environments. Taking rationality as a constant permits one to attribute variations in state be havior to various characteristics of the international system."29

Neoclassical realists, in contrast, argue that the notion of a smoothly functioning mechanical transmission belt 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.30 What this means in practice is that the translation of capabilities into national behavior is often rough and capricious over the short and medium term.

Friedberg found that in turn-of-the-century Britain "official assess ments did not adjust steadily, but neither did they shift dramatically and decisively as the result of external shocks. . . . [C]hange went for

ward as the result of gradual, diffuse intellectual developments that were consolidated and accelerated by periodic crises."31 The process of assessment, moreover, was fragmented along bureaucratic and func tional lines within the British government, with debates over relative power centering on simple numerical indicators of capability which often held sway because of their familiarity or cognitive appeal rather than their substantive appropriateness. As a result, the actual British policy response to relative decline was significantly more halting, incon sistent, and "nonstrategic" than a simple structural model would predict.

Every neoclassical realist makes a similar point, and some put per ceptions at the heart of their work. In The Elusive Balance, for example,

28 Friedberg (fn. 24), 13. 29 Keohane, "Theory of World Politics," in Keohane (fn. 1), 167. 30Following Robert Jervis's Perception and Misperception in International Politics (Princeton: Prince

ton Princeton University Press, 1976), Wohlforth first dwelt on the implications of this point in "The Perception of Power: Russia in the Pre-1914 Balance," World Politics 39 (April 1987).

31Friedberg (fn. 24), 288.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 159

Wohlforth supports the general thrust of Leffler's argument while looking at cold war dynamics from the Soviet side. World War II may have eliminated the Axis, he points out, but it did little to establish a clear hierarchy among the victorious allies and thus set the stage for endless disputes in the decades afterward. At base, he contends, the re current cycles of superpower tension from the 1940s to the 1980s were quite similar, and all were rooted in the ambiguities of relative power and policymakers' perceptions of it: "Each [cycle of tension] was shaped by a change in the power relationship differently interpreted by the two sides.... In the wake of each shift, each side tried to maximize its own position. Unwilling to go to war to test the power distribution, they reached stalemates after crises, posturing and signaling until a new per ceived shift led to another round" (pp. 301-2).

For Wohlforth, therefore, the cold war is best understood not as a sta

ble bipolar arrangement in which the superpowers acted as sensible du opolists but rather as an ongoing dispute between the U.S. and USSR over who had how much power and what influence over the interna tional system they were thus entitled to exercise. The Soviet Union, he argues, constantly struggled to gain a share of the international spoils? influence abroad, control over international institutions, general prestige and deference?commensurate with its perceived power capabilities. The United States, perceiving its own power capabilities to be greater and more diversified, struggled to deny the Soviets such a global role. Periodically these tensions came to a boil, with the episodes displaying a familiar pattern: "a perceived shift in power, publicly acknowledged by both sides; a new Soviet drive for increased prestige; positive early feed back on the new policy; sharp crises that eventually revealed the contra dictions between the two sides' interpretations of the political implications of the power shift"; and an eventual relaxation of tensions based on mutual acceptance of a stalemate (p. 182). Wohlforth argues that during 1983-85 the last cold war cycle began to wind down and

would probably have ended with a new mini-d?tente ratifying the status quo circa, say, 1970. In 1985, however, Gorbachev's reforms altered the

picture irrevocably, leading (albeit unintentionally) to the shedding of the Soviet empire and then the dissolution of the USSR itself.

Together Leffler and Wohlforth provide a comprehensive view of the two superpowers' foreign policies from the beginning of the cold

war to its end, with changing relative power ultimately driving threat perceptions at each key point. Tracing the connections between power and policy, however, is more difficult than it might seem?because, as

Wohlforth says, "rapid shifts in behavior may be related to perceived

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

160 WORLD POLITICS

shifts in the distribution of power which are not captured by typical measures of capabilities." Over the decades U.S. and Soviet perceptions of power "followed a broad pattern ... [that] was connected to changes in real capabilities . . . [but] it would be impossible to choose a single indicator or composite index [of power capabilities] that would predict the precise perceptual pattern without prior knowledge" (pp. 294, 302).32

