Arms Control

Arms Control as Wedge Strategy
Arms Control as Timothy W. Crawford

and Khang X. Vu
Wedge Strategy
How Arms Limitation Deals
Divide Alliances
Downloaded from http://direct.mit.edu/isec/article-pdf/46/2/91/1968534/isec_a_00420.pdf by MARSHALL CENTER RESEARCH LIBRARY user on 19 January 2022
S
trategic arms control
is in crisis.1 The United States and Russia have retreated from agreements
that formed the framework for post–Cold War arms cuts and strategic stabil-
ity. In 2002, the United States withdrew from the 1972 Anti-Ballistic Missile
(ABM) Treaty. In 2007, Russia suspended the 1990 Treaty on Conventional
Armed Forces in Europe (CFE) and formally renounced it in 2015. The U.S.
Department of Defense’s 2018 Nuclear Posture Review depicted an increasingly
“uncertain security environment,” characterized by new nuclear arms races,
proliferation, and great power competition, which made it imperative that the
United States drop arms control agreements that were not beneªcial to its se-
curity, veriªable, and enforceable.2 The United States withdrew from the 1987
Intermediate-Range Nuclear Forces Treaty in 2019 and from the Open Skies
Treaty in 2020. The only strategic arms control agreement between the United
States and Russia (i.e., the 2011 New Strategic Arms Reduction Treaty [New
START]) expires in 2026. The time for building a successor framework for stra-
tegic arms control is running out. Whether a political basis for one can be
found is now a pressing matter of international security.
Two trends are driving today’s arms control crisis. First, nuclear stability is
weakened by new technology: improving missile accuracy, remote sensing,
antisubmarine capabilities, and a precision revolution in conventional weap-
Timothy W. Crawford is Associate Professor of Political Science at Boston College. He is the author of The
Power to Divide: Wedge Strategies in Great Power Competition (Ithaca, N.Y.: Cornell University
Press, 2021). Khang X. Vu is a Ph.D. candidate in political science at Boston College from Hanoi, Vietnam.
The authors thank Andrew Bowen, Michael Glosny, Viet Phuong Nguyen, Leor Sapir, and
Chengzhi Yin, as well as their fellow panelists at the 2020 American Political Science Association
annual meeting, Vasabjit Banerjee and Qi Zhang, and the anonymous reviewers.
1. On this theme, see Rose Gottemoeller, “Rethinking Nuclear Arms Control,” Washington Quar-
terly, Vol. 43, No. 3 (2020), pp. 139–159, doi.org/10.1080/0163660X.2020.1813382; Steven E. Miller,
“A Nuclear World Transformed: The Rise of Multilateral Disorder,” Daedalus, Vol. 149, No. 2
(Spring 2020), pp. 17–36, doi.org/10.1162/daed_a_01787; Linton F. Brooks, “The End of Arms Con-
trol?” Daedalus, Vol. 149, No. 2 (Spring 2020), pp. 84–100, doi.org/10.1162/daed_a_01791; Alexey
Arbatov, “Mad Momentum Redux? The Rise and Fall of Nuclear Arms Control,” Survival, Vol. 61,
No. 3 (2019), pp. 7–38, doi.org/10.1080/00396338.2019.1614785; and Götz Neuneck, “The Deep
Crisis of Nuclear Arms Control and Disarmament: The State of Play and the Challenges,” Jour-
nal for Peace and Nuclear Disarmament, Vol. 2, No. 2 (2019), pp. 439–444, doi.org/10.1080/
25751654.2019.1701796.
2. Ofªce of the Secretary of Defense, Nuclear Posture Review (Washington, D.C.: U.S. Department of
Defense, February 2018), pp. 5–14.
International Security, Vol. 46, No. 2 (Fall 2021), pp. 91–129, https://doi.org/10.1162/isec_a_00420
© 2021 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.
91
International Security 46:2 92
ons.3 Such changes, uncertainty about how they will be used, and competitive
pressures to counter them have obsolesced the existing framework of arms
control.4 Second, the political forceªeld that sustained the old framework has
been altered by China’s rise and the deterioration of U.S.-Russia relations.5
Russia and China now contest U.S. primacy with military modernization, chal-
lenges to U.S. alliances, and opposing world order visions.6 Can strategic arms
control be revitalized amidst great power competition and nuclear disorder?
Can the United States still use it to improve its security?
To answer these questions, we examine the fault lines in arms control
thought that reºect debates about why states prioritize arms control and how
power politics ªt in. Here, we develop a theory of arms control wedge strate-
gies. Wedge strategies use “diplomacy and statecraft to move or keep a poten-
tial adversary out of an opposing alliance”; they seek to “prevent, break up, or
weaken [an opposing] alliance at an acceptable cost.”7 Along with other mo-
tives for arms control (e.g., to preserve stability, reduce costs, or secure military
advantages), states may use it in a wedge strategy to weaken alignment be-
tween adversaries. When states do so successfully, arms control does not just
reºect power relations, it alters them. Our theory explains how arms control
can drive wedges by affecting adversaries’ threat perceptions, their beliefs
about the costs and beneªts of formal commitments, and their degree of trust
in each other. Because these factors can weaken adversaries’ alignments, the
motive to drive a wedge may not only shape how states conduct arms control
diplomacy but also inºuence the terms that they put in the ªnal deal. We apply
our theory to three landmark strategic arms control negotiations: the Five-
3. Keir A. Lieber and Daryl G. Press, “The New Era of Counterforce: Technological Change and
the Future of Nuclear Deterrence,” International Security, Vol. 41, No. 4 (Spring 2017), pp. 18–46,
doi.org/10.1162/ISEC_a_00273; Heather Williams, “Asymmetric Arms Control and Strategic Sta-
bility: Scenarios for Limiting Hypersonic Glide Vehicles,” Journal of Strategic Studies, Vol. 42, No. 6
(2019), pp. 793–801, doi.org/10.1080/01402390.2019.1627521; and Elizabeth Sherwood-Randall,
“The Age of Strategic Instability: How Novel Technologies Disrupt the Nuclear Balance,” Foreign
Affairs, July 21, 2020, https://www.foreignaffairs.com/articles/russia-fsu/2020-07-21/age-strategic-
instability.
4. Neuneck, “Deep Crisis of Nuclear Arms Control.”
5. Miller, “Nuclear World Transformed”; Ofªce of the Secretary of Defense, Nuclear Posture Review,
pp. 5–14; and Eugene Rumer, “A Farewell to Arms . . . Control,” Carnegie Endowment for Interna-
tional Peace, April 17, 2018, https://carnegieendowment.org/2018/04/17/farewell-to-arms-.-.-.-
control-pub-76088.
6. Alexander Cooley and Daniel Nexon, Exit from Hegemony: The Unraveling of the American Global
Order (New York: Oxford University Press, 2020), pp. 80–109; and Matthew Kroenig, The Return of
Great Power Rivalry: Democracy versus Autocracy from the Ancient World to the U.S. and China (New
York: Oxford University Press, 2020).
7. Timothy W. Crawford, The Power to Divide: Wedge Strategies in Great Power Competition (Ithaca,
N.Y.: Cornell University Press, 2021), p. 1; and Timothy W. Crawford, “Preventing Enemy Coali-
tions: How Wedge Strategies Shape Power Politics,” International Security, Vol. 35, No. 4 (Spring
2011), p. 156, doi.org/10.1162/ISEC_a_00036.
Arms Control as Wedge Strategy 93
Power Treaty and the Four-Power Treaty at the 1921–22 Washington Naval
Conference, the 1963 Limited Nuclear Test Ban Treaty (LTBT), and the 1972
Strategic Arms Limitation Talks (SALT I). We show how the wedge motive in-
formed these negotiations and inºuenced great power relations.8
Our research has two signiªcant policy implications. First, despite recent
setbacks, strategic arms control can continue to play a critical role in U.S. secu-
rity policy. Though great power competition strains the old frameworks, the
United States can use strategic arms control—especially the platform of New
START renewal—to create an advantage in the strategic triangle with Russia
and China. Second, the United States may also be challenged by other states’
use of arms control to divide opponents. In particular, North Korea’s missile
and nuclear developments could prompt a deal to cap its arsenal before it per-
fects capabilities to attack the U.S. mainland. However, North Korea might
demand concessions that would weaken the U.S.-South Korea alliance.
The article proceeds as follows. First, we review the main lines of arms con-
trol thought. We show how our theory augments conventional accounts of the
motives and effects of arms control. Second, we detail three mechanisms of di-
vision through which states use arms control to drive wedges. Third, we exam-
ine three major cases of arms control wedge strategies, using process-tracing to
reveal the processes of division at work in each case.9 Finally, we conclude by
explaining why those processes are relevant to current U.S. nuclear arms con-
trol policies.
