The Origins of Major Wnr
ern world pi ior to 1900. The first three cases demonstrate how fragile bipolar systems can be: in each case, war broke out even though the declining state was only roughly equal to the rising state in military power. In three of the four multipolar cases, it was a declining state with significant military superiority that brought on systemic war.. Even in the one anomalous case, the Seven Years War, fears of decline were instrumental in pushing Austria to organize a war of elimination against Prussia. The case does suggest that under a narrow set of conditions—the rising state is an aggressive upstart with little territory—major war might break out in multipolarity among states with roughly equal military power. Overall, however, the weight of history strongly upholds the logic of dynamic differentials theory.
The Implications of the Argument
This book has sought to provide a dynamic realist theory of major war that represents a Lakatosian progressive problem shift within the realist paradigm. The evidence of the empirical chapters suggests that the theory has moved toward this goal: it explains the empirical facts covered by existing theories, while also accounting for evidence left unexplained by these theories.1 The theory reaches this goal by synthesizing the systemic strengths of current realist arguments, while avoiding the tendency to dip down to the unit level to explain individual cases. This approach helps us to reexamine Kenneth Waltz's popular distinction between theories of international political outcomes and theories of foreign policy. For Waltz, systemic theories explain continuities in outcomes within bipolar or multipolar systems; to understand differences in behavior across states and over time, he argues, one must go down to the unit level.2 This view has encouraged most scholars to assume that systemic theory can only establish the broad constraints on state behavior, and that for greater explanatory power one must automatically incorporate domestic- and individual-level variables.
The book shows the limitations of this perspective. At any point in time, states face specific systemic constraints reflecting their unique trends and differentials of power. This fact allows us to make predictions about how individual states will act without necessary consideration of their unit-level characteristics.3 Indeed, strong predictions on outcomes like major war can be made only with a theory that predicts when and why particular states will initiate actions making such outcomes likely. A good systemic theory of foreign policy, therefore, is a prerequisite for a good theory of international political outcomes. Thus while Waltz's neorealism does not seek to explain when major wars will break out, but only why they might recur, this book offers falsifiable predictions about when states will either initiate
The Origins of Major War
The Implications of the Argument
major wars or take the hostile steps that greatly increase the risks of such a war. Moreover, as the evidence shows, the key major wars, as well as "near-misses" such as the Berlin and Cuban missile crises, were indeed driven by the dynamics of relative power. Unit-level forces can be significant, as I discuss below. But a crucial first step in theory building is establishing a deductively consistent systemic argument—the goal being to determine to what extent behavior and international outcomes can be explained solely with reference to systemic constraints. Such a theory will then set the context and boundaries for the causal role of unit-level variables.
This chapter first considers some of the implications of the argument for both realist and liberal international relations theory. I then examine the book's practical implications for the great power dynamic that will likely dominate the next three decades or more—the relationship between the United States and China.
Realism and Major War
By fusing the strengths of current realist theories, this book makes three main contributions to realist theory. First, it examines the importance of power shifts across bipolar and multipolar systems. Classical realism and neorealism emphasize polarity; hegemonic stability theory and preventive war arguments stress the importance of dynamic power trends. Yet polarity and the problems of decline have not been brought together in one theory. Dynamic differentials theory shows that polarity affects whether declining states will take actions that could lead to major war. In multi-polarity, the declining state must have a significant level of military superiority to consider risking major war. In bipolarity, it may launch major war or crises threatening such a war whether superior or merely equal to the rising state—indeed, it may do so even if somewhat inferior, as we saw with the two Berlin crises.
Considering polarity and power trends simultaneously helps to eliminate anomalies in the existing theories. Classical realism has trouble explaining why war would break out in bipolar systems like Sparta-Athens, Carthage-Rome, and France-Hapsburgs, when both states were essentially equal (that is, when there was a "balance of power"). For neorealism, the very fact that war occurred in bipolarity is surprising. Hegemonic stability theory, with its view that superiority makes for peace, cannot explain why in each of the major wars from 1600 to 1945—with the exception of the Seven Years War—-war would be initiated by a state with marked military superiority.' The fact that decline occurred in a multipolar context provides the answer.