As a case study of how relative material capabilities, perceptions of them, and other factors combine to shape historical developments,

Wohlforth offers Soviet policy at the end of the cold war. Gorbachev was spurred to launch his campaign of domestic renewal, he shows, by external stalemate, internal assessments that Soviet capabilities had greatly deteriorated, and a conviction that appropriate reforms could undo the damage. It was the combination of concern over perceived relative decline and confidence that he could reverse it, in other words,

that led Gorbachev to embark on the far-reaching changes which ulti mately brought his entire system crashing down. The full and devastating extent of Soviet weaknesses became clear only as the reforms progressed, however, and so by the time the Soviet Union's external and internal col lapses suddenly loomed they were practically faits accomplis.33

Such vagaries are a common feature of neoclassical realist analysis. In their books Zakaria and Christensen both note the importance of perceptual "shocks," in which single events suddenly make decision makers aware of the cumulative effects of gradual long-term power trends.34 Elsewhere Christensen argues that European leaders have often misread both the distribution of capabilities and the efficacy of

32The waters are further muddied, he argues, by a host of other problems: "Power cannot be tested; different elements of power possess different utilities at different times; the relation of perceived power to material resources can be capricious; the mechanics of power are surrounded by uncertainty; states possess different conversion ratios and comparative advantages; the perceived prestige hierarchy and the military distribution may not coincide for prolonged periods; states adopt asymmetrical strategies to maximize their positions and undercut rivals; signals get confused among allies, rivals, and domes tic audiences" (pp. 306-7).

33In addition to his article "Realism and the End of the Cold War" (fn. 27), Wohlforth has teamed

up with Schweller for a further neoclassical realist take on this subject; see Randall L. Schweller and William C. Wohlforth, "Power Test: Updating Realism in Response to the End of the Cold War," Se

curity Studies (forthcoming). On the difficulties in drawing clear theoretical lessons from these events, however, see William C. Wohlforth, "Reality Check Revising Theories of International Politics in Re sponse to the End of the Cold War," World Politics 50 (July 1998).

34Zakaria finds that American "statesmen's perceptions of national power shift[ed] suddenly, rather than incrementally, and [were] shaped more by crises and galvanizing events like wars than by statisti cal measures" (p. 11). Christensen argues that it was only the sudden awareness, in 1947, of the extent of British decline that shocked the Truman administration into recognizing the true distribution of power and triggered the shift toward active containment (pp. 32ff).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 161

offensive and defensive military strategies, and thus acted in ways con trary to what pure systemic theories would predict.35 And Schweller, fi nally, argues in Deadly Imbalances that it was actually a misperception of the distribution of power which drove the foreign policy of one of the poles of the international system at the beginning of World War II.

Given the true state of affairs, he writes,

it would have been far better for the Soviets to have balanced against, rather than bandwagoned with, Germany [in 1939]. In that case Stalin would have presented Hitler with the prospect of a two-front war, seriously undermining the F?hrers strategy and perhaps causing its abandonment. But because he mis takenly perceived Europe as a tripolar, not a bipolar, system with France and Britain as the third pole, Stalin expected a war of attrition in the West. The fall of France abruptly ended Stalins dream of easy conquests in a postwar period when the rest of Europe would be exhausted, (p. 168)

Bringing the State Back In

The second intervening variable emphasized by neoclassical realists? especially Zakaria and Christensen?is the strength of a country's state apparatus and its relation to the surrounding society. Gross assessments of the international distribution of power are inadequate, they contend, because national leaders may not have easy access to a country's total material power resources. Once raised, the notion that international power analysis must take into account the ability of governments to ex tract and direct the resources of their societies seems almost obvious, and

in fact it simply involves incorporating into international relations the ory variables that are routine in other subfields of political science.36 And yet this represents an important and powerful development in realist

35 Thomas J. Christensen, "Perceptions and Alliances in Europe, 1865-1940," International Organi zation 51 (Winter 1997).