8. Previous scholars noted wedge aspects of these cases but did not connect them to a general con-
cept or mechanisms of wedge strategies. See Harold Sprout and Margaret Sprout, Toward a New
Order of Sea Power: American Naval Policy and the World Scene, 1918–1922 (Princeton, N.J.: Princeton
University Press, 1940), p. 92; Emily O. Goldman, Sunken Treaties: Naval Arms Control between the
Wars (University Park, Pa.: Penn State University Press, 1994), p. 226; Gordon H. Chang, Friends
and Enemies: The United States, China, and the Soviet Union, 1948–1972 (Redwood City, Calif.: Stan-
ford University Press, 1990), pp. 229–232; Lorenz M. Lüthi, The Sino-Soviet Split: Cold War in the
Communist World (Princeton, N.J.: Princeton University Press, 2008), p. 271; and Jonathan Haslam,
Russia’s Cold War: From the October Revolution to the Fall of the Wall (New Haven, Conn.: Yale Uni-
versity Press, 2011), p. 263.
9. In the twentieth century, there were at least four other major attempts to use arms control as
wedge strategies, including: the 1912 Anglo-German naval talks, in which Germany tried to en-
courage British neutrality in a future European conºict; the 1935 Anglo-German Naval Agree-
ment, through which Germany fragmented the “Stresa Front” counter-balancers (i.e., Britain,
France, and Italy); the 1955 Soviet disarmament proposal that called for deep conventional and nu-
clear cuts to weaken and divide the Western allies; and the 1974–1979 U.S.-Soviet strategic arms
control talks (leading to SALT II), which the Soviets used to stave off U.S.-China collusion. See, for
example, Sean M. Lynn-Jones, “Détente and Deterrence: Anglo-German Relations, 1911–1914,”
International Security, Vol. 11, No. 2 (Fall 1986), pp. 127–128; Barry R. Posen, The Sources of Military
Doctrine: France, Britain, and Germany between the World Wars (Ithaca, N.Y.: Cornell University Press,
1984), p. 202; Matthew Evangelista, “Cooperation Theory and Disarmament Negotiations in
the 1950s,” World Politics, Vol. 42, No. 4 (July 1990), pp. 506–507, 520–521; and Robert S. Ross, The
Indochina Tangle: China’s Vietnam Policy, 1975–1979 (New York: Columbia University Press, 1988),
pp. 57, 82–83, 144, 174, 200.
Existing Theories of Strategic Arms Control
Modern arms control centers on formal, treaty-based measures to limit the
numbers, types, uses, and destructiveness of weapons.10 Many such measures
are general, broad, regime-like schemes. Others are more “rivalry speciªc,”
focused on two or a handful of powers (or alliances), crafted to address mili-
tary aspects of their competition.11 The latter kind we call “strategic arms
control.”12 Within its scope are many important pre-nuclear and nuclear agree-
ments from the interwar period (e.g., 1921 Four-Power Treaty), the early 1970s
(e.g., SALT I and ABM Treaty), and the post–Cold War era (e.g., CFE Treaty,
1991 Strategic Arms Reduction Treaty [START], and New START). The ªrst
scope condition for our theory is strategic arms control, which concerns high-
stakes military forces that have great implications for the balance of power.
The second scope condition for our theory is the strategic triangle,
wherein the motive to prevent, weaken, or break up a threatening alliance
may play an important role in driving arms control negotiations and out-
comes. Thus, our theory hinges on a fundamental characteristic of strategic
arms control in a strategic triangle—it is struck between rivals. That idea of co-
operation between rivals has, since the 1950s, fueled progress and debates in
arms control thought. To elucidate our theory, we ªrst give an overview of
those developments.
the standard model of strategic arms control

We begin with the model rooted in the pioneering work of Thomas Schelling,
Donald Brennan, and Hedley Bull.13 The fundamental premise of the standard
model is that adversaries can have “important interests of military policy in
common” and design measures to “realize the cooperation which their inter-
10. Jennifer L. Erickson, “Arms Control,” in Alexandra Gheciu and William C. Wohlforth, eds.,
Oxford Handbook of International Security (New York: Oxford University Press, 2018), pp. 399–414,
doi.org/10.1093/oxfordhb/9780198777854.013.26; and Coit D. Blacker and Gloria Duffy, eds.,
International Arms Control: Issues and Agreements, 2nd ed. (Redwood City, Calif.: Stanford Univer-
sity Press, 1984).
11. Timothy W. Crawford, “Arms Control and Arms Race,” International Encyclopedia of the Social
Sciences, 2nd ed. (Farmington Hills, Mich.: Macmillan Reference, 2008), p. 176.
12. Hedley Bull, “Strategic Arms Limitation: The Precedent of the Washington and London Naval
Treaties,” in Morton A. Kaplan, ed., SALT: Problems and Prospects (Morristown, N.J.: General
Learning, 1973), pp. 26–30.
13. Donald G. Brennan, “Setting and Goals of Arms Control,” in Donald G. Brennan, ed., Arms
Control, Disarmament, and National Security (New York: George Braziller, 1961), pp. 19–42;
Thomas C. Schelling, “Reciprocal Measures for Arms Stabilization,” in Brennan, ed., Arms Control,
Disarmament, and National Security, pp. 167–186; Hedley Bull, The Control of the Arms Race: Disarma-
ment and Arms Control in the Missile Age (New York: Praeger, 1961); and Thomas C. Schelling and
Morton H. Halperin, Strategy and Arms Control (New York: Twentieth Century Fund, 1961).
ests actually entail.”14 The counterintuitive point is that states can achieve
three common objectives (i.e., increased stability, limited war, and reduced ex-
penses) as direct effects of arms control measures15 through “military collabo-
ration with potential enemies.”16
The ªrst direct effect is increased stability, or a reduced likelihood of war.
Proponents of the standard model see arms control as promoting three levels
of stability. Arms control can promote “strategic stability” by helping states
“construct a military balance that in itself dissuades [them] from thinking
that they can use force effectively” against each other.17 It can promote
“arms race stability” by “tranquilizing” the action-reaction dynamic of com-
petitive buildups.18 And it can promote “crisis stability” through schemes that
“create ‘transparency’ and rules of the road that will reduce chances of acci-
dental escalation in a crisis confrontation.”19 More broadly, arms control talks
and agreements may foster stability by reducing states’ uncertainties about
“current and likely future forces, practices, and intentions.”20
The second direct effect of arms control is to limit a war’s scope and vio-
lence. States may forge agreements to constrain, eliminate, or reduce the collat-
eral damage from more destructive weapons, or to help end wars quickly.21
Saving money is the third direct effect, which states can accomplish by avoid-
ing an arms race that, absent an agreement, would otherwise occur.22 What all
three objectives have in common is to ªnd positive sum measures from which
both sides can directly gain. Veriªcation and transparency measures, such as
on-site inspections and notiªcation rules, can further cooperation. At a mini-
mum, methods for disclosing information make it easier for rivals to cooperate
14. Robert Jervis, “Arms Control, Stability, and Causes of War,” Political Science Quarterly, Vol. 108,
No. 2 (Summer 1993), p. 241, doi.org/10.2307/2152010.
15. Our theory focuses on motives that inform states’ decisions to pursue and conclude arms con-
trol. Other theories focus on exogenous factors that permit or preclude dealmaking. See, for exam-
ple, Charles L. Glaser, “Realists as Optimists: Cooperation as Self-Help,” International Security,
Vol. 19, No. 3 (Winter 1994/95), pp. 65–67, doi.org/10.2307/2539079; and Andrew J. Coe and Jane
Vaynman, “Why Arms Control Is So Rare,” American Political Science Review, Vol. 114, No. 2 (May
2020), pp. 342–355, doi.org/10.1017/S000305541900073X.
16. Schelling, “Reciprocal Measures,” p. 169; and Bull, Control of the Arms Race, p. 10.
17. Richard K. Betts, “Systems for Peace or Causes of War? Collective Security, Arms Control, and
the New Europe,” International Security, Vol. 17, No. 1 (Summer 1992), p. 30, doi.org/10.2307/
2539157.
18. Schelling and Halperin, Strategy and Arms Control, p. 169; and Bull, Control of the Arms Race,
pp. 4–12.
19. Betts, “Systems for Peace,” p. 38.
20. Albert Carnesale, Joseph S. Nye, and Graham T. Allison, “An Agenda for Action,” in Graham
T. Allison, Albert Carnesale, and Joseph S. Nye, eds., Hawks, Doves, and Owls: An Agenda for
Avoiding Nuclear War (New York: W.W. Norton, 1985), p. 243.
21. Schelling and Halperin, Strategy and Arms Control, pp. 17–24.
22. Bull, Control of the Arms Race, pp. 12–20; and Bernard Brodie, “On the Objectives of Arms Con-
trol,” International Security, Vol. 1, No. 1 (Summer 1976), p. 19, doi.org/10.2307/2538574.
despite persistent distrust. But such policies also may “play a broader and
more dynamic role than detecting violations” by signaling reciprocal reassur-
ances and constituting concrete channels for stabilizing behavior.23
the skeptic’s model of strategic arms control

During the heyday of superpower arms control between the 1960s and 1980s,
hawks not only opposed cooperating with the Soviets but also derided the
standard model’s elision of the more competitive (i.e., zero-sum) and political
motives of strategic arms control.24 Rivals, skeptics argue, use strategic arms
control to obtain or preserve relative military and political advantages over
one another.25 They use it to make opponents complacent and vulnerable, and
then to “negotiate from strength” for further gains.26 All this makes substantial
strategic arms control hard to achieve and sustain.27 For skeptics, veriªcation
is as much a problem as it is a solution for cooperation given underlying an-
tagonism and distrust. Rivals use transparency measures to pursue or protect
competitive advantages. They exploit veriªcation schemes to obtain sensitive
information about their enemy’s arsenal and promote deception about their
own.28 When arms control is needed, such as between serious rivals, it is often
unobtainable because transparency measures to monitor compliance could ex-
23. Schelling and Halperin, Strategy and Arms Control, p. 93; and Viet Phuong Nguyen and Man-
Sung Yim, “Building Trust in Nonproliferation: Transparency in Nuclear-Power Development,”
Nonproliferation Review, Vol. 24, No. 5–6 (2017), pp. 509–526, doi.org/10.1080/10736700.2018
.1448036.