[236]
The second contribution is the building of an argument which can explain when and why a state might take hard-line measures that increase the probability of war through inadvertent means. Many realists emphasize the problem of declining power. Crisis and security-dilemma theorists, including defensive neorealists, underscore the risk that hard-line policies can provoke inadvertent spiraling. No established theory, however, combines these two dimensions into a comprehensive decision-making model.
By developing such a model, this book can make predictions of why, even in the nuclear age, states would ever get themselves into cold wars or crises that risk all-out destruction. States will accept such risks only when continuing with established policies will not stem decline but stronger actions hold out the promise of stabilizing their power position. Yet in moving to harder-line policies, they will weigh the risks of further decline against the risks of provoking an inadvertent spiral to major war. This argument helps us to explain not only why great power crises with a high risk of major war are so rare, but also why they occasionally occur. The two Berlin crises and the Cuban missile crisis only broke out when one of the two superpowers believed that internal measures alone would not reverse decline, but that crisis initiation might achieve this objective.
The book's third contribution is its analysis of three different forms of decline and their varying effects on declining states. Entrenched relative stagnation has been studied by countless scholars. Less well studied are the problems of power oscillations and of disjunctures in economic/potential power and military power. Negative power oscillations occur when the other state's policies are relatively more successful over at least the short term. The declining state will not only worry about the loss of power, but will anticipate that its subsequent effort to catch up may push the other state into preventive policies, perhaps even war. Depending on the size of the oscillation, strong action now to ameliorate decline can be rational. This dynamic was at work, as we saw, in the three cold war crises. Most realists ignore power oscillations or do not integrate them with the risks of inadvertent spiraling. Consequently, they do not offer complete explanations for the occasional but dangerous risk-taking witnessed in the cold war.
The separation of economic /potential power from military power is particularly valuable in explaining cases where states embark on the most extreme measures to uphold their security. Probably the most destabilizing situation in world politics is one where a state is militarily superior, but inferior in economic and especially potential power. This was Germany's problem prior to both world wars: it possessed marked military superiority, yet it faced a Russia with three times its population and forty times its land mass. Without war, the German civilian and military leaders believed that Russia would eventually overwhelm Europe; -with war, Germany might not only eliminate the threat, but it could grab the territory needed
[237I
The Origins of Major War
for long-term security. Inferiority in potential power was also at the heart of the problem in a number of other important cases, particularly Carthage-Rome and France-Hapsburgs and to a lesser degree in the ancient Greek case and the Napoleonic Wars.
Current realist theories, by overlooking the importance of disjunctures in economic/potential power and military power, provide less comprehensive systemic explanations across the various cases. Classical realists and neorealists recognize Germany's insecure geographic position in the center of Europe, but they tend to fall back on Hitler's personality and Nazi hy-pernationalism to explain the specific motives for the Second World War. Classical realists have trouble explaining the First World War, given the balance of power between the two alliance blocs. Neorealists invoke miscalculation in multipolarity to account for war in 1914. Yet multipolarity is a constant that cannot explain the changing incentive for major war over time. More to the point, World War I was not a war of miscalculation; Berlin wanted war for preventive reasons and did everything necessary to bring it on under the best possible conditions.
In sum, realist theories as they stand remain disconnected and incomplete. Classical realism rightly emphasizes power differentials; neorealism, polarity and the security dilemma; hegemonic stability theory and preventive war theories, the problem of dynamic trends. By synthesizing and reformulating these elements, dynamic differentials theory provides an argument with greater explanatory power across the full range of cases.
