Theory of International Politics by Kenneth N. Waltz is a foundational work in the field of international relations theory, published in 1979 by Addison-Wesley Publishing Company as part of the Addison-Wesley Series in Political Science. The book represents a systematic attempt to construct a rigorous, scientific theory of international politics that can explain enduring patterns of state behavior and international outcomes.
The work is organized as a scholarly monograph that combines philosophical reflection on the nature of scientific inquiry with substantive theoretical construction and critical analysis of existing approaches. Waltz draws on his extensive academic career, acknowledging support from the National Science Foundation, the Institute of International Studies and Committee on Research at the University of California, Berkeley, the Center for International Affairs at Harvard, the Guggenheim Foundation, and the Institute for the Study of World Politics. The philosophy department of the London School of Economics provided an environment for relating international-political theory to broader issues in the philosophy of science.
The book’s intellectual debts are evident throughout, engaging with thinkers ranging from classical political theorists to contemporary social scientists. Waltz acknowledges colleagues who read drafts, including Robert Jervis and John Ruggie, as well as Robert Art and Glenn Snyder who made comments. The manuscript was completed in July 1978 in Harborside, Maine, with the author declaring it done “without a sense of completion” but with relief and gratitude.
The work stands as a major statement of the structural realist or neorealist approach to international politics, offering both a critique of existing theories and an original theoretical framework. It addresses fundamental questions about how to study international politics scientifically, what causes war and peace, and how the structure of the international system shapes state behavior. The book has become a standard reference point in international relations scholarship, generating extensive debate and influencing generations of scholars.
Context
The book emerged during a period of significant intellectual ferment in international relations scholarship. The behavioral revolution had transformed political science in the preceding decades, bringing new methodological sophistication but also creating confusion about the proper relationship between description, explanation, and theory. Waltz’s work responds to what he saw as fundamental confusions about the nature of theory and the appropriate methods for studying international politics.
The intellectual context includes several overlapping developments. First, the systems theory movement had gained prominence across the social sciences, with scholars like David Easton, Morton Kaplan, and others attempting to apply systems concepts to political phenomena. Waltz engages critically with this tradition, particularly the work of Richard Rosecrance, Stanley Hoffmann, and Morton Kaplan, arguing that their systems approaches fail to meet the requirements of genuine systems theory.
Second, the debate between reductionist and systemic explanations was central to social science methodology. Reductionist approaches sought to explain international outcomes through the attributes and interactions of states, while systemic approaches emphasized the constraints imposed by the structure of the international system itself. Waltz’s work represents a sustained argument for the necessity and possibility of systemic theory.
Third, the Cold War context shaped the substantive concerns of the book. The bipolar structure of post-World War II international politics, the nuclear revolution, and the stability of the postwar order are recurring themes. Waltz’s analysis of bipolarity as a source of stability reflects and engages with contemporary debates about the nature of the Cold War international system.
Fourth, the book participates in longstanding debates about the nature of scientific explanation in social science. Waltz draws on philosophy of science literature, including the work of Karl Popper, Imre Lakatos, and others, to argue for a particular conception of theory and theory testing. His critique of inductivism and his insistence on the importance of theory for guiding empirical inquiry reflect broader methodological debates in the social sciences.
The book also engages with Marxist and neo-Marxist theories of imperialism, responding to the tradition from Hobson and Lenin through later theorists of neocolonialism and dependency. Waltz’s critique of economic reductionism in the explanation of international politics addresses a major alternative theoretical tradition.
Finally, the work appears at a moment when the discipline of international relations was seeking to establish itself as a rigorous scientific enterprise. The book offers both a diagnosis of why previous efforts had failed and a prescription for how to proceed. Its combination of methodological reflection, critical analysis, and theoretical construction made it a landmark contribution to the field.
Structure of the Book
The book is organized into chapters that develop Waltz’s argument progressively. The opening chapter, “Laws and Theories,” establishes the philosophical foundations for the entire work. Waltz distinguishes laws from theories, arguing that laws establish relations between variables while theories explain why those relations hold. He critiques inductivist approaches that seek to build theory through accumulation of data, arguing that “data never speak for themselves” and that theory must precede meaningful empirical inquiry. The chapter also introduces the distinction between reductionist and systemic theories that structures much of the book’s analysis.
The second chapter examines reductionist theories of international politics, focusing particularly on economic theories of imperialism. Waltz analyzes the Hobson-Lenin theory as “the best of such approaches” and evaluates its strengths and weaknesses. He traces the development of Marxist theories of imperialism from Hobson through Lenin to later theorists of neocolonialism, arguing that these theories fail because they cannot explain the persistence of imperialist behavior across different economic systems and because they confuse the attributes of units with the structure of the system.
The third chapter continues the critique of reductionist approaches, examining what Waltz calls “reductionist and systemic theories.” This chapter analyzes the work of systems theorists including Richard Rosecrance, Stanley Hoffmann, and Morton Kaplan, arguing that despite their claims to systemic analysis, these thinkers ultimately fall back on reductionist explanations. Rosecrance’s framework is criticized for lacking a genuine systems-level concept, Hoffmann’s approach for failing to distinguish changes within systems from changes between systems, and Kaplan’s for using systems language without developing a genuine structural theory.
The fourth chapter presents the core argument for a systemic approach. Waltz argues that low-level explanations of international outcomes repeatedly fail because outcomes remain similar even when the attributes and interactions of agents vary widely. This disjunction indicates that constraints are operating at the system level. The chapter develops the analogy between microeconomic theory and international-political theory, arguing that just as market structure constrains firm behavior, international structure constrains state behavior.
The fifth chapter develops the concept of political structure in detail. Waltz defines structure as the arrangement or ordering of a system’s parts, distinguished from the attributes and interactions of units. He identifies three elements of structure: the ordering principle (anarchy versus hierarchy), the specification of functions of units, and the distribution of capabilities. For international politics, the ordering principle is anarchy, the units are functionally undifferentiated (like units), and the structure is defined by the distribution of capabilities across states.
The sixth chapter applies the structural approach to the analysis of international systems, examining how the shift from multipolarity to bipolarity changed the character of international politics. Waltz uses the example of Western European cooperation after World War II to show how structural change made possible outcomes that had previously been impossible. The chapter also addresses the relationship between structures and processes, arguing that structures constrain and dispose while processes involve the interactions of units.
The seventh chapter, “Structural Causes and Economic Effects,” explores the economic dimensions of international structure. Waltz discusses collective action problems, the “tyranny of small decisions,” and the distinction between micro- and macro-theories. He argues that structural constraints cannot be wished away and that states face fundamental dilemmas in pursuing their interests within an anarchic system.
The eighth chapter, “Structural Causes and Military Effects,” applies the theory to questions of war and peace. Waltz examines how the structure of the international system affects the likelihood of conflict, arguing that balance-of-power politics prevail wherever the order is anarchic and populated by units wishing to survive. The chapter distinguishes between a theory of international politics and a theory of foreign policy, arguing that structural theory explains outcomes produced by the uncoordinated actions of states rather than the particular policies of individual states.