24. On the skeptics’ perspective, see James H. Lebovic, Flawed Logics: Strategic Nuclear Arms Con-
trol from Truman to Obama (Baltimore, Md.: Johns Hopkins University Press, 2013); Keith L.
Shimko, Images and Arms Control: Perceptions of the Soviet Union in the Reagan Administration (Ann
Arbor: University of Michigan Press, 1992), pp. 123–148; Douglas Seay, “What Are the Soviets’ Ob-
jectives in Their Foreign, Military, and Arms Control Policies?” in Lynn Eden and Steven E. Miller,
eds., Nuclear Arguments: Understanding the Strategic Nuclear Arms and Arms Control Debates (Ithaca,
N.Y.: Cornell University Press, 1989), pp. 47–108; and Andrew Kydd, “Arms Races and Arms Con-
trol: Modeling the Hawk Perspective,” American Journal of Political Science, Vol. 44, No. 2 (April
2000), pp. 228–244, doi.org/10.2307/2669307.
25. Paul H. Nitze, “The 1985 Alastair Buchan Memorial Lecture: The Objectives of Arms Control,”
Survival, Vol. 27, No. 3 (1985), pp. 98, 104, doi.org/10.1080/00396338508442236; and Colin S. Gray,
House of Cards: Why Arms Control Must Fail (Ithaca, N.Y.: Cornell University Press, 1992), pp. 126,
141.
26. Sean M. Lynn-Jones, “Lulling and Stimulating Effects of Arms Control,” in Albert Carnesale
and Richard N. Haass, eds., Superpower Arms Control: Setting the Record Straight (Cambridge, Mass.:
Ballinger, 1987), p. 224; Schelling and Halperin, Strategy and Arms Control, p. 6; Carnesale, Nye,
and Allison, “An Agenda for Action,” p. 244; and Coral Bell, Negotiation from Strength: A Study in
the Politics of Power (New York: Alfred A. Knopf, 1963).
27. The paradox is that arms control is either “impossible” (between serious rivals) or “irrelevant”
(between friendly states): Gray, House of Cards, pp. 16–19.
28. Coe and Vaynman, “Why Arms Control Is So Rare”; Alexander Glaser, “Unmaking the Bomb:
Verifying Limits on the Stockpiles of Nuclear Weapons,” AIP Conference Proceedings, Vol. 1898,
No. 1 (2017), pp. 2–8, doi.org/10.1063/1.5009211; and Christine Comley et al., Conªdence, Security,
and Veriªcation: The Challenge of Global Nuclear Weapons Arms Control (Reading, UK: Atomic
Weapons Establishment, 2000), pp. 5–6.
pose the parties to increased security risks.29 When arms control negotiations
and deals do occur, skeptics expect power politics, rather than more ªne-
grained technical issues, to drive negotiations.30
Both models emphasize that direct effect motives—whether cooperative or
competitive—shape the arms control “game” itself.31 This afªnity between the
two models becomes overt in synthetic accounts of strategic arms control that
depict rivals simultaneously seeking to harness both cooperative and competi-
tive direct effects.32 Our theory extends the idea of arms control as a way to
achieve competitive advantage, but it shifts the focus from two states and di-
rect effects to three states and indirect effects.
Applying Wedge Strategy Theory to Arms Control
Scholars from both the standard and the skeptical perspectives recognize that
strategic arms control can create indirect effects in other areas of the states’ rela-
tionships that are “outside” the military game. Some have shown how states use
negotiations to simultaneously link nonmilitary issues and arms control goals to
overcome deadlocks by shifting the game from zero-sum to positive sum.33 Oth-
ers have explained how states use arms control to send diplomatic signals, facili-
tate broader political cooperation, and reinforce their systemic privileges.34 In
our theory, the intent to reap such indirect effects can be a potent driver of ne-
gotiations and deals.35
29. Coe and Vaynman, “Why Arms Control Is So Rare.”

30. Arms control “will be negotiated to suit a political agenda and very much in political terms
[by] real live politicians, who at the very highest of levels neither know nor care about the ªfty-
seven or more varieties of stability.” Gray, House of Cards, p. 118. On the primacy of political calcu-
lations in arms control, see also Goldman, Sunken Treaties, p. 7; Robert Gordon Kaufman, Arms
Control during the Pre-Nuclear Era: The United States and Naval Limitation between the Two World Wars
(New York: Columbia University Press, 1990), pp. 4, 196; Marshall D. Shulman, “Arms Control in
an International Context,” in Franklin A. Long and George W. Rathjens, eds., Arms, Defense Policy,
and Arms Control (New York: W.W. Norton, 1976), pp. 54–55; Bull, Control of the Arms Race, pp. 65–
76; and Salvador de Madariaga, Disarmament (New York: Coward-McCann, 1929), p. 109, also
pp. 38–56.
31. Schelling and Halperin, Strategy and Arms Control, p. 6.
32. See, especially, John D. Maurer, “The Purposes of Arms Control,” Texas National Security
Review, Vol. 2, No. 1 (2018), pp. 8–27, doi.org/10.26153/tsw/870.
33. George W. Downs, David M. Rocke, and Randolph M. Siverson, “Arms Races and Coopera-
tion,” World Politics, Vol. 38, No. 1 (October 1985), pp. 127–132, doi.org/10.2307/2010353.
34. Brennan, “Setting and Goals,” pp. 39–41. Also see Schelling and Halperin, Strategy and Arms
Control, p. 44; Hedley Bull, “Arms Control and World Order,” International Security, Vol. 1, No. 1
(Summer 1976), pp. 6–8, doi.org/10.2307/2538573; Andrew J. Coe and Jane Vaynman, “Collusion
and the Nuclear Proliferation Regime,” Journal of Politics, Vol. 77, No. 4 (October 2015), pp. 983–
997, doi.org/10.1086/682080; Patrick M. Morgan, “Elements of a General Theory of Arms Con-
trol,” in Paul R. Viotti, ed., Conºict and Arms Control: An Uncertain Agenda (New York: Routledge,
1986), p. 286; and Charles L. Glaser, Analyzing Strategic Nuclear Policy (Princeton, N.J.: Princeton
University Press, 1990), pp. 316, 329–337.
35. Brennan, “Setting and Goals,” p. 39. On U.S. pursuit of “linkage” in détente-era arms control,
Policymakers may employ a form of statecraft to obtain an indirect effect,
and arms control, like any other security policy, also can be used in this way.
Thus, although states’ concerns about stability and arms racing may provide
some of the impetus for arms control talks, leaders may only conclude impor-
tant agreements because the negotiating process enables them to pursue and
obtain highly valued nonmilitary payoffs. It is common to ªnd not only that
multiple motives lie behind important decisions but also that the desire to
achieve indirect effects weighs heavily in those choices.36 Prime Minister
Winston Churchill’s dramatic decision in July 1940 to attack the French ºeet in
Oran, Algeria, for example, was meant not only to ensure that the Germans
did not capture the ships but also to “make a profound impression in every
country” (the United States especially) of Britain’s determination to ªght on.37
In short, a policy’s indirect effects may be as signiªcant as the direct ones in the
overall calculus.38
Cold War analysts highlighted an indirect effect of arms control that is espe-
cially pertinent to this point. They warned that U.S.-Soviet arms control could
precipitate ªssures in the Western alliance because weaker North Atlantic
Treaty Organization (NATO) allies would fear both U.S.-Soviet collusion
and the erosion of the U.S. commitment to European defense.39 Hawks sur-
mised that Soviet arms control diplomacy thus had an additional grand
strategic motive—to “exacerbate dissension within the Western alliance.”40
Their arguments were not based in a general concept or theory of arms control
wedge strategies.41 Here, our theory intervenes.
see Dan Caldwell, American-Soviet Relations: From 1947 to the Nixon-Kissinger Grand Design
(Westport, Conn.: Greenwood, 1981), pp. 86–90; and John Lewis Gaddis, Strategies of Containment:
A Critical Appraisal of American National Security Policy during the Cold War, rev. ed. (New York:
Oxford University Press, 2005), pp. 292, 309–310.
36. For examples, see Jon Western, “Sources of Humanitarian Intervention: Beliefs, Information,
and Advocacy in the U.S. Decisions on Somalia and Bosnia,” International Security, Vol. 26, No. 4
(Spring 2002), pp. 112–142, doi.org/10.1162/016228802753696799; Derek Chollet, The Road to the
Dayton Accords: A Study of American Statecraft (New York: Palgrave Macmillan, 2005), p. 5; and
Gaddis, Strategies of Containment, pp. 37, 65.
37. Winston S. Churchill, Their Finest Hour (Boston: Houghton Mifºin, 1949), p. 238. Thanks to
Robert Jervis for suggesting this example.
