Abstract

Does neorealism offer a convincing account of great power balancing behavior? Many scholars argue that it does not. This conclusion rests on a misunderstanding of neorealist theory and an erroneous reading of the evidence. Properly specified, neorealism holds that great powers place an overriding emphasis on the need for self-help. This means that they rely relentlessly both on arming and on imitating the successful military practices of their peers to ensure their security. At the same time, they rarely resort to alliances and treat them with skepticism. There is abundant historical evidence to support these claims. Since 1816, great powers have routinely achieved an effective balance in military capabilities with their relevant competitors and promptly copied the major military innovations of the period. Case studies show that these outcomes are the product of states' efforts to ensure security against increasingly capable rivals. Meanwhile, the diplomatic record yields almost no examples of firm peacetime balancing coalitions over the past 200 years. When alliances have formed, great powers have generally doubted the reliability of their allies and of their opponents' allies. Thus neorealism provides a solid foundation for explaining great power balancing behavior.

Introduction

Do great powers balance the way neorealism predicts they will?1 Critics answer no. John Vasquez asserts that “states simply do not engage in balancing with the kind of regularity that … [neorealism] assumes,” and dismisses efforts to explain why as evidence of a degenerating research program.2 Paul Schroeder concludes that neorealist claims about the “prevalence of balancing in international politics do not withstand historical scrutiny,” and advises “international historians not to adopt the neo-realist paradigm.”3 Daniel Nexon judges that “[b]alance of power theory, at least in its stronger [neorealist] variants, cannot survive the combined weight of arguments and evidence” in the literature, though he concedes that there may be “a case for preserving a weak balance of power theory.”4 Even self-identified realists agree that neorealism's apparent explanatory failures warrant revision of the theory. Randall Schweller declares that states balancing too little and too late is “a common historical occurrence … common throughout space and time.” To explain this finding, he develops a domestic political argument.5 William Wohlforth offers an account of state behavior that relies on perceptions of the balance of power because he sees scant evidence that the balance itself has a significant effect on states’ actions: “If the balance of power has laws, then they are laws with loopholes big enough to drive a superpower through. If states show a tendency to balance against power, it is a weak one.”6

In contrast, we argue that neorealism provides a compelling explanation of great power balancing behavior. Properly understood, neorealism contends that states place an overriding emphasis on the need for self-help.7 Not knowing others’ intentions and aware that there is no higher authority to protect them, great powers understand that they must provide for their own security. Consequently, states rely relentlessly both on arming and on imitating the successful military practices of peer competitors and rarely resort to alliances for their security. In other words, they constantly balance through internal means and seldom through external means.8 Even when pushed into external balancing, states lack confidence in its effectiveness. International politics being an uncertain, high-stakes business, great powers put stock in capabilities they own, but little in those others pledge.9

Although these arguments follow from neorealism's assumptions and are consistent with its core logic, they represent a minority view. Neorealists themselves have not developed the claim that great powers conceive of internal and external balancing differently and behave accordingly. The father of neorealism, Kenneth Waltz, merely comments that “[f]aced with unbalanced power states try to increase their own strength or they ally with others to bring the international distribution of power into balance.”10 Likewise, John Mearsheimer contends that internal and external balancing are simply two different forms of the same behavior.11 Others read neorealism the same way. James Morrow declares that states regard arming and alliances as alternate means to achieving security.12 Schroeder asserts that the neorealist conception of self-help includes states acting alone as well as in concert with others.13 Most analysts take a narrower view and focus only on external balancing. As Jack Levy and William Thompson note, “[M]ost of the balance of power literature conceives of balancing in terms of counterbalancing alliances.”14

Our study, however, reveals abundant evidence to support a neorealist theory of balancing. Great powers have routinely engaged in internal balancing since 1816, arming and imitating the successful military practices of others to counter the capabilities of their rivals. In approximately 80 percent of the cases we examined, they achieved an effective balance in military capabilities with relevant competitors and promptly copied the major military innovations of the period (the Prussian system, battlefleet warfare, blitzkrieg, carrier warfare, strategic bombing, and nuclear weapons). Case studies suggest that these outcomes are the product of states’ efforts to ensure security against increasingly capable peers. Yet the failure to establish these facts has obscured neorealism's explanatory power.15

At the same time, great powers have seldom engaged in or relied on external balancing. Our examination of the diplomatic record yields five clear-cut cases of external balancing in wartime and one in peacetime over the past 200 years. Moreover, states have generally doubted the reliability of their allies and of their opponents’ allies. Such evidence will be familiar to analysts of great power balancing; but contrary to critics’ claims, it validates neorealist logic, which holds that great powers put little faith in the help of others.16 Together, these findings on internal and external balancing show that neorealism's proponents have understated its successes while its detractors have overstated its failures.

The remainder of this article proceeds as follows. In the first section, we lay out a neorealist theory of balancing and address potential objections. The second section tests the theory using quantitative and qualitative evidence on great power arming, imitation, and alliance behavior since 1816. The final section summarizes our arguments and findings, and draws out implications for theory and policy.

A Neorealist Theory of Balancing

Neorealism argues that great powers seek security in a self-help world. The best means for protection are internal balancing and external balancing. Great powers trust internal balancing because it minimizes reliance on others. Therefore, they will persistently and promptly respond to advances in their peers’ capabilities by augmenting their armed forces and imitating others’ successful military practices. Conversely, great powers distrust external balancing because it involves dependence on others. As a result, they will seldom enter into alliances, and they will treat them with skepticism.

We make several assumptions about great powers and their environment.17 To begin, we assume that great powers are the most important actors in international politics, that they are rational and unitary, that they possess some military capability, and that they give priority to their security. By rational, we mean that states are sensitive to costs. Unitary indicates that they are black boxes that differ chiefly in size, not in composition. States have some information about their peers’ capabilities, but they are not privy to their intentions.18 Great powers prioritize their security because it is a prerequisite for achieving other goals. Finally, military capability allows states to hurt and possibly destroy one another.

Next we assume that the international system is anarchic; that is, there is no central authority above states. We also assume that military capability projects variably across space because of two factors: geographical distance and barriers. Great powers suffer a loss of strength gradient when they deploy capabilities over long ranges.19 Barriers can take many forms, including rivers, mountains, jungles, and cities. For present purposes, the most important distinction is between land and sea. Specifically, states can project military might more invasively over land than at sea.20

Under these circumstances, states closely monitor their military capabilities relative to those of their peers. Without a central authority to protect them, knowing that others have the ability to hurt them, but unsure about others’ intentions, great powers conclude that they must procure the means to defend themselves. More simply, they understand that they operate in a self-help world.21 Yet acquiring military capability creates a the security dilemma; when one state tries to make itself more secure by increasing its military might, others are rendered less secure.22 To guard against harm, states work hard to increase their military capabilities, leading to competitive spirals. There is no escaping the horns of this dilemma: building capability cannot guarantee security, but failing to do so can guarantee insecurity.

In this competition, states are strongly constrained by their power, which is to say the resources they have available to convert into military capability.23 Resource-rich great powers search for a margin for error, a buffer of military superiority, to enhance their security. Their less powerful peers seek to minimize their military inferiority.24 States that do not have resources on a par with the most powerful states in the world also engage in self-help, but in attenuated and more diverse forms. Only when they break into the ranks of the great powers, usually as a function of uneven growth rates, do their relations with other great powers become fiercely rivalrous.25

To generate military capability, states engage in balancing behavior. Balancing refers to a state's efforts to amass military might so as to deter another's aggression or prevail in a conflict should deterrence fail.26 Great powers can adopt two kinds of balancing policies: internal and external. Internal balancing occurs when states increase their military might. One way they do this is by arming: bolstering military spending, weapons production, troop levels, or some combination thereof.27 Another is by imitating the successful military practices of others, especially their doctrinal or technological innovations.28 Great powers balance externally when they form alliances—firm military commitments to deter or defend against common rivals.29

Internal balancing is most desirable because it is true self-help.30 Great powers will react routinely, promptly, and proportionately to threats, which are products of military capability and geography.31 States will internally balance in a reciprocal dynamic with threatening peers—constrained only by the sum of their resources—but less so with weak and distant competitors. This entails varying levels of effort: great powers are likely to work harder to enhance their military might when they possess comparatively fewer resources or reside closer to other great powers.

External balancing is least desirable because it requires the help of others. States are reluctant to rely on alliances because they have little assurance that their partners will honor their commitments.32 Still, any help may improve a state's position. Thus, great powers are always on the lookout for security hedges, and loose commitments are likely to be ubiquitous. If balancing means combining military capabilities to deter and defeat peers, however, then faithful reliance on external balancing will be rare. As Waltz points out, true external balancing will occur “only under the pressure of [great power] war.”33 That is to say, external balancing is an act of desperation by states in extremis. Consequently, great powers will not merely discount their partners’ promises; they will also doubt the reliability of their opponents’ allies.34 It follows that alliances are slow to form and slow to respond to the threats or actions of rivals.

For great powers, bandwagoning and buck-passing are not viable alternatives to external balancing. States bandwagon when they ally with a formidable peer and allow their partner a disproportionate share of the spoils.35 Although bandwagoning is conventionally presented as a viable option, great powers will shun it.36 In a neorealist world, any move that creates a more capable rival reduces a state's security.37 Buck-passing, too, is often cited as a genuine alternative to balancing.38 Ostensibly, buck-passing involves a state sitting on the sidelines and getting another state to incur the costs of balancing against a threatening power. As Mearsheimer notes, aspiring buck passers build up their capabilities and avoid alliance commitments.39 Yet this is no more than ordinary great power behavior.

Geography has implications for arming, imitation, and alliance decisions.40 Land and sea powers can be expected to employ different levels of defense effort. Great powers have more to fear from close continental competitors than from remote maritime states. Safer behind their moats, sea powers need not devote as great a proportion of their resources as land powers to developing military might. In addition, geography drives how states allocate their assets. Insular powers are more responsive to sea-based military innovations, whereas land powers are more responsive to land-based military innovations because these jeopardize the foundations of their security.41 Finally, land and sea powers have different attitudes toward alliances. Being more secure, sea powers will be even less likely to enter into alliance commitments than land powers.42

Our neorealist theory is structural and determinate. The theory is structural because it highlights the unintended consequences of great power behavior. Systemic outcomes cannot be reduced to the efforts and aims of states.43 It is the system that causes states to try to tilt the distribution of military capabilities in their favor, but responses to such shifts frustrate states, homogenize behavior, and promote unintended outcomes. Although great powers seek advantage, balances recur; although they seek security, insecurity reigns.44

Many analysts argue that neorealism cannot explain great power behavior because structure shapes states’ actions rather than determining them.45 As one prominent work on the subject declares, “Over the long term, international political outcomes generally mirror the actual distribution of power among states. In the shorter term, however, the policies states pursue are rarely objectively efficient or predictable based upon a purely systemic analysis.”46 We argue, however, that structure has causal force even over short timespans and that in international politics it forces balancing. Indeed, it is contradictory to claim, as neorealists do, that the brooding shadow of violence is ever present and then conclude that rational states do not typically respond to it.47

Nevertheless, we are not structural determinists. Our theory is purely structural; that is, it admits no elements at the individual or domestic level of analysis, not because such elements are trivial—they are not—but because focusing on structure alone is a demanding way to explore causal priority and the limits of various analytical approaches. All theories have anomalies; none is fully determinate; but the strongest factors create tendencies that allow reasonable suppositions. There remains room for volition in our theory. Negative feedback is likely to be negligible or neutral in response to many actions in world politics, but balancing behavior, we argue, is not among them. Because security is a paramount concern, states have strong incentives to balance, but choice is ineradicable.

In sum, neorealism holds that the default course of action for great powers is incessant internal balancing. Further, they will enter into firm alliances only when conflict is imminent, and often not even then. Weaker states will try to catch up to more powerful states. Land powers will put forth more defense effort than sea powers and will seek to match the arms buildups and innovations of other land powers. Similarly, sea powers will expend less effort than land powers on military measures and will peg their arming and imitation to other sea powers. Although both land and sea powers will infrequently enter into alliance commitments, the latter will do so less often than the former.

Arming, Imitation, and Alliance Behavior

Correctly specified, neorealism powerfully predicts great power balancing behavior. Here we use the historical record to show that great powers arm and imitate promptly against their peers; that such internal balancing behavior occurs for the reasons neorealism stipulates; that great powers rarely opt for external balancing; and that they view alliances as risky propositions.

Before turning to these tasks, we clarify the scope of the analysis. The following are great powers: Austria-Hungary, 1816–1918; Britain, 1816–1945; France, 1816–1940; Prussia (Germany), 1816–1945; Russia (the Soviet Union), 1816–1990; Italy, 1861–1943; Japan, 1895–1945; and the United States, 1898-present.48 The analysis begins in 1816, when reliable data become widely available and it is fair to assert that nation-states have entered into force. It ends in 1990, when the Soviet Union collapsed, leaving the United States as the sole great power in the world and rendering analysis of traditional balancing temporarily moot.49

We code land power versus sea power according to geographic fact except where power overwhelms geography.50 This means that Austria-Hungary, France, Prussia (Germany), Italy, and Russia (the Soviet Union) are land powers. Because they are relatively close to Eurasia, Britain and Japan are partially continental sea powers. Much farther away, the United States was a fully insular sea power before World War II. After the collapse of Western Europe's great powers, however, the United States deployed formidable forces immediately adjacent to the Soviet Union, thereby becoming a fully continental sea power.51 Structural imperatives drove this outcome: a Soviet Union that dominated Eurasia could develop military capabilities far superior to those of the United States.52

INTERNAL BALANCING: ARMING

We claim that great powers seek to create or reduce margins of military superiority vis-à-vis rivals.53 This logic is especially relevant among states in the same neighborhood, because they have the greatest opportunity to hurt each other. Moreover, we expect great powers to engage in prompt internal balancing. Hence, if states develop a marked advantage, they seldom maintain it for long. Instead, an effective military balance will prevail most of the time among competitors.54 The same is true of sea powers, because they view other sea powers as their most relevant rivals. In addition, states with resources inferior to those of their neighbors are likely to devote more effort to arming. Conversely, sea powers are less threatened than land powers by virtue of geography, so they are likely to devote less effort to arming.55

By military capabilities, we mean expenditures and personnel, and draw our data from the Correlates of War (COW) project.56 To generate a capability measure, we first took the great powers’ military expenditures in a given year and gave the state with the highest spending a score of 100. Others were assigned scores reflecting their spending as a percentage of the lead state's expenditures. We then did the same for military personnel. Finally, we averaged each power's expenditure and personnel scores, and the result was its military capability for a given year.