36 As Zakaria points out, everyone knows Charles Tilly's mantra that "war made the state and the state made war"; it is just that heretofore the implications of the first clause have received far more at tention than those of the second. See Zakaria, From Wealth to Power, 39-40; Tilly, "Reflections on the

History of European State-Making," in Charles Tilly, ed., The Formation of National States in Western Europe (Princeton: Princeton University Press, 1975), 42; and Christensen, Useful Adversaries, 20ff. A similar stress on the role of state structure is a characteristic of some recent Innenpolitik theories as well, although the two schools differ over the nature and importance of this variable and the interpretation of many cases; for an overview of this work, see Evangelista (fn. 8). For pioneering examinations of the role of the state in the formation and implementation of foreign policy, see Peter J. Katzenstein, ed., Between Power and Plenty (Madison: University of Wisconsin Press, 1978); Stephen Krasner, Defend ing the National Interest (Princeton: Princeton University Press, 1978); Ikenberry et al. (fn. 20); and

Michael Mastanduno et al., "Toward a Realist Theory of State Action," International Studies Quarterly 33 (December 1989).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

162 WORLD POLITICS

theorizing, for it brings analysis significantly closer to the real world without abandoning the paradigm's core concepts and assumptions.

Zakaria observes that historians note the expansion of American foreign policy in the years before World War I and ask why it occurred. Yet for a while, even after the United States had become perhaps the richest country in the world, most opportunities to expand American influence abroad were rejected?and even when it did become active later on, the U.S. lagged behind its European counterparts. "For a po litical scientist," therefore, "viewing the country's power and expansion in comparative perspective, the more puzzling question is why America did not expand more and sooner" (p. 5).

Zakaria concludes, after testing dozens of opportunities for expan sion against propositions derived from different theories, that Ameri can external behavior depended on the means at the disposal of national decision makers. He thus affirms the logic that capabilities shape intentions but finds it necessary to introduce state strength as an intervening variable between national capabilities and officials' behav ior: "Foreign policy is made not by the nation as a whole but by its gov ernment. 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 de cisionmakers can achieve their ends" (p. 9). His story of American for eign policy during these years, therefore, includes a discussion of the emergence of the administrative state:

The decades after the Civil War saw the beginning of a long period of growth in Americas material resources. But this national power lay dormant beneath a weak state, one that was decentralized, diffuse, and divided. The presidents and their secretaries of state tried repeatedly to convert the nations rising power into influence abroad, but they presided over a federal state structure and a tiny cen tral bureaucracy that could not get men or money from the state governments or from society at large_The 1880s and 1890s mark the beginnings of the mod ern American state, which emerged primarily to cope with the domestic pres sures generated by industrialization. . . . This transformation of state structure complemented the continuing growth of national power, and by the mid-1890s the executive branch was able to bypass Congress or coerce it into expanding

American interests abroad. Americas resounding victory in the Spanish-Amer ican War crystallized the perception of increasing American power . . . [and]

America expanded dramatically in the years that followed, (pp. 10-11)

Zakaria explicitly tests propositions drawn from defensive realism against his cases and finds that such a security-based approach is only

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 163

sporadically supported by the evidence. According to defensive realism, he claims, nations are supposed to exert themselves on the international scene "in times of insecurity, against powerful nations with aggressive intentions." Instead, "when confronted by real threats . . . the United States usually opted to contract its interests." Conversely, "greater secu rity bred greater activism and expansion" (pp. 11-12).

Christensen, meanwhile, notes the widespread consensus among scholars that Sino-American cooperation from 1972 onward is best ex plained by a shared realist desire to balance against the Soviet Union? and the equally widespread consensus that earlier Sino-American tensions are best explained by Innenpolitik variables (such as ideological differences, domestic political pressures, or leaders' psychology). He sets out to show that the latter proposition is not strictly true and that

American and Chinese behavior had its real source in the international

system even during the late 1940s and 1950s. Christensen argues in Useful Adversaries that at critical points during

these years both the American and the Chinese leadership felt com pelled to mobilize national resources in order to respond to perceived shifts in the international balance of power?to engage, that is, in what

Waltz describes as "internal balancing" (p. 245).37 Christensen stresses, however, just how difficult it is for countries to execute such operations, especially when they are accompanied by major shifts in national policy.