The book concludes with an appendix that provides supporting data and a bibliography compiled with research assistance. Throughout, the structure moves from philosophical foundations through critical analysis of existing approaches to the construction and application of an original theory.
Main Arguments
The central argument of the book is that international politics can and should be understood through a systemic theory that explains how the structure of the international system shapes state behavior and international outcomes. Waltz argues that previous approaches have failed because they are reductionist, seeking causes at the individual or national level when the most important causes operate at the system level.
The first major argument concerns the nature of theory itself. Waltz contends that theories are not collections of laws but explanations that show why laws hold. Laws establish relations between variables, but theories explain those relations by identifying the mechanisms that produce them. He argues that students of international politics have been committed to induction, examining cases in hopes that connections will emerge representing “reality that is out there,” but this approach cannot succeed because “data never speak for themselves.” Reality is “an elaborate conception constructed and reconstructed” through theory.
The second major argument is that reductionist theories of international politics fail. Waltz examines economic theories of imperialism, particularly the Hobson-Lenin theory, and argues that while these are impressive as theories, they cannot explain the persistence of imperialist behavior across different economic systems. The root cause of imperialism is “the imperialism of great power,” which operates across differently organized economies. The economic organization that will “cause” imperialism is whatever form proves most effective at the time. Waltz traces the development of neocolonial theory as an attempt to salvage Lenin’s theory when capitalist states fail to pursue colonial policies, arguing that this involves changing the definition of imperialism to accommodate what happened rather than anticipating facts.
The third major argument is that systems theories of international politics have failed to be genuinely systemic. Waltz examines the work of Rosecrance, Hoffmann, and Kaplan, arguing that each ultimately falls back on reductionist explanations. Rosecrance has outlined a framework rather than developed a theory, with no systems-level notion conditioning state behavior. Hoffmann’s approach cannot distinguish changes within systems from changes between systems because he mingles unit-level and systems-level elements in his definition of structure. Kaplan’s systems theory fails because he cannot ask how the system’s structure constrains actors; his explanations only concern the system itself, and his approach is reductionist labeled as systemic.
The fourth major argument is that a genuine systems theory is both necessary and possible. Waltz argues that the failure of low-level explanations indicates that causes have been incorrectly or incompletely specified. When the same effects follow from different causes, constraints must be operating on the independent variables. These constraints cannot be treated as just another independent variable at the same level, because they may operate on all variables and do so differently as systems change. A systems theory would explain regularities of behavior and lead one to expect outcomes within specified ranges, but the behavior of states and statesmen is indeterminate.
The fifth major argument concerns the definition of political structure. Waltz argues that structure is the arrangement or ordering of a system’s parts, not a collection of institutions. For domestic politics, structure is hierarchically ordered, with units standing in relations of super- and subordination. For international politics, the ordering principle is anarchy, the units are functionally undifferentiated, and the structure is defined by the distribution of capabilities. This definition abstracts from every attribute of states except capabilities and from relations of states, producing a “positional picture.”
The sixth major argument is that the balance of power is a structural phenomenon. Waltz argues that balance-of-power politics prevail wherever two requirements are met: the order is anarchic and populated by units wishing to survive. The theory explains results produced by uncoordinated actions of states, makes assumptions about interests and motives rather than explaining them, and explains constraints that confine all states. It gives general answers to what a state will have to react to, explaining why similar behavior is expected from similarly situated states.
The seventh major argument concerns the relationship between structure and change. Waltz argues that a structural change is a revolution, whether violent or not, because it gives rise to new expectations about outcomes produced by units whose placement varies with structure. Across systems, a theory explains change. The shift from a multipolar to a bipolar system changed the character of international politics, making possible outcomes that had previously been impossible.
The eighth major argument is that structural constraints cannot be wished away. Waltz argues that states face fundamental dilemmas in pursuing their interests within an anarchic system. Rational behavior, given structural constraints, does not lead to wanted results. With each country constrained to take care of itself, no one can take care of the system. The advice to place international interest above national interests is meaningless because states must do whatever they think necessary for their own preservation.
Key Concepts
The book develops several distinctive concepts that structure its theoretical argument. The most fundamental is the distinction between laws and theories. Laws establish relations between variables, taking the form “If a, then b,” where a is one or more independent variables and b is the dependent variable. Theories explain why those relations hold. Waltz argues that theories do not emerge from establishing laws; without at least a sketchy theory, one cannot know what needs explanation or what data count as evidence.
The concept of reductionism is central to the book’s critical argument. Reductionist theories seek to understand the whole by knowing the attributes and interactions of its parts. In international politics, reductionist theories locate causes at the individual or national level. Waltz argues that reductionism fails because outcomes remain similar and repetitive even when the attributes and interactions of agents vary widely. This disjunction indicates that constraints must be operating at the system level.
The concept of structure is the book’s most important theoretical contribution. Structure is the arrangement or ordering of a system’s parts, not a collection of institutions. Waltz identifies three elements of structure: the ordering principle (anarchy versus hierarchy), the specification of functions of units, and the distribution of capabilities. For international politics, the ordering principle is anarchy, the units are functionally undifferentiated, and the structure is defined by the distribution of capabilities across states.
The concept of anarchy is fundamental to the theory. Anarchy means that the international system has no overarching authority; states must cope with internal and external problems by themselves. From anarchy one infers broad expectations about state behavior. Distinguishing anarchic structures of different type permits narrower, more precise definitions.
The concept of like units is closely related. Calling states “like units” means each is like all others in being an autonomous political unit—another way of saying states are sovereign. States are alike in tasks faced, not abilities to perform them; differences are of capability, not function. Each state duplicates others’ activities: agencies for making, executing, interpreting laws, raising revenues, defending itself.
The concept of the distribution of capabilities defines the structure of the international system. International-political systems are distinguished only by number of great powers. Structure changes with distribution of capabilities. While capabilities are attributes of units, their distribution across units is a system-wide concept, not a unit attribute.
The concept of the balance of power is developed as a structural phenomenon. Balance-of-power politics prevail wherever the order is anarchic and populated by units wishing to survive. The theory explains results produced by uncoordinated actions of states, makes assumptions about interests and motives rather than explaining them, and explains constraints that confine all states.
The concept of systemic and subsystemic causes distinguishes between causes operating at the level of the international system and causes operating at the level of individual states. Waltz argues that unit-level and systems-level causes interact, so explanation at the unit level alone is misleading. Some causes are at the unit level, but since variations in presumed causes do not correspond closely to variations in outcomes, some causes must be at a different level.