Regime-Type, Liberal Theory, and Major War
The book's argument also has implications for liberal arguments, particularly those that stress the role of domestic-level causes of major war. The theoretical chapters held such unit-level factors constant to isolate the role of shifts in the differentials of power. By relaxing that assumption, we can examine how certain domestic forces might operate under properly specified systemic conditions. Overall, the empirical chapters showed that unit-level factors were less important in causing major wars and crises than is commonly supposed. Yet this book does not deny the importance of such factors in history. They can be expected to have two separate effects independent of relative power changes.
First, even though the theory's power conditions are almost always necessary conditions for major war, they may not be sufficient: chapter 2 noted that declining states will sometimes initiate major conflict for aggressive nonsecurity motives. One cannot ignore the Genghis Khans of history who seem to be propelled more by glory and greed than by concerns about ris-
The Implications of the Argument
ing neighbors. It is worth reiterating, however, that such examples do not falsify the theory, only qualify its salience.1 Moreover, the fact that so many of the key cases of the twentieth century and of previous centuries were driven primarily or exclusively by fears of decline shows the strong salience of the dynamic differentials argument.
Second, domestic forces may affect the probability of major war through variations on the regime-type of the rising state. In the theoretical chapters, I assumed that the declining state was fundamentally uncertain of the future intentions of the rising state. That is, the declining state assumed either that the other was just as likely as not to attack later at its peak (a 50-50 chance), or that the other's propensity to attack was a function of how far it rose. This assumption allowed us to isolate the interactive effects of other more systemic causal factors, while showing how conflict might arise even when all states sought only their own security. A declining state's analysis of the other's regime-type, however, should have some influence on its estimate of the other's likelihood of attacking later (the third parameter from chapter 2). A declining authoritarian state will probably be just as suspicious of a rising democracy as of a rising nondemocracy. Declining democracies, however, are likely to place some importance on the domestic characteristics of the rising state—but not always in ways consistent with the hypotheses of the liberal "democratic peace" literature.
By the logic of the democratic peace, a declining democracy should be less likely to attack a rising democracy at time f0 if it has strong reason to believe that the rising state will remain democratic after it peaks at time t,. At f„ the (formerly) rising state should be disinclined to attack because of its respect for the other's democratic ways and because of legislative constraints on its ability to make war.6 Anticipating this fact, the declining state at t0 will lower its estimate of the rising state's likelihood of attacking later and thus be less inclined to preventive war.
Note, however, that there is a key condition underpinning the logic here: the relative stability of the rising state's regime-type. What really matters is not whether the rising state is currently democratic, but whether it will still be democratic years down the road, after it peaks. If Russia were rising today, for example, the fact that it is democratic would be of little comfort; given Russia's fragile democratic structure, it seems hard to predict its regime-type in ten or fifteen years. This suggests an important implication of this book's argument for liberal democratic peace theory, at least as this theory relates to great powers and major war: in dynamic power environments, peace will be robust primarily between established, stable democracies.
This discussion suggests that to strengthen liberal theory, we need to combine the dynamics of domestic politics with the dynamics of systemic power changes. Since what matters to a declining state is its estimate of the
[239]
The Origins of Major War
The Implications of the Argument
future "democratic-ness" of the rising state, the former will base this estimate primarily on the internal trends within the latter, if the rising state is a democracy showing increasing signs of instability, there will be less confidence about its future type. If, however, the rising state is currently authoritarian but undergoing democratization, this should give the declining state greater confidence that by the time the other peaks, it will have become democratic enough to be relatively peaceful.7 This leads to a second and surprising implication of dynamic differentials theory: declining and democratic states may be more likely to initiate preventive moves against unstable democracies than against authoritarian states that show a strong trend toward democracy.8
In sum, systemic realist arguments do not need to reject the insights of liberal theory. Regime-type can matter. Yet if declining power is as critical a causal force as the empirical chapters suggest, liberal theories need to address the twin problems of domestic instability and future intentions. Declining states know that the rising state, regardless of its regime-type, has little reason to attack while still rising. But they worry about the other's intentions years into the future once it is more powerful. In the context of relative decline, therefore, the stability of the rising state's current regime-type should provide one important means for estimating these future intentions.9
Practical Implications
What are the argument's implications for the prospects for peace in the post-cold war world? Neorealists who predict instability in Europe and Asia due to the emergence of global and regional multipolarity are unnecessarily pessimistic.10 Multipolar systems are less likely to fall into major war than bipolar ones, since the conditions for war are less permissive." As we have seen, major wars can occur in bipolarity when states are either equal or unequal in military power, but in multipolarity the initiator requires marked superiority to consider attacking the system. Hence, the presence of many regional powers in Europe and Asia will help moderate the behavior of any particular state. Even a resurgently nationalistic Germany would be deterred from taking on the European system again, since costly bilateral wars would harm its relative power position versus third parties. The same holds for China or Japan in the Far East.