To be sure, data on personnel and expenditures are contested.57 Furthermore, these indicators do not include factors such as force postures and doctrine that render military forces more or less effective, and they are therefore crude reflections of states’ military might. Yet a thorough assessment of our theory involves quantitative data on arming; state behavior is most likely to be influenced by assessments of manpower and money; and the COW project is accepted as the most comprehensive and accurate source of such information.

Specifying an effective military balance is inescapably controversial. There is no consensus on how much military capability a state needs to be secure, but international relations scholars suggest that an attacker is likely to win if it has more than a 2:1 advantage in military might.58 As for the meaning of “prompt,” a review of the literature reveals a consensus of around five years.59 Accordingly, we count as an anomaly for our theory every year beyond four years in which a great power had more than a 2:1 advantage over another.60

The evidence offers strong support for the claim that effective military balances prevail most of the time among great powers. Pairing every land power with every one of its peers in every year from 1816 to 1990 and doing the same for all the sea powers produces 1,651 military ratios.61 An effective balance was evident in 84 percent of them.62 Most of the anomalies involve liminal great powers, that is, states with resources that put them on the great power/minor power borderline. If we exclude Prussia from 1816 to 1870 and Italy from 1861 to 1918, when they were the weakest of the powers, neorealism's success rate increases to 92 percent.

Crucially, capability balances cannot be reduced to power balances. The reason is that great powers dedicated different amounts of effort to arming. To measure effort, we collected data on each state's military personnel as a percentage of its male population aged twenty to thirty-nine, as well as its military expenditures as a percentage of its gross national product.63 Then we indexed the two measures—the leading state in each category earned a score of 100—and averaged the indices to establish an effort score.

One manifestation of differing levels of effort is variation between land and sea powers. No state made less effort than the fully insular United States from its entry into the dataset in 1898 to World War II. As for the partially continental sea powers, Britain consistently made less effort than Europe's land powers. Indeed, it ranked last in terms of effort approximately half the time, and second from last a further quarter of the time. In Asia, Japan routinely made less effort than the Soviet Union.

More revealing are the differences within categories. Among land powers, great powers with fewer resources devoted more effort to building up their military capabilities than their more powerful competitors. In every single year of the quarter century before World War I, France and Russia expended more effort generating military capability than Germany.64 France almost always worked harder than Russia. During the late 1930s, Germany and France made more effort than the Soviet Union. The Soviet Union devoted more effort to arming than the United States throughout the Cold War. Among sea powers, the United States always made less effort than Britain and Japan. When forced into a continental commitment after 1945, the United States expended huge resources on defense.

A caveat is in order: liminal great powers expended less effort than neorealism suggests they will. Austria-Hungary and Italy did not work as hard as Germany, France, and Russia in the late nineteenth and early twentieth centuries. In all probability, this is because they would have had to exert prohibitively high levels of effort to match their more resource-rich peers, and too much arming can have ruinous consequences.65 A better option is to arm to a sustainable level and supplement these efforts with diplomacy.

In broad gauge, neorealism explains arming well. Great powers counter the arming efforts of their peers within five years. On land or at sea, effective military balances prevail in the preponderance of cases. Yet there is variation in the level of effort: less powerful states work harder than their more powerful neighbors; land powers make more effort than sea powers; and sea powers apply themselves harder the closer they are to land powers.

INTERNAL BALANCING: IMITATION

Our theory contends that great powers will imitate the successful military innovations of peer competitors and do so promptly.66 Geography plays a key role: proximate states are more likely to imitate than remote states; land powers are more likely to copy land-based innovations; and sea powers are more likely to copy sea-based innovations. To generate a roster of military innovations, we examined eight accounts of military innovation since 1816 and selected innovations that appeared in more than half of them.67 This procedure yields six innovations: the Prussian system; battlefleet warfare; blitzkrieg; carrier warfare; strategic bombing; and nuclear weapons.

For each innovation, we posit a demonstration point, a date when the innovation's transformative effect on states’ military capabilities became known in the international system through an action by the innovating state.68 We then establish how long it took for other great powers to imitate the innovation. If imitation took less than five years, we call this a success for neorealism; five or more we call a failure. Judging when imitation has occurred is difficult: the development and deployment of novel organizational forms, technologies, and tactics is usually a gradual process. Innovations often combine multiple elements, and imitating states seldom copy every aspect of an innovation. Our approach is to determine when an aspiring imitator has replicated the essentials—the features that have radically enhanced the innovator's capability.

prussian system. By integrating the breech-loading rifle, the railroad, and the telegraph with a general staff system, Prussia surpassed other states in its ability to deploy a large, firepower-oriented, fast-moving, and highly coordinated army on the battlefield.69 Although some analysts cite the Austro-Prussian War (1866) as the demonstration point for this innovation, there are reasons to question their coding.70 For instance, Prussia's mobilization did not go as planned. Moreover, the superiority of Prussian methods was not clear to observers at the time, who attributed Prussia's victory to the vaunted needle gun and Austria-Hungary's incompetence. Thus, the demonstration point for the Prussian system is the Franco-Prussian War (1870-71).71

France imitated Prussia's innovation rapidly. In fact, it employed a basic Prussian system in 1870: it had a general staff; its troops were armed with the Chassepot breech-loading rifle; and it conducted its second railway mobilization—the first was in 1859—using an extensive rail network. Immediately following defeat, the French imitated the Prussian system more closely, reforming the general staff, creating the Commission Supérieure des Chemins de Fer in 1872 to manage France's railroads, and expanding the rail network.72 Russia had also incorporated some elements of the Prussian system by the time of the Franco-Prussian War, having switched to breechloaders in 1867 and having created a main staff in 1868.73 It did not, however, move quickly to build a competitive rail network.74 Italy's imitation efforts were also partial. The Italians carried out a rail mobilization as early as 1859 (with the French) and adopted the Carcano breechloader in 1867, but they did not create a genuine general staff until the 1880s.75 In contrast, Austria-Hungary did not begin to imitate the Prussian system until 1881.76 As one would expect of a sea power confronted by a land-based innovation, Britain incorporated the Prussian system slowly.77

battlefleet warfare. Britain's battlefleet warfare innovation revolutionized the way navies could mass force to destroy enemy fleets and control the seas.78 At its base lay a new technology: all-big-gun, heavily armored, fast propulsion battleships, which rendered existing ships obsolete.79 The core task facing navies—winning decisive battles—remained the same, as did naval doctrine and strategy, but the means at their disposal were transformed.80 Unlike most other innovations that were demonstrated in the context of war, the demonstration point for battlefleet warfare was the peacetime launch of the first of these new ships, the HMS Dreadnought, in February 1906.81 The other great sea powers quickly imitated Britain's innovation: the United States began to build dreadnoughts in 1906; Japan followed in 1907.82 The great land powers also imitated Britain's example swiftly: Germany, Russia, France, and Italy were all building dreadnoughts by 1910, and Austria-Hungary began doing so in 1911.83

blitzkrieg. Although several great powers sought to exploit the potential of the tank after World War I, it was Germany that developed the most effective approach to armored warfare in the late 1930s. Merging tanks, radios, and aircraft with combined arms doctrine and organization along divisional lines, blitzkrieg enabled the German army to achieve tactical breakthroughs and convert them into operational penetration of enemy positions.84 Although elements of blitzkrieg were used during the invasion of Poland in the fall of 1939, observers at the time attributed Germany's victory to Polish weakness. The system did not demonstrate its revolutionary effects until the Fall of France in mid-1940.85

In the European theater, the remaining great power combatants imitated Germany's blitzkrieg methods with great speed. The Soviet Union was practicing the art by 1942, and Britain and the United States caught up soon after opening fronts on the continent late in the war.86 Because Italy's geography was ill suited to blitzkrieg, the Italians did not adopt it before capitulating to the Allies in 1943 and exiting the ranks of the great powers.87 In the Pacific theater, Japan had little need for blitzkrieg. Once the United States joined the battle, the geography of the island-hopping campaign was incompatible with deep-penetration armored warfare.88

carrier warfare. The introduction of carrier warfare—that is, the combined use of multiple aircraft carriers supported by destroyer screens and logistical ships—transformed the fighting power of its practitioners by enabling them to launch offensive strikes against enemy navies across great distances.89 Its demonstration point is generally considered to be the Battle of Midway in June 1942, though some analysts cite Pearl Harbor in December 1941.90 Regardless, all the sea powers were practicing carrier warfare by the end of World War II. Japan responded with such speed to its defeat at Midway that it reorganized its fleet around multicarrier task forces a year before the United States did. Indeed, a good case can be made for the United States and Japan as joint innovators and imitators.91 Meanwhile, Britain adopted carrier warfare in 1945, mostly as a consequence of performing combined operations with the United States Navy.92 Predictably, the land powers, Germany and the Soviet Union, did not make a meaningful effort to adopt carrier warfare.93

strategic bombing. The advent of strategic bombing—the selection and destruction of industrial targets by long-range bombers supported by fighter escorts—increased states’ prospects for victory in war by allowing them to degrade their adversaries’ productive capacity and military capability. The demonstration point for strategic bombing is 1944, when the United States and Britain used air power to destroy German industries and transportation links. By the spring of 1945, Germany no longer had the means to continue fighting.94 Two potential imitators, Germany and Japan, were defeated less than a year after the demonstration point and cannot be used to test the theory. The Soviet Union, on the other hand, reacted quickly. In 1948 Joseph Stalin established the Soviet air force as a separate service on equal footing with the army and navy. In addition, he called for better air defense capability, the creation of a rocket force, and the development of a strategic bomber force.95

nuclear weapons. The sheer destructive power of nuclear weapons means that the mere adoption of the technology, independent of any organizational change, is enough to vastly enhance a state's military capability.96 The United States demonstrated the effects of nuclear weapons when it dropped the atomic bomb on Hiroshima in August 1945. The only candidate imitator, the Soviet Union, was not far behind the U.S. example, instituting a crash atomic weapons program almost immediately and detonating its own device in 1949.97 The Soviets were only ten months behind the Americans in developing the hydrogen bomb, and the two countries were racing neck-and-neck to develop ballistic missiles in the mid-1950s.98

In brief, neorealism provides a compelling account of great power imitation. In all but one of the cases we examined, candidate imitators reproduced the essential elements of the innovation in less than five years—faster if the new method was unveiled during a total war. All the anomalies arise in the Prussian system case, and even here only one possible imitator failed outright. Put concretely, there were fourteen candidate imitators, one clear case of failure (Austria-Hungary with the Prussian system), and two cases of partial imitation (Russia and Italy, also with the Prussian system), which is a success rate no lower than 79 percent.99

INTERNAL BALANCING: QUALITATIVE CASES

The evidence shows that effective balances prevail among peer competitors and that successful military practices diffuse rapidly through the great power system. According to neorealism, these outcomes occur because states worry about their security, monitor one another's military capabilities, and repair any erosion of their relative position.100 To support this logic, we describe the arming and imitation behavior of France, Germany, and Russia from 1871 to 1945.101

arming. The period between the Franco-Prussian War and World War II was one of peer competitors constantly arming against each other. With its decisive defeat, France became acutely aware of German superiority and was determined to overcome it. In July 1871, President Adolphe Thiers told the National Assembly: “You will see that we are going to devote all our energies to the reorganization of the French army.”102 Left and right agreed on the importance of obligatory military service, and a raft of reforms followed.103 In 1876 Léon Gambetta warned, “After having watched the maneuvers of every branch of her [Germany's] army … we do not possess any forces which can compare with the troops I have just seen.”104 There was little dissent that France must be on a par with Germany.105

Although Russia and Germany were united in the Three Emperors’ League, Russia focused on self-help after 1870. Universal conscription was adopted in 1874 to counter Germany's superior military capability.106 Reforms under Russia's war minister, Dmitry Miliutin, made progress, but its leaders knew their country was still no match for Germany, and so redoubled their efforts.107 In an 1885 memorandum that would guide Russian military policy for more than a decade, Gen. N.N. Obruchev argued for deploying more forces to Russia's western provinces to compensate for the faster mobilization schedules of Germany and Austria-Hungary.108 Conscripts’ service terms were dramatically reduced to increase rolls and reserves, and improved education boosted literacy threefold from 1865 to 1900.109 Tsar Alexander III, who would cast a long shadow in the late imperial period, succinctly stated: “Russia only has two allies: the army and the navy.”110

Nor was Germany resting on its laurels. After victory, it expanded the army and improved its war machine.111 By the mid-1880s, senior German military planners had become alarmed by Russian progress.112 Russia staged maneuvers on its western border and began building major rail lines at the same time that France was increasing its armed forces.113 Hajo Holborn describes how “in spite of the alliances, [German Chancellor Otto von] Bismarck felt it imperative to expand the German army … most drastically, in 1887, while France and Russia particularly were building up their armies.”114 Speaking on foreign affairs publicly for the last time, Bismarck, the consummate diplomat, “repudiated the value of all alliances, and he declared that Germany must rely on her own strength. ‘The pike in the European carp-pond prevent us from becoming carp.'”115