38. Robert Jervis, System Effects: Complexity in Political and Social Life (Princeton, N.J.: Princeton
University Press, 1997), pp. 29–33; and Fred Charles Iklé, How Nations Negotiate (New York:
Praeger, 1964), pp. 42–43, 55–58.
39. Edward N. Luttwak, “SALT and the Meaning of Strategy,” Washington Quarterly, Vol. 1, No. 2
(1978), p. 26, doi.org/10.1080/01636607809450219; Schelling and Halperin, Strategy and Arms Con-
trol, pp. 46–47; and Jane M.O. Sharp, “Arms Control and Alliance Commitments,” Political Science
Quarterly, Vol. 100, No. 4 (Winter 1985/86), p. 650, doi.org/10.2307/2151545.
40. Seay, “What Are the Soviets’ Objectives,” pp. 96, 105.
41. On developing policy-applicable theory, see Alexander L. George and Andrew Bennett, Case
Studies and Theory Development in the Social Sciences (Cambridge: Massachusetts Institute of Tech-
nology Press, 2005), pp. 263–285; and Daniel Byman and Matthew Kroenig, “Reaching beyond the
Ivory Tower: A How To Manual,” Security Studies, Vol. 25, No. 2 (2016), pp. 289–319, doi.org/
10.1080/09636412.2016.1171969.
Our theory starts with a basic means-end calculus—A’s direct policy toward
B is meant to have an indirect effect on C.42 It posits that A may pursue strate-
gic arms control with B in part to advance a wedge strategy that isolates C by
weakening alignment between B and C. What we mean by “weaken align-
ment” is captured in Glenn H. Snyder’s “expectations” concept, which defines
alignments as “patterns of expectations and intentions [among states] regard-
ing future support or opposition” in “disputes or wars with…other states.”43
Thus, a wedge strategy weakens alignment between B and C when it lowers
expectations that they will support each other in future conflict.
The wedge strategy motive does not always inform arms control decisions
and outcomes. Our theory posits that it sometimes does in signiªcant ways.
Different states approach arms control with their own motives and goals. Even
if a state seeks a deal to drive a wedge, the deal will not happen if the other(s)
don’t want it for their own reasons. And even if the wedge effect looms large
as a beneªt to that state, it is just a part of the overall bargained outcome con-
cerning that state’s security.44 For the wedge motive to be signiªcant in a case,
then, at least one party pushing the process (A) must seek to promote a wedge
effect. That indirect motive need not be A’s only or most important goal in the
process, but it must be important enough to lead A’s decision-makers to favor
the process and deal.45
Understanding strategic arms control’s potential to divide opponents im-
proves our grasp of wedge strategy in power politics.46 In general, wedge
strategies arise in any triangular competition in which one state’s strength or
survival is endangered by the collusion of two others. In such circumstances,
“states seek to reduce the menace to themselves by sowing discord among po-
tential adversaries” because they “gain security as well as bargaining power
when [their] main adversaries are themselves divided.”47 Although wedge
strategies can be either accommodative or coercive, this article focuses on the
ªrst category. When facing two potential adversaries simultaneously, it is often
better to accommodate one instead of confronting both.48 And arms control is
42. Jervis, System Effects, pp. 31–32.

43. Glenn H. Snyder, “Process Variables in Neorealist Theory,” Security Studies, Vol. 5, No. 3 (1996),
p. 174, doi.org/10.1080/09636419608429279; and Glenn H. Snyder, “Alliance Theory: A Neorealist
First Cut,” Journal of International Affairs, Vol. 44, No. 1 (Spring/Summer 1990), p. 105, https://
www.jstor.org/stable/24357226.
44. Brennan, “Setting and Goals,” pp. 34–37.
45. Downs, Rocke, and Siverson, “Arms Races and Cooperation,” pp. 130–132.
46. On investigating wedge mechanisms, see Stacie E. Goddard and Daniel H. Nexon, “The
Dynamics of Global Power Politics: A Framework for Analysis,” Journal of Global Security Studies,
Vol. 1, No. 1 (February 2016), pp. 13–15, doi.org/10.1093/jogss/ogv007.
47. Jervis, System Effects, pp. 177–179, 282.
48. Crawford, “Preventing Enemy Coalitions,” pp. 160–162; Yasuhiro Izumikawa, “To Coerce or
Reward? Theorizing Wedge Strategies in Alliance Politics,” Security Studies, Vol. 22, No. 3 (2013),
a way to accommodate rather than coerce. Selective accommodation “singles
out” a target state for “preferential treatment” to keep or lure it away from
alignment with a more dangerous one.49 Thus, in any selective accommoda-
tion scenario, there are at least three actors: the divider (A), and two others
(B and C) that are (or are becoming) aligned against A. One of them (B) is the
target of accommodation; the other (C) is the adversary to be isolated. Strategic
arms control—as the standard model implies—can be an important channel
for such accommodation, insofar as B values the military and political beneªts
that a deal with A may provide.
In general, an accommodative wedge strategy can generate division in three
ways.50 The ªrst way involves a speciªc, contingent exchange: A offers bene-
ªts to B if B will, in return, isolate C. The second way exacerbates conºicts:
A’s accommodative bid creates or elevates issues in dispute between B and C.
The third way involves reassurance: A accommodates B to dilute common
threat perceptions that cause B and C to align against A.51 These three ways of
dividing are not, in general, mutually exclusive, but when they are distilled
into speciªc mechanisms of arms control wedge strategy, we expect each
mechanism to dominate in a strategic context that is unconducive to the
other two.
Next, we show how these three ways of dividing adversaries translate into
more speciªc mechanisms of arms control wedge strategy. We show how A can
use arms control to weaken B–C alignment through a contingent exchange
(speciªc linkage), or by exacerbating conºict between B and C (catalyze con-
ºict), or by changing B’s threat perception (peace offensive). Table 1 depicts
these arms control wedge strategy mechanisms and highlights the conceptual
ties between these mechanisms and more general theoretical images of cohe-
sion. Each image captures a distinct reason why alliances—or, less formally,
alignments—form or deform, and therefore implies a different wedge mecha-
nism. Thus, the three wedge mechanisms differ in their internal logics and in
the kinds of strategic situations in which they tend to occur.
pp. 498–531, doi.org/10.1080/09636412.2013.816121; and Daniel Treisman, “Rational Appease-

ment,” International Organization, Vol. 58, No. 2 (Spring 2004), pp. 345–373, https://www.jstor.org/
stable/3877861.
49. George Liska, Nations in Alliance: The Limits of Interdependence (Baltimore, Md.: Johns Hopkins
Press, 1962), pp. 188–189; Frederick H. Hartmann, The Conservation of Enemies: A Study in Enmity
(Westport, Conn.: Greenwood, 1982), p. 147; Glenn H. Snyder, Alliance Politics (Ithaca, N.Y.: Cornell
University Press, 1997), p. 337; and Crawford, Power to Divide, pp. 9–11.
50. Crawford, Power to Divide, p. 5.
51. Jeffrey W. Knopf, “Varieties of Assurance,” Journal of Strategic Studies, Vol. 35, No. 3 (2012),
383–387, doi.org/10.1080/01402390.2011.643567.
Table 1. Arms Control Wedge Strategy Mechanisms
Cohesion Alignment
Mechanisms Images Trajectories Processes of Division
speciªc bidding B and C extending A’s bid alters B’s alignment with C
linkage war status quo via direct exchange
catalyze team of B and C A’s bid elevates conºicting
conºict rivals diverging interest between B and C
peace balance B and C A’s bid promotes discrepant threat
offensive of threat converging perceptions between B and C
speciªc linkage mechanism

This ªrst mechanism through which arms control can generate division entails
a direct exchange: A offers relatively attractive rewards to B (i.e., arms limita-
tion) contingent upon B’s compliance with A’s demand that B alters its align-
ment with C (see table 1). This mechanism stems from a “bidding war” image
of cohesion, which depicts alignments as products of delicate calculations of
costs and beneªts that are sensitive to competitive bids. Here, “a small shift in
. . . cost-beneªt ratios can turn neutrals into allies, allies into neutrals, and al-
lies into enemies,” and a credible, well-timed “side-payment may determine
with which of two potential . . . coalitions a state may line up.”52 Absent such
bids, the image suggests that alignments will maintain the status quo. But in
the context of tight alignment competition, states’ positions reºect attempts to
obtain the best payoff from at least two alignment options. A more competitive
offer can sway a state’s allegiance. Speciªc linkage has three key indicators.
First, B and C are on a path to extend the status quo of their partnership.
Second, A’s agreement to limit its arms or military posture relative to B is—or
is central to—the beneªt to B’s interest in arms reduction on which the deal
pivots. Also, the exchange is explicit: A’s demand that B reciprocate by for-
mally altering its alignment with C is embedded in the negotiations leading to
the arms control agreement. Third, B’s leaders comply with A’s demand to se-
cure the arms limit beneªts embodied in the deal.