He therefore introduces the concept of "national political power," which he defines as "the ability of state leaders to mobilize their nation's human and material resources behind security policy initiatives." Like Zakaria's "state power," this acts as "a key intervening variable between the international challenges facing the nation and the strategies adopted by the state to meet those challenges" (pp. 11, 13). Because

American and Chinese statesmen lacked sufficient "national political power" to do exactly as they pleased, Christensen argues, they had to use domestically popular but unnecessary policies in a secondary arena (conflict with each other) as a cover for unpopular but necessary poli cies in a primary arena (mobilization against the Soviet Union):

Viewing basic changes in the international balance of power, Truman in 1947 and Mao in 1958 decided to mobilize their nations around long-term strategies designed to respond to those shifts. In both cases, the strategies adopted re quired significant public sacrifice in peacetime, so the leaders faced difficulties in selling those strategies to their respective publics. The manipulation or exten

37Cf. Waltz (fn.l), 168.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

164 WORLD POLITICS

sion of short-term conflict with the other nation, while not desirable on straightforward international or domestic grounds, became useful in gaining and maintaining public support for the core grand strategy, (p. 6)38

Christensen does not take a position in the theoretical debate be tween defensive realists and their critics over whether active foreign policies are usually driven by increased power or increased threat; his model is general enough to incorporate both. From his tone and choice of cases, however, one gets a picture of far-sighted elites sensitive to the consequences of changing relative power, yoked to penny-pinching publics who respond only to obvious, short-term military threats. The elites tend to get what they want in the end but have to make conces sions to their publics along the way?with the result that foreign policy is linked to systemic incentives but not wholly determined by them.

Other neoclassical realists advocate exploring the influence of addi tional intervening variables on foreign policy. Friedberg captures their general attitude when he writes that "neorealists are probably right that, all other things being equal, multipolar systems are intrinsically unsta ble. In the real world, however, everything else is not equal, and non structural factors can serve either to exacerbate or to mitigate the tendencies that are inherent in a system's structure."39 Schweller argues in Deadly Imbalances that a full theory of foreign policy should include the nature of states' goals or interests, which he operationalizes as the degree to which they are status quo or revisionist?satisfied or dissatis fied with the existing distribution of international spoils, "the prestige, resources, and principles of the system" (p. 24) .40 By combining degrees of relative power and revisionism, he conjures up an international bes tiary and shows how each country played to its predicted type before and during World War II: strongly revisionist great powers such as Nazi Germany acted like "wolves," moderately revisionist great powers such as the Soviet Union acted like "foxes," indifferent great powers such as

38 In some respects Christensen follows here in the footsteps of revisionist historians such as Richard M. Freeland; see Freeland, The Truman Doctrine and the Origins ofMcCarthyism: Foreign Pol icy, Domestic Politics, and Internal Security, 1946?48 (New York: Schocken Books, 1974). Unlike revi sionist analyses of Truman's China policy, however, Christensen downplays the role of economic

motives in American behavior and sees the Truman administration as using domestic anticommunism rather than creating it, and being in control of it rather than being controlled by it.

39 Aaron L. Friedberg, "Ripe For Rivalry: Prospects for Peace in a Multipolar Asia," International Security 18 (Winter 1993-94), 11. See also Randall L. Schweller, "Domestic Structure and Preventive

War: Are Democracies More Pacific?" World Politics 44 (January 1992). For a neoclassical realist analy sis of how domestic-level variables can be incorporated into realist theories, see Jennifer Sterling Folker, "Realist Environment, Liberal Process, and Domestic-Level Variables," International Studies Quarterly 47 (1997).

40 For Scwheller's discussion of revisionism, see pp. 19-26; and idem, "Bandwagoning for Profit: Bringing the Revisionist State Back In," in Brown et al.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 165

the United States acted like "ostriches," revisionist lesser powers such as Italy and Japan acted like "jackals," and so forth.