The concept of the “tyranny of small decisions” is borrowed from Alfred E. Kahn to explain how individual rational choices can produce collectively undesirable outcomes. If one hundred consumers choose option x, causing the market to make decision X, it is not necessarily true that those consumers would have voted for that outcome if the large decision had been presented for their explicit consideration. So long as the structure is left unaffected, changes in intentions and actions of particular actors cannot produce desirable outcomes.
The concept of micro- and macro-theories is used to clarify the relationship between theories of foreign policy and theories of international politics. A microeconomic theory is a theory of the market built from assumptions about individual behavior. A macrotheory is a theory about the national economy built on supply, income, and demand as systemwide aggregates. A macrotheory of international politics would show how the international system is moved by system-wide aggregates, but no agencies with comparable capabilities exist internationally.
The concept of the “inductivist illusion,” borrowed from Levi-Strauss, describes the belief that truth is won through accumulating more data and examining more cases. Waltz argues that gathering more data yields only more data and larger correlations; “data never speak for themselves.”
The concept of “organized complexity,” borrowed from Warren Weaver, describes the nature of the subject matter that requires a systemic approach. Complex and organized subjects preclude traditional investigation methods.
The concept of “due process of inquiry,” borrowed from Martin Landau, requires following the logic and procedures one’s methodology prescribes. Most students of international politics have failed to observe this due process, concerning themselves with methods rather than the logic of their use.
Themes
A recurring theme in the book is the relationship between description and explanation in social science. Waltz argues that statistics are descriptions in numerical form, not explanations. They describe a universe through manipulation of samples, but statistical operations cannot bridge the gap between description and explanation. The pursuit of ever more data and ever larger correlations cannot substitute for theoretical understanding. This theme reflects a broader concern with the nature of scientific inquiry and the proper methods for studying social phenomena.
Another theme is the persistence of international outcomes despite changes in the attributes and interactions of states. Waltz observes that over centuries states have changed, but the quality of international life has remained much the same. States may seek reasonable ends but cannot figure out how to reach them. The problem is not stupidity or ill will; intelligence and goodwill cannot discover and act on adequate programs. This theme points to the power of structural constraints and the limits of human agency in international politics.
The theme of unintended consequences runs throughout the book. In a purely competitive economy, everyone’s profit-seeking drives the profit rate down, eventually to zero under static conditions. Yet it would be absurd to infer that everyone seeks to minimize profit. Similarly, in international politics, rules inferred from the results of state interactions are often prescribed to actors as conditions of system maintenance, but an orderliness abstracted from behavior cannot guide behavior. Outcomes are produced by the uncoordinated actions of states, not by anyone’s design.
The theme of the relationship between structure and agency is central to the book’s argument. Structures constrain and dispose; they do not determine. A systems theory explains continuities, recurrences, and repetitions within a system, but not particular events. The behavior of states and statesmen is indeterminate. Structural concepts explain big, important, enduring patterns, but they are no good on detail. This theme reflects a sophisticated understanding of the relationship between structural constraints and human freedom.
The theme of the distinctiveness of international politics is developed through the contrast with domestic politics. Domestic politics is hierarchically ordered, with units standing in relations of super- and subordination. International politics consists of like units duplicating one another’s activities. National politics consists of differentiated units performing specified functions; international politics consists of like units in an anarchic order. This contrast underlies the argument for a distinct theory of international politics.
The theme of the limits of cooperation in an anarchic system is explored through the analysis of collective action problems. Some courses of action cannot be sensibly followed by an individual unless others also follow them, and individuals cannot follow them unless many others are likely to do so. The result of individual rational choices may be outcomes that most people do not want. The only remedies for strong structural effects are structural changes.
The theme of the relationship between economic and political explanations runs throughout the book. Waltz engages extensively with economic theories of imperialism, arguing that they fail to explain the persistence of imperialist behavior across different economic systems. He also draws analogies between economic markets and international politics, arguing that market structure constrains firm behavior just as international structure constrains state behavior. The relationship between economic and political capabilities is explored in the discussion of whether the nation-state is “just about through as an economic unit.”
The theme of the Cold War and bipolarity is present throughout the book. Waltz argues that the emergence of the Russian and American superpowers permitted wider, more effective cooperation among Western European states. The shift from a multipolar to a bipolar system changed the character of international politics. The stability of the postwar order is attributed to structural factors rather than to the wisdom or virtue of statesmen.
The theme of the tragedy of international politics is implicit in the book’s argument. States face fundamental dilemmas in pursuing their interests within an anarchic system. Rational behavior, given structural constraints, does not lead to wanted results. With each country constrained to take care of itself, no one can take care of the system. The advice to place international interest above national interests is meaningless because states must do whatever they think necessary for their own preservation.
The Author’s Conclusions
Waltz concludes that a systems theory of international politics is both necessary and possible. The failure of reductionist theories to explain international outcomes indicates that causes operate at the system level as well as at the unit level. A theory of international politics must be systemic, explaining how the structure of the international system constrains and disposes state behavior.
The author concludes that the structure of the international system is defined by three elements: the ordering principle (anarchy), the specification of functions of units (which drops out for anarchic systems of like units), and the distribution of capabilities. This definition abstracts from every attribute of states except capabilities and from relations of states, producing a “positional picture” of the international system.
Waltz concludes that balance-of-power politics prevail wherever the order is anarchic and populated by units wishing to survive. The theory explains results produced by uncoordinated actions of states, makes assumptions about interests and motives rather than explaining them, and explains constraints that confine all states. It gives general answers to what a state will have to react to, explaining why similar behavior is expected from similarly situated states.
The author concludes that structural constraints cannot be wished away. States face fundamental dilemmas in pursuing their interests within an anarchic system. Rational behavior, given structural constraints, does not lead to wanted results. With each country constrained to take care of itself, no one can take care of the system. The only remedies for strong structural effects are structural changes.
Waltz concludes that a theory of international politics is not a theory of foreign policy. The two are distinct enterprises, answering different questions. A theory of international politics explains results produced by uncoordinated actions of states, while a theory of foreign policy would lead to different expectations about tendencies and styles of different countries’ policies. Confusing the two leads to errors in both theory and practice.
The author concludes that structural change is revolutionary, whether violent or not, because it gives rise to new expectations about outcomes produced by units whose placement varies with structure. Across systems, a theory explains change. The shift from a multipolar to a bipolar system changed the character of international politics, making possible outcomes that had previously been impossible.
Waltz concludes that the international system is a self-help system. States must do whatever they think necessary for their own preservation, since no one can be relied on to do it for them. The advice to place international interest above national interests is meaningless because no one has shown how nations can resolve the tension between pursuing their own interests and acting for the system.
The author concludes that his theory explains continuities, recurrences, and repetitions within a system, but not change within a system. Structures appear static because they endure, but they are dynamic in altering behavior and outcomes. Structural effects are overlooked because they are repeatedly the same. Structure is not empty; it gains elegance and power from being general, though it is no good on detail. Structural concepts explain big, important, enduring patterns.