52. Julian R. Friedman, “Alliance in International Politics,” in Julian R. Friedman, Christopher

Bladen, and Steven Rosen, eds., Alliance in International Politics (Boston: Allyn and Bacon, 1970),
p. 18; and Bruce Russett, “Components of an Operational Theory of Alliance Formation,” in Fried-
man, Bladen, and Rosen, eds., Alliance in International Politics, p. 252. Also see John A. Vasquez,
Contagion and War: Lessons from the First World War (Cambridge: Cambridge University Press,
2018), pp. 32–36; and Scott Wolford, The Politics of Military Coalitions (Cambridge: Cambridge Uni-
versity Press, 2015), pp. 52–70.
catalyze conºict mechanism
In this second mechanism, A negotiates an arms control agreement with B to
exacerbate tensions between B and C (see table 1). This mechanism stems from
an image of cohesion that depicts allies as “intimate enemies,” “warring
friends,” or a team of rivals.53 Here, even as they organize against others, part-
ners try to shirk and shift burdens of cooperation, and they try to control,
restrain, and compete against one another. In such alliances, we ªnd both com-
pelling reasons to cooperate and combustible conºicting interests. Recognizing
the latter, a divider may, as Robert Jervis notes, manipulate issues that “create
conºict among those who might otherwise come together.”54 A may press for
an arrangement to a dispute that undermines a prior bargain between B and C
by advantaging one more than the other; or, A may endorse a position that is
preferred by B but opposed by C. The catalyze conºict arms control mecha-
nism thus implies three observable indicators. First, with frictions already
straining alignment between B and C, A pursues arms control with B, intend-
ing to worsen the divergence. Second, A tries to fashion the deal with B in
ways that will achieve this divergence. A may, for example, advance an
arms control deal with B that “takes its side” in a dispute with C, prompting
C’s backlash against a deal “made at [its] expense.”55 Even if the terms of the
A–B deal do not directly impinge C’s military priorities, the terms may reduce
A’s military pressure on B such that B is freer to focus on its conºicts with C.
Either way, C will perceive B’s cooperation with A as betrayal. Third, the con-
duct of the A–B arms control process and/or sealing of a deal does, in fact, ac-
celerate growing antagonism between B and C.
peace offensive mechanism

The premise of this third mechanism is that the accommodative function of
arms control can help lower the parties’ threat perceptions. Its causal logic
springs from the well-known “balance of threat” image of cohesion.56 Keeping
to the strategic triangle, the cause of alignment between B and C is the danger
that A poses to them both. This implies the pathway to dis-unity—discrepancy
53. See Patricia A. Weitsman, “Intimate Enemies: The Politics of Peacetime Alliances,” Security
Studies, Vol. 7, No. 1 (1997), pp. 156–193, doi.org/10.1080/09636419708429337; Jeremy Pressman,
Warring Friends: Alliance Restraint in International Politics (Ithaca, N.Y.: Cornell University Press,
2008). Also see Paul W. Schroeder, “Alliances 1815–1945: Weapons of Power and Tools of Manage-
ment,” in David Wetzel, Robert Jervis, and Jack S. Levy, eds., Systems, Stability, and Statecraft: Essays
on the International History of Modern Europe (New York: Palgrave Macmillan, 2004), pp. 195–222;
Evan N. Resnick, Allies of Convenience: A Theory of Bargaining in U.S. Foreign Policy (New York: Co-
lumbia University Press, 2019); and Thomas J. Christensen, Worse than a Monolith: Alliance Politics
and Problems of Coercive Diplomacy in Asia (Princeton, N.J.: Princeton University Press, 2011).
54. Jervis, System Effects, p. 190, also p. 189, n. 44.
55. Iklé, How Nations Negotiate, p. 57.
56. Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987), p. 32.
between B’s and C’s perceptions of the threat posed by A.57 Thus, with a
“peace offensive,” A uses conciliatory diplomacy to alleviate B’s insecurity in
a way that weakens B’s alignment with C. Likewise, with an arms control
peace offensive, A can use its agreement to limit arms with B to diminish B’s
perception of the danger that A poses (see table 1).58 This creates (or reinforces)
a gap between B’s perception of A’s threat and C’s perception of A’s threat. For
the peace offensive mechanism, process-tracing should reveal three indicators.
First, B and C are converging against A because of a shared perception of A’s
threat. Second, that convergence endangers A’s security or interests. Third,
A recognizes or anticipates this process and defuses it by pursuing a strategic
arms control agreement that is beneªcial to B’s security.
Case Study Method and Research Design
We address three key matters in each of our three case studies. First, we show
that wedge motive was an important ex ante objective of the arms control ne-
gotiation for decision-makers on one side of the arms control process. A’s lead-
ers sought through strategic arms control with B to weaken alignment between
B and C. Second, we demonstrate that A’s leaders implement the strategy fol-
lowing one of the three mechanisms described in table 1. Third, we describe
the wedge effect, showing that the strategic arms control deal weakened ex-
pectations of mutual support between B and C, via the process speciªed by the
primary mechanism in each case.
Owing to the nature of both the theory itself and the theory-development
enterprise, process-tracing is essential in our case studies.59 States may pur-
sue strategic arms control for several reasons simultaneously. In such circum-
stances of equiªnality (i.e., when multiple motives or mechanisms lead to
the same outcome), process-tracing allows us to differentiate leaders’ motives
(e.g., to drive wedges or secure certain political outcomes), to show how those
motives inºuence leaders’ approaches to the strategic arms control process,
and to observe a wedge effect (i.e., weakened alignment expectations). Finally,
process-tracing is essential to developing a typological theory that delineates
different arms control wedge strategy conªgurations.60
The three landmark cases of twentieth-century strategic arms control in this
57. Ole R. Holsti, “Alliances: Political,” in Neil J. Smelser and Paul B. Bates, eds., International
Encyclopedia of the Social and Behavioral Sciences (Oxford: Pergamon, 2001), p. 399.
58. Lynn-Jones, “Lulling and Stimulating Effects.”
59. Jack S. Levy, “Case Studies: Types, Designs, and Logics of Inference,” Conºict Management and
Peace Science, Vol. 25, No. 1 (2008), pp. 5–6, doi.org/10.1080%2F07388940701860318.
60. On process-tracing and relational mechanisms, see George and Bennett, Case Studies and The-
ory Development, pp. 205–262; and Charles Tilly, “Mechanisms in Political Processes,” Annual
Review of Political Science, Vol. 4 (2001), pp. 24–26, doi.org/10.1146/annurev.polisci.4.1.21.
article span pre-nuclear and nuclear eras and include the full range of B–C
alignment trajectories: extending the status quo, diverging, or converging (see
table 1). Thus, each logic of division is demonstrated through causal process
observations in a corresponding case.
Washington Naval Conference and the Anglo-Japanese Alliance
In December 1921, the United States convened the Washington Naval

Conference, where it led negotiations that produced the Five-Power Washington
Naval Treaty in which Britain, Japan, and the United States (along with France
and Italy as signatories) agreed to stop battleship building for ten years and to
maintain (among the ªrst three powers) the 5:5:3 ratio that then characterized
the balance of capital ship strength among them. Most accounts of the origins
of the naval treaty emphasize direct effect motivations (e.g., avoid an arms
race, save money, and preserve stability) for the major participants.61 A combi-
nation of conducive conditions made it possible to complete the agreement.
First, a highly motivated broker with strong leverage led the negotiations.
Second, all parties expected that unilateral veriªcation of compliance with the
main terms would be easy and unthreatening.62 Certainly for the Warren G.
Harding administration, avoiding a costly naval arms race and the associated
instability were major goals. The administration faced domestic demands to
cut defense spending and to increase international peace efforts.63 Some schol-
ars suggest that the Washington Naval Conference prevented a naval arms
race and built a framework of great power cooperation that fostered stability
in East Asia.64 In contrast, we argue that the Harding administration had a
wedge motive, which was one key reason why it promoted and embraced the
Five-Power Treaty.
This case shows the United States pursuing a wedge goal through the spe-
ciªc linkage mechanism. While the Five-Power Treaty was a necessary focal
point of the Washington Naval Conference, the deal hinged on broader agree-
ments on the management of great power relations.65 The United States did
want the direct beneªts of arms limits. It also wanted to dismantle the Anglo-
61. Brodie, “Objectives of Arms Control,” p. 34; and Christopher Hall, Britain, America, and Arms
Control, 1921–37 (Basingstoke, UK: Macmillan, 1987), p. 218.
62. Coe and Vaynman, “Why Arms Control Is So Rare,” p. 350; and Kaufman, Arms Control,
pp. 56–57, 96–108.
63. Thomas H. Buckley, The United States and the Washington Conference, 1921–1922 (Knoxville: Uni-
versity of Tennessee Press, 1970), pp. 9–19.
64. John H. Maurer, “Arms Control and the Washington Conference,” in Erik Goldstein and John
Maurer, eds., The Washington Conference, 1921–22: Naval Rivalry, East Asian Stability, and the Road to
Pearl Harbor (Portland, Oreg.: Frank Cass, 1994), pp. 267–268, 286–289. Also see Blacker and Duffy,
International Arms Control, pp. 89–92; and Kaufman, Arms Control, p. 47.