Multipolarity in the twenty-first century will not be the problem. The problem is the risk of significant shifts in the power balance. Major wars and destabilizing crises occur when dominant states anticipate deep decline. Dynamic differentials theory thus focuses our attention on powers that are most likely to rise against the established states. Concerns about
the relative rise of China are particularly evident, and for good reason. Since the early 1980s, China's economy has been growing consistently at an annual rate of two to three times that of any other great power. Simple extrapolation suggests that China could catch up to America in total GNP within a couple of decades. The worry, of course, is that China's rising economic strength will eventually be translated into the kind of military power that could threaten U.S. security. To the extent that China emerges as the only challenger in what would become a strongly bipolar world, this worry will be intensified.12
Will China's relative growth undermine the stability the global system now enjoys? Behind this issue lurk the questions of why exactly the system has been essentially stable since the early 1990s (with stability defined as a low probability of major war) and which state is most likely to initiate a new round of cold wars and crises.
For liberals, the stability since 1991 has a number of roots: the end of an ideological battle between capitalism and socialism;13 the spread of democracy;14 increasing economic interdependence;15 and a growing web of economic and political institutions.16 This book does not dismiss the potential effects of such unit-level and nonpower systemic variables. Yet just as liberal theories miss one of the core causes of conflict over the millennia—the fear of long-term decline—they also overlook what may be both the necessary and sufficient conditions for the recent peace: that unlike in previous eras, there is no reason to believe that the dominant great power, the United States, is declining deeply and inevitably. Should such a belief arise and gain widespread acceptance, history indicates that all of the liberal causes of peace combined would have little restraining value.
The United States, compared to potential rivals, currently occupies a unique historical position. Unlike Britain in the nineteenth century, which was superior in at most industrial production, the United States is supremely dominant in all three dimensions of power: military, economic, and potential.17 With the once-vaunted Red Army now in tatters, the United States is clearly the world's only remaining conventional global superpower. Although Russia still retains a large strategic-missile force, its economy is so weak that the main fear is not Russian attack, but the government's inability to maintain control of its nuclear weapons. China's nuclear arsenal is still a small fraction of America's, as are the British and French stockpiles.
Economically, China's historically strong growth rate is a concern. But to explain the stability from 1991 to 2000 and the stability that will likely continue for at least the next decade,, we should note the crucial differences between recent Chinese growth and the rise of Russia after 1945. After World War II, U.S. leaders had good reason to fear Soviet economic growth. Russia, owing to reforms starting in the nineteenth century, was by 1945 poised
[240]
[241]
The Origin* of Major War
to become an industrial superpower. Moreover, the years after 1945 were a period when total industrial output, particularly in "heavy" goods like steel and machinery, was still a critical basis for economic and military power. Thus the fear that the Soviets might become the Americans' economic equal (or better) was a reasonable one.