Waltz concludes that the theory has explanatory and predictive power and elegance. Elegance means explanations and predictions are general. A theory of international politics explains why war recurs and indicates conditions making war more or less likely, but does not predict particular wars. Within a system, a theory explains continuities, recurrences, and repetitions, not change. Across systems, a theory explains change.
The author concludes that the book represents a “Copernican revolution” in the study of international politics, showing how much of states’ actions and outcomes can be explained by forces at the system level. The theory describes the range of likely outcomes and how the range varies as systems change, and what pressures and possibilities are posed by different structures, but not how effectively units will respond. Structurally one can describe pressures, but not predict reactions without knowledge of internal dispositions.
Significance
The book’s significance lies in its systematic attempt to place the study of international politics on a rigorous theoretical foundation. Waltz’s distinction between reductionist and systemic theories, his definition of political structure, and his argument for the necessity and possibility of systems theory have shaped subsequent debate in international relations. The book offers both a diagnosis of why previous approaches
Detailed Section-by-Section Summary
Laws and Theories
Kenneth N. Waltz opens his Theory of International Politics by establishing the foundational distinction between laws and theories, a necessary preliminary for his stated aims: to examine existing theories of international politics, to construct a new theory that remedies their defects, and to explore the theory’s applications. He begins by defining a law as a relation between variables, expressed in the form “If a, then b,” where the relationship is found repeatedly and gives an expectation of future occurrence. Laws can be absolute or probabilistic, with the latter implying a strong imputation of necessity in the natural sciences but a lesser one in the social sciences. Waltz contrasts this with a common but inadequate definition of theory as merely a collection or set of laws pertaining to a particular behavior. Such a view makes theories only quantitatively more complex than laws, not different in kind.
To illustrate the pitfalls of this approach, Waltz recounts Karl Deutsch’s example of Homer’s description of Troy’s walls, which Heinrich Schliemann tested empirically by digging. Deutsch regards simple if-then theories as “special theories” that may later become embedded in a grand theory, shifting from a yes-or-no question to a how-much question. Waltz critiques this reliance on correlation, arguing that a high coefficient of correlation does not warrant a causal relation. He notes that squaring the coefficient only accounts for a percentage of variance among “dots on a piece of paper,” and the relation inferred may be spurious or genuine. The example of a push imparted to a cart illustrates that the same observation can be explained radically differently by Aristotle, Galileo, or Newton. Statistics, Waltz insists, are descriptions in numerical form, not explanations, and they cannot bridge the gap between description and explanation. Following Deutsch’s advice to formulate propositions in terms of probability would lead students of international politics to behave like Aristotelian physicists, and they have “disappointingly little to show” for such efforts.
Waltz then introduces the “inductivist illusion,” a term from structural anthropologist Levi-Strauss, which is the belief that truth is won through accumulating more data and examining more cases. Gathering more data yields only more data and larger correlations, for “data never speak for themselves.” He quotes C. S. Peirce: “direct experience is neither certain nor uncertain, because it affirms nothing—it just is.” Ross Ashby’s example of astrophysicists seeking to explain star clusters illustrates the problem: the beginner wants trajectories of components, but this knowledge would fill volumes of numerical tables, and he would realize he did not really want all that. The problem is finding what we want without being overwhelmed with useless detail. Today’s students of politics are committed to induction, examining cases hoping connections will emerge representing “reality that is out there,” but one can never say with assurance that an inductively arrived-at state of affairs corresponds to something objectively real. Reality is an elaborate conception constructed and reconstructed.
On Methodology and Theory Testing
Waltz argues that the object of study in international politics requires a systemic approach rather than traditional modes, using Warren Weaver’s term “organized complexity” to suggest that complex and organized subjects preclude traditional investigation methods. Martin Landau’s concept of “due process of inquiry” requires following the logic and procedures one’s methodology prescribes, and most students of international politics have failed to observe this, concerning themselves with methods rather than the logic of their use. Once a methodology is adopted, method choice becomes merely tactical, and theoretical questions must be raised at the outset of inquiry. Waltz then lists seven requirements for testing theories: state the theory being tested; infer hypotheses from it; subject hypotheses to experimental or observational tests; use the theory’s own definitions of terms; eliminate or control perturbing variables; devise distinct and demanding tests; and ask whether failure means the theory flunks completely, needs repair, or requires narrowing its explanatory claims. Since hypotheses derived from theories are tested, a wrong hypothesis should prompt reexamination of steps two and seven. No theory can ever be proved true; credibility depends on the variety and difficulty of tests.
Waltz critiques a study by Singer, Bremer, and Stuckey (1972) that attempts to evaluate “equally plausible, but logically incompatible, theoretical formulations” about conditions associated with peace and stability or war and instability. Their three independent variables are concentration of capability among major powers, changes of that concentration, and changes of capability among powers, comparing “parity-fluidity” and “preponderance-stability” models. Waltz finds serious deficiencies: the theories being tested are not stated; hypotheses are not explained as inferred from theories; variables are not defined as theories would define them. The authors identify war with instability without explaining how expectations accord with any particular theory, and they introduce “concentration of power” as a key variable though no known theory employs it—existing theories refer to numbers of great powers or polarities, with “polarity” variously defined. Rules one, two, and four are ignored. The findings differ between nineteenth and twentieth centuries, yet the authors offer only “barest speculation” about omitted variables and none about how variables were originally defined, unheeding rule five. The model consists of “three highly similar and arbitrarily chosen variables,” yet no further tests are devised, and the seventh rule is ignored. Waltz cautions that rigorous testing of vague theory is “an exercise in the use of methods rather than a useful effort to test theory,” and demanding tests may cause poorly developed theories to be discarded prematurely. One should ask what the theory leads one to expect, check expectations against historical observations before precise refinements, and gear test rigor to the precision or generality of expectations.
Reductionist Theories and the Hobson-Lenin Theory of Imperialism
Waltz notes that theories do not emerge from establishing laws; without at least a sketchy theory, one cannot know what needs explanation or what data count as evidence. Proceeding without theory is “like shooting a gun in the general direction of an invisible target”—even hitting the bull’s-eye would go unrecognized. He observes “the small gain in explanatory power” from recent decades of international-political studies; “nothing seems to accumulate, not even criticism.” Theories are divided into reductionist, where causes are at the individual or national level, and systemic, where causes are at the international level. With reductionism, “the whole is understood by knowing the attributes and the interactions of its parts,” as in the classic case of understanding organisms by disassembling them and applying physical and chemical knowledge. The reductionist urge in international politics derives more from failures at the international-political level than from successes of other disciplines, and a practical reason is that national decisions seem to account for most of what happens in the world—defense spending, nuclear weapons choices, fighting or seeking peace.
