Balancing in Neorealism

Balancing in Neorealism
Balancing in Joseph M. Parent and

Sebastian Rosato
Neorealism
D
o great powers bal-
ance 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 regular-
ity that . . . [neorealism] assumes,” and dismisses efforts to explain why as evi-
dence 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-identiªed realists agree that neorealism’s apparent explana-
Joseph M. Parent is Associate Professor of Political Science at the University of Miami. Sebastian Rosato is
Associate Professor of Political Science at the University of Notre Dame.
Earlier versions of this article were presented at Columbia University, Dartmouth College, Har-
vard 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.
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 impor-
tant 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, speciªcally 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 Sad-
dle 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 Pro-
grams: An Appraisal of Neotraditional Research on Waltz’s Balancing Proposition,” American Po-
litical 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.
International Security, Vol. 40, No. 2 (Fall 2015), pp. 51–86, doi:10.1162/ISEC_a_00216
© 2015 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
51
International Security 40:2 52
tory 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 ªnding, he develops a
domestic political argument.5 William Wohlforth offers an account of state be-
havior that relies on perceptions of the balance of power because he sees scant
evidence that the balance itself has a signiªcant 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 know-
ing 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 suc-
cessful 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 balanc-
ing, states lack conªdence 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 inter-
nal and external balancing differently and behave accordingly. The father of
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 Univer-
sity 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.
Balancing in Neorealism 53
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 dif-
ferent 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 litera-
ture conceives of balancing in terms of counterbalancing alliances.”14
Our study, however, reveals abundant evidence to support a neorealist the-
ory 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, battleºeet warfare, blitzkrieg, carrier warfare,
strategic bombing, and nuclear weapons). Case studies suggest that these out-
comes 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 exter-
10. Kenneth N. Waltz, “Evaluating Theories,” American Political Science Review, Vol. 91, No. 4 (De-
cember 1997), p. 915. See also Kenneth N. Waltz, “The Origins of War in Neorealist Theory,” Jour-
nal of Interdisciplinary History, Vol. 18, No. 4 (April 1988), p. 625.
11. Mearsheimer, The Tragedy of Great Power Politics, pp. 156–157. Mearsheimer brieºy 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 Or-
ganization, 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 bal-
ancing. The canonical texts contain only a handful of empirical examples. See Waltz, Theory of In-
ternational 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 Sys-
tem,” 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 of-
fers a comprehensive account of great power internal balancing behavior.
nal balancing. Our examination of the diplomatic record yields ªve 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 To-
gether, these ªndings on internal and external balancing show that neo-
realism’s proponents have understated its successes while its detractors have
overstated its failures.
The remainder of this article proceeds as follows. In the ªrst section, we lay
out a neorealist theory of balancing and address potential objections. The sec-
ond section tests the theory using quantitative and qualitative evidence on
great power arming, imitation, and alliance behavior since 1816. The ªnal sec-
tion summarizes our arguments and ªndings, 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. There-
fore, 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 be-
cause 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 inter-
national politics, that they are rational and unitary, that they possess some mil-
16. For similar ªndings, 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 Mys-
terious 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 Com-
petition 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 Univer-
sity Press, 2014), pp. 31–36.
itary 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 chieºy in size, not in composition. States have some informa-
tion 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 de-
stroy 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 barri-
ers. Great powers suffer a loss of strength gradient when they deploy capa-
bilities over long ranges.19 Barriers can take many forms, including rivers,
mountains, jungles, and cities. For present purposes, the most important dis-
tinction is between land and sea. Speciªcally, 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’ in-
tentions, 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 secu-
rity, but failing to do so can guarantee insecurity.
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,” Interna-
tional Security, Vol. 39, No. 3 (Winter 2014/15), pp. 52–53.
19. Kenneth E. Boulding, Conºict 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 Vil-
lage 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; Mear-
sheimer, The Tragedy of Great Power Politics, pp. 83–84; and Daniel H. Deudney, Bounding Power: Re-
publican 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.
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 mini-
mize 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 attenu-
ated 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 ªercely rivalrous.25
To generate military capability, states engage in balancing behavior. Bal-
ancing refers to a state’s efforts to amass military might so as to deter another’s
aggression or prevail in a conºict should deterrence fail.26 Great powers
can adopt two kinds of balancing policies: internal and external. Internal bal-
ancing 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—ªrm military com-
mitments to deter or defend against common rivals.29
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 en-
sure their short-term growth and security without jeopardizing their long-term growth and secu-
rity. 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 Conºict 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 deªnition of alliances, see Glenn H. Snyder, Alli-
ance Politics (Ithaca, N.Y.: Cornell University Press, 1997), p. 4. For a deªnition that includes looser
arrangements, see Walt, The Origins of Alliances, pp. 12–13.
Internal balancing is most desirable because it is true self-help.30 Great pow-