Behind Russian economic strides lay the state's potential power: a huge reserve of raw materials and a population equal to that of the United States. Most important, however, were Russian technological strengths. The Russians not only matched the Manhattan Project within four years, but were the first to launch an ICBM and a satellite. Moreover, from 1945 to 1962 U.S. second-strike capability was not necessarily assured. Potential Soviet technological achievements, backed by massive spending in basic research, thus represented a real threat to long-term U.S. security.18
The potential rising Chinese threat has a number of important differences from the situation in the early postwar years. First, China, despite recent gains, is still in many ways a developing nation. Eighty percent of its population still works in agriculture, using highly labor-intensive technology. China must now move beyond an expertise in light manufacturing into the second stage of development: advanced industrial production. Yet with the United States immersed in the third stage of high-tech production, China will have to scramble to compete in the information age.19
It is in the dimension of potential power that China's real strength remains the most ambiguous. With a territory equal to the United States, China has an adequate resource base. Moreover, such a large area permits the dispersion of nuclear missiles. Such a dispersion is a critical condition for becoming a nuclear superpower, since it allows the nation to absorb a coun-terforce first-strike without necessarily destroying itself in the process.20
In other aspects of potential power, however, China faces important obstacles. Although historically a large population brings increased economic strength, in China's case it is clearly too big for its own good. Super-powerdom requires a per capita surplus over basic consumption that can be devoted to ongoing research and investment, global power projection capability, and high-tech weaponry. China's economic growth may give it a total GNP equal to or greater than that of the United States within a decade or so. But with four times the population, per capita income will remain a fraction of America's for some time to come. This places clear restrictions on China's ability to shift greater funds to the military prerequisites of superpowerdom.
China also remains inferior in probably the most important element of potential power in the modern world: technology. The United States (and Japan) lead the world in almost all technological areas. China is far from narrowing the current gap, including that in military technology. So while in the 1950s U.S. leaders could rightly worry that Russia might overtake
1
The Implications of the Argument
America in technological know-how—a superiority that might be translated into military superiority—there seems little reason at present to think China could achieve this goal. Moreover, the United States now possesses a large and secure second strike. Given the diversity of the U.S. bomber, submarine, and missile forces, only a breakthrough on an anti-missile defense system could undermine the U.S. nuclear deterrent. Yet the Americans remain second to none in such technology.
This analysis helps explain the global stability of the 1990s and why the subsequent few years should also be relatively peaceful.21 Note that liberal arguments have difficulty accounting for this stability, at least in terms of the key great power relationship: the United States and China. China remains authoritarian, so democratic peace arguments do not apply. By the turn of the new century, China was still outside the institutional framework that supposedly fosters peace (Organization for Security and Cooperation in Europe, NATO, Partnership for Peace, WTO).22
The issue at hand, therefore, is under what conditions we would expect a deterioration in U.S.-Chinese relations over the next two decades.23 Since rising states wish to avoid conflict, the theory would expect China to continue to be relatively peaceful (especially since economic engagement is helping to fuel its growth). If destabilizing policies are to be initiated, the perpetrator will likely be a declining United States.24 Chapter 2 outlined three variables and three parameters that help predict when a declining state will shift from peaceful engagement to hard-line strategies. Two of the variables focus on the depth and inevitability of decline in the absence of strong action. We have seen that Washington over the last few years has had little reason to believe that decline would be either deep or inevitable. Given this, and given the real risks involved in reigniting a new cold war through containment, engagement has made sense.
Over the next two decades, however, intense debates between hawks and doves over the depth and inevitability of decline—and what the United States can do about it—will likely increase, especially if China can sustain high growth rates.25 As I argued in chapter 2, however, both groups will likely agree on the basic causal logic, disagreeing only over estimates of variable/parameter values. Hawks will tend to reject engagement because they are more pessimistic about the depth of the U.S. fall. Downplaying arguments as to why Chinese economic growth should peter out, such individuals will focus on China's potential to become the world's largest economy and to translate this economic power into military strength. While acknowledging that hard-line strategies will likely produce a new cold war, hawks will probably have lower estimates than doves of the likely spiraling effects (parameter two). The risks of letting China rise would therefore be greater than the risks of inadvertent war. Two other parameters from chapter 2 will also animate the hawks-doves
The Origins of Major War
debate. The first is the likelihood of"Chirm attacking the United States later at its peak, if permitted to grow. Hawks undoubted]}' will perceive China, given its Communist leadership, as a future threat (and democratic-peace theorists should agree). Moderates and doves will point to the existence of" nuclear weapons and America's secure second strike as a powerful deterrent even if China becomes preponderant.26 The other parameter is the extent to which a hard-line strategy short of war can actually overcome U.S. decline. Agreement is likely here that through U.S. adoption of containment, Chinese economic growth can be moderated: given China's need for trade and investment, CoCom-like restrictions would clearly hurt. But the value of such restrictions must be balanced against the real risk of unleashing a new cold war that would increase the chance of inadvertent war.