Waltz identifies the economic theory of imperialism developed by Hobson and Lenin as “the best of such approaches”—not necessarily correct but “most impressive as theory.” It is elegant, powerful, simply stated, incorporating few elements, claiming to explain imperialism, most modern wars, and conditions permitting peace. The theory’s standing hinges on whether the economic theory is valid, whether conditions envisioned held in most imperialist countries, and whether most countries with those conditions were imperialist. Exceptions fail to invalidate a theory if their occurrence can be satisfactorily explained—”a wind that wafts a falling leaf does not call Newton’s theory of universal gravitation into question.” Hobson’s sixth chapter of Part I of his Imperialism (1902) contains “all of the elements of later economic explanations of imperialism from Lenin to Baran and Sweezy,” with key concepts including “overproduction,” “surplus capital,” “maldistribution of consuming power,” “recurrent gluts,” and “consequent depressions.”
Imperialism and Great Power Dynamics
Waltz argues that if capitalist states—the most advanced of their day—did not affect others more than others affected them, and did not occasionally engage in outwardly imperialist activity, that would be odd. The absence of imperialism in the face of unbalanced power would “sorely require explanation.” Weakness invites control; strength tempts one to exercise it, even for others’ “good.” The phenomenon is more general and older than the theory offered to explain it, and the root cause operating across differently organized economies is “the imperialism of great power.” The economic organization that will “cause” imperialism is whatever form proves most effective at the time. Waltz compares this to Newton’s gravitational force, which worked earlier though not fully identified; the causes of imperialism, present in advanced capitalism, were present earlier, though identifying capitalism with imperialism has obscured this. After World War I, Lenin and his followers tested their thesis in strongest form: capitalism produces imperialism, and the leading capitalist state will be the fiercest imperialist. Trotsky foresaw America becoming the world’s most imperialist nation, touching off “military collisions” of “unprecedented scale.” Not only must the leading capitalist state be most imperialistic, but its policies must be the major cause of war.
Joseph Schumpeter’s contrary essay argued “precapitalist elements, survivals, reminiscences, and power factors” propel states into imperialist ventures. Military classes, once needed for consolidation and extension of states, do not disappear upon completing tasks; they live on, seeking employment and prestige, supported by others imbued with their spirit. Such atavistic forces give rise to imperialist tendencies, though Schumpeter conjectured the United States would “exhibit the weakest imperial trend.” Like Veblen, Schumpeter assigns war’s causes to continued vogue of outmoded militarism, believing Germany and Japan—where capitalist forces haven’t fully supplanted feudal elements—constitute the greatest war danger. Judged by prediction accuracy, Veblen and Schumpeter “carry the day,” but prediction is insufficient for validity. They posed the problem latter-day Marxists faced: salvaging Lenin’s theory when capitalist states fail to pursue colonial policies—when none clings to colonies. The solution was neocolonialism, developed from the early 1950s onward, which separates imperialism from empires. Lenin had defined imperialism as an internal condition of certain states, “the monopoly stage of capitalism,” not a policy or result, but for Lenin that condition necessarily found political expression—imperialism originated privately but expressed itself publicly, requiring soldiers and sailors. Empires without colonies, and imperialist policies requiring little force, were unimaginable to Lenin.
The Failure of Kaplan’s Systems Theory
The author’s critique of Morton Kaplan’s systems theory begins by identifying a fundamental problem: Kaplan finds establishing the international system’s identity “equally troublesome” as he does other foundational questions. Kaplan views states not only as the environment of the system but also as subsystems within it, meaning disturbances must arise from the actors themselves. He writes that the international system “tends toward the subsystem dominant pole,” citing his “balance-of-power” system as “subsystem dominant” because its essential rules are subordinate to the rules of individual national systems.
The author calls Kaplan’s language “loose and imprecise,” noting that he writes of subsystems “sharing dominance” and entering “into an equilibrium somewhat like that of the oligopolistic market.” This is a critical error: “subsystems’ dominance” negates a systems approach entirely. An oligopolistic market is not one where firms dominate but where influence is indeterminate. Unlike economics, Kaplan never develops a distinct, operational definition of the environment of states—one linked to yet distinct from the states forming it—so the extent of system versus subsystem influence cannot be investigated.
Kaplan defines a system as a set of related variables, requiring one to describe its successive states and locate variable changes, but his criteria for fixing identity are “weak and incomplete.” He emphasizes the question of what makes a set a set rather than answering it; his “models” are mere collections of variables, not true models. Citing Abraham Kaplan, the author notes that “when one system is a model of another they resemble one another in form and not in content,” and structural properties are “highly abstract.” Morton Kaplan fails to address the problem of form, instead ranging systems “along a scale of integrative activity,” which explains his claim that systems theory permits the integration of variables from different disciplines. Borrowing across fields is legitimate only if fields are homologous; borrowing variables is an empirical question, so Kaplan cannot gain the advantage of applying similar theories to different realms.
The consequence is reductionism. Kaplan cannot ask how the system’s structure constrains actors; his explanations only concern the system itself. He fails to distinguish interaction of units from their arrangement, putting relations at the systemic level. Though he claims systems deal with “any states whatsoever,” his vague system means answers depend on what states are like—making the approach reductionist in both attributes and interrelations. His propositions are about decision-making units and rules, not about the effect of different systems on units.
The author also critiques Kaplan’s use of “feedback,” borrowed from cybernetics, which operates only within a hierarchic organization. The thermostat-furnace example implies a controller and controlled instrument, but in international relations “nothing” corresponds to this; Kaplan’s usage conveys only that states change policies in response to others’ moves, offering nothing new.
Three salient difficulties emerge. First, Kaplan identifies his balance-of-power system with the historical condition of contention among five or so great powers, obscuring that balance-of-power theory applies to any self-help system with two or more units. Second, he fails to keep attributes and interactions of units out of the definition of structure; he denies attributes a place but smuggles them in through rules. Third, he does not develop the structural component; describing international systems as “subsystem dominant” means “a subsystem dominant system is no system at all,” making his approach reductionist labeled as systemic.
Students of international politics fall into two categories: those using “system” as jargon, and those patterning work on the general-systems model. The general-systems model is a “misfortune” because international politics lacks the articulated order and hierarchic arrangement needed; it is not a full-blown organization with differentiated parts performing specified functions.
The Case for a Systemic Approach
Low-level explanations of international outcomes repeatedly fail because outcomes remain similar and repetitive even when the attributes and interactions of agents vary widely. This disjunction indicates that causes have been incorrectly or incompletely specified. When the same effects follow from different causes, constraints must be operating on the independent variables. These constraints cannot be treated as just another independent variable at the same level, because they may operate on all variables and do so differently as systems change. Therefore, reduction is inadequate, and an analytic approach must yield to a systemic one. Some causes are at the unit level, but since variations in presumed causes do not correspond closely to variations in outcomes, some causes must be at a different level. Unit-level and systems-level causes interact, so explanation at the unit level alone is misleading.