In 1889 France responded to Germany's 1888 Septennat bill to increase military spending.116 Charles de Freycinet, the French war minister, wrote Joseph de Miribel, his chief of the general staff, lamenting that Germany's “considerable augmentations” put the French “in a situation of inferiority too dangerous for us to resign ourselves to it…. We cannot remain faced with such an increase of forces without taking similar measures.” Miribel concurred: “We ought to have forces equivalent to those of the Germans.”117 Freycinet's reforms duly overhauled conscription and canceled exemptions.118

A German army bill presented in 1892 aimed to increase the country's strength in response to improvements in French and Russian forces.119 Bismarck's successor, Leo von Caprivi, explained that diplomacy would not reduce German defense spending: “Diplomacy alone cannot help. There was Frederick the Great whom almost all of Europe had attacked in 1756, although he was considered as one of the best diplomatists of his day. There is an international race in armaments in Europe and no one can disarm. No power can fall behind the others in armaments; even a halt in the race would be equivalent to a partial disarmament.”120 Caprivi's fears predated the consolidation of the Franco-Russian alliance.121

Russian reform efforts were not keeping up. In 1903 a secret war ministry report revealed that Russia was “insufficiently prepared for defense not only against Germany, but even against Austria-Hungary.”122 Crushing losses in the Russo-Japanese War in 1905 confirmed that assessment and initiated another round of reforms. A 1906 law shortened active service for the first time since 1874, closed recruitment loopholes, and raised trained reservist numbers by 25 percent.123 Under the purview of Ministers A.F. Rödiger and V.A. Sukhomlinov, the Russian military overhauled regimental purchasing, boosted pay, enhanced pensions, tightened performance reviews, streamlined mobilization, redeployed forces, and updated equipment.124

Germany responded, and the cycle accelerated. In 1911 Chief of the German General Staff Gen. Helmuth von Moltke lobbied for large troop increases as a “precondition for survival,” causing the army to swell over the next several years.125 He declared, “Germany must be strong enough to rely on its own power” and had to expand its forces immediately.126 Reichstag politicians concurred: the Socialist leader, August Bebel, thought the balance of forces was worse than the war ministry reported, and Catholic Center Party Chairman Matthias Erzberger explained: “The principal cause of the present danger lay in the changed situation in the army in Russia; against this there was no other means than the [military appropriations] bill.”127 Moltke fretted in February 1914, “Russia's preparedness for war has made gigantic progress since the Russo-Japanese war, and it is now much greater than ever in the past.”128

German actions triggered major military increases by France and Russia.129 By 1912 France was drafting 70–80 percent of available manpower, and Germany was still raising troop numbers.130 The French resolved to close this gap by extending service terms to three years, anticipating that frontline strength would be equal by 1915 or 1916.131 A massive military equipment credit followed to match a similar one that had just passed the Reichstag.132 Meanwhile, the Russian military was increasing troop levels by 500,000 and struggling to spend all the funds it was receiving.133 When war broke out in 1914, Prussian War Minister Gen. Erich von Falkenhayn had begun a successful lobbying effort for universal conscription to close the gap with Russia.134

Arming against Germany remained a cornerstone of French foreign policy during the interwar period. The Treaty of Versailles set stringent limits on Germany's military, capping its army at 100,000 men and restricting its weaponry, but France feared the future. As a French Senate report put it, Germany was “powerless today, but capable of becoming formidable again tomorrow.”135 Soon after the treaty was signed, French Prime Minister Georges Clemençeau declared, “France must have lots of children. If not, you can put what you like into the treaty—France will be done for.”136 Peter Jackson chronicles France's energetic diplomacy between the wars, but even he finds that traditional internal balancing remained central to French security and that at no time were diplomatic solutions dominant in France.137 French forces were so fearsome that Britain was often more concerned about French military action than it was about German action.138 When Germany rearmed, France followed suit, and it was not until 1938–39 that the number of German forces exceeded the number of French forces.139 French arming efforts were so successful that when Germany attacked France in 1940, “not a single general expected victory to result.”140

France's emphasis on internal balancing reflected a belief that states operate in a self-help world. Recounting a conversation with French Prime Minister Édouard Daladier in February 1939, the U.S. ambassador observed: “He [Daladier] considered Chamberlain a desiccated stick; the King a moron … Eden a young idiot…. He felt that England had become so feeble and senile that the British would give away every possession of their friends rather than stand up to Germany and Italy.”141 French Foreign Minister Georges Bonnet expressed similar misgivings about the Soviet Union, complaining to his British counterpart, Viscount Halifax, that “[t]he Russians needed watching.”142

With its defeat in World War I, the Soviet Union took steps to enhance its military capability. Stalin identified the key weakness: “[O]ld Russia … was ceaselessly beaten for her backwardness. For military backwardness … for industrial backwardness. We must make good this lag in ten years. Either we do it or they crush us.”143 Accordingly, he sought to lift Russia “by her bootstraps to a higher level of economic and military development.”144 With German rearmament, the task became pressing. Before World War II, Soviet Foreign Minister Vyacheslav Molotov reported that Stalin “had felt that only by 1943 could we meet the Germans on an equal footing.”145 In response to German rearmament from 1933 to 1939, Soviet military spending more than doubled and personnel more than quadrupled.146

Germany between the wars is also a story of amassing military capability.147 During the 1920s, Germany evaded many of the restrictions imposed by the Allies in 1919 and built a highly disciplined force more than three time larger than the 100,000-man army mandated by the Versailles settlement.148 After Adolf Hitler came to power in 1933, Germany rearmed openly and massively. Hitler feared that such a policy was unsustainable, however: the Soviet Union possessed vast resources that could be translated into overwhelming military might. In 1936 he warned, “One has only to compare the Red Army as it actually exists today with the assumptions of military men 10 or 15 years ago to realize the menacing extent of this development. Only consider the results of a further development over 10, 15, or 20 years and think what conditions will be like then!”149 Hitler resolved on preventive war before the distribution of military capabilities shifted decisively in the Soviet Union's favor.150

imitation. The period under discussion featured two land-based innovations: the Prussian system and blitzkrieg. In both cases, Germany's advantage was transitory because France and Russia quickly imitated Germany's example to the best of their abilities.

France followed the Prussian system nearly to the letter.151 As the minister of war, Gen. François du Barail, acknowledged in 1873: “We imitated much more than we created.”152 German figures set the standards for French targets, and reforms followed: new fortifications, new technology, new organization, and new education.153 Participants explained their actions explicitly in such terms. Freycinet overhauled the French general staff, declaring it would be “analogous to the one which had permitted Marshal von Moltke to achieve such great results.”154 Chairman of the Comité de l'Artillerie Gen. Gaëtan de Rochebouët declared, “The opinion of nearly all the infantry generals is that we cannot have a caliber inferior to that of the Germans.”155 In 1870 France had four rail lines to the Franco-German border while Germany had nine; by 1886 France had twelve to Germany's nine; and by 1913, France had sixteen to Germany's thirteen.156

Following the Franco-Prussian War, the Russian General Staff called for bold new measures to deal with Germany's advantages.157 In 1872 Col. P.L. Lobko and Miliutin wrote complementary memoranda suggesting solutions to Germany's new way of war.158 Russian forces went from largely stationary to more mobile and flexible, with the addition of more cavalry, more rail lines, and the creation of fortress infantry, which freed up regular infantry.159 The Russian army raised training requirements; increased the quality of its rifles and artillery; adopted German exercises such as war games, staff rides, and strategic studies; and established a prestigious general staff.160

Russia faced hard material constraints in its efforts to copy the railroad element of the Prussian system. Building a competitive rail network was difficult given Russia's vast geography and economic inferiority.161 Nevertheless, the Russian government quadrupled the percentage of state spending on infrastructure projects such as railroads from 1891 to 1903.162 As a result, Russia made great strides in closing the quantitative, if not the qualitative, gap with Germany: Germany had more than six times the length of track that Russia did in 1860; by 1880 that figure had fallen to one and a half times; and twenty years later they were almost equivalent.163

During World War II, the Soviet Union promptly imitated Germany's blitzkrieg methods.164 Russian investment in tanks had kept pace with German investment, and some advocates of Russian armored doctrine were sympathetic to blitzkrieg tactics before the war, but they were an uninfluential minority.165 When Germany invaded in June 1941, Russian unpreparedness and doctrinal backwardness cost the country dearly. As the Russian war economy ramped up production and the state was wrenchingly reorganized, military officials struggled to adapt to the new tactics.166 Fundamental changes were initiated in the first year; and no later than the offensives of November 1942, Russian forces combined the core elements of blitzkrieg, and Russian offensives reversed German gains.167

In sum, France, Germany, and Russia armed and imitated successful practices to counter increases in their competitors’ capabilities. This is not to rule out other causes of arming and imitation, including leader preferences or bureaucratic politics. Nevertheless, the timing of these states’ actions and the reasoning behind them lend substantial support to neorealist arguments that great powers arm and imitate in response to adverse shifts in the distribution of military capabilities.168

EXTERNAL BALANCING: ALLIANCES

Neorealism holds that states put little stock in external balancing because it requires reliance on others. As a result, great powers rarely engage in external balancing, view it as a risky proposition when they do, and seldom respond to rival alliances because they suspect that opponents' allies are also unreliable.

RARE. In keeping with conventional usage, if two or more great powers sign an alliance treaty and the terms of that treaty commit them to common defense against a peer competitor or great power coalition, then they are externally balancing.169 To be sure, alliances are only a proxy for external balancing. Great powers may establish them without intending to abide by their commitments. They may plan to support each other against third parties but elect not to sign a treaty. The content of treaty commitments is not necessarily a true reflection of signatories’ thinking. Nonetheless, alliances are the best indication of states coordinating their defense efforts, and the most objective source is the Alliance Treaty Obligations and Provisions (ATOP) dataset, which provides details of all interstate alliances from 1816 to 2003.170

By this metric, great powers rarely balance externally. This lack of external balancing does not reflect an aversion to diplomacy; the ATOP dataset identifies 309 alliances involving great powers from 1816 to 2003. Yet just 73 were agreements between great powers, and only 43 of those were targeted at peer competitors.171 Of these, 29 did not commit the signatories to mutual defense. Hence, no more than 14 of the 73 great power alliances fit a common definition of external balancing.

Five cases of external balancing occurred during wartime, a finding that supports the neorealist contention that great powers resort to firm alliances in extremis. Indeed, four of the five were contracted during wars whose outcomes promised to shift substantially the distribution of power: Britain, France, Italy, and Russia during World War I; Germany, Italy, and Japan during World War II; Britain and the Soviet Union during World War II; and Britain, the Soviet Union, and the United States during World War II.172

Of the nine peacetime alliances, only the Franco-Russian Alliance (1891–1917) can be considered a clear-cut example of external balancing.173 The German Confederation (1820–66) and the Austro-Prussian Alliance (1851–54) were designed as much to manage Austro-Prussian relations and maintain domestic tranquility as to coordinate capabilities against foreign rivals.174 The text of the St. Petersburg Convention (1873–78) between Germany and Russia read as a defensive alliance; five months after ratification of the convention, however, it was “not entirely clear whether or not it was considered to be still valid.”175 Nor was the Dual Alliance (1879–1918) an exemplar of external balancing. Bismarck was more interested in restraining Austria-Hungary than in building a counterweight to Russia. In fact, he drew closer to Russia, first reviving the Three Emperors' League (1881–87), and then concluding the Reinsurance Treaty (1887–90).176 As for the Triple Alliance (1882–1915), it involved only one unambiguous balancing pledge: Germany's promise to defend Italy in the unlikely event of an attack by France.177 Despite Germany and Austria having been allied for thirty-five years, Holborn relates how, at the outbreak of World War I, “[n]o formal military convention had ever been concluded between Germany and Austria-Hungary, nor had the two allies worked out a common war plan. The Austrians were not even given general information on the Schlieffen Plan.”178 None of the twentieth-century alliances we examined can be considered firm mutual commitments to deter or defend against other great powers. The Treaty of Björkö (1905) was unrealized; France and the Soviet Union showed a weak commitment to the Franco-Soviet Treaty (1935–39); and the Pact of Steel (1939–43) was neither defensive nor reliable.179

RISKY. To establish whether or not great powers view external balancing as risky, we identified instances in which great powers in an alliance fought on the same side in a war against another great power or powers.180 Given that the parties to the treaty ultimately made great sacrifices in a common cause, one might expect them to have been confident in their partners’ reliability before war broke out.181 For this reason, these cases constitute a tough test for neorealism's claim that great powers put little stock in external balancing. A strict application of this procedure yields nine candidate cases, but only four are suitable tests of neorealism: Germany and Austria-Hungary, Britain and France, and France and Russia before World War I; and Germany and Italy before World War II.182

An examination of these four most-likely cases reveals a great deal of uncertainty. France was not confident that Britain would support it in a war against Germany. Britain's foreign policy, especially its rapprochement with Germany beginning in 1912, had “created substantial margin for doubt” by the spring of 1914.183 These doubts persisted throughout the July Crisis. Nor were they the product of the French imagination; Germany, too, was unsure about Britain's commitment to France until the verge of war.184

France was also concerned about Russia's reliability. In the years before World War I, French policymakers were deeply impressed by Russia's growing capability. Rather than welcome this situation, however, they worried that a strengthened Russia might pursue an independent foreign policy rather than support France against Germany. Some feared a Russo-German rapprochement.185 Right up to the outbreak of hostilities, French officials suspected that Russia was not truly committed to fighting the Central Powers.186

The opposing side also demonstrated little confidence in external balancing. German policymakers did not believe that Austria-Hungary would stand by Germany unless its own interests were directly threatened. As German Chancellor Theobald von Bethmann-Hollweg put it during the July Crisis, Austria-Hungary would not enter “a war as our ally on behalf of a German cause.” Moreover, he was concerned that if Germany did not offer its steadfast support, then Austria-Hungary would “approach the western powers, whose arms are open, and we lose our last reasonable ally.”187 Meanwhile, Austria-Hungary doubted Germany's reliability, especially given Berlin's recent history of appeasing Russia in Eastern Europe. This was understandable; as late as June 1914, Kaiser Wilhelm refused to answer Emperor Franz Joseph when he asked the kaiser if Austria-Hungary could count unconditionally on Germany in the future.188 Christopher Clark summarizes European great power politics on the eve of World War I: “[I]t is worth noting how fragile and flux-prone many of the key decision-makers felt the alliance system was.”189

With good reason, Germany had serious misgivings about Italy's commitment to the Pact of Steel on the eve of World War II. Italy was the model of unreliability, negotiating with Britain, France, and the Soviet Union during the late 1930s. Even during the Anglo-French staff conversations in August 1939, negotiators entertained the possibility that they might be able to persuade Italy to remain neutral in the event of war.190 Little wonder, then, that Hitler referred to Italy's leader, Benito Mussolini, as his “disloyal Axis partner.”191

What has been missing from the discussion so far is equally illuminating. Except for Britain, all of the cases above involve continental powers. Britain is a partially continental sea power, and it appears because of its reputation as an unreliable ally. Unsurprisingly, insular powers are geopolitically sheltered and have less need for external balancing. It is exclusively states in the riskiest positions at the riskiest times that are compelled to seek firm alliances.