Quite apart from the vividness of its presentation there is clearly something to this idea, but unfortunately Schweller slights discussion of the sources of revisionism and so fails to integrate the concept or ganically into his broader systemic argument. Sometimes he implies that revisionism is a domestic pathology?that is, a purely unit-level variable. Yet at other times he implies that revisionism can emerge sim ply because changes in the systemic superstructure (the distribution of international spoils) do not keep pace with changes in the systemic base (the distribution of power capabilities). This latter revisionism would not be a unit-level factor at all and would not require the positing of a true difference in state interest, abstractly conceived: it would be a product of the dynamics of the system, rather than following from the character of the revisionist state itself. One of the chief contributions of

Gilpin, Kennedy, and Mandelbaum, in fact, was to show just such a process at work time and again and to illustrate how much history could be accounted for by the simple story of "differentials in growth rates and technological change, leading to shifts in the global economic balances, which in turn gradually impinge upon the political and mili tary balances."41 Contra Schweller, therefore, revisionism may well cre ate more trouble?and require more accommodation?in the practical realm than in the theoretical one.

Designing Social Inquiry

A distinct methodological perspective flows from neoclassical realism's theoretical argument: analysts wanting to understand any particular case need to do justice to the full complexity of the causal chain linking relative material power and foreign policy outputs. Realism, in this view, is a theoretical hedgehog: it knows one big thing, that systemic forces and relative material power shape state behavior. People who ig nore this basic insight will often waste their time looking at variables that are actually epiphenomenal. Yet people who cannot move beyond the system will have difficulty explaining most of what happens in in ternational relations. Waltz himself captured this dynamic best when he wrote: "The third image describes the framework of world politics, but without the first and second images there can be no knowledge of the forces that determine policy; the first and second images describe

Kennedy (fn. 22), xx.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

166 WORLD POLITICS

the forces in world politics, but without the third image it is impossible to assess their importance or predict their results."42

Neoclassical realists therefore think that neither spare game-theo retic modeling nor pure "thick description" are good approaches to for eign policy analysis. They favor beginning intellectually at the systemic level but then taking care to trace precisely how, in actual cases, relative power is translated and operationalized into the behavior of state ac tors.43 To some extent, they agree with Robert O. Keohane that "the de bate between advocates of parsimony and proponents of contextual subtlety resolves itself into a question of stages, rather than either-or choices. We should seek parsimony first, then add on complexity while monitoring the effects this has on the predictive power of our theory: its ability to make significant inferences on the basis of limited infor

mation."44 A major dilemma they confront, however, is their appreciation of the degree to which their central, parsimonious independent variable needs to be studied in conjunction with & variety of messy contextual fac tors in order to say much of interest about their subject matter. For neo classical realism, to paraphrase Clausewitz, explaining foreign policy is usually very simple, but even the simplest explanation is difficult. While many in the field have come to favor a formal, universalist ap

proach to political phenomena, neoclassical realists stubbornly insist that significant area expertise is critical for an accurate understanding of countries' foreign policy behavior. The theory's basic concepts "are simple and generalizable across cultures and political systems," they contend, but "the application of the approach to any given country re quires a great deal of knowledge about the nation in question" (Chris tensen, 248). To investigate how perceptions matter, for example, one has to get inside the heads of key state decision makers, something that often requires foreign language capabilities and/or archival research. And to incorporate state structure as an intervening variable, one has to know a decent amount about how different countries' political insti tutions work, both in theory and in practice. Accordingly, the volumes by Wohlforth and Christensen (like those by Friedberg and Leffler) are

42 Kenneth N. Waltz, Man, the State and War (New York: Columbia University Press, 1959), 238. 43 On the use of "process tracing," see Alexander L. George, "Case Studies and Theory Develop

ment: The Method of Structured, Focused Comparison," in Paul Gordon Lauren, ed., Diplomacy: New Approaches in History, Theory, and Policy (New York: Free Press, 1979); and Alexander L. George and Timothy J. McKeown, "Case Studies and Theories of Organizational Decisionmaking," in Advances in Information Processing in Organizations, vol. 2 (]M Press, 1985). For an argument that Innenpolitik rather than systemic variables deserve to be the starting point for such a method, see Moravcsik (fn. 8), 541ff.

44 Keohane (fn. 29), 187-88, emphasis in original.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 167

based on extensive and ground-breaking archival work, while both Schweller and Zakaria incorporate significant amounts of primary source material and historical nuance into their analyses.