To construct a theory, one must abstract from reality, leaving aside most of what is seen and experienced. Students of international politics have tried to get closer to reality and increase empirical content, while natural science has advanced by moving away from everyday reality and lowering the degree of empiricism. Natural scientists look for simplicities; students of politics complicate their studies with more variables. Despite differences in subject matter, one must bound the domain, organize it, simplify materials, concentrate on central tendencies, and single out the strongest propelling forces.
The theory to be constructed must be systemic. Theory explains regularities of behavior and leads one to expect outcomes within specified ranges, but the behavior of states and statesmen is indeterminate. The dominant behavioral approach frames propositions about the behavior, strategies, and interactions of states, but unit-level propositions do not account for systems-level phenomena. Since the variety of actors and actions is not matched by the variety of outcomes, systemic causes are in play, so a systems theory is needed and possible. This requires conceiving of an international system’s structure and showing how it works its effects—a “Copernican revolution” showing how much of states’ actions and outcomes can be explained by forces at the system level.
To “explain” means to say why the range of expected outcomes falls within certain limits; why patterns of behavior recur; why events repeat themselves, including events few actors like. The structure of a system acts as a constraining and disposing force, so systems theories explain and predict continuity within a system. They show why changes at the unit level produce less change of outcomes than expected without systemic constraints.
A theory has explanatory and predictive power and elegance. Elegance means explanations and predictions are general. A theory of international politics explains why war recurs and indicates conditions making war more or less likely, but does not predict particular wars. Within a system, a theory explains continuities, recurrences, and repetitions, not change. Structural approaches are sometimes said to be disappointing because structure is supposedly static and nearly empty. Structures appear static because they endure, but they are dynamic in altering behavior and outcomes. Structural effects are overlooked because they are repeatedly the same. Structure is not empty; it gains elegance and power from being general, though it is no good on detail. Structural concepts explain big, important, enduring patterns.
Structures may suddenly change. A structural change is a revolution, whether violent or not, because it gives rise to new expectations about outcomes produced by units whose placement varies with structure. Across systems, a theory explains change. A theory succeeds only if political structures are defined to identify their causal effects and show how those effects vary as structures change. From anarchy one infers broad expectations; distinguishing anarchic structures of different type permits narrower, more precise definitions.
The Shift from Multipolar to Bipolar
Consider the shift from a multipolar to a bipolar system for European states. When European states were the world’s great powers, unity among them could only be dreamt of. Politics among them tended toward a zero-sum game; each power viewed another’s loss as its own gain. When some great powers moved toward cooperation, they did so to oppose other powers. The emergence of the Russian and American superpowers permitted wider, more effective cooperation among Western European states. They became “consumers of security.” For the first time in modern history, the determinants of war and peace lay outside the arena of European states, and the means of preservation were provided by others.
This made possible the “upgrading of the common interest”—all working together to improve everyone’s lot rather than obsessing over the division of benefits. The fear that one’s greater advantage would be translated into military force was removed. Britain, France, Germany, and Italy saw that war among them would be fruitless and soon believed it impossible. Because their security depended on others’ policies, unity could be worked for, though not easily achieved.
Once the possibility of war disappears, states can more freely risk relative loss. Enterprises more beneficial to some parties can be engaged in, hoping other activities will reverse the balance and believing the enterprise is valuable. Economic gains may be granted for expected political advantages, including strengthening European cooperation. Removal of security worries does not end conflict but changes its content. Hard bargaining within the European Economic Community shows governments still care about who gains more or less. Conflicts of interest remain, but not the expectation that force will resolve them. Politics among European states changed in quality after World War II because the system changed from multipolar to bipolar. The limited progress toward Western European unity cannot be understood without considering this structural change. The example shows what a theory can and cannot tell: it describes the range of likely outcomes and how the range varies as systems change, and what pressures and possibilities are posed by different structures, but not how effectively units will respond.
Structurally one can describe pressures, but not predict reactions without knowledge of internal dispositions. A systems theory explains changes across systems, not within them. Important discontinuities occur within a system; if they occur within a system that endures, their explanation requires a different level of analysis.
Defining Political Structure
Defining political systems structurally is uncommon, though anthropologists, economists, and political scientists already use structural notions for their own subjects. These disciplines deliberately set aside the qualities, motives, and interactions of actors—not because these are unimportant, but because they want to see how such factors are affected by structure.
Structure is the arrangement or ordering of a system’s parts, not a collection of institutions. A constitution describes only part of the arrangement; actual political structures diverge from formal constitutions. The first question is by what principle the parts are arranged. Domestic politics is hierarchically ordered: units stand in relations of super- and subordination. The ordering principle gives the first bit of structural information. The second bit is the specification of functions of formally differentiated parts. Congress supplies military forces, the President commands them; Congress makes laws, the executive enforces them, agencies administer, judges interpret. The British Prime Minister and Parliament differ in roles from the American President and Congress. The third bit is the distribution of capabilities across units.
The Tyranny of Small Decisions
The passage opens by examining a fundamental problem of collective action: certain courses of action cannot be sensibly pursued by an individual unless many others are likely to do the same. Two everyday examples illustrate this. Many people choose to drive private cars rather than take trains because cars offer flexibility in scheduling and destination, yet railway passenger service remains a wanted convenience, especially in bad weather. Similarly, many people shop at supermarkets rather than corner grocery stores because supermarkets offer larger stocks and lower prices, yet corner grocery stores providing credit and delivery are also a wanted convenience. The cumulative result of most people driving and shopping at supermarkets is to reduce passenger service and decrease the number of corner grocery stores. These outcomes may not be what most people want, and they may be willing to pay to prevent these services from disappearing, but individuals acting alone can do nothing to affect the outcome. Increased patronage would do it, but not by one person and a few others.
These instances are described as examples of what Alfred E. Kahn calls “large” changes brought about by the accumulation of “small” decisions. People are victims of the “tyranny of small decisions”: if one hundred consumers choose option x, causing the market to make decision X (where X equals 100 x), it is not necessarily true that those consumers would have voted for that outcome if the large decision had been presented for their explicit consideration. If the market does not present the large question, individuals are doomed to making decisions that are sensible within narrow contexts even while knowing they bring about a result most do not want. Either that, or they must organize to change the market’s structure—for example, by bringing consumer units roughly up to the size of the units making producers’ decisions. So long as the structure is left unaffected, changes in the intentions and actions of particular actors cannot produce desirable outcomes. Structures may be changed by altering the distribution of capabilities across units, or by imposing requirements where people previously decided for themselves. If some merchants sell on Sunday, others may have to do so to remain competitive even though most prefer a six-day week. Most can do as they please only if all are required to keep comparable hours. The only remedies for strong structural effects are structural changes.