65. Downs, Rocke, and Siverson, “Arms Races and Cooperation,” pp. 130–132.
Japanese Alliance (formed in 1902), which was up for renewal in 1921.66 This
wedge motive was one of the key “political considerations” that, “far more
than economic necessity or the demands of strategy and weapons technology,”
drove U.S. diplomacy at the Washington Naval Conference.67 In late May 1921,
the British government, responding to hints from the U.S. Secretary of State,
Charles E. Hughes, encouraged the United States to propose an effort to
arrange a tripartite common Paciªc policy. Britain hoped that under those aus-
pices they might draw the United States into a refashioned Anglo-Japanese-
U.S. alliance.68 Instead, the United States convened a conference that joined
Paciªc policy and naval arms control and exploited the linkage between them
to abolish the Anglo-Japanese Alliance.69
motive
Although the United States and Japan fought on the same side in World War I,
its end left them with serious conºicts. The Anglo-Japanese Alliance seemed
to blame. Because of this alliance, Japan gained from the war control over
Germany’s Paciªc islands north of the equator. Its alliance with Britain had
also buffered U.S. opposition to Japan’s coercive diplomacy against China
in the 1915 “Twenty-One Demands” crisis.70 From the United States’ perspec-
tive, the Anglo-Japanese Alliance emboldened Japan’s adventurism in the Far
East and its designs for U.S. positions in the Philippines, Guam, and Hawaii.71
In the months before the Washington Naval Conference, U.S. ofªcials thus
“strove with growing intensity of purpose to drive a wedge into the Anglo-
Japanese Alliance.”72
The United States was unambiguous about its opposition to renewal
66. On Anglo-Japanese negotiations about renewing the alliance, see Documents on British Foreign
Policy, 1919–1939 (hereafter DBFP), First Series, Vol. 14, in Rohan Butler and J.P.T. Bury, eds., Far
Eastern Affairs, April 1920–February 1922 (London: Her Majesty’s Stationary Ofªce, 1966), pp. 287–
322.
67. Roger Dingman, Power in the Paciªc: The Origins of Naval Arms Limitation, 1914–1922 (Chicago:
University of Chicago Press, 1976), p. 140.
68. Ira Klein, “Whitehall, Washington, and the Anglo-Japanese Alliance, 1919–1921,” Paciªc His-
torical Review, Vol. 41, No. 4 (1972), p. 477, n. 54, doi.org/10.2307/3638396; Malcolm D. Kennedy,
The Estrangement of Great Britain and Japan, 1917–35 (Berkeley: University of California Press, 1969),
p. 54; and Marquess Curzon of Kedleston to Sir C. Eliot (Tokyo), July 8, 1921, DBFP, pp. 331–334.
69. Nicholas J. Spykman, America’s Strategy in World Politics: The United States and the Balance of
Power (New York: Harcourt, 1942), p. 173.
70. On the crisis, see Timothy W. Crawford, “The Strategy of Coercive Isolation,” in Kelly M.
Greenhill and Peter Krause, eds., Coercion: The Power to Hurt in International Politics (New York: Ox-
ford University Press, 2018), pp. 234–238.
71. Raymond L. Buell, The Washington Conference (New York: D. Appleton, 1922), pp. 103, 120–122;
Mark Sullivan, The Great Adventure at Washington: The Story of the Conference (Garden City, N.Y.:
Doubleday, Page, 1922), p. 230; and Zoltán I. Búzás, “The Color of Threat: Race, Threat Perception,
and the Demise of the Anglo-Japanese Alliance (1902–1923),” Security Studies, Vol. 22, No. 4 (2013),
pp. 573–606, doi.org/10.1080/09636412.2013.844514.
72. Sprout and Sprout, Toward a New Order, pp. 92–93; Klein, “Whitehall,” pp. 462, 468–469; and
of the Anglo-Japanese Alliance. As Aukland Geddes, the British Ambassador
to the United States, warned British Foreign Secretary Alfred Curzon on
June 6, Hughes thought that renewal would be “disastrous.”73 Later that
month, Hughes elaborated this point in a conversation with Geddes. The
United States “had very clear policies in the Far East.” It had embraced
“the ‘Open Door’ policy and the integrity of China, and now . . . the integrity
of Russia.” Consequently, “if Great Britain and Japan had any arrangement by
which Great Britain was to support the special interests of Japan, the latter
might be likely . . . to take positions which” challenged U.S. interests. Such
a situation, Hughes warned, “would be fraught with mischief.” It would
be far better “for the United States to ªnd complete support [from] Great
Britain.” This attitude was not “antagonistic to Japan,” Hughes averred, be-
cause it “would be in her interests as in the interests of the peace of the
world.”74 In the context of this exposition, Hughes suggested that a three-way
understanding of common Paciªc policies would be better than a renewed
Anglo-Japanese Alliance.75
In early July, signals of U.S. opposition intensiªed. Hughes’ conversations
with Geddes made it clear that “the mere fact of renewal, in whatever form,
would create a very unfavorable impression” with the U.S. public and govern-
ment.76 Under instructions from Harding and Hughes, the U.S. ambassador to
Britain, George Harvey, conªrmed to Curzon that “without fail [the] United
States government would view with regret the renewal or establishment of
any special relation of cooperation or partnership between the British Empire
and Japan.” In this context of strident warnings, Geddes advised Curzon that
“opinion on the subject of naval disarmament has tended to crystallize round
the termination of Anglo-Japanese Alliance.” Furthermore, Geddes observed
that a certain idea of how to end the alliance was forming in Washington—to
replace the alliance with a “tripartite declaration of common policy in the
Alfred Whitney Griswold, The Far Eastern Policy of the United States (New Haven, Conn.: Yale Uni-
versity Press, 1938), p. 275.
73. Sir A. Geddes (Washington) to Earl Curzon, June 6, 1921, DBFP, pp. 300–301.
74. Memorandum of a Conversation between the Secretary of State and the British Ambassador
(Geddes), June 23, 1921, Foreign Relations of the United States (FRUS), 1921, Vol. 2 (Washington,
D.C.: Government Printing Ofªce [GPO], 1936), pp. 314–315. Geddes’s reports to Curzon about
this conversation were incisive. There could be “no doubt” that the United States regarded “even
the thought of renewal of [the Anglo-Japanese] treaty as evidence” of unfriendly British inten-
tions. Hughes “undoubtedly believe[d] that existence of an Anglo-Japanese alliance makes Japan
less tractable and therefore tends to make an American-Japanese war less easy to avoid.” Sir A.
Geddes (Washington) to Earl Curzon, June 24, 1921, DBFP, pp. 310–311; and Sir A. Geddes (Wash-
ington) to Earl Curzon, June 24, 1921, DBFP, p. 312.
75. Memorandum of a Conversation between the Secretary of State and the British Ambassador
(Geddes), June 23, 1921, FRUS, 1921, Vol. 2, p. 315.
76. Sir A. Geddes (Washington) to the Marquess Curzon of Kedleston, July 2, 1921, DBFP, p. 320;
and Sir A. Geddes (Washington) to the Marquess Curzon of Kedleston, July 6, 1921, DBFP, p. 326.
Paciªc” to which a deal on “limitation of armaments in [the] Paciªc might
be added.”77
speciªc linkage mechanism

Indeed, the United States intended to directly link the arms control and the tri-
partite common Paciªc policy projects. Hughes informed Curzon that confer-
ences for both projects “ought to be held in Washington [and] that both were
indispensable parts of the same whole (emphasis added).”78 The Harding ad-
ministration’s position regarding naval disarmament, which was the focus of
the Washington Naval Conference negotiations, afforded the United States
“extraordinary bargaining power” to advance other political objectives. As
Harold and Margaret Sprout explain, this leverage derived “mainly from [its]
possession of sixteen ‘post-Jutland’ capital ships, built and building. . . . Com-
pletion of the ships on schedule would give the United States by far the strong-
est battle line in the world for a considerable number of years at least. . . . With
the United States thus setting the pace of naval construction . . . any proposal
put forward by the American delegation at the [Naval] conference was bound
to receive serious consideration.”79
Britain wanted a way out of the looming naval arms race, and so did
Japan.80 “Our most urgent object,” said Churchill in June 1921, “was to avoid a
naval rivalry between Great Britain and the United States.”81 To keep pace
with the U.S. buildup (which Britain felt it must do absent an agreement)
“would require a ªnancial outlay which Great Britain’s war-weakened and se-
riously depressed economy was then in no condition to bear.”82 Britain was
struggling to recover economically from World War I, and it would soon need
77. Sir A. Geddes (Washington) to the Marquess Curzon of Kedleston, July 7, 1921, DBFP, p. 328.
78. Marquess Curzon of Kedleston to Sir A. Geddes (Washington), July 14, 1921, DBFP, p. 342.
79. Sprout and Sprout, Toward a New Order, pp. 138–139.
80. On British urgency, see Erik Goldstein, “The Evolution of British Diplomatic Strategy for the
Washington Conference,” in Erik Goldstein and John Maurer, eds., The Washington Conference,
1921–22: Naval Rivalry, East Asian Stability, and the Road to Pearl Harbor (Portland, Oreg.: Frank
Cass, 1994), pp. 14–15; and John Maurer, “Arms Control and the Washington Conference,” in
Goldstein and Maurer, eds., The Washington Conference, 1921–22, pp. 272–274. On Japan’s interest,
see Sadao Asada, “From Washington to London: The Imperial Japanese Navy and the Politics of
Naval Limitation, 1921–30,” in Goldstein and Maurer, eds., The Washington Conference, 1921–22,
p. 152; Morinosuke Kajima, ed., The Diplomacy of Japan, 1894–1922, Vol. 3: First World War, Paris
Peace Conference, Washington Conference (Tokyo: Kajima Institute of International Peace, 1980),
pp. 444–445, 459–460; Ichihashi Yamato, The Washington Conference and After, A Historical Survey
(Stanford, Calif.: Stanford University Press, 1928), p. 4, ns. 7, 19; and Sir C. Eliot (Tokyo) to the
Marquess Curzon of Kedleston, August 12, 1921, DBFP, pp. 372–373.