UNRESPONSIVE. The ATOP dataset yields abundant evidence that great powers are slow to reciprocate the countervailing efforts of others. Of the nine examples of peacetime external balancing mentioned above, the only potential instance of tit-for-tat behavior is the Triple Alliance and the Franco-Russian Alliance. Even in this case, France waited almost a decade before signing an entente with Russia in 1891; it did not cement the alliance until 1894; and it did not ratify military protocols regarding the European balance of power until 1901.192 Each offered mostly verbal support for the other in crises in North Africa and the Far East, and made halting progress to coordinate strategies after 1912.193 The Triple Alliance did little to counter increased Franco-Russian cooperation and lacked coordinated plans when war broke out.194

There are no other examples of reciprocation. Britain, France, and Russia did not respond to the creation of the German Confederation or the Austro-Prussian Alliance. As has often been noted, the German great powers were considered no more formidable a threat in tandem than they were on their own.195 The St. Petersburg Convention and the Treaty of Björkö were not regarded as meaningful alliances and therefore elicited no response. Rather than respond to the Franco-Soviet Treaty by forming a countervailing coalition, Nazi Germany exploited the unreliability of external balancing to drive a wedge between the allies, establishing a consultation pact with France (the 1938 Franco-German Declaration) and a nonaggression treaty with the Soviet Union (the 1939 Molotov-Ribbentrop Pact).196 Lastly, the Pact of Steel made little difference to British Prime Minister Neville Chamberlain, who doubted Italy's attachment to Germany, worked to decouple the two, and did not accelerate ongoing Anglo-French cooperation.197

In short, great powers negotiate alliances for many reasons, but coordinating defenses against other great powers is seldom one of them. Nor do they view their own or others’ allies as reliable. Thus, even the firmest alliances show deep cracks, and there is only one instance of great powers reciprocating the moves of a countercoalition.

Conclusion

Self-help is at the heart of neorealism. States know that even when the world is calm, it is not safe, and absent a higher authority they must protect themselves. Thus, great powers place their trust in internal balancing to deal with adverse shifts in the balance of military capabilities. Power and geography mean that different states exert different levels of effort and imitate different innovations, but overall internal balancing is relentless. The same is not true of external balancing. Allies can issue assurances that they will contribute capabilities to combat threats, but states cannot be sure that their partners will honor their commitments. As a result, not only do great powers rarely balance externally, but they also doubt the firmness of their own and their rivals’ alliances.

The historical record supports these arguments—great powers balance as neorealism predicts. Since 1816 they have routinely armed and imitated in response to their peers’ increases in capability. Some of them have gained a military advantage by increasing recruitment and spending, but such mismatches have been exceptional and short lived. Balances have been the norm, because every great power has carefully monitored its peer competitors and calibrated its efforts accordingly. Similarly, innovations have granted a military advantage to early adopters, but not for long. Land powers have imitated land-based innovations promptly; sea powers have done the same with sea-based innovations. All have imitated especially quickly in the cauldron of war.

In contrast, great powers have rarely balanced externally. In wartime, they have had little choice but to enter into alliances, but not so in peacetime. Even when alliances have formed, great powers have doubted the reliability of their partners and their adversaries’ partners. In theory and in fact, great powers view internal balancing as reliable, practice it routinely, and reciprocate the internal efforts of others. Meanwhile, they view external balancing as risky, practice it rarely, and seldom respond to the external moves of their adversaries.

Whatever one's theoretical inclination, a neorealist theory of balancing is a tool with wide utility. Critics can use it as a baseline to tout the value added of alternative views. Proponents can use it as a foundation to develop more complex models of state behavior or as an inspiration to extend its logic to different outcomes. Although our inquiry into internal balancing has been preliminary, the results are encouraging and merit exploration. How do states set the ceiling on their arming efforts? Why do states sometimes fail in their internal balancing efforts? What is the decisionmaking calculus of liminal great powers? As for external balancing, firm alliances among great powers may be rare, but looser agreements remain valuable tools of statecraft, and neorealism has few explanations for why states choose the diplomatic policies they do.198

A neorealist theory of balancing also has enduring practical relevance. It offers a parsimonious explanation of the prevailing situation and a serviceable guide to the future. The reason commentators have detected a lack of balancing against the United States since the end of the Cold War is that there have been no other great powers in the international system.199 If uneven growth rates persist, however, China will become a peer competitor of the United States.200 As that happens, both countries will compete feverishly for security, increase military spending, develop new military platforms, and react swiftly to the other's capabilities. As other great powers emerge, pundits will lavish attention on diplomacy, but the major players still will not rely on external balancing for their security. In the timeless cycles of power politics, salvation lies within.

Acknowledgments

Earlier versions of this article were presented at Columbia University, Dartmouth College, Harvard University, the University of Northern Illinois, the University of Notre Dame, and the 2014 annual convention of the International Studies Association. In addition to participants at these seminars, the authors thank Daniel Altman, Joshua Baron, Nilda Garcia, Stacie Goddard, Matt Hales, Michael Horowitz, Paul MacDonald, Maria del Pilar Quintana, and especially John Mearsheimer. For able research assistance, they are grateful to Mariana Gaviria, Joseph Karas, Morgan Lee, and W.P. Quinlan. Finally, they owe an enormous debt of gratitude to the late Kenneth Waltz.

Notes

1. 

Neorealism is a school of international relations theories that privileges structural factors and relegates domestic and individual-level factors to lesser roles in causal priority. The most important examples are Kenneth N. Waltz, Theory of International Politics (New York: McGraw-Hill, 1979); and John J. Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001). We distinguish between balances and balancing, and are interested primarily in the latter. Balances describe an outcome, specifically an equilibrium of military capability in the international system. Balancing refers to state behavior designed to augment military capabilities, alone or in concert, for deterrence or defense. See Colin Elman, “Introduction: Appraising Balance of Power Theory,” in John A. Vasquez and Elman, eds., Realism and the Balancing of Power: A New Debate (Upper Saddle River, N.J.: Prentice Hall, 2003), pp. 8–9; and T.V. Paul, “Introduction: The Enduring Axioms of Balance of Power Theory and Their Contemporary Relevance,” in Paul, James J. Wirtz, and Michel Fortmann, eds., Balance of Power: Theory and Practice in the 21st Century (Stanford, Calif.: Stanford University Press, 2004), p. 2.

2. 

John A. Vasquez, “The Realist Paradigm and Degenerative versus Progressive Research Programs: An Appraisal of Neotraditional Research on Waltz's Balancing Proposition,” American Political Science Review, Vol. 91, No. 4 (December 1997), p. 910.

3. 

Paul Schroeder, “Historical Reality vs. Neo-realist Theory,” International Security, Vol. 19, No. 1 (Summer 1994), pp. 124, 148.

4. 

Daniel H. Nexon, “The Balance of Power in the Balance,” World Politics, Vol. 61, No. 2 (April 2009), p. 353.

5. 

Randall L. Schweller, Unanswered Threats: Political Constraints on the Balance of Power (Princeton, N.J.: Princeton University Press, 2006), p. 16.

6. 

William C. Wohlforth, The Elusive Balance: Power and Perceptions during the Cold War (Ithaca, N.Y.: Cornell University Press, 1993), p. 299. Other realists who claim that structure alone cannot explain great power balancing include Thomas J. Christensen and Jack Snyder, “Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity,” International Organization, Vol. 44, No. 2 (Spring 1990), pp. 140–147; Jeffrey W. Taliaferro, Steven E. Lobell, and Norrin M. Ripsman, “Introduction: Neoclassical Realism, the State, and Foreign Policy,” in Lobell, Ripsman, and Taliaferro, eds., Neoclassical Realism, the State, and Foreign Policy (New York: Cambridge University Press, 2009), pp. 1–31; and Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987), pp. 263–265.

7. 

On self-help, see Waltz, Theory of International Politics, pp. 91, 105–107, 111; and Mearsheimer, The Tragedy of Great Power Politics, p. 33.

8. 

On internal and external balancing, see Waltz, Theory of International Politics, p. 118; and Mearsheimer, The Tragedy of Great Power Politics, pp. 156–157.

9. 

The terms are from Samuel P. Huntington, “Arms Races: Prerequisites and Results,” Public Policy, Vol. 8, No. 1 (1958), p. 46.

10. 

Kenneth N. Waltz, “Evaluating Theories,” American Political Science Review, Vol. 91, No. 4 (December 1997), p. 915. See also Kenneth N. Waltz, “The Origins of War in Neorealist Theory,” Journal of Interdisciplinary History, Vol. 18, No. 4 (April 1988), p. 625.

11. 

Mearsheimer, The Tragedy of Great Power Politics, pp. 156–157. Mearsheimer briefly makes a claim similar to ours, but it is not central to his theory. See ibid., p. 33.

12. 

James D. Morrow, “Arms versus Allies: Trade-Offs in the Search for Security,” International Organization, Vol. 47, No. 2 (Spring 1993), pp. 207–208, 213–217.

13. 

Schroeder, “Historical Reality vs. Neo-realist Theory,” p. 116. Waltz implies the same. See Anna Cornelia Beyer, Kenneth Waltz's Life and Thought: An Interview (Raleigh: Lulu.com, 2015), pp. 84–85; and Waltz, Theory of International Politics, pp. 111, 163.

14. 

Jack S. Levy and William R. Thompson, “Balancing on Land and at Sea: Do States Ally against the Leading Global Power?” International Security, Vol. 35, No. 1 (Summer 2010), p. 23.

15. 

Neorealists have provided little evidence to substantiate their contentions about internal balancing. The canonical texts contain only a handful of empirical examples. See Waltz, Theory of International Politics, pp. 125, 127; and Mearsheimer, The Tragedy of Great Power Politics, pp. 286, 309, 313, 315, 318, 326. For further supporting evidence, see Christopher Layne, “The Unipolar Illusion: Why New Great Powers Will Rise,” International Security, Vol. 17, No. 4 (Spring 1993), pp. 20–32; Colin Elman, “The Logic of Emulation: The Diffusion of Military Practices in the International System,” Ph.D. dissertation, Columbia University, 1999; and João Resende-Santos, Neorealism, States, and the Modern Mass Army (New York: Cambridge University Press, 2007). None of these works offers a comprehensive account of great power internal balancing behavior.

16. 

For similar findings, see Jack S. Levy and William R. Thompson, “Hegemonic Threats and Great Power Balancing in Europe, 1495–1999,” Security Studies, Vol. 14, No. 1 (January/March 2005), pp. 23–29; Richard Rosecrance and Chih-Cheng Lo, “Balancing, Stability, and War: The Mysterious Case of the Napoleonic International System,” International Studies Quarterly, Vol. 40, No. 4 (December 1996), pp. 497–498; and Marc Trachtenberg, “The Question of Realism: A Historian's View,” Security Studies, Vol. 13, No. 1 (Autumn 2003), pp. 189–190.

17. 

For similar lists, see Charles L. Glaser, Rational Theory of International Politics: The Logic of Competition and Cooperation (Princeton, N.J.: Princeton University Press, 2010), pp. 28–33; Sean M. Lynn-Jones and Steven E. Miller, preface to Michael E. Brown, Lynn-Jones, and Miller, eds., The Perils of Anarchy: Contemporary Realism and International Security (Cambridge, Mass.: MIT Press, 1995), pp. ix-x; and Nuno P. Monteiro, Theory of Unipolar Politics (New York: Cambridge University Press, 2014), pp. 31–36.

18. 

To clarify, we distinguish between goals—what states want—and intentions, which are plans for achieving goals. See Sebastian Rosato, “The Inscrutable Intentions of Great Powers,” International Security, Vol. 39, No. 3 (Winter 2014/15), pp. 52–53.

19. 

Kenneth E. Boulding, Conflict and Defense: A General Theory (New York: Harper, 1962), pp. 230–231, 245–247. What some call geography, others call geopolitics. See Patrick Porter, The Global Village Myth: Distance, War, and the Limits of Power (Washington, D.C.: Georgetown University Press, 2015), pp. 23–28.