Structural Constraints and the International Interest
Structural constraints cannot be wished away. In every age and place, units of self-help systems—nations, corporations, or whatever—are told that the greater good requires them to act for the system, not for their own narrowly defined advantage. In the 1950s, as fear of nuclear war grew, some concluded the alternative to world destruction was world disarmament. In the 1970s, with rapid growth of population, poverty, and pollution, some concluded that states must meet the needs of the political ecosystem in its global dimensions or court annihilation. The international interest must be served, meaning national interests are subordinate to it. Problems are found at the global level; solutions depend on national policies. No one has shown how nations can resolve the tension between pursuing their own interests and acting for the system. Rational behavior, given structural constraints, does not lead to wanted results. With each country constrained to take care of itself, no one can take care of the system. A footnote states that states face a “prisoners’ dilemma”: if each of two parties follows his own interest, both end up worse off than if each acted to achieve joint interests.
A strong sense of peril may lead to a clear definition of ends, but their achievement is not thereby made possible. Effective action depends on providing necessary means and on conditions permitting nations to follow appropriate policies. World-shaking problems cry for global solutions, but there is no global agency to provide them. Necessities do not create possibilities; wishing that final causes were efficient ones does not make them so. Great tasks can be accomplished only by agents of great capability, which is why states, especially major ones, are called on to act for the world’s survival. But states must do whatever they think necessary for their own preservation, since no one can be relied on to do it for them. The advice to place international interest above national interests is meaningless.
Micro-Theory and Macro-Theory
The distinction between micro- and macro-theories explains this meaninglessness. Among economists the distinction is well understood; among political scientists it is not. A microeconomic theory is a theory of the market built from assumptions about individual behavior, showing how actions and interactions of units form and affect the market and how the market affects them. A macrotheory is a theory about the national economy built on supply, income, and demand as systemwide aggregates. In economics, both deal with large realms; the difference is in approach, not size of objects. A macrotheory of international politics would show how the international system is moved by system-wide aggregates—amount of world GNP, world imports and exports, deaths in war, defense spending, migration. It would look like a macroeconomic theory in the style of John Maynard Keynes, though it is hard to see how international aggregates would make sense. A macrotheory of international politics would lack the practical implications of macroeconomic theory: national governments can manipulate system-wide economic variables, but no agencies with comparable capabilities exist internationally. Even were such a theory available, we would still be stuck with nations as the only agents capable of acting to solve global problems, and would have to revert to a micropolitical approach.
Some have hoped that changes in awareness, purpose, organization, and ideology of states would change the quality of international life. Over centuries states have changed, but the quality of international life has remained much the same. States may seek reasonable ends but cannot figure out how to reach them. The problem is not stupidity or ill will. Intelligence and goodwill cannot discover and act on adequate programs.
Errors in International-Political Theory
The passage identifies two closely related errors in international politics theory: assuming a necessary correspondence between motive and result, and inferring rules for actors from observed results. An economic analogy clarifies this: in a purely competitive economy, everyone’s profit-seeking drives the profit rate down, eventually to zero under static conditions. Inferring that everyone seeks to minimize profit, or that competitors must adopt this as a rule, would be absurd. Yet in international politics, rules inferred from the results of state interactions are often prescribed to actors as conditions of system maintenance. S. F. Nadel is quoted: “an orderliness abstracted from behaviour cannot guide behaviour.” Analytic reasoning where a systems approach is needed leads to laying down arbitrary prerequisites for balances of power and world stability (e.g., number of great powers exceeding two, a balancer role, stable military technology, abiding by rules). But balances form without these, and since 1945 the world has been stable and major powers peaceful despite nonconformity to theorists’ stipulations. Balance-of-power politics prevail wherever only two requirements are met: the order is anarchic and populated by units wishing to survive.
The Fallacy of Unit-Level Explanation
If results require someone to want and work for them, explanation turns on what separate states are like, making national-level theories sufficient. If equilibrium is maintained by states abiding by rules, one needs an explanation of agreement on rules, not a balance-of-power theory—balances would result from behavior explained by national psychology or bureaucratic politics. A balance-of-power theory would have nothing to explain; the notion becomes merely a framework organizing accounts of what happened. A construction starting as a theory ends as a set of categories, multiplying rapidly, turning the quest for explanatory power into a search for descriptive adequacy.
The Theory’s Scope and Confusions
Balance-of-power theory is criticized for not explaining particular policies of states (e.g., why state X made a move last Tuesday). This is like expecting the theory of universal gravitation to explain the wayward path of a falling leaf—a theory at one level of generality cannot answer questions at a different level. Another error is mistaking a theory of international politics for a theory of foreign policy. Confusion is rooted in uncertainty about the distinction between national and international politics, or denial that the distinction should be made. Morgenthau believes problems of predicting foreign policy make international-political theories difficult, if not impossible—but this only holds if the latter reduces to the former. Graham Allison’s three “models” purport to offer alternative approaches to international politics, but only Model I is such; Models II and III are approaches to foreign policy. Offering bureaucratic politics as an alternative to the state-as-actor is like saying a theory of the firm is an alternative to a theory of the market—a mistake no competent economist would make. A footnote notes Herbert Simon wrongly believes classical economists tried “to predict the behavior of rational man without making an empirical investigation of his psychological properties.” Balance-of-power theory explains results produced by uncoordinated actions of states, makes assumptions about interests and motives rather than explaining them, and explains constraints that confine all states. It gives general answers to what a state will have to react to, explaining why similar behavior is expected from similarly situated states—similar, not identical. A theory of foreign policy would lead to different expectations about tendencies and styles of different countries’ policies.
Testing the Theory
The author constructed a systems theory in the previous chapter; this chapter states balance-of-power theory as a further development. The next three chapters refine it by showing how expectations vary with changes in structure. Before tests, one asks if the theory is internally consistent and tells us things of interest we would not know otherwise. Many prefer falsifying tests, following Karl Popper. Confirmations are rejected because confirming cases may be offered while confounding cases are avoided, and numerous confirmations are negated by one falsifying instance. The conception of theory from Chapter 1 opens the possibility of devising confirmatory tests: compare features of the observed domain with the picture the theory has limned. Structural theories gain plausibility if similarities of behavior are observed across realms different in substance but similar in structure, and differences where realms are similar in substance but different in structure. International-political theory gains credibility from confirmations in economics, sociology, anthropology. Testing is difficult due to interdependence of fact and theory. One sticks with the theory that reveals most, even if its validity is suspect. Tests are easy to think up but hard to carry through. We should exploit all ways of testing: falsification, hard confirmatory tests, comparing real and theoretical worlds, comparing behaviors in realms of similar and different structure. A theory becomes plausible if many hypotheses inferred from it are successfully subjected to tests.
Two Expectations and Their Tests
Two expectations arise: balances of power recurrently form, and states tend to emulate successful policies of others.