ers 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 faith-
ful reliance on external balancing will be rare. As Waltz points out, true exter-
nal 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. Con-
sequently, great powers will not merely discount their partners’ promises; they
will also doubt the reliability of their opponents’ allies.34 It follows that alli-
ances 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 alterna-
tives to external balancing. States bandwagon when they ally with a formida-
ble peer and allow their partner a disproportionate share of the spoils.35
Although bandwagoning is conventionally presented as a viable option, great
30. Waltz, Theory of International Politics, p. 168; and Mearsheimer, The Tragedy of Great Power Poli-
tics, 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 deªnition,
only internal balancing is possible in bipolarity.
31. For more expansive deªnitions of threat, see Thomas J. Christensen, Useful Adversaries: Grand
Strategy, Domestic Mobilization, and Sino-American Conºict, 1947–1958 (Princeton, N.J.: Princeton
University Press, 1996), pp. 27–28; and Walt, The Origins of Alliances, pp. 25–26.
tics, 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 Relation-
ship with Conºict,” 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 Alli-
ances, pp. 17, 19–21; and Mearsheimer, The Tragedy of Great Power Politics, pp. 162–163.
powers will shun it.36 In a neorealist world, any move that creates a more capa-
ble rival reduces a state’s security.37 Buck-passing, too, is often cited as a genu-
ine 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 develop-
ing military might. In addition, geography drives how states allocate their
assets. Insular powers are more responsive to sea-based military innova-
tions, whereas land powers are more responsive to land-based military inno-
vations because these jeopardize the foundations of their security.41 Finally,
land and sea powers have different attitudes toward alliances. Being more se-
cure, 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 capabili-
36. Randall L. Schweller, “Bandwagoning for Proªt: Bringing the Revisionist State Back In,” Inter-
national Security, Vol. 19, No. 1 (Summer 1994), pp. 72–107.
tics, 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 Uni-
versity 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.
ties in their favor, but responses to such shifts frustrate states, homogenize
behavior, and promote unintended outcomes. Although great powers seek ad-
vantage, 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, interna-
tional political outcomes generally mirror the actual distribution of power
among states. In the shorter term, however, the policies states pursue are
rarely objectively efªcient or predictable based upon a purely systemic analy-
sis.”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 struc-
tural; that is, it admits no elements at the individual or domestic level of analy-
sis, 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 de-
terminate; but the strongest factors create tendencies that allow reasonable
suppositions. There remains room for volition in our theory. Negative feed-
back 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 secu-
rity 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 ªrm alliances only
when conºict 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 ef-
fort than sea powers and will seek to match the arms buildups and innova-
44. On this point, see also Robert Jervis, “Security Regimes,” in Stephen D. Krasner, ed., Interna-
tional 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.
tions 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 speciªed, neorealism powerfully predicts great power balancing be-