20. 

See, for example, Levy and Thompson, “Balancing on Land and at Sea,” pp. 16–19; Mearsheimer, The Tragedy of Great Power Politics, pp. 83–84; and Daniel H. Deudney, Bounding Power: Republican Security Theory from the Polis to the Global Village (Princeton, N.J.: Princeton University Press, 2007), pp. 42–43, 118–121.

21. 

Waltz, Theory of International Politics, pp. 91, 105–107, 111; and Mearsheimer, The Tragedy of Great Power Politics, p. 33.

22. 

John H. Herz, “Idealist Internationalism and the Security Dilemma,” World Politics, Vol. 2, No. 2 (January 1950), p. 157; and Robert Jervis, “Cooperation under the Security Dilemma,” World Politics, Vol. 30, No. 2 (January 1978), pp. 169–170.

23. 

Thus, we distinguish between power or resources, on the one hand, and military capability or might, on the other. The former are the material for and can be converted into the latter.

24. 

See Hans J. Morgenthau, Politics among Nations: The Struggle for Power and Peace, 4th ed. (New York: Alfred A. Knopf, 1967), pp. 201–202. This is not to say that great powers devote all of their resources to building military capability. We argue that states devote as much as they need to ensure their short-term growth and security without jeopardizing their long-term growth and security. That is, the proportion of resources devoted to military power is limited by a guns versus butter trade-off and an inter-temporal security trade-off.

25. 

On uneven growth, see Robert Gilpin, War and Change in World Politics (New York: Cambridge University Press, 1981), pp. 93–94, 158–159, 178–179, 201–202, 210; and Paul Kennedy, The Rise and Fall of the Great Powers: Economic Change and Military Conflict from 1500 to 2000 (New York: Vintage, 1989), pp. xv-xx, xxii-xxiv.

26. 

Previous neorealist work on balancing privileges gross power rather than military capabilities. See, for example, Waltz, Theory of International Politics, p. 118; and Mearsheimer, The Tragedy of Great Power Politics, p. 157. We privilege military capabilities rather than gross power.

27. 

On arming, see Waltz, Theory of International Politics, p. 118; and Mearsheimer, The Tragedy of Great Power Politics, p. 157.

28. 

On imitation, see Waltz, Theory of International Politics, pp. 124, 127–128; Emily O. Goldman and Richard B. Andres, “Systemic Effects of Military Innovation and Diffusion,” Security Studies, Vol. 8, No. 4 (Summer 1999), pp. 82–83; and Resende-Santos, Neorealism, States, and the Modern Mass Army, pp. 9–13.

29. 

On external balancing, see Waltz, Theory of International Politics, p. 118; and Mearsheimer, The Tragedy of Great Power Politics, p. 156. For a similar definition of alliances, see Glenn H. Snyder, Alliance Politics (Ithaca, N.Y.: Cornell University Press, 1997), p. 4. For a definition that includes looser arrangements, see Walt, The Origins of Alliances, pp. 12–13.

30. 

Waltz, Theory of International Politics, p. 168; and Mearsheimer, The Tragedy of Great Power Politics, pp. 156–157. Polarity plays a limited role in our theory, dictating the options available to great powers: internal and external balancing are both possible in multipolarity, whereas, by definition, only internal balancing is possible in bipolarity.

31. 

For more expansive definitions of threat, see Thomas J. Christensen, Useful Adversaries: Grand Strategy, Domestic Mobilization, and Sino-American Conflict, 1947–1958 (Princeton, N.J.: Princeton University Press, 1996), pp. 27–28; and Walt, The Origins of Alliances, pp. 25–26.

32. 

Waltz, Theory of International Politics, p. 165; and Mearsheimer, The Tragedy of Great Power Politics, p. 33. For an exploration of why states have reason to doubt great power allies, see Michael Beckley, “The Myth of Entangling Alliances: Reassessing the Security Risks of U.S. Defense Pacts,” International Security, Vol. 39, No. 4 (Spring 2015), pp. 17–22.

33. 

Waltz, Theory of International Politics, p. 167.

34. 

This may explain why alliances designed for deterrence do not reliably deter their targets. See Brett V. Benson, “Unpacking Alliances: Deterrent and Compellent Alliances and Their Relationship with Conflict,” Journal of Politics, Vol. 73, No. 4 (October 2011), pp. 1111–1127.

35. 

On bandwagoning, see Waltz, Theory of International Politics, p. 126; Walt, The Origins of Alliances, pp. 17, 19–21; and Mearsheimer, The Tragedy of Great Power Politics, pp. 162–163.

36. 

Randall L. Schweller, “Bandwagoning for Profit: Bringing the Revisionist State Back In,” International Security, Vol. 19, No. 1 (Summer 1994), pp. 72–107.

37. 

Waltz, Theory of International Politics, p. 126; and Mearsheimer, The Tragedy of Great Power Politics, p. 163.

38. 

On buck-passing, see Mearsheimer, The Tragedy of Great Power Politics, pp. 157–159; Barry R. Posen, The Sources of Military Doctrine (Ithaca, N.Y.: Cornell University Press, 1984), pp. 63, 74, 232; and Waltz, Theory of International Politics, pp. 164–165.

39. 

Mearsheimer, The Tragedy of Great Power Politics, pp. 158–159.

40. 

On this point, see also Mearsheimer, The Tragedy of Great Power Politics, pp. 44, 271–272; and Walt, The Origins of Alliances, pp. 23–24. For the claim that, by incorporating geography into his theory, Mearsheimer is able to say more about state behavior than Waltz, see Richard Little, The Balance of Power in International Relations: Metaphors, Myths, and Models (New York: Cambridge University Press, 2007), pp. 238–239. Note that geography does not appear in or follow from Mearsheimer's assumptions.

41. 

On this point, see also Huntington, “Arms Races,” p. 52.

42. 

For a similar argument, see Mearsheimer, The Tragedy of Great Power Politics, pp. 267–269, 271–272.

43. 

The idea originates with Waltz, Theory of International Politics, p. 107. See also Jeffrey W. Taliaferro, “Security Seeking under Anarchy: Defensive Realism Revisited,” International Security, Vol. 25, No. 3 (Winter 2000/01), p. 133.

44. 

On this point, see also Robert Jervis, “Security Regimes,” in Stephen D. Krasner, ed., International Regimes (Ithaca, N.Y.: Cornell University Press, 1983), p. 185.

45. 

See, for example, Randall L. Schweller, Deadly Imbalances: Tripolarity and Hitler's Strategy of World Conquest (New York: Columbia University Press, 1998), pp. 3, 6; and Brian Rathbun, “A Rose by Any Other Name: Neoclassical Realism as the Logical and Necessary Extension of Structural Realism,” Security Studies, Vol. 17, No. 2 (April 2008), pp. 304–307.

46. 

Taliaferro, Lobell, and Ripsman, “Introduction,” p. 4.

47. 

The shadow of violence metaphor is associated with Waltz, Theory of International Politics, p. 102. Nevertheless, Waltz maintains that neorealism cannot be a theory of state behavior. See ibid., pp. 121–122; and Kenneth N. Waltz, “International Politics Is Not Foreign Policy,” Security Studies, Vol. 6, No. 1 (Autumn 1996), pp. 54–57.

48. 

Because our definition of a great power—a state that has resources on a par with the most powerful states in the world—is similar to theirs, this coding follows Mearsheimer, The Tragedy of Great Power Politics, p. 414; and Waltz, Theory of International Politics, p. 162. For lists that use more expansive criteria to identify great powers, see Vesna Danilovic, When Stakes Are High: Deterrence and Conflict among Major Powers (Ann Arbor: University of Michigan Press, 2002), p. 46; and Jack S. Levy, War in the Modern Great Power System, 1495–1975 (Lexington: University Press of Kentucky, 1983), p. 47.

49. 

See, for example, William C. Wohlforth, “The Stability of a Unipolar World,” International Security, Vol. 24, No. 1 (Summer 1999), pp. 5–41.

50. 

For a sea power versus land power distinction that relies on naval capabilities rather than geography, see Levy and Thompson, “Balancing on Land and at Sea,” pp. 25–26.

51. 

This procedure allows us to classify the great powers as European land powers, Asian land powers, and/or sea powers. Austria-Hungary, Britain, France, Italy, Prussia (Germany), Russia (Soviet Union), and the United States (1946–90) are European land powers. Japan, Russia (Soviet Union), and the United States (1946–90) are Asian land powers. Britain, Japan, and the United States are sea powers. Note that Britain and Japan are coded as both sea and land powers and are therefore compared to both sea powers and the relevant land powers in our analysis.

52. 

On this point, see James McAllister, No Exit: America and the German Problem, 1943–1954 (Ithaca, N.Y.: Cornell University Press, 2002), pp. 251–252; and Gregory Mitrovich, Undermining the Kremlin: America's Strategy to Subvert the Soviet Bloc, 1947–1956 (Ithaca, N.Y.: Cornell University Press, 2009), pp. 1–2, 7–8.

53. 

All data referred to in this article can be accessed via http://dx.doi.org/10.7910/DVN/LDDARY.

54. 

Waltz makes a similar point: “A self-help system is one in which those who do not help themselves … will suffer. Fear of such unwanted consequences stimulates states to behave in ways that tend toward the creation of balances of power.” See Waltz, Theory of International Politics, p. 118.

55. 

Although Levy and Thompson focus on sea powers in their analysis of balancing, they do not test the claims listed here. See Levy and Thompson, “Balancing on Land and at Sea,” pp. 19–23.

56. 

Correlates of War (COW), “National Material Capabilities Data (v. 4.0),” www.correlatesofwar.org. The dataset is described in J. David Singer, “Reconstructing the Correlates of War Dataset on Material Capabilities of States, 1816–1985,” International Interactions, Vol. 14, No. 2 (April 1988), pp. 115–132.

57. 

Compare, for example, The Statesman's Year-Book (London: Macmillan, various years); and COW, “National Material Capabilities Data (v4.0).”

58. 

For examples of this kind of reasoning, see Michael C. Desch, Power and Military Effectiveness: The Fallacy of Democratic Triumphalism (Baltimore, Md.: Johns Hopkins University Press, 2008), pp. 31–37; John J. Mearsheimer, “Numbers, Strategy, and the European Balance,” International Security, Vol. 12, No. 4 (Spring 1988), pp. 175–177; and Schweller, Deadly Imbalances, pp. 17–19. Military analysts have long argued that attackers require a 3:1 local advantage over defenders to prevail. See John J. Mearsheimer, “Assessing the Conventional Balance: The 3:1 Rule and Its Critics,” International Security, Vol. 13, No. 4 (Spring 1989), pp. 56–65.

59. 

Critics fault neorealism for failing to explain state behavior “over the short to medium term.” See Gideon Rose “Neoclassical Realism and Theories of Foreign Policy,” World Politics, Vol. 51, No. 1 (October 1998), p. 147. Although these scholars do not indicate what they mean by short to medium term, many cite the alleged failure of the other great powers to balance against and ultimately to deter Nazi Germany as an exemplar of neorealism's explanatory weakness. See, for example, Norrin M. Ripsman, Jeffrey W. Taliaferro, and Steven E. Lobell, “Conclusion: The State of Neoclassical Realism,” in Lobell, Ripsman, and Taliaferro, Neoclassical Realism, the State, and Foreign Policy, p. 284; and Schweller, Unanswered Threats, p. 2. This implies that the short to medium term is less than seven years: Adolf Hitler came to power in 1933, and Germany attacked France in 1940. Accordingly, we define short to medium term as less than five years and code states that act within this time frame as doing so promptly.

60. 

Given data limitations and the imperfect relationship between our measure of military capability and actual military capabilities, we avoid false precision and use whole number ratios (1:1, 2:1, 3:1). Our substantive findings remain the same regardless of procedure.

61. 

The analysis in this section excludes Germany from 1919 to 1933, when its army was artificially capped at 100,000 men.

62. 

Of the 1,651 cases, the capability ratio was 1:1 in 861 cases (52%), 2:1 in 533 cases (32%), and 3:1 or greater in 257 cases (16%). Defining “prompt” as three years yields a total of 1,699 cases. Of these, the capability ratio was 1:1 in 772 cases (45%), 2:1 in 595 cases (35%), and 3:1 or greater in 332 cases (20%).

63. 

For the data, see Brian R. Mitchell, International Historical Statistics: Europe, 1750–1993 (New York: Macmillan, 1998), pp. 12–47; Brian R. Mitchell, International Historical Statistics: Africa, Asia, and Oceania (New York: Macmillan, 1998), pp. 19–27, 31; Brian R. Mitchell, International Historical Statistics: The Americas, 1750–2000 (New York: Palgrave Macmillan, 2003), pp. 11–23, 31; John M. Hobson, “The Military-Extraction Gap and the Wary Titan: The Fiscal Sociology of British Defence, 1870–1913,” Journal of European Economic History, Vol. 22, No. 3 (Winter 1993), pp. 478–479; Jari Eloranta, “Why Did the League of Nations Fail?” Cliometrica, Vol. 5, No. 1 (January 2011), p. 35; Robert J. Lieber, Power and Willpower in the American Future: Why the United States Is Not Destined to Decline (New York: Cambridge University Press, 2012), pp. 123–125; and Noel E. Firth and James H. Noren, Soviet Defense Spending: A History of CIA Estimates, 1950–1990 (College Station: Texas A&M University Press, 1998), pp. 129–130.

64. 

Our analysis of effort begins in 1885, because this is the first year in which the relevant data are available for all the great powers.

65. 