Some might question whether the result is truly a theory of foreign policy at all. Hard-line positivists and historians, for example, might both point to the lack of precise predictions generated by neoclassical realism, the stress it places on detailed historical analysis, and so on, and claim that this approach should not really be called social science. (Pos itivists would say this disapprovingly, of course, while historians would say it approvingly.) Still, they would not be correct. "A theory," Waltz reminds us, "indicates that some factors are more important than oth ers and specifies relations among them_A theory arranges phenom ena so that they are seen as mutually dependent; it connects otherwise disparate facts; it shows how changes in some of the phenomena neces sarily entail changes in others."45 Whether or not Waltz himself would agree, the neoclassical realist theory of foreign policy does most of these things, linking clearly specified independent, intervening, and depen dent variables in a direct causal chain. "It does not simply state that do mestic politics matter in foreign policy, but specifies the conditions under which they matter" (Christensen, 252).

Thus neoclassical realism predicts that an increase in relative mate rial power will lead eventually to a corresponding expansion in the am bition and scope of a country's foreign policy activity?and that a decrease in such power will lead eventually to a corresponding contrac tion. It also predicts that the process will not necessarily be gradual or uniform, however, because it will depend not solely on objective mate rial trends but also on how political decision makers subjectively per ceive them. And it predicts that countries with weak states will take longer to translate an increase in material power into expanded foreign policy activity or will take a more circuitous route there.

It is true nonetheless that neoclassical realism has a decidedly non mechanistic feel. It recognizes, in keeping with recent theoretical de velopments elsewhere in the physical and social sciences, that sometimes small choices can have big consequences and that foreign policy behavior may look "clocklike" only from a distance and over the long term; on close inspection and over the short to medium term, "cloudlike" activity may be the norm.46 Furthermore, neoclassical real

45 Waltz (fn.l), 8-10. 46 For the cloud/clock distinction and its implications, see Gabriel A. Almond with Stephen Genco,

"Clouds, Clocks, and the Study of Politics," in Almond, A Discipline Divided: Schools and Sects in Polit ical Science (Newbury Park, Calif.: Sage, 1990).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

168 WORLD POLITICS

ism does not claim that power-related factors will drive all aspects of a state's foreign policy, only that they will affect its broad contours. Crit ics might charge that all these qualifications make the theory harder (though not impossible) to falsify and thus discredit. Adherents would have to concede the point and acknowledge this as a serious weakness; they might counter, however, that neoclassical realism has compensat ing advantages, particularly in the opportunities it offers for building satisfying comprehensive explanations of foreign policy without aban doning the theory's core assumptions. Its very looseness, in other words, makes it a useful framework for carrying out the kind of midrange the orizing that so often is the best social science can hope to achieve.

Conclusion: The Road Ahead

On the evidence of its works to date, the neoclassical realist school has much to offer students of foreign policy. Theoretically, it retains signif icant abstraction and parsimony in its basic form while providing clear guidelines for those interested in achieving greater richness and fit.

Methodologically, it calls for an emphasis on theoretically informed narratives that trace how relative material power is translated into the

behavior of actual political decision makers. Its adherents have shown that this approach can illuminate the behavior of countries in many re gions of the world during many historical periods. By this point, how ever, it should be old news that relative power matters. Future work in this vein should therefore focus on continuing to specify the ways in tervening unit-level variables can deflect foreign policy from what pure structural theories might predict.

For example, despite the best efforts of neoclassical realists, the link between objective material power capabilities and policymakers' subjec tive assessment of them remains murky. Critics might see the school's emphasis on perceptions as a giant fudge factor, useful for explaining away instances where foreign policy and material power realities di verge. Precise theoretical development in this area would be helpful, ex plicating just how various psychological, ideational, and cultural factors may affect how political actors perceive their own and others' capabili ties and how such perceptions are translated into foreign policy.47

47 Two recent examples of how psychological insights can successfully be brought into foreign pol icy analysis are Yuen Foong Khong, Analogies at War: Korea, Munich, Dien Bien Phu, and the Vietnam

Decisions of 1985 (Princeton: Princeton University Press, 1992); and Jonathan Mercer, Reputation and International Politics (Ithaca, N.Y.: Cornell University Press, 1996); a useful survey of recent work in this area is James M. Goldgeier, "Psychology and Security," Security Studies 6 (Summer 1997).