havior. Here we use the historical record to show that great powers arm and
imitate promptly against their peers; that such internal balancing behavior oc-
curs 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 fol-
lowing 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 avail-
able 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 pow-
ers. Because they are relatively close to Eurasia, Britain and Japan are partially
continental sea powers. Much farther away, the United States was a fully in-
sular sea power before World War II. After the collapse of Western Europe’s
great powers, however, the United States deployed formidable forces immedi-
ately adjacent to the Soviet Union, thereby becoming a fully continental sea
48. Because our deªnition 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 Conºict 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 Secu-
rity, 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 ge-
ography, see Levy and Thompson, “Balancing on Land and at Sea,” pp. 25–26.
power.51 Structural imperatives drove this outcome: a Soviet Union that domi-
nated 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 superi-
ority 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 balanc-
ing. 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. Con-
versely, sea powers are less threatened than land powers by virtue of geogra-
phy, 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 ªrst took the great powers’ military expenditures in a given year
and gave the state with the highest spending a score of 100. Others were as-
signed scores reºecting their spending as a percentage of the lead state’s ex-
penditures. We then did the same for military personnel. Finally, we averaged
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 them-
selves . . . 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.
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 Further-
more, these indicators do not include factors such as force postures and doc-
trine that render military forces more or less effective, and they are therefore
crude reºections of states’ military might. Yet a thorough assessment of our
theory involves quantitative data on arming; state behavior is most likely to be
inºuenced 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 ªve 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 bal-
ances 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 bal-
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 Se-
curity, Vol. 12, No. 4 (Spring 1988), pp. 175–177; and Schweller, Deadly Imbalances, pp. 17–19. Mili-
tary 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 ulti-
mately to deter Nazi Germany as an exemplar of neorealism’s explanatory weakness. See, for ex-
ample, 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 deªne short to medium term as less than ªve 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 ªndings remain the same regardless of procedure.
61. The analysis in this section excludes Germany from 1919 to 1933, when its army was
artiªcially capped at 100,000 men.
ance 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, neo-
realism’s success rate increases to 92 percent.
Crucially, capability balances cannot be reduced to power balances. The rea-
son 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 per-
centage of its male population aged twenty to thirty-nine, as well as its mili-
tary 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 continen-
tal sea powers, Britain consistently made less effort than Europe’s land pow-
ers. 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 al-
ways worked harder than Russia. During the late 1930s, Germany and France
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%). Deªning “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 ªrst year in which the relevant data are
available for all the great powers.
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 pow-
ers, 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 neo-
realism suggests they will. Austria-Hungary and Italy did not work as hard as
Germany, France, and Russia in the late nineteenth and early twentieth centu-
ries. In all probability, this is because they would have had to exert prohibi-
tively 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 ªve years. On land or at sea, effective mili-
tary 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 ap-
ply themselves harder the closer they are to land powers.
internal balancing: imitation

Our theory contends that great powers will imitate the successful military in-
novations of peer competitors and do so promptly.66 Geography plays a key
role: proximate states are more likely to imitate than remote states; land pow-
ers 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 innova-
tions, we examined eight accounts of military innovation since 1816 and se-
lected innovations that appeared in more than half of them.67 This procedure
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 uni-
form. 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 Interna-
tional 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 Mili-
tary 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
yields six innovations: the Prussian system; battleºeet warfare; blitzkrieg; car-
rier 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 ªve years, we call this a success
for neorealism; ªve or more we call a failure. Judging when imitation has oc-
curred is difªcult: 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 inno-
vator’s capability.
prussian system. By integrating the breech-loading riºe, the railroad, and
the telegraph with a general staff system, Prussia surpassed other states in its
ability to deploy a large, ªrepower-oriented, fast-moving, and highly coordi-
nated army on the battleªeld.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
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 In-
novation 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 Em-
ily 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 Nine-
teenth 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.
Prussian system in 1870: it had a general staff; its troops were armed with
the Chassepot breech-loading riºe; and it conducted its second railway
mobilization—the ªrst was in 1859—using an extensive rail network. Im-
mediately 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 breech-
loaders 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
battleºeet warfare. Britain’s battleºeet warfare innovation revolution-
ized the way navies could mass force to destroy enemy ºeets and control the
seas.78 At its base lay a new technology: all-big-gun, heavily armored, fast pro-
pulsion 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 Un-
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 Riºe Revolution: Information, Military Inno-
vation, 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.
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 Inºuence 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 Japa-
nese Navy, 1887–1941 (Annapolis: Naval Institute Press, 1997), p. 152. See also Horowitz, The Diffu-
sion of Military Power, pp. 138–139. Although wireless telegraphy is sometimes mentioned as a
technological component of the battleºeet innovation, it had a negligible effect. See Boot, War
Made New, p. 197.
80. Horowitz, The Diffusion of Military Power, pp. 144–145.
like most other innovations that were demonstrated in the context of war, the
demonstration point for battleºeet warfare was the peacetime launch of
the ªrst 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 do-
ing 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 effec-
tive 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 ele-
ments 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 practic-
ing 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
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 ªrst true dread-
noughts 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 Innova-
tion 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 Hit-
ler (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
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 Paciªc the-
ater, Japan had little need for blitzkrieg. Once the United States joined the bat-
tle, 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 com-
bined use of multiple aircraft carriers supported by destroyer screens and
logistical ships—transformed the ªghting power of its practitioners by en-
abling them to launch offensive strikes against enemy navies across great dis-
tances.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 ºeet around multicarrier task forces a year be-
fore 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 com-
bined operations with the United States Navy.92 Predictably, the land powers,
Germany and the Soviet Union, did not make a meaningful effort to adopt car-
rier warfare.93
strategic bombing. The advent of strategic bombing—the selection and
destruction of industrial targets by long-range bombers supported by ªghter
escorts—increased states’ prospects for victory in war by allowing them to de-
grade their adversaries’ productive capacity and military capability. The dem-
onstration point for strategic bombing is 1944, when the United States and
Britain used air power to destroy German industries and transportation links.
87. Samuel W. Mitcham Jr., Blitzkrieg No Longer: The German Wehrmacht in Battle, 1943 (Mechanics-
burg, 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: Car-
rier 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 Dif-
fusion 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.
By the spring of 1945, Germany no longer had the means to continue ªght-
ing.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 estab-
lished 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 develop-
ing 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 ªve 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 fail-
ure (Austria-Hungary with the Prussian system), and two cases of partial imi-
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 offen-
sives.” 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 trans-
portation network meant that Germany was unable “to put weapons and supplies in the hands of
the divisions actually ªghting.” 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 Revolu-
tions 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.
tation (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 de-
cisive 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 im-
portance of obligatory military service, and a raft of reforms followed.103 In
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 battleºeet 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 ªve-year
window (the only additional anomaly being the Soviet acquisition of atomic weapons).
100. Levy and Thompson claim that it is difªcult 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 (Au-
gust 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 ªgures, 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 Uni-
versity 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.
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 sched-
ules of Germany and Austria-Hungary.108 Conscripts’ service terms were dra-
matically 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
104. Quoted in Carroll, French Public Opinion and Foreign Affairs, 1870–1914, p. 74.
105. Allan Mitchell, Victors and Vanquished: The German Inºuence 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 Uni-
versity 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.
planners had become alarmed by Russian progress.112 Russia staged maneu-