See, for example, Aaron L. Friedberg, In the Shadow of the Garrison State: America's Anti-Statism and Its Cold War Grand Strategy (Princeton, N.J.: Princeton University Press, 2000), pp. 3–8.

66. 

Other theories predict imitation, but only neorealism claims that it will be prompt and uniform. See Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 82–98; and Michael C. Horowitz, The Diffusion of Military Power: Causes and Consequences for International Politics (Princeton, N.J.: Princeton University Press, 2010), pp. 20–22, 30–60.

67. 

The sources are Max Boot, War Made New: Technology, Warfare, and the Course of History, 1500 to Today (New York: Gotham, 2006); Eliot A. Cohen, “A Revolution in Warfare,” Foreign Affairs, Vol. 75, No. 2 (March/April 1996), pp. 37–54; Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 98–101; Barton C. Hacker, “The Machines of War: Western Military Technology 1850–2000,” History and Technology, Vol. 21, No. 3 (September 2005), pp. 255–300; Horowitz, The Diffusion of Military Power, p. 61; Andrew F. Krepinevich, “Cavalry to Computer: The Pattern of Military Revolutions,” National Interest, Fall 1994, pp. 30–42; Williamson Murray, “Thinking about Revolutions in Military Affairs,” Joint Forces Quarterly, Summer 1997, pp. 69–76; and Alex Roland, “Technology and War,” paper presented at the “Summary Conference on the Study of War,” Triangle Institute for Security Studies, June 1997, http://www.unc.edu/depts/diplomat/AD_Issues/amdipl_4/roland.html.

68. 

On demonstration points, see Horowitz, The Diffusion of Military Power, p. 24.

69. 

Boot, War Made New, pp. 124–130, 196; Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 116–117; Geoffrey L. Herrera and Thomas G. Mahnken, “Military Diffusion in Nineteenth Century Europe: The Napoleonic and Prussian Military Systems,” in Emily O. Goldman and Leslie C. Eliason, eds., The Diffusion of Military Technology and Ideas (Stanford, Calif.: Stanford University Press, 2003), pp. 217–225; Krepinevich, “Cavalry to Computer,” pp. 34–35; and Robert L. O'Connell, Of Arms and Men: A History of War, Weapons, and Aggression (New York: Oxford University Press, 1989), pp. 202–204.

70. 

See, for example, Waltz, Theory of International Politics, p. 127.

71. 

Boot, War Made New, pp. 116–145; Bernard Brodie and Fawn M. Brodie, From Crossbow to H-Bomb (New York: Dell, 1962), pp. 137–138; Herrera and Mahnken, “Military Diffusion in Nineteenth Century Europe,” pp. 221–222; William H. McNeill, The Pursuit of Power: Technology, Armed Force, and Society since A.D. 1000 (Chicago: University of Chicago Press, 1982), p. 250; and O'Connell, Of Arms and Men, pp. 204, 207. For the claim that other European states did not learn the lessons of the Austro-Prussian War, see Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” p. 116.

72. 

Herrera and Mahnken, “Military Diffusion in Nineteenth Century Europe,” pp. 229–232; and Martin van Creveld, Technology and War: From 2000 b.c. to the Present (New York: Free Press, 1989), pp. 158–159.

73. 

Herrera and Mahnken, “Military Diffusion in Nineteenth Century Europe,” pp. 234–235.

74. 

Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 116–117; and Geoffrey L. Herrera, “Inventing the Railroad and Rifle Revolution: Information, Military Innovation, and the Rise of Germany,” Journal of Strategic Studies, Vol. 27, No. 2 (June 2004), p. 261.

75. 

Brodie and Brodie, From Crossbow to H-Bomb, p. 138; and Geoffrey Wawro, The Austro-Prussian War: Austria's War with Prussia and Italy in 1866 (New York: Cambridge University Press, 1997), p. 285.

76. 

Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 116–117; and Herrera and Mahnken, “Military Diffusion in Nineteenth Century Europe,” p. 228.

77. 

Herrera and Mahnken, “Military Diffusion in Nineteenth Century Europe,” pp. 236–238.

78. 

Horowitz, The Diffusion of Military Power, p. 138. See also Robert J. Art, The Influence of Foreign Policy on Seapower: New Weapons and Weltpolitik in Wilhelminian Germany (London: Sage, 1973), pp. 32, 39; and George Modelski and William R. Thompson, Seapower in Global Politics, 1494–1993 (Seattle: University of Washington Press, 1988), pp. 75–76.

79. 

David C. Evans and Mark R. Peattie, Kaigun: Strategy, Tactics, and Technology in the Imperial Japanese Navy, 1887–1941 (Annapolis: Naval Institute Press, 1997), p. 152. See also Horowitz, The Diffusion of Military Power, pp. 138–139. Although wireless telegraphy is sometimes mentioned as a technological component of the battlefleet innovation, it had a negligible effect. See Boot, War Made New, p. 197.

80. 

Horowitz, The Diffusion of Military Power, pp. 144–145.

81. 

The Admiralty went to great lengths to hide the capabilities of the Dreadnought, which was not completed until the end of the year. Therefore, February 1906 is the earliest possible demonstration point. See Horowitz, The Diffusion of Military Power, pp. 138, 142–143, 150–151.

82. 

Siegfried Breyer, Schlachtschiffe und Schlachtkreuzer, 1905–1970 (Munich: J.F. Lehmanns, 1970), pp. 208, 350. David Evans and Mark Peattie claim that Japan did not lay down its first true dreadnoughts until 1909. See Evans and Peattie, Kaigun, pp. 159–160. Conversely, O'Connell claims: “The Japanese … responded almost immediately to the Dreadnought with their own Satsuma.” See O'Connell, Of Arms and Men, p. 228.

83. 

Breyer, Schlachtschiffe und Schlachtkreuzer, pp. 276, 393, 412, 431, 438. O'Connell claims that all the great powers had laid down a dreadnought by 1910. See O'Connell, Of Arms and Men, p. 228.

84. 

Boot, War Made New, pp. 220–224; Cohen, “A Revolution in Warfare,” pp. 46–47; Thomas G. Mahnken, “Beyond Blitzkrieg: Allied Responses to Combined-Arms Armored Warfare during World War II,” in Goldman and Eliason, The Diffusion of Military Technology and Ideas, pp. 243–253; and van Creveld, Technology and War, pp. 179–180.

85. 

Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” p. 120; and Mahnken, “Beyond Blitzkrieg,” pp. 244, 253.

86. 

Boot, War Made New, pp. 237–238; Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” p. 121; and Mahnken, “Beyond Blitzkrieg,” pp. 253–266. On the Soviet Union, see also David M. Glantz and Jonathan House, When Titans Clashed: How the Red Army Stopped Hitler (Lawrence: University Press of Kansas, 1995), pp. 286–289; and Richard Overy, “Total War II: The Second World War,” in Charles Townshend, ed., The Oxford Illustrated History of Modern War (New York: Oxford University Press, 1997), pp. 127–128. On the United States, see also David E. Johnson, Fast Tanks and Heavy Bombers: Innovation in the U.S. Army, 1917–1945 (Ithaca, N.Y.: Cornell University Press, 1998), p. 226.

87. 

Samuel W. Mitcham Jr., Blitzkrieg No Longer: The German Wehrmacht in Battle, 1943 (Mechanicsburg, Pa.: Stackpole, 2010), p. 262; and van Creveld, Technology and War, p. 180. For the claim that none of the combatants were able to implement blitzkrieg in Italy, see Alan Axelrod, Encyclopedia of World War II, Vol. 2 (New York: Facts on File, 2007), p. 458.

88. 

Krepinevich, “Cavalry to Computer,” p. 38.

89. 

Cohen, “A Revolution in Warfare,” p. 44; Emily O. Goldman, “Receptivity to Revolution: Carrier Air Power in Peace and War,” in Goldman and Eliason, The Diffusion of Military Technology and Ideas, pp. 272–274; and Horowitz, The Diffusion of Military Power, pp. 67–70, 77, 84.

90. 

Boot, War Made New, p. 267; Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” p. 120; Goldman, “Receptivity to Revolution,” pp. 267, 273; and Horowitz, The Diffusion of Military Power, p. 69.

91. 

Goldman, “Receptivity to Revolution,” pp. 281–288; and Horowitz, The Diffusion of Military Power, pp. 65, 70.

92. 

Goldman, “Receptivity to Revolution,” pp. 276–281, 288, 301; and Horowitz, The Diffusion of Military Power, pp. 77, 86–87.

93. 

On Germany, see Goldman, “Receptivity to Revolution,” pp. 293–297.

94. 

Boot, War Made New, pp. 278–279; Stephen Peter Rosen, Winning the Next War: Innovation and the Modern Military (Ithaca, N.Y.: Cornell University Press, 1994), pp. 156–163; and Overy, “Total War II,” p. 131. Robert A. Pape argues that the bombing of Germany's industry and transportation network “hardly mattered” and did not leave “the German army less able to resist Allied offensives.” Nevertheless, he admits that German armament production was halved between July 1944 and March 1945 and that the fuel situation was “more serious.” In addition, bombing of the transportation network meant that Germany was unable “to put weapons and supplies in the hands of the divisions actually fighting.” See Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996), pp. 276–283, at pp. 281, 282, 279.

95. 

William T. Lee and Richard F. Staar, Soviet Military Policy since World War II (Stanford, Calif.: Hoover Institution Press, 1986), pp. 10, 12–15; and Geoffrey Roberts, Stalin's Wars: From World War to Cold War, 1939–1953 (New Haven, Conn.: Yale University Press, 2006), p. 363.

96. 

Horowitz, The Diffusion of Military Power, pp. 99–102; and Murray, “Thinking about Revolutions in Military Affairs,” p. 70.

97. 

Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” p. 121; Horowitz, The Diffusion of Military Power, pp. 102, 112, 125; and William C. Potter, “The Diffusion of Nuclear Weapons,” in Goldman and Eliason, The Diffusion of Military Technology and Ideas, p. 162.

98. 

Huntington, “Arms Races,” p. 73.

99. 

The candidate imitators and time to imitation are as follows: for the Prussian system, France (2 years), Russia (partial), Italy (partial), and Austria-Hungary (failure); for battlefleet warfare, the United States (1 year) and Japan (1 year); for blitzkrieg, the Soviet Union (3 years), the United States (4 years), Britain (4 years), and Italy (not applicable); for carrier warfare, Japan (0 years) and Britain (3 years); for strategic bombing, the Soviet Union (3 years); and for nuclear weapons, the Soviet Union (4 years). Although the United States and Britain imitated blitzkrieg within 4 years of its demonstration point, they did so within a year of landing in Europe and becoming continental sea powers. Hence our theory performs almost as well using a three-year window as a five-year window (the only additional anomaly being the Soviet acquisition of atomic weapons).

100. 

Levy and Thompson claim that it is difficult to distinguish internal balancing against threats from other sources of arms buildups, such as bureaucratic politics or the vested interests of the military. See Levy and Thompson, “Hegemonic Threats and Great Power Balancing in Europe, 1495–1999,” p. 14. Our case study attempts to address this problem.

101. 

We selected these cases for examination because they are well predicted by the foregoing quantitative analysis. On this approach, see Evan S. Lieberman, “Nested Analysis as a Mixed-Method Strategy for Comparative Research,” American Political Science Review, Vol. 99, No. 3 (August 2005), pp. 435–452.

102. 

Quoted in E. Malcolm Carroll, French Public Opinion and Foreign Affairs, 1870–1914 (New York: Century, 1931), p. 49. For similar statements by other prominent figures, see Richard D. Challener, The French Theory of the Nation in Arms, 1866–1939 (New York: Russell and Russell, 1965), pp. 20, 32; David B. Ralston, The Army of the Republic: The Place of the Military in the Political Evolution of France, 1871–1914 (Cambridge, Mass.: MIT Press, 1967), p. 36; Paul-Marie de La Gorce, The French Army: A Military-Political History, trans. Kenneth Douglas (New York: George Braziller, 1983), p. 8; and Douglas Porch, The March to the Marne: The French Army, 1871–1914 (New York: Cambridge University Press, 1981), p. 10.

103. 

Jean-Marie Mayeur and Madeleine Rebérioux, The Third Republic from Its Origins to the Great War, 1781–1914, trans. J.R. Foster (New York: Cambridge University Press, 1984), p. 15.

104. 

Quoted in Carroll, French Public Opinion and Foreign Affairs, 1870–1914, p. 74.

105. 

Allan Mitchell, Victors and Vanquished: The German Influence on Army and Church in France after 1870 (Chapel Hill: University of North Carolina Press, 1984), pp. 105–109; and Porch, The March to the Marne, pp. 24–25.

106. 

Robert F. Baumann, “Universal Service Reform: Conception to Implementation, 1873–1883,” in David Schimmelpenninck van der Oye and Bruce W. Menning, eds., Reforming the Tsar's Army: Military Innovation in Imperial Russia from Peter the Great to the Revolution (New York: Cambridge University Press, 2004), p. 12; W. Bruce Lincoln, The Great Reforms: Autocracy, Bureaucracy, and the Politics of Change in Imperial Russia (DeKalb: Northern Illinois University Press, 1990), p. 156; and Alexander Polunov, Russia in the Nineteenth Century: Autocracy, Reform, and Social Change, 1814–1914 (New York: M.E. Sharpe, 2005), p. 158.

107. 

Robert F. Baumann, “The Russian Army, 1853–1881,” in Frederick W. Kagan and Robin Higham, eds., The Military History of Tsarist Russia (New York: Palgrave, 2002), pp. 143–147. On Russia's threat perception of Germany, see David Alan Rich, The Tsar's Colonels: Professionalism, Strategy, and Subversion in Late Imperial Russia (Cambridge, Mass.: Harvard University Press, 1998), p. 5; and William C. Fuller Jr., Strategy and Power in Russia, 1600–1914 (New York: Free Press, 1992), p. 384.