How does a president, in the midst of such tradeoffs, choose a policy to maximize the nation's long-run security? The president's task is to establish the best estimates for the variables and parameters described above, and then to balance off the benefits, costs, and risks for each option along the soft-line/hard-line spectrum. Over time, as more information is received, a rational president will update his or her estimates and adjust policy accordingly. Currently, China's long-term rise to superpower status may not seem inevitable if engagement continues. Within the next decade, however, if China's relative growth continues, estimates will have to be revised and policy can be predicted to gravitate toward the hard-line end of the spectrum.
In short, a wait-and-see policy is rational now given the risks of a new cold war, but in another decade U.S. leaders may find themselves back in Truman's dilemma of 1945. Truman, despite warnings that a hard-line strategy would spark a destabilizing rivalry, moved to restrict Soviet growth. In ten years' time, U.S. policy-makers will likely face a similarly profound choice. Whether they move to containment will depend less on China's friendliness, and more on the updated estimates of the depth and inevitability of decline, the degree to which decline can be averted by strong action, and the likelihood of war as a result of such action.
This book's theory is a theory on the effects of decline, not on its causes per se.27 It thus offers predictions on state behavior for different future scenarios, but it cannot predict which scenario will transpire. China might have an internal revolution tomorrow that halts its growth, whereas Russia might get its economic act together to become the new rising state. By establishing the variables and parameters affecting a state's expected probability of survival, however, the theory shows the interaction between the key causal factors that determine a state's rational policy. Moreover, by isolating the effects of differentials and trends in relative power, the theory helps leaders understand the systemic framework for their policies before they plunge into all the complications of the unit level.
[244]
The Implications of the Argument
Existing realist theories remain at odds because of limitations in their deductive structures. They thus offer poHcv-makers few predictions that arc-not challenged by opposing views within the realist camp.-5 Dynamic differentials theory, by synthesizing their strengths into one causal logic, helps to resolve this intrarealist conflict. No systemic argument will explain and predict everything; rational leaders will still want to consider domestic and personality factors in their decision-making process. Yet given the evidence demonstrating how strongly and how often systemic variables override such factors, leaders can ignore systemic constraints only at their own peril.
This book has shown the explanatory and predictive value of taking a dynamic realist approach to the analysis of great power behavior and major war. Unit-level factors no doubt still play important causal roles. Yet by emphasizing dynamic trends in the power balance, this study reinforces the tragic dimension of world politics. Even good security-seeking states will be inclined to hostile acts in the face of deep and inevitable decline. Moreover, the problem for declining states is less the other's present characteristics than its future characteristics once it grows to preponderance. And since rising states have an incentive to project peaceful intentions, declining states have difficulty estimating future intentions based on the other's current behavior.
The intractable problems of decline suggests new research agendas for both realist and liberal scholars. Realists have to abandon the sterile debate over whether equality or inequality between great powers is destabilizing. As we have seen, both may be problematic, depending on polarity and on the depth and inevitability of the dominant state's projected decline. The real question is therefore under what conditions does equality or inequality lead to war. This book helps to answer this question.
Realists must also go beyond their primary focus on relative power as an exogenous force that actors simply accept, and act upon. Dynamic differentials theory provides a framework allowing actors to adopt hard-line policies short of war that stand a chance of reversing decline. This framework offers important theoretical advantages. By facilitating predictions on when cold war rivalries and great power crises will occur, it permits predictions on the probability of major war as a continuous variable. Valid arguments on the risks of spiraling to inadvertent war within rivalries and crises can thus be integrated with a broader realist theory of major war.