vers 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 impera-
tive to expand the German army . . . most drastically, in 1887, while France and
Russia particularly were building up their armies.”114 Speaking on foreign af-
fairs publicly for the last time, Bismarck, the consummate diplomat, “repudi-
ated 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 be-
coming carp.’”115
In 1889 France responded to Germany’s 1888 Septennat bill to increase mili-
tary spending.116 Charles de Freycinet, the French war minister, wrote Joseph
de Miribel, his chief of the general staff, lamenting that Germany’s “consider-
able 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 in-
crease of forces without taking similar measures.” Miribel concurred: “We
ought to have forces equivalent to those of the Germans.”117 Freycinet’s re-
forms 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 inter-
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 (Provi-
dence: 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
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.
national 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 equiva-
lent 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 re-
port revealed that Russia was “insufªciently prepared for defense not only
against Germany, but even against Austria-Hungary.”122 Crushing losses in
the Russo-Japanese War in 1905 conªrmed that assessment and initiated an-
other round of reforms. A 1906 law shortened active service for the ªrst 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 mobili-
zation, 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
120. Quoted in Huntington, “Arms Races,” p. 46 n. 8. See also ibid., pp. 65–66; and Craig, The Poli-
tics of the Prussian Army, 1640–1945, p. 243. The Germans did, however, fall behind in ªeld guns.
France deployed a new 75-millimeter ªeld 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 Im-
perial 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 Univer-
sity, 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 Conºict 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.
power” and had to expand its forces immediately.126 Reichstag politicians con-
curred: 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 success-
ful 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 to-
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 ªnished 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 Conºict 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.
morrow.”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 chron-
icles France’s energetic diplomacy between the wars, but even he ªnds 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 ac-
tion than it was about German action.138 When Germany rearmed, France fol-
lowed 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 suc-
cessful that when Germany attacked France in 1940, “not a single general ex-
pected victory to result.”140
France’s emphasis on internal balancing reºected 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
135. Quoted in Arnold Wolfers, Britain and France between Two Wars: Conºicting 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 Rich-
ard D. Challener, “The French Foreign Ofªce: 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 Unªnished
Peace after World War I: America, Britain, and the Stabilisation of Europe, 1919–1932 (New York: Cam-
bridge 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 Uni-
versity 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 Mar-
tin S. Alexander, “In Defence of the Maginot Line: Security Policy, Domestic Politics, and the Eco-
nomic 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 fortiªcations, see An-
thony 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 prepared-
ness and modernization.” See Doughty, The Seeds of Disaster: The Development of French Army Doc-
trine, 1919–1939 (Hamden, Conn.: Archon, 1985), p. 2.
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 identiªed the key weakness: “[O]ld Russia . . . was
ceaselessly beaten for her backwardness. For military backwardness . . . for in-
dustrial 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 re-
armament, 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 re-
armament 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 over-
whelming 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
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 For-
eign 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 Conºicts (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.
before the distribution of military capabilities shifted decisively in the Soviet