The Economic Foundations of Bipolar Stability
The United States, by virtue of its size and resources, possesses a remarkably low level of economic dependence on other nations. The costs and even the odds of losing a trading partner are minimal for America, while the reverse is true for its partners, who depend far more on the US than it does on them. Sustained economic sanctions against the US would amount to “economic self-mutilation” for those who attempted them. The argument that a rival power could foreclose American trade and investment in successive parts of the world and “quietly strangle” it is dismissed as apocalyptic thinking rather than political analysis. If some countries deal less with the US, others will inevitably move economically closer to it, and the US retains the capacity to grant or withhold favors in trade, aid, loans, and military security. The Soviet Union is even less economically dependent on the outside world than the US, but it possesses less economic and political leverage. The sheer size of the two great powers insulates them from the effects of other states’ behavior, producing an equilibrium at a low level of interdependence. This picture contrasts sharply with that of “transnationalists and interdependers,” who cling to an “economic version of the domino theory.” The world is not closely coupled; the discrepancy is wide between the homogeneity suggested by “interdependence” and the heterogeneity of a world full of bristling nationalisms. Considering the world a unit and calling it “interdependent” is logically wrong and politically obscurantist. Diplomacy is compared to chess, and the chessboard must be accurately described. The chapter concludes that smaller numbers are better than larger for states at the top, but economic interdependence did not establish which small number is best.
Structural Causes and Military Effects
The chapter aims to show why two is the best of small numbers, using military interdependence. To establish the virtues of two-party systems, systems of different numbers must be compared. A stable international-political system means it remains anarchic and no consequential variation occurs in the number of principal parties. The multipolar system lasted three centuries; the bipolar system has lasted three decades. Not all changes of number are changes of system; the defining difference between bipolar and multipolar systems is found in balancing behavior. With two powers, imbalances are righted only by internal efforts, while with more than two, shifts in alignment provide an additional means of adjustment. Three and four are threshold numbers; systems of three have the unfortunate characteristic that two powers can gang up on the third, while four is the lowest acceptable number for multipolar systems. Five is thought to be another threshold number, allowing a balancer, but beyond five, complications accelerate as numbers grow. Until 1945, the nation-state system was multipolar, always with five or more powers. The conventional wisdom holds that a balance-of-power system is unstable with only two powers and requires four, with a fifth serving as balancer. However, the notion of a balancer is a historical generalization from Britain’s position in the 18th and 19th centuries, dependent on narrowly defined, historically unlikely conditions. The flexibility of alliances is doubted; the old balance-of-power system resembles the new collective-security system of the League and the United Nations, both depending on the neutrality of alignment at the moment of serious threat. The difficulty of states swinging to the side of the threatened has two parts: the free-rider problem, where members of a group sharing a common interest may not act to further it, and the uncertainties about who threatens whom that accelerate as the number of states increases.
The Persistence of Political Forces
The idea that a new world began with the atomic bombing of Hiroshima is dismissed, as perennial political forces matter more than new military technology in shaping state behavior. States remain the primary vehicles of ideology. Historical international movements—the international brotherhood of autocrats after 1815, cosmopolitan liberalism, international socialism, and international communism—were in each case captured by individual nations, with adherents harnessed to national interest. Thus, the Soviet Union in crisis became Russian, and American policy became realistically and cautiously constructed despite liberal rhetoric. Economists have long known that time makes peaceful coexistence among major competitors easier; they learn to interpret one another’s moves. In small-number systems, as a group shrinks, members face fewer choices about whom to deal with, and the smallest groups manage affairs most easily. With more than two parties, solidarity is at risk because parties can combine, and interdependence breeds hostility and fear. With just two members, choice disappears; each can deal only with the other, with no appeal to third parties. Tension is high, but the pressure to moderate behavior is heavy. On disarmament, two conditions can make agreement possible: first, if the would-be winner of an arms race curtails its program, as the United States did in the 1920s, making the Washington Naval Arms Limitation Treaty possible; and second, if two powers consider mutual interests without much thought to others’ capabilities, as in the 1972 treaty limiting antiballistic missiles. The simplicity of a bipolar world makes the two great powers conservative, but structure does not explain everything; one must look at capabilities, actions, interactions, and structure. The United States and Soviet Union, like duopolists, are learning to cope but will not avoid crises. Relations changed perceptibly in the 1960s and ’70s, but worries about American-Russian agreements at others’ expense are exaggerated. Mutual opposition may require rather than preclude adjustment of differences, but first steps toward agreement do not lead to second and third steps; they mingle with other events keeping tension high. This pattern was set by the Test Ban Treaty of 1963. Both countries are wary of joint ventures, cannot know benefits will be equal, and cannot be certain arrangements will bind; rules of reciprocity and caution prevail. They are called frere ennemi and adversary partners. A small-number system can be disrupted by a Hitler and the reactions of a Chamberlain, but leaders of great states are not free agents; pressures encourage sensible response. The durability of the bipolar world is examined, with national rankings changing slowly and economic growth rate differences not large enough to alter standings except in the long run.
The Utility of Force and the Security Dilemma
Military strength is compared to a miser’s money that may depreciate in value, but force may be most useful when it dissuades attack without battle. For a status-quo power, nonuse of force signals strength. Thinking in “cataclysmic terms” and dreading all-out war makes cataclysm less likely. John Herz coined “security dilemma”—states unsure of one another’s intentions arm for security, setting a vicious circle where the means to anyone’s security threatens someone else. The dilemma is produced by situations, not wills; it cannot be solved, only dealt with. Second-strike nuclear forces are the principal means. Nuclear forces are useful precisely because their use is forestalled; great powers’ forces are most useful and least costly if priced only in money, not blood. Great powers are never “masters with free hands”—always “Gullivers,” more or less tightly tied. They fight more wars than lesser states, and their wars arise from position in the international system, not national character. At or near the top they fight; as they decline they become peaceful. Military force is divisible, especially for states that can afford much. Nuclear weapons deter nuclear weapons and limit escalation. For over three decades, force has been used in controlled ways for conscious political purposes, from garrisoning Berlin to the war in Vietnam. Since World War II we have seen political organization and pervasion of power, not cancellation of force by nuclear stalemate. Identifying power with control is an elementary error; force is always on the side of the governed, and no foreign body can rule a country unable to rule itself. Military force establishes control over territory, not within it; it affects external behavior, dissuades aggression, and meets aggression if dissuasion fails. Dissuasion is easier than “compellence.” In Vietnam, the United States faced compelling action plus promoting political order. Success or failure in peripheral places matters less materially now. American successes are overlooked: securing Western Europe, defending South Korea. The United States and Soviet Union have more reason to be satisfied with the status quo than earlier great powers. For the United States, the stake in Vietnam was internally generated honor and credibility; little was at stake globally. Victory would not create American hegemony; defeat would not create Russian hegemony—the duopoly would endure. Two sources of confusion are identified: first, the usefulness of force is mistakenly identified with its use; second, the American definition equates power with control. An agent is powerful to the extent he affects others more than they affect him. International politics is necessarily a small-number system; advantages of arriving at two are decisive. Bipolar great powers are more self-sufficient; interdependence loosens between them.