The dilemma of decline also requires a shift in focus for liberal scholars and those employing modern game theory. Too much emphasis has been placed upon comparative statics—on examining snapshots of actor characteristics and of the information each side possesses regarding those characteristics. This book indicates that the problem is less one of determining the other's present type than of estimating its future type in environments of
[245]
The Origins of Major War
shifting power. Large-N quantitative studies of the behavior of democratic versus authoritarian states have not captured this. Moreover, to the extent that games of incomplete information focus on how costly signals reveal the other's current type, they miss the core problem: the declining state's anxiety regarding the other's type many years down the road. This future type is something about which the rising state's signaling practices, even before one considers its incentive to misrepresent, can reveal little.
I end with a call for theoretical cooperation between realist and liberal paradigms. For too long these paradigms have been unnecessarily at odds with each other, with realists stressing the primacy of power while liberals counter with domestic factors, international norms, and psychological pathologies. Yet as I suggested in chapter 2, there exists an underlying causal logic on which both camps can agree. Since both groups want to help security-seeking states (in particular, the United States) make rational decisions, both can recognize the problem that decline poses but also the problem of the rising state's future type and of the risk that hostile policies can bring on an inadvertent war. Dynamic differentials theory, in order to build a strong systemic realist theory, has isolated the role of changing power on a rational state's decision-making process. Relaxing the theory's parameters and assumptions, however, illuminates how shifts in non-power variables should affect state behavior within any particular dynamic power environment. Realists and liberals can then dispense with debates over whether power or nonpower variables "matter." They can move to the more fruitful question of the conditions under which they matter, and to what extent. They can also examine when and to what extent power variables will work with nonpower variables to create their effects, and when in fact power might override other factors (or vice versa).
Beginning with this common framework, empirical analyses would then not seek to score definitive coups against the other paradigm. Rather, since empirical counterexamples can always be identified for every argument, the task would be to show how often power factors trumped nonpower variables (or the converse), and how often only a mixture of variables explains the events in question. This book has demonstrated the significant influence of declining power on state behavior across time. Yet to provide guidance on issues such as the rise of China, scholars must offer leaders coherent arguments for how power differentials and trends interact with • other parameters to shape a state's rational policy. Dynamic differentials theory offers one such argument.
Appendix
[246]
Table A.i reproduces the figures on national strength during the first years of war from Jacek Kugler and William Domke's article "Comparing the Strength of Nations," Comparative Political Studies 19 (April 1986): 39-70. The authors construct an overall index of actualized power that takes into account not only a nation's economic resource base (as measured by GNP) but also its political capacity to mobilize this resource base.
Kugler and Domke's figures of "national strength" are a function of a nation's "internal capabilities" plus its "external capabilities." Internal capabilities in turn equal the "societal resource base" (GNP) multiplied by the "relative political capacity" of a state (a separate index measured as a ratio of the state's actual extraction of resources versus its expected extraction). External capabilities equal the foreign aid the state receives to fight a war multiplied by its relative political capacity.
Kugler and Domke do not provide separate figures for the Soviet Union 1939-41 and Britain 1941-42. Neither do they provide separately Germany's total national strength for 1941 and 1942. I have calculated those German figures by taking the authors' total for the Axis alliance in 1941 and 1942 and subtracting from it the power figures for Italy and Japan. On the eastern front in particular, Germany's index of strength was 153.2 in 1941 and 165.5 in 1942.
Tables A.2-A.4 provide each state's percentage share of resources over five different indices of power in the Correlates of War data set (University of Michigan). While the data to calculate these relative balances were not necessarily available to the leaders of any of these states at the time and are often distorted by questionable national-accounting practices and fluctuating exchange rates, the figures do provide a rough check on the accuracy of
1247]