Union’s favor.150
imitation. The period under discussion featured two land-based innova-
tions: the Prussian system and blitzkrieg. In both cases, Germany’s advantage
was transitory because France and Russia quickly imitated Germany’s exam-
ple 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 ªgures set the standards for French tar-
gets, and reforms followed: new fortiªcations, new technology, new organiza-
tion, 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
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 His-
tory, 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 riºes were so
effective during the Franco-Prussian War that the Germans distributed 600,000 captured Chasse-
pots to their troops immediately and crafted a new German model in 1871. On the riºe 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, Febru-
ary 1999, pp. 168–169, 175–176. See also Allan Mitchell, “Private Enterprise or Public Service? The
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 ºexible, 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 riºes
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 ele-
ment of the Prussian system. Building a competitive rail network was difªcult
given Russia’s vast geography and economic inferiority.161 Nevertheless, the
Russian government quadrupled the percentage of state spending on infra-
structure 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 ªgure had fallen to one and a half times; and twenty years
later they were almost equivalent.163
Eastern Railway Company and the French State in the Nineteenth Century,” Journal of Modern His-
tory, 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 Uni-
versity 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.
and Herrera, “Inventing the Railroad and Riºe 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 Ri-
vals: 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 Univer-
sity 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.
During World War II, the Soviet Union promptly imitated Germany’s blitz-
krieg methods.164 Russian investment in tanks had kept pace with German in-
vestment, and some advocates of Russian armored doctrine were sympathetic
to blitzkrieg tactics before the war, but they were an uninºuential 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 ofªcials
struggled to adapt to the new tactics.166 Fundamental changes were initiated in
the ªrst year; and no later than the offensives of November 1942, Russian
forces combined the core elements of blitzkrieg, and Russian offensives re-
versed German gains.167
In sum, France, Germany, and Russia armed and imitated successful prac-
tices 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 re-
quires reliance on others. As a result, great powers rarely engage in external
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 Univer-
sity 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.
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 reºection
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 reºect an aversion to diplomacy; the ATOP dataset identi-
ªes 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 ªt a common deªnition
of external balancing.
Five cases of external balancing occurred during wartime, a ªnding that
supports the neorealist contention that great powers resort to ªrm alliances in
extremis. Indeed, four of the ªve were contracted during wars whose out-
comes 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
169. Our approach to identifying cases of external balancing is similar to that of Levy and Thomp-
son in “Balancing on Land and Sea,” pp. 28–30.
170. Alliance Treaty Obligations and Provisions Project, “Alliance Treaty Obligations and Provi-
sions Data (v3.0),” www.atop.rice.edu. For details of the dataset, see Brett Ashley Leeds et al., “Al-
liance 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 an-
other great power if it refers to a great power by name, “great powers” in general, “all other pow-
ers,” or “all other states.” Increasing the number of possible cases in this way makes it harder for
us to ªnd that external balancing is rare.
172. The other wartime alliance comprised Britain, France, and Austria-Hungary during the Cri-
mean War.
173. See, for example, Patricia A. Weitsman, Dangerous Alliances: Proponents of Peace, Weapons of
German Confederation (1820–66) and the Austro-Prussian Alliance (1851–54)