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 169

It should be possible, moreover, to explore further the notion that national power capabilities must be "usable" to have an impact on for eign policy. Future work could analyze how different state structures constrain or facilitate the extraction and deployment of national power by state leaders. And other variables serving as power "multipliers" or "dividers" might be uncovered, operating by themselves or in conjunc tion with those already described. Stephen Peter Rosen and Kenneth

M. Pollack, for example, have recently argued that cultural variables have shaped Indian and Arab institutions, respectively, in ways detri mental to military effectiveness; such hybrid theoretical linkages could easily be incorporated into a neoclassical realist framework without straying too far from the model's basic power-related argument.48

The impact of changing relative power on other factors could also be an important field of inquiry. Thus instead of viewing ideas as either purely independent or purely dependent variables, future neoclassical realists could explore how, in conjunction with relative power, they could play both roles simultaneously. From the Founding onward, for example, Americans have generally agreed that their domestic institu tions should be disseminated to others but have disagreed over the form this ideological transmission should take. "Exemplars" have believed that the nation should rest content with setting an example for the

world, while "crusaders" have believed the nation should take a more direct and activist role in shaping political developments abroad in ac cordance with American ideals.49 During most of the nineteenth cen tury the more modest version generally prevailed, typified by John

Quincy Adams's admonition that the country should go not abroad "in search of monsters to destroy." By the twentieth century the ambitious version had gained the upper hand, as Woodrow Wilson took the na tion to war to make the world "safe for democracy."50 From a neoclassi cal realist perspective, the first place to look in explaining such a shift

On ideas, see Sheri Berman, "Ideas, Norms, and Culture in Political Analysis" (Paper delivered at Work shop on Ideas and Culture in Political Analysis, Princeton University, May 1998). A recent sampler of foreign policy-related cultural analysis is Peter Katzenstein, ed., The Culture of National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996).

48 Rosen, Societies and Military Power: India and Its Armies (Ithaca, N.Y.: Cornell University Press, 1996); and Pollack, "The Influence of Arab Culture on Arab Military Effectiveness" (Ph.D. diss., MIT, 1996).

49 See Robert W. Tucker, "Exemplar or Crusader? Reflections on Americas Role," National Interest, no. 5 (Fall 1986).

50 The Adams quotation can be found in his "Address of July 4,1821," in Walter LaFeber, ta., John Quincy Adams and American Continental Empire (Chicago: Times Books, 1965), 45; the Wilson quota tion can be found in his "Address Recommending the Declaration of a State of War," April 2,1917, President Wilsons Foreign Policy: Messages, Addresses, Papers, ed. James B. Scott (New York: Oxford Uni versity Press, 1918), 287.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

170 WORLD POLITICS

would not be intellectual history or presidential psychology, but the massive increase in relative power the country had experienced between Adams and Wilson. One might still need to know the content of American political ideology, however, in order to understand the spe cific policy choices officials made in either era.

Neoclassical realism should also redirect our attention toward critical

issues such as what drives the waxing and waning of material power ca pabilities in the first place. Factors such as differential growth rates, it argues, will end up dictating the roles countries can play in world poli tics. For this reason if for no other, as Wohlforth says, "Any realist dis cussion of international change must combine the domestic and international levels of analysis. A [purely structural] realist explanation cannot offer a comprehensive account of precisely why a given state's domestic political, social, and economic institutions decline in compar ison to those of competing powers."51

If neoclassical realists continue to incorporate unit-level intervening variables into their basic power-oriented argument, ironically, they might find themselves bumping into chastened Innenpolitikers coming from the other direction. For as Matthew Evangelista has noted, "Per haps the most promising development in the field is the recognition among scholars inclined toward domestic explanations for foreign pol icy that these explanations are inadequate. Many scholars understand that they must incorporate factors at the level of the international sys tem into their explanations and, moreover, that they must do so in ways that are more systematic than the mere assertion that everything mat ters."'52

Future work, finally, should also develop neoclassical realism's dis tinct perspective on policy issues. Offensive realists generally predict that the future of international relations will resemble its conflict

ridden past. Defensive realists and Innenpolitikers often disagree, argu ing that great power conflict is likely to emerge if and only if military technology favors preemption or domestic pathologies drive countries to

51 Wohlforth (fn. 27), 19. This does not mean, of course, that easy answers to such questions are available. Paul Krugman was recently asked, "What are the great puzzles economists are trying to solve these days?" He replied, "The biggest question of all is still, Why are some countries rich and some countries poor?' Long ago, Bob Solow?the father of growth theory in economics?said that when it comes down to the question of why some countries do well over the long term and some do badly, you always end up in a blaze of amateur sociology. We're a little bit past that, but not much." Wired, May 1998,146.