108. 

Alex Marshall, The Russian General Staff and Asia, 1800–1917 (New York: Routledge, 2006), pp. 68, 95.

109. 

Jacob W. Kipp, “The Imperial Russian Navy, 1696–1900: The Ambiguous Legacy of Peter's ‘Second Arm,'” in Kagan and Higham, The Military History of Tsarist Russia, p. 170.

110. 

Quoted in Ariel Cohen, “Russia's Military on the March in Asia,” National Interest, July 25, 2013, http://nationalinterest.org/commentary/russias-military-the-march-asia-8772.

111. 

Hajo Holborn, A History of Modern Germany, Vol. 3: 1840–1945 (Princeton, N.J.: Princeton University Press, 1982), p. 251; and Gordon A. Craig, The Politics of the Prussian Army, 1640–1945 (New York: Oxford University Press, 1964), pp. 217–222.

112. 

Martin Kitchen, A Military History of Germany: From the Eighteenth Century to the Present Day (Bloomington: Indiana University Press, 1975), p. 140.

113. 

Volker R. Berghahn, Imperial Germany, 1871–1914: Economy, Society, Culture, and Politics (Providence: Berghahn, 1994), p. 265.

114. 

Holborn, A History of Modern Germany, Vol. 3, p. 251. See also Kitchen, A Military History of Germany, p. 150.

115. 

A.J.P. Taylor, Bismarck: The Man and the Statesman (Gloucester, U.K.: Sutton, 2003), p. 229.

116. 

David Stevenson, Armaments and the Coming of War: Europe, 1904–1914 (New York: Oxford University Press, 1996), p. 57.

117. 

Freycinet, quoted in Jack Snyder, The Ideology of the Offensive: Military Decision Making and the Disasters of 1914 (Ithaca, N.Y.: Cornell University Press, 1984), p. 55; and Miribel, quoted in Allan Mitchell, “The Freycinet Reforms and the French Army, 1883–1893,” Journal of Strategic Studies, Vol. 4, No. 1 (February 1981), p. 21. Even one of Freycinet's most tenacious critics, Capt. Georges Gilbert, agreed that the criterion must be “the 60,000 [soldiers] we lack in order to be on a par with Germany.” See Mitchell, “The Freycinet Reforms and the French Army, 1883–1893,” p. 21.

118. 

For statistics, see Challener, The French Theory of the Nation in Arms, 1866–1939, p. 60. On French conscription laws, see Eugen Weber, Peasants into Frenchmen: The Modernization of Rural France, 1870–1914 (Stanford, Calif.: Stanford University Press, 1976), pp. 293–294, 298.

119. 

Holborn, A History of Modern Germany, Vol. 3, p. 320.

120. 

Quoted in Huntington, “Arms Races,” p. 46 n. 8. See also ibid., pp. 65–66; and Craig, The Politics of the Prussian Army, 1640–1945, p. 243. The Germans did, however, fall behind in field guns. France deployed a new 75-millimeter field gun in 1897, and Germany did not respond until 1905, mainly because of bureaucratic politics. See David G. Herrmann, The Arming of Europe and the Making of the First World War (Princeton, N.J.: Princeton University Press, 1996), pp. 17–19; and Robert M. Ripperger, “The Development of the French Artillery for the Offensive, 1890–1914,” Journal of Military History, Vol. 59, No. 4 (October 1995), pp. 599–601.

121. 

George F. Kennan, The Fateful Alliance: France, Russia, and the Coming of the First World War (New York: Pantheon, 1984), pp. 235, 250, 253; and Georges Michon, The Franco-Russian Alliance, 1891–1917, trans. Norman Thomas (London: Allen and Unwin, 1929), p. 68.

122. 

Quoted in Fuller, Strategy and Power in Russia, 1600–1914, p. 382. See also ibid., p. 395.

123. 

John Bushnell, “Peasants in Uniform: The Tsarist Army as a Peasant Society,” Journal of Social History, Vol. 13, No. 4 (Summer 1980), p. 565; and Bruce W. Menning, Bayonets before Bullets: The Imperial Russian Army, 1861–1914 (Bloomington: Indiana University Press, 1992), p. 226.

124. 

John W. Steinberg, All the Tsar's Men: Russia's General Staff and the Fate of the Empire, 1898–1914 (Baltimore, Md.: Johns Hopkins University Press, 2010), pp. 154, 186–190; Menning, Bayonets before Bullets, pp. 221–232; Bruce W. Menning, “Mukden to Tannenberg: Defeat to Defeat, 1905–1914,” in Kagan and Higham, The Military History of Tsarist Russia, pp. 215–217; Bruce W. Menning and John W. Steinberg, “Lessons Learned: The Near-Term Military Legacy of 1904–05 in Imperial Russia,” in Steven Ericson and Allen Hockley, eds., The Treaty of Portsmouth and Its Legacies (Hanover, N.H.: Dartmouth College Press, 2008), p. 78; W. Thomas Wilfong, “Rebuilding the Russian Army, 1905–14: The Question of a Comprehensive Plan for National Defense,” Ph.D. dissertation, Indiana University, 1977, pp. 69, 71, 73, 76–77, 85–87, 151–153, 159, 165; Fuller, Strategy and Power in Russia, 1600–1914, p. 410; William C. Fuller, Civil-Military Conflict in Imperial Russia, 1881–1914 (Princeton, N.J.: Princeton University Press, 1985), pp. 194–195; and Walter M. Pintner, “Russian Military Thought: The Western Model and the Shadow of Suvorov,” in Peter Paret, ed., Makers of Modern Strategy: From Machiavelli to the Nuclear Age (Princeton, N.J.: Princeton University Press, 1986), p. 364.

125. 

Quoted in Kitchen, A Military History of Germany, p. 186.

126. 

Quoted in Herrmann, The Arming of Europe and the Making of the First World War, p. 181.

127. 

Quoted in ibid., p. 183. See also Stevenson, Armaments and the Coming of War, pp. 210, 224, 227.

128. 

Quoted in Norman Stone, The Eastern Front 1914–1917 (New York: Penguin, 1998), p. 37. Moltke went on, “In two or three years Russia would have finished arming. Our enemies’ military power would then be so great that he did not know how he could deal with it. Now we were still more or less a match for it.” Quoted in Dominic Lieven, The Aristocracy in Europe, 1815–1914 (New York: Columbia University Press, 1992), pp. 226–227.

129. 

Herrmann, The Arming of Europe and the Making of the First World War, p. 191; and Eugen Weber, The Nationalist Revival in France, 1905–1914 (Berkeley: University of California Press, 1959), chap. 12.

130. 

Holborn, A History of Modern Germany, Vol. 3, p. 343; Herrmann, The Arming of Europe and the Making of the First World War, p. 185; and Volker R. Berghahn, Germany and the Approach of War in 1914, 2nd ed. (New York: St. Martin's, 1993), pp. 16, 125.

131. 

A.J.P. Taylor, The Struggle for Mastery in Europe, 1848–1918 (New York: Oxford University Press, 1977), pp. 500–501; and Holborn, A History of Modern Germany, Vol. 3, p. 342.

132. 

Stevenson, Armaments and the Coming of War, p. 279.

133. 

Fuller, Civil-Military Conflict in Imperial Russia, 1881–1914, p. 227; Wilfong, “Rebuilding the Russian Army, 1905–14,” p. 130; Peter Gatrell, Government, Industry, and Rearmament in Russia, 1900–1914: The Last Argument of Tsarism (New York: Cambridge University Press, 1994), pp. 133–134; and Stevenson, Armaments and the Coming of War, pp. 2, 145.

134. 

Herrmann, The Arming of Europe and the Making of the First World War, p. 220.

135. 

Quoted in Arnold Wolfers, Britain and France between Two Wars: Conflicting Strategies of Peace since Versailles (Hamden, Conn.: Archon, 1963), p. 33. The translation from French is ours.

136. 

Quoted in Philippe Bernard and Henri Dubief, The Decline of the Third Republic, 1914–1938 (New York: Cambridge University Press, 1985), p. 146. For similar statements at the time, see Richard D. Challener, “The French Foreign Office: The Era of Philippe Berthelot,” in Gordon A. Craig and Felix Gilbert, eds., The Diplomats, 1919–1939 (Princeton, N.J.: Princeton University Press, 1953), p. 77; Peter Jackson, Beyond the Balance of Power: France and the Politics of National Security in the Era of the First World War (New York: Cambridge University Press, 2013), p. 425; and Anthony P. Adamthwaite, The Lost Peace: International Relations in Europe, 1918–1939 (New York: Palgrave Macmillan, 1981), p. 80.

137. 

Jackson, Beyond the Balance of Power, pp. 6–7, 514. See also Patrick O. Cohrs, The Unfinished Peace after World War I: America, Britain, and the Stabilisation of Europe, 1919–1932 (New York: Cambridge University Press, 2006), p. 49; and Zara Steiner, The Lights That Failed: European International History 1919–1933 (New York: Oxford University Press, 2005), pp. 458, 469.

138. 

Robert Boyce, The Great Interwar Crisis and the Collapse of Globalization (New York: Palgrave Macmillan, 2009), pp. 124–125; and Steiner, The Lights That Failed, p. 232.

139. 

Mearsheimer, The Tragedy of Great Power Politics, pp. 317–319; and Williamson Murray, The Change in the European Balance of Power, 1938–1939: The Path to Ruin (Princeton, N.J.: Princeton University Press, 1984), pp. 20–21. On French defensive measures, see J.E. Kaufmann and H.W. Kaufmann, The Maginot Line: None Shall Pass (Westport, Conn.: Praeger, 1997), pp. 12–19; and Martin S. Alexander, “In Defence of the Maginot Line: Security Policy, Domestic Politics, and the Economic Depression in France,” in Robert Boyce, ed., French Foreign and Defence Policy, 1918–1940: The Decline and Fall of a Great Power (New York: Routledge, 1998), pp. 164–194. For the claim that the French borrowed much of the design for the Maginot forts from German fortifications, see Anthony Kemp, The Maginot Line: Myth and Reality (New York: Military Heritage, 1982), p. 27.

140. 

Ernest R. May, Strange Victory: Hitler's Conquest of France (New York: Hill and Wang, 2000), pp. 7, 455, at p. 7. According to Robert Allan Doughty, “French forces were a model of preparedness and modernization.” See Doughty, The Seeds of Disaster: The Development of French Army Doctrine, 1919–1939 (Hamden, Conn.: Archon, 1985), p. 2.

141. 

Quoted in Anthony Adamthwaite, France and the Coming of the Second World War, 1936–1939 (Totowa, N.J.: Frank Cass, 1977), p. xv. See also Robert J. Young, In Command of France: French Foreign Policy and Military Planning, 1933–1940 (Cambridge, Mass.: Harvard University Press, 1978), pp. 221–226.

142. 

Quoted in Adamthwaite, France and the Coming of the Second World War, 1936–1939, p. 312.

143. 

Quoted in Isaac Deutscher, Stalin: A Political Biography (New York: Oxford University Press, 1949), p. 328.

144. 

Theda Skocpol, States and Social Revolutions: A Comparative Analysis of France, Russia, and China (New York: Cambridge University Press, 1979), pp. 223–224.

145. 

Quoted in Alex Weisiger, Logics of War: Explanations for Limited and Unlimited Conflicts (Ithaca, N.Y.: Cornell University Press, 2013), p. 115.

146. 

Roberts, Stalin's Wars, p. 16; and Weisiger, Logics of War, p. 115.

147. 

Gordon A. Craig, Germany, 1866–1945 (New York: Oxford University Press, 1978), p. 697.

148. 

Craig, The Politics of the Prussian Army, 1640–1945, pp. 382–426.

149. 

Quoted in Weisiger, Logics of War, p. 115.

150. 

Dale C. Copeland, The Origins of Major War (Ithaca, N.Y.: Cornell University Press, 2000), pp. 118–145; and Weisiger, Logics of War, pp. 105–122.

151. 

Michael Howard, “The Armed Forces,” in F.H. Hinsley, ed., The New Cambridge Modern History, Vol. 11: Material Progress and World-Wide Problems (New York: Cambridge University Press, 1967), p. 211; Mitchell, Victors and Vanquished, pp. 47–48; Porch, The March to the Marne, pp. 24–25; and Allan Mitchell, “‘A Situation of Inferiority’: French Military Reorganization after the Defeat of 1870,” American Historical Review, Vol. 86, No. 1 (February 1981), p. 50. We omit Germany because it cannot imitate its own system, but it certainly improved by copying others. French rifles were so effective during the Franco-Prussian War that the Germans distributed 600,000 captured Chassepots to their troops immediately and crafted a new German model in 1871. On the rifle race, see Rachel Chrastil, Organizing for War: France 1870–1914 (Baton Rouge: Louisiana State University Press, 2010), p. 4.

152. 

Quoted in Ralston, The Army of the Republic, p. 50.

153. 

J.F.V. Keiger, France and the World since 1870 (London: Arnold, 2001), p. 52; Ralston, The Army of the Republic, p. 87; Stefan T. Possony and Etienne Mantoux, “Du Picq and Foch: The French School,” in Edward Mead Earle, ed., Makers of Modern Strategy: Military Thought from Machiavelli to Hitler (New York: Atheneum, 1967), p. 218; Porch, The March to the Marne, pp. 39–43; Meredith Perry Gilpatrick, “Military Strategy on the Western Front from 1871 to 1914,” Ph.D. dissertation, University of Chicago, 1957, p. 166; and Mitchell, “‘A Situation of Inferiority,’” p. 59.

154. 