81. Quoted in Goldstein, “Evolution of British Diplomatic Strategy,” p. 15.
82. Sprout and Sprout, Toward a New Order, p. 123. In May 1921, Churchill warned the cabinet that
a naval arms race with the United States “would pit America against Japan, draw in the British
because of their intention to maintain a one-power standard, and then escalate because the Ameri-
cans would build against an Anglo-Japanese combination.” Klein, “Whitehall,” p. 475.
to repay war debts to the United States. It was obvious, then, that Britain
needed the United States “to make a concession and to abandon her intention
to build a great navy” (as the cabinet put it in July 1921). “In all this business,”
remarked Churchill, “the United States have a great deal to give and a great
deal more to withhold.”83 One last facet of the political context ampliªed U.S.
leverage: Britain feared looking desperate and losing prestige if it appeared to
initiate the effort to negotiate naval limits. Instead, it wanted the United States
to call for an arms conference and organize formal talks.84
This delicate political context enabled the United States to set the confer-
ence agenda and impose its speciªc linkage strategy. Its approach was, as one
British observer put it, “to accept no [naval arms] treaty unless the [Anglo-
Japanese] alliance was scrapped and to out-build all others in naval arma-
ments if no treaty was forthcoming.”85 As the Sprouts explain, the linkages in
this bargaining position were indeed credible: “[R]enewal of the Alliance . . .
would certainly have given fresh impetus to navalism in America and just
as certainly would have put naval limitation beyond the sphere of practi-
cal politics.”86
On July 5, Curzon encouraged Harding to convene a conference of “powers
directly concerned . . . to consider all essential matters bearing upon Far East
and Paciªc Ocean with a view to arriving at a common understanding de-
signed to assure settlement by peaceful means, the elimination of naval war-
fare, consequent elimination of naval arms, etc.”87 On July 7, Harvey informed
Curzon that the United States would welcome any British suggestions “which
have as their object, the replacement of [the] Anglo-Japanese Alliance by [a]
British-American-Japanese declaration of common policy.”88 At the 1921
Imperial Conference held later that month in London, Curzon explained to the
Dominions’ ministers that the United States meant “to put us in a position
where we could, without disloyalty to our Allies, terminate the Anglo-
Japanese Agreement and substitute something for it, probably of a tripar-
tite character.”89
In September, Hughes reinforced to the British that it would be essential to
83. Cabinet and Churchill quotes found in Goldstein, “Evolution of British Diplomatic Strategy,”
pp. 15, 17.
84. Sprout and Sprout, Toward a New Order, p. 124.
85. Kennedy, Estrangement of Great Britain, p. 54.
87. Ambassador in Great Britain (Harvey) to the Secretary of State, July 8, 1921, FRUS, 1921, Vol. 1
(Washington, D.C.: GPO, 1936), pp. 19–21.
88. Sir A. Geddes (Washington) to the Marquess Curzon of Kedleston, July 7, 1921, DBFP, p. 328.
89. Memorandum by Mr. Lampson as to whether the Anglo-Japanese Alliance should be directly
discussed at the Washington Conference, August 18, 1921, DBFP, p. 381.
deal with the Anglo-Japanese Alliance at the Washington Naval Conference.90
The critical issue was the obligations of military support entailed by the alli-
ance. In October, a British Foreign Ofªce memorandum assessed the solution
favored by the United States: if “a Tripartite Agreement is to be substituted for
the Anglo-Japanese Alliance Agreement it will be necessary to eliminate
from the new instrument military commitments . . . for otherwise the United
States can never be induced to become a party. Stripped of military clauses, the
Anglo-Japanese Agreement loses its character as an alliance and becomes
merely a declaration of policy . . . of a somewhat anodyne nature.”91
En route to the conference, the British head delegate, Alfred Balfour, re-
viewed his agenda for Prime Minister Lloyd George:
The ultimate aim . . . [is] to secure the largest possible limitation of armaments
consistent with the safety of the British Empire. It is clear, however, that if sat-
isfactory and durable results are to be achieved in regard to naval disarma-
ment . . . an agreement must be reached in regard to certain political problems
which have arisen in China and the Paciªc. First and foremost among the latter
problems is that of the Anglo-Japanese Alliance. . . . [A]dherence to the
Alliance in its present form will be very unpopular in the [United States], and
will render the conclusion of [an] arrangement for the limitation of armaments
extremely difªcult to negotiate.
Given those constraints, Balfour proposed to suggest a “draft tripartite agree-
ment,” one object of which was “to bring the existing Anglo-Japanese Alliance
to an end.”92
Before and during the conference, Hughes controlled the format and negoti-
ated the agenda to reach that end. First, he stymied a British initiative to hold a
separate preliminary conference in London to work out the terms of the com-
mon Paciªc policy agreement with just the British Empire, Japan, China, and
the United States before the Washington naval talks. By separating the two ne-
gotiations, Curzon had hoped to set the terms of the successor to the Anglo-
Japanese Alliance before facing the United States’ considerable leverage at the
Washington Naval Conference.93 In response to Hughes’s insistence that
the two conferences must be linked in Washington, Curzon declared on
July 14 that he did not “see how [they] could possibly be parts of the same
90. Memorandum by the Secretary of State of a Conversation with the British Ambassador
(Geddes), September 20, 1921, FRUS, 1921, Vol. 1, pp. 72–74.
91. Foreign Ofªce Memorandum respecting a Tripartite Agreement, October 22, 1921, DBFP,
p. 448.
92. Mr. Balfour (Washington Delegation) to Mr. Lloyd George, November 11, 1921, DBFP, pp. 467–
468.
93. Memorandum by the Marquess Curzon of Kedleston on the situation re proposed Conference
at Washington, July 24, 1921, DBFP, pp. 345–351.
whole.”94 By early August, Curzon relented, deciding that “the best course
[was] that [the] American Government having assumed exclusive responsibil-
ity for the conference, should prepare the agenda themselves.”95 Hughes im-
posed a tight contingency between the results of both the contemplated naval
limits treaty (which would entail U.S. concessions to Britain and Japan) and
the common Paciªc policy arrangement.96
Hughes’s scheme for the Washington Naval Conference included a large
multilateral forum that would focus on naval arms limitation and a parallel,
secret subgroup of British, Japanese, and U.S. ministers who would negotiate
the common Paciªc policy agreement.97 It was in these latter private talks
that “it was decided to bury the alliance inside” the Four-Power Treaty.98
One of seven treaties to emerge from the Washington Naval Conference,
this treaty bound the trio (plus France) in various promises to confer and
consult on Paciªc affairs and delivered the coup de grâce to the Anglo-
Japanese Alliance.
Its terms were crafted by Hughes to do so.99 The Four-Power Treaty did not
entail any positive alliance obligations. The only concrete action that it dictated
was negative—Article IV stipulated that, once ratiªed by the four principals,
the Anglo-Japanese alliance agreement would terminate. Along with that for-
mal device, an explicit political contingency cemented the linkage to the naval
arms negotiations. Full ratiªcation of the Four-Power Treaty by the principals
was made a prerequisite for any U.S. vote to ratify the Five-Power Naval
Limits Treaty.100
94. Marquess Curzon of Kedleston to Sir A. Geddes (Washington), July 14, 1921, DBFP, p. 343.
Curzon noted later that when the cabinet had ªrst suggested “a Conference on Paciªc matters,” it
had not “contemplate[d] the conjunction of such a Conference with a Conference on the even
larger topic of disarmament.” Memorandum by the Marquess Curzon of Kedleston on the situa-
tion re proposed Conference at Washington, July 24, 1921, DBFP, doc. 337, p. 345; and Marquess
Curzon of Kedleston to Sir C. Eliot (Tokyo), August 5, 1921, DBFP, p. 370.
95. Marquess Curzon of Kedleston to Sir A. Geddes (Washington), August 3, 1921, DBFP, pp. 366–
367.
96. Sprout and Sprout, Toward a New Order, pp. 132–134; J. Chal Vinson, “The Drafting of the Four-
Power Treaty of the Washington Conference,” Journal of Modern History, Vol. 25, No. 1 (March
1953), pp. 40–47, doi.org/10.1086/237563; and Secretary of State to the Ambassador in Great Brit-
ain (Harvey), July 13, 1921, FRUS, 1921, Vol. 1, pp. 28–29.
97. Sir A. Geddes (Washington) to the Marquess of Curzon of Kedleston, September 21, 1921,
DBFP, p. 401; and Letter from the Marquess of Curzon of Kedleston to Sir A. Geddes (Washing-
ton), DBFP, pp. 407–408.