52 Evangelista (fn. 4), 202; see also Harald M?ller and Thomas Risse-Kappen, "From the Outside In and the Inside Out: International Relations, Domestic Politics, and Foreign Policy," in Skidmore and Hudson (fn. 8), 29-32.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

NEOCLASSICAL REALISM 171

flagrant excess.53 Neoclassical realists, in contrast to both, emphasize the contingency of history and the importance of how foreign policy is actually conducted, because they see certain situations as particularly "ripe for rivalry."

For example, they view China's recent attempts to throw its weight around as normal and entirely predictable?just the sort of behavior one would expect from "a thoroughly traditional great power."54 "Fast rising powers are almost invariably troublemakers," Friedberg writes, "if only because they are reluctant to accept institutions, border divisions, and hierarchies of political prestige put in place when they were com paratively weak. Emerging powers seek to change, and in some cases to overthrow, the status quo and to establish new arrangements that more accurately reflect their own conception of their place in the world."55

Moreover, because rapid growth often produces social turmoil, because accommodation is tricky, and because China is emerging onto the scene in a multipolar regional environment lacking most of the ele

ments that can mitigate conflict, the future of East Asian (as opposed to European) international politics seems especially problematic.

Yet for neoclassical realists these are tendencies, not inexorable laws;

whether the region actually erupts into conflict, they argue, will depend in large part on how the United States, China, and other important

Asian powers decide to manage their ambivalent relationships. The starting point for policy advice, the school's adherents believe, is to rec ognize that the United States today is a status quo hegemon, whose fortunate situation is captured by the statement of a frank British offi cial in 1934: "We are in the remarkable position of not wanting to quarrel with anybody because we have got most of the world already or the best parts of it and we only want to keep what we have got and pre vent others from taking it away from us" (cited in Schweller, 24). In this view, one of the main tasks of American policymakers over the next sev eral years will be to analyze the nature and extent of Chinese revisionism and determine at what point accommodation is no longer wise. After

53 The degree of optimism or pessimism about the future among Innenpolitikers and defensive real ists, therefore, depends in part on how likely they think it is that at least one important power will suc cumb to a domestic pathology. For constrasting offensive and defensive realist views about future European security, see Mearsheimer, (fn. 9); Stephen Van Evera, "Primed for Peace," International Se curity 15 (Winter 1990-91); and Jack Snyder, "Averting Anarchy in the New Europe," International Security 15 (Winter 1990-91).

54 Fareed Zakaria, "Speak Softly, Carry a Veiled Threat," New York Times Magazine, February 18, 1996, 36.

55 Aaron Friedberg, "Warring States: Theoretical Models of Asia Pacific Security," Harvard Inter national Review 18 (Spring 1996), 13.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms

172 WORLD POLITICS

all, Schweller reminds us, the difference between a modestly and a strongly revisionist power is the difference between Weimar and Nazi

Germany, which obviously merited different policy responses (p. 32). In the end, neoclassical realism's relative modesty about its ability to

provide tidy answers or precise predictions should perhaps be seen not as a defect but rather as a virtue, stemming as it does from a judicious appraisal of its object of inquiry. As Aristotle noted, the "actions which political science investigates, admit of much variety and fluctuation_

We must be content, then, in speaking of such subjects ... to indicate the truth roughly and in outline ... for it is the mark of an educated man to look for precision in each class of things just so far as the na ture of the subject admits."56

56 Aristotle, Nicomachean Ethics 1:3, in Richard McKeon, ed., The Basic Works of Aristotle (New York: Random House, 1941), 936.

This content downloaded from 76.109.204.44 on Sun, 12 Aug 2018 23:50:44 UTC All use subject to https://about.jstor.org/terms