were designed as much to manage Austro-Prussian relations and maintain do-
mestic 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; ªve months after ratiªcation 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, ªrst
reviving the Three Emperors’ League (1881–87), and then concluding the
Reinsurance Treaty (1887–90).176 As for the Triple Alliance (1882–1915), it in-
volved only one unambiguous balancing pledge: Germany’s promise to de-
fend Italy in the unlikely event of an attack by France.177 Despite Germany and
Austria having been allied for thirty-ªve 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 ªrm 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
War (Stanford, Calif.: Stanford University Press, 2004), pp. 101–111. Even this case is open to ques-
tion. 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, Diplo-
macy, 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 Strug-
gle 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.
risky. To establish whether or not great powers view external balancing

as risky, we identiªed 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 sacriªces in a common cause,
one might expect them to have been conªdent in their partners’ reliability be-
fore 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 be-
fore World War II.182
An examination of these four most-likely cases reveals a great deal of uncer-
tainty. France was not conªdent 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 rapproche-
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 de-
tails 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 conªdence 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 conªdent that
Japan would come to their aid in a European war in 1914. A ªnal case involves short-term offen-
sive preparations: Italy and Prussia formed their alliance just weeks before the Austro-Prussian
War with the express purpose of ªghting 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.
ment.185 Right up to the outbreak of hostilities, French ofªcials suspected that

Russia was not truly committed to ªghting the Central Powers.186
The opposing side also demonstrated little conªdence in external balanc-
ing. 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 his-
tory 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
ºux-prone many of the key decision-makers felt the alliance system was.”189
With good reason, Germany had serious misgivings about Italy’s commit-
ment to the Pact of Steel on the eve of World War II. Italy was the model of un-
reliability, 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 re-
ferred 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
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 re-
liability, but also little evidence of Russian conªdence. Certainly, conªdence in alliance commit-
ments varies across states and time, but the general ªnding of limited conªdence 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.
and have less need for external balancing. It is exclusively states in the riskiest
positions at the riskiest times that are compelled to seek ªrm alliances.
unresponsive. The ATOP dataset yields abundant evidence that great pow-
ers are slow to reciprocate the countervailing efforts of others. Of the nine ex-
amples 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 af-
ter 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 re-
garded 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 lit-
tle difference to British Prime Minister Neville Chamberlain, who doubted
Italy’s attachment to Germany, worked to decouple the two, and did not accel-
erate 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 ªrmest alliances show
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 Al-
liances, 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.
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 them-
selves. 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 capa-
bilities 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 ªrmness 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 re-
sponse to their peers’ increases in capability. Some of them have gained a mili-
tary 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 advan-
tage to early adopters, but not for long. Land powers have imitated land-based
innovations promptly; sea powers have done the same with sea-based innova-
tions. 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 in-
ternal efforts of others. Meanwhile, they view external balancing as risky, prac-
tice 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 dif-
ferent outcomes. Although our inquiry into internal balancing has been pre-
liminary, the results are encouraging and merit exploration. How do states set
the ceiling on their arming efforts? Why do states sometimes fail in their inter-
nal balancing efforts? What is the decisionmaking calculus of liminal great
powers? As for external balancing, ªrm 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 of-
fers a parsimonious explanation of the prevailing situation and a serviceable
guide to the future. The reason commentators have detected a lack of balanc-
ing 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 bal-
ancing for their security. In the timeless cycles of power politics, salvation
lies within.
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.
Copyright of International Security is the property of MIT Press and its content may not be
copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for
individual use.

Balancing in Neorealism - 748

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Balancing in Neorealism - 748

Uploaded by

Copyright:

Available Formats

Balancing in Joseph M. Parent and

neorealism, Kenneth Waltz, merely comments that “[f]aced with unbalanced

A Neorealist Theory of Balancing

In this competition, states are strongly constrained by their power, which is

Internal balancing is most desirable because it is true self-help.30 Great pow-

Arming, Imitation, and Alliance Behavior

Correctly speciªed, neorealism powerfully predicts great power balancing be-

internal balancing: arming

ance was evident in 84 percent of them.62 Most of the anomalies involve

internal balancing: imitation

internal balancing: qualitative cases

planners had become alarmed by Russian progress.112 Russia staged maneu-

before the distribution of military capabilities shifted decisively in the Soviet

external balancing: alliances

German Confederation (1820–66) and the Austro-Prussian Alliance (1851–54)

risky. To establish whether or not great powers view external balancing

Wandycz, The Twilight of French Eastern Alliances, 1926–1936: French-Czechoslovak-Polish Relations

ment.185 Right up to the outbreak of hostilities, French ofªcials suspected that

You might also like