98. Ian Nish, “Japan in Britain’s View of the International System, 1919–37,” in Ian Nish, ed.,
Anglo-Japanese Alienation, 1919–1952: Papers of the Anglo-Japanese Conference on the History of the
Second World War (Cambridge: Cambridge University Press, 1982), p. 30.
99. Vinson, “Drafting of the Four-Power Treaty,” p. 47.
effect
U.S. arms control diplomacy helped kill the Anglo-Japanese Alliance in
1921.101 In a subsequent judgment by Robert Vansittart, who was Curzon’s pri-
vate secretary in 1921, Britain “abrogated the Anglo-Japanese Alliance to
please America.”102 But the notion that Britain dropped the alliance to please
the United States does not explain why it was so important for Britain to do
so. The United States agreed to restrain a naval buildup in the Paciªc that oth-
erwise would have been ruinous for Britain.
The expected wedge effects of terminating the alliance were partially borne
out: Anglo-Japanese alignment expectations cratered, but Japan was not re-
strained in the way the United States expected it would be. In 1931–32, Japan
attacked and then conquered Manchuria, without any support from Great
Britain. Likewise, Japan had no formal allies when it expanded the war against
China in July 1937.
LTBT and the Sino-Soviet Split
Conventional accounts of the 1963 LTBT emphasize that the United States
joined the agreement, despite some risk of Soviet cheating, because it would
reinforce its nuclear superiority in key areas, diminish the tide of proliferation,
discourage the nuclear arms race, reduce atmospheric fallout, and stabilize a
rocky political relationship with the Soviet Union.103 Looming behind all of
these concerns was the massive Soviet H-bomb test of October 1961, and the
frighteningly close brush with nuclear war one year later in the Cuban mis-
sile crisis, which focused minds on the search for cooperative solutions to the
shared nuclear danger.104 From this perspective, the United States was not
interested in aggravating the Sino-Soviet split that was emerging as test ban
101. Klein, “Whitehall,” p. 468.

102. Robert G. Vansittart, Roots of the Trouble (London: Hutchinson, 1941), p. 26.
103. Glenn T. Seaborg, Kennedy, Khrushchev, and the Test Ban (Berkeley: University of California
Press, 1981), pp. 30–34, 188–199, 209–227, 249; Marc Trachtenberg, A Constructed Peace: The Making
of the European Settlement, 1945–1963 (Princeton, N.J.: Princeton University Press, 1999), pp. 379–
400; Andreas Wenger and Marcel Gerber, “John F. Kennedy and the Limited Test Ban Treaty:
A Case Study of Presidential Leadership,” Presidential Studies Quarterly, Vol. 29, No. 2 (June 1999),
p. 460, https://www.jstor.org/stable/27551999; Ronald J. Terchek, The Making of the Test Ban Treaty
(The Hague, Netherlands: Martinus Nijhoff, 1970), pp. 8–10; Ivo H. Daalder, “The Limited Test Ban
Treaty,” in Carnesale and Haass, eds., Superpower Arms Control, pp. 10–16; Avis Bohlen, “The Rise
and Fall of Arms Control,” Survival, Vol. 45, No. 3 (2003), pp. 8–9, doi.org/10.1080/0039.2003
.10071605; and Justin Key Canªl, “Trump’s Nuclear Test Would Risk Everything to Gain Nothing,”
War on the Rocks blog, July 8, 2020, https://warontherocks.com/2020/07/trumps-nuclear-test-
would-risk-everything-to-gain-nothing/.
104. Seaborg, Kennedy, Khrushchev, and the Test Ban, pp. 175–176, 199.
talks progressed. Our analysis, by contrast, shows how this indirect effect
contributed in important ways to the value that the United States placed on
the agreement.
In its pursuit of the LTBT, we ªnd that the John F. Kennedy administration
embraced the catalyze conºict mechanism to weaken alignment between the
Soviet Union and the People’s Republic of China. Although the Sino-Soviet
split had an ideological origin and had existed since the mid-1950s, the LTBT
accelerated it by elevating the Soviet Union and China’s conºict over the lat-
ter’s nuclear program, policies toward the West, and the leadership of the in-
ternational Communist movement.105
motive
Exacerbating the split had been a goal of the Kennedy administration early on.
In August 1960, Soviet specialist George F. Kennan argued, “The main target of
our diplomacy should be to heighten the divisive tendencies within the Soviet
Bloc. The best means to do this lies in the improvement in our relations with
Moscow.”106 Kennan’s thinking shaped the administration’s postelection ap-
proach to the Soviets.107 When the Kennedy administration obtained informa-
tion about China’s nuclear program in 1961, the president regarded a potential
Chinese nuclear test “as likely to be historically the most signiªcant and worst
event of the 1960s.”108 Also in 1961, the Policy Planning Council recommended
“tak[ing] advantage of the present Sino-Soviet split in the interests of widen-
ing it further or of otherwise exploiting it.”109 Capitalizing on a disintegrating
Sino-Soviet alliance, Kennedy wanted to isolate China to counter its nuclear
program by cooperating with the Soviet Union over a test ban treaty.110 If
105. On the Sino-Soviet split, see Danhui Li and Yafeng Xia, “Jockeying for Leadership: Mao and
the Sino-Soviet Split, October 1961–July 1964,” Journal of Cold War Studies, Vol. 16, No. 1 (Winter
2014), pp. 24–60, doi.org/10.1162/JCWS_a_00430; Vladislav Zubok, “‘Look What Chaos in the
Beautiful Socialist Camp!’ Deng Xiaoping and the Sino-Soviet Split, 1957–1963,” Cold War In-
ternational History Project Bulletin, No. 10 (March 1998), pp. 152–162, https://www.wilsoncenter
.org/sites/default/ªles/media/documents/publication/CWIHP_Bulletin_10.pdf; and Odd Arne
Westad, ed., Brothers in Arms: The Rise and Fall of the Sino-Soviet Alliance, 1945–1963 (Stanford, Calif.:
Stanford University Press, 1998), pp. 165–188.
106. Bernard J. Firestone, “Kennedy and the Test Ban: Presidential Leadership and Arms Control,”
in Douglas Brinkley and Richard T. Grifªths, eds., John F. Kennedy and Europe (Baton Rouge: Louisi-
ana State University Press, 1999), p. 86.
107. Chang, Friends and Enemies, pp. 229–232.
108. James Fetzer, “Clinging to Containment: China Policy,” in Thomas G. Paterson, ed., Kennedy’s
Quest for Victory: American Foreign Policy, 1961–1963 (New York: Oxford University Press, 1989),
p. 182.
109. Draft Paper Prepared in the Policy Planning Council, October 26, 1961, FRUS, 1961–1963:
Northeast Asia, Vol. 22 (Washington, D.C.: GPO, 1996), p. 167.
110. Noam Kochavi, A Conºict Perpetuated: China Policy during the Kennedy Years (Westport, Conn.:
Praeger, 2002), p. 219.
China persisted with its nuclear program, that would deepen China’s isolation
and kill the Sino-Soviet alliance.
catalyze conºict mechanism

In April 1961, the United States estimated that as China’s power in the
Communist bloc grew, the Soviet Union would ªnd arms control measures
more attractive because they would delay China’s nuclear program—to
China’s dismay.111 The Kennedy administration was aware of Sino-Soviet ten-
sions in 1961 and was keen to exploit them to further weaken Sino-Soviet
alignment. Kennedy thus sought help from the Soviet Premier Nikita
Khrushchev to stop the Chinese nuclear program. Before the June 1961 Vienna
Summit, the United States expected that a Kennedy-Khrushchev meeting
would “exacerbate Soviet-Chinese relations.”112 When he met Khrushchev at
the summit, Kennedy argued that it was in their shared interest to lower the
risk of a general nuclear war by preventing China from acquiring nuclear
capability.113 Khrushchev, however, was not at that time comfortable with
the idea of a Sino-Soviet split and U.S.-Soviet cooperation against a Soviet
partner.114 The summit thus ended without signiªcant progress. Kennedy
speculated that Khrushchev’s lack of interest in a test ban was a result of
Chinese pressure.115
Inside the Communist bloc, Sino-Soviet relations worsened. In October
1961, the 22nd Congress of the Communist Party of the Soviet Union
showed that the perceived unity between China and the Soviet Union was vul-
nerable.116 Still, despite signiªcant differences over the leadership of the
Communist movement, Khrushchev’s “peaceful coexistence,” and China’s nu-
clear program, they tried to manage the split and avoided overtly disparaging
one another.117 To the United States, those differences signaled opportunity. In
March 1962, administration ofªcials discussed how to exacerbate the split.
One of the proposals was to “sound out the Soviets regarding a rapproche-
ment on political issues of mutual interest” while avoiding making deals
111. National Intelligence Estimate, April 6, 1961, FRUS, 1961–1963: Arms Control and Disarma-
ment, Vol. 7 (Washington, D.C.: GPO, 1995), p. 37. Also see Schelling and Halperin, Strategy and
Arms Control, p. 44.
112. Telegram from the Embassy in the Soviet Union to the Secretary of State, May 4, 1961, FRUS,
1961–1963: Soviet Union, Vol. 5 (Washington, D.C.: GPO, 1998), p. 131.
113. Chang, Friends and Enemies, pp. 231–232.
114. Ibid.
115. Memorandum of Conversation, June 