Quoted in Mitchell, “The Freycinet Reforms and the French Army, 1883–1893,” p. 20. See also Mitchell, Victors and Vanquished, pp. 105–109.

155. 

Quoted in Mitchell, Victors and Vanquished, p. 67. On French artillery, see also Ralston, The Army of the Republic, p. 88; Stevenson, Armaments and the Coming of War, pp. 176–177, 217; and Porch, The March to the Marne, chap. 12.

156. 

David Stevenson, “War by Timetable? The Railway Race before 1914,” Past & Present, February 1999, pp. 168–169, 175–176. See also Allan Mitchell, “Private Enterprise or Public Service? The Eastern Railway Company and the French State in the Nineteenth Century,” Journal of Modern History, Vol. 69, No. 1 (March 1997), pp. 18–41.

157. 

Rich, The Tsar's Colonels, pp. 4, 84, 89, 95–96, 218–219.

158. 

Baumann, “The Russian Army, 1853–1881,” p. 142.

159. 

Menning, Bayonets before Bullets, pp. 24—29, 113–116; and Felix Patrikeeff and Harold Shukman, Railways and the Russo-Japanese War: Transporting War (New York: Routledge, 2007), p. 2. On the increases in reserves, see Menning, Bayonets before Bullets, pp. 29, 100, 108–109, 112; and John L.H. Keep, Soldiers of the Tsar: Army and Society in Russia 1462–1874 (New York: Oxford University Press, 1985), pp. 376, 378.

160. 

Menning, Bayonets before Bullets, pp. 104, 107; Fuller, Strategy and Power in Russia, 1600–1914, p. 283; and Matitiahu Mayzel, “The Formation of the Russian General Staff, 1880–1917: A Social Study,” Cahiers du Monde Russe et Soviétique, Vol. 16, Nos. 3–4 (July/December 1975), p. 297.

161. 

Goldman and Andres, “Systemic Effects of Military Innovation and Diffusion,” pp. 116–117; and Herrera, “Inventing the Railroad and Rifle Revolution,” p. 261.

162. 

Peter Gatrell, The Tsarist Economy, 1850–1917 (New York: St. Martin's, 1986), p. 221. Railroads were the largest expense in this category, but it included other items.

163. 

For the data, see Mitchell, International Historical Statistics: Europe, 1750–1993, pp. 673–674, 676–677. On Russian railroads, see Menning, Bayonets before Bullets, p. 19; S.C.M. Paine, Imperial Rivals: China, Russia, and Their Disputed Frontiers (New York: M.E. Sharpe, 1996), p. 257; and Rich, The Tsar's Colonels, pp. 98–99. Although Russia imitated the Prussian system, it did not do so with great success. On the cultural and institutional factors that impeded adaptation, see Mark L. von Hagen, “Autocracy Defeats Military Reform on Eve of First World War,” Russian History, Vol. 38, No. 1 (January 2011), p. 152; Allan K. Wildman, The End of the Russian Imperial Army: The Old Army and the Soldiers’ Revolt (March-April 1917) (Princeton, N.J.: Princeton University Press, 1980), pp. 64–70; Marc Ferro, Nicholas II: Last of the Tsars, trans. Brian Pierce (New York: Oxford University Press, 1993), pp. 4, 42, 70; and Dominic Lieven, Nicholas II: Twilight of the Empire (New York: St. Martin's, 1993), p. 70. On the Russian tendency to build rail lines along economic rather than strategic corridors, see Marshall, The Russian General Staff and Asia, 1800–1917, p. 186; and D.N. Collins, “The Franco-Russian Alliance and Russian Railways, 1891–1914,” Historical Journal, Vol. 16, No. 4 (December 1973), pp. 779–780.

164. 

France was defeated at the demonstration point; therefore the French reaction to blitzkrieg cannot be used to test the theory.

165. 

Donald Cameron Watt, Too Serious a Business: European Armed Forces and the Approach to the Second World War (Berkeley: University of California Press, 1975), pp. 65–66.

166. 

Mark Harrison, Soviet Planning in Peace and War, 1938–1945 (New York: Cambridge University Press, 1985), pp. 81–83, 100, 111, 114, 119.

167. 

John Erickson, Stalin's War with Germany, Vol. 1: The Road to Stalingrad (New York: Harper and Row, 1975), pp. 302, 374–376, 458; Glantz and House, When Titans Clashed, pp. 62, 66, 99–103, 156; Condoleezza Rice, “The Making of Soviet Strategy,” in Paret, Makers of Modern Strategy, pp. 672–673; Arthur J. Alexander, Armor Development in the Soviet Union and the United States (Santa Monica, Calif.: RAND Corporation, 1976), pp. 35–38; Walter S. Dunn, Soviet Blitzkrieg: The Battle for White Russia, 1944 (London: Lynne Rienner, 2000), pp. 2–11; Bryan I. Fugate and Lev Dvoretsky, Thunder on the Dnepr: Zhukov-Stalin and the Defeat of Hitler's Blitzkrieg (Novato, Calif.: Presidio, 1997), pp. 304–307; and David M. Glantz, Colossus Reborn: The Red Army at War, 1941–1943 (Lawrence: University Press of Kansas, 2005), pp. 23–24, 31, 34.

168. 

Although it is less plentiful, there is good evidence that the same dynamics are at work at sea as on land. See Theodore Ropp, The Development of a Modern Navy: French Naval Policy 1871–1904 (Annapolis: Naval Institute Press, 1987), pp. 235, 352; Donald W. Mitchell, A History of Russian and Soviet Sea Power (New York: Macmillan, 1974), pp. 280–282; and Arne Røksund, The Jeune École: The Strategy of the Weak (Boston: Brill, 2007), pp. ix-xiii, 227–229.

169. 

Our approach to identifying cases of external balancing is similar to that of Levy and Thompson in “Balancing on Land and Sea,” pp. 28–30.

170. 

Alliance Treaty Obligations and Provisions Project, “Alliance Treaty Obligations and Provisions Data (v3.0),” www.atop.rice.edu. For details of the dataset, see Brett Ashley Leeds et al., “Alliance Treaty Obligations and Provisions, 1815–1944,” International Interactions, Vol. 28, No. 3 (July 2002), pp. 237–260.

171. 

This count is probably overstated, because we code an alliance as being directed against another great power if it refers to a great power by name, “great powers” in general, “all other powers,” or “all other states.” Increasing the number of possible cases in this way makes it harder for us to find that external balancing is rare.

172. 

The other wartime alliance comprised Britain, France, and Austria-Hungary during the Crimean War.

173. 

See, for example, Patricia A. Weitsman, Dangerous Alliances: Proponents of Peace, Weapons of War (Stanford, Calif.: Stanford University Press, 2004), pp. 101–111. Even this case is open to question. Schweller notes that France's and Russia's motives were not entirely defensive. See Schweller, Unanswered Threats, p. 2.

174. 

To the extent that the Austro-Prussian Alliance had an international dimension, it merely guarded Austria-Hungary's Italian holdings. See Taylor, The Struggle for Mastery in Europe, 1848–1918, pp. 43–44; Paul W. Schroeder, The Transformation of European Politics, 1763–1848 (New York: Oxford University Press, 1994), pp. 714–715, 788; Henry Kissinger, Diplomacy (New York: Simon and Schuster, 1994), pp. 80–81; Norman Rich, Great Power Diplomacy, 1814–1914 (New York: McGraw-Hill, 1992), pp. 20–21; and Fritz Stern, Gold and Iron: Bismarck, Bleichröder, and the Building of the German Empire (New York: Vintage, 1979), chap. 3.

175. 

William L. Langer, European Alliances and Alignments, 1871–1890 (New York: Alfred A. Knopf, 1931), pp. 23–25. See also Taylor, The Struggle for Mastery in Europe, 1848–1918, pp. 219–220.

176. 

Taylor, The Struggle for Mastery in Europe, 1848–1918, pp. 259–260, 270–271; Kissinger, Diplomacy, pp. 159–160, 165; Rich, Great Power Diplomacy, 1814–1914, pp. 244—245; Stern, Gold and Iron, pp. 315–317; and Weitsman, Dangerous Alliances, pp. 53–59, 66–76.

177. 

Kissinger, Diplomacy, p. 160; Rich, Great Power Diplomacy, 1814–1914, p. 233; Taylor, The Struggle for Mastery in Europe, 1848–1918, p. 276; and Weitsman, Dangerous Alliances, pp. 79–87.

178. 

Holborn, A History of Modern Germany, Vol. 3, p. 434. See also Stone, The Eastern Front 1914–1917, pp. 70–76. For the claim that there was some military coordination beginning in 1909, see Weitsman, Dangerous Alliances, p. 78.

179. 

On the Treaty of Björkö, see Taylor, The Struggle for Mastery in Europe, 1848–1918, pp. 432–434. On the Franco-Soviet Treaty, see Anthony Adamthwaite, Grandeur and Misery: France's Bid for Power in Europe, 1914–1940 (London: Arnold, 1995), p. 197; Kissinger, Diplomacy, p. 296; Piotr S. Wandycz, The Twilight of French Eastern Alliances, 1926–1936: French-Czechoslovak-Polish Relations from Locarno to the Remilitarization of the Rhineland (Princeton, N.J.: Princeton University Press, 1988), pp. 395–398, 404, 409, 427–428, 452; and Young, In Command of France, pp. 148–149, 236–240.

180. 

The war data are from COW, “Inter-state War Data (v4.0),” www.correlatesofwar.org. For details of the dataset, see Meredith Reid Sarkees and Frank W. Wayman, Resort to War: A Data Guide to Inter-State, Extra-State, Intra-State, and Non-State Wars, 1816–2007 (Washington, D.C.: CQ Press, 2010).

181. 

Our analysis covers the twelve months before each war broke out.

182. 

Two cases are unsuitable because they involve dual loyalties: France and Russia would have had little confidence in Italy before World War I given its simultaneous alliance with Germany. Two more cases are geographically remote: Britain and Russia would not have been confident that Japan would come to their aid in a European war in 1914. A final case involves short-term offensive preparations: Italy and Prussia formed their alliance just weeks before the Austro-Prussian War with the express purpose of fighting Austria-Hungary. It is unlikely that they would have harbored many misgivings in such circumstances.

183. 

Christopher Clark, The Sleepwalkers: How Europe Went to War in 1914 (London: Allen Lane, 2012), p. 351.

184. 

Ibid., pp. 314–364, 488–554; Zara S. Steiner and Keith Neilson, Britain and the Origins of the First World War, 2nd ed. (New York: Palgrave Macmillan, 2003), pp. 100–116, 229–257; and C.J. Lowe and M.L. Dockrill, The Mirage of Power: British Foreign Policy, 1902–1914, Vol. 1 (London: Routledge, 1972), pp. 107–154.

185. 

T.G. Otte, “A ‘Formidable Factor in European Politics': Views of Russia in 1914,” in Jack S. Levy and John A. Vasquez, eds., The Outbreak of the First World War: Structure, Politics, and Decision-Making (New York: Cambridge University Press, 2014), pp. 94–98; and Clark, The Sleepwalkers, pp. 312–313, 321.

186. 

Sean McMeekin, July 1914: Countdown to War (London: Icon, 2013), pp. 146–147; and Clark, The Sleepwalkers, pp. 422–423. There is little evidence that Russia was concerned about France's reliability, but also little evidence of Russian confidence. Certainly, confidence in alliance commitments varies across states and time, but the general finding of limited confidence is robust.

187. 

Quoted in Clark, The Sleepwalkers, pp. 422–423, at p. 422.

188. 

Ibid., pp. 321, 400.

189. 

Ibid., p. 321.

190. 

Young, In Command of France, p. 243.

191. 

Quoted in Zara Steiner, The Triumph of the Dark: European International History 1933–1939 (New York: Oxford University Press, 2011), p. 999. See also ibid., p. 794.

192. 

Kissinger, Diplomacy, p. 181; Rich, Great Power Diplomacy, 1814–1914, pp. 260–261; Taylor, The Struggle for Mastery in Europe, 1848–1918, pp. 335–336, 338–339; and Snyder, Alliance Politics, p. 269.

193. 

Fuller, Strategy and Power in Russia, 1600–1914, pp. 430, 442–443; and Ropp, The Development of a Modern Navy, pp. 201, 240–241.

194. 

Holborn, A History of Modern Germany, Vol. 3, p. 434; and Stone, The Eastern Front 1914–1917, pp. 70–76. On the more general lack of cohesion in the Triple Alliance, see Weitsman, Dangerous Alliances, pp. 91–97.

195. 

Kissinger, Diplomacy, p. 81; and Rich, Great Power Diplomacy, 1814–1914, p. 21.

196. 

Adamthwaite, Grandeur and Misery, p. 216; and Kissinger, Diplomacy, pp. 345–348.

197. 

Steiner, The Triumph of the Dark, p. 794.

198. 

For work in this vein, see Yasuhiro Izumikawa, “To Coerce or Reward? Theorizing Wedge Strategies in Alliance Politics,” Security Studies, Vol. 22, No. 3 (August 2013), pp. 498–531; and Kyle Haynes, “Decline and Devolution: The Sources of Strategic Military Retrenchment,” International Studies Quarterly, Vol. 59, No. 3 (September 2015), pp. 490–502.

199. 

Keir A. Lieber and Gerard Alexander, “Waiting for Balancing: Why the World Is Not Pushing Back,” International Security, Vol. 30, No. 1 (Summer 2005), pp. 115–125.

200. 

For contrasting views on China's growth, see Linda Yueh, China's Growth: The Making of an Economic Superpower (New York: Oxford University Press, 2013); and Michael Beckley, “China's Century? Why America's Edge Will Endure,” International Security, Vol. 36, No. 3 (Winter 2011/12), pp. 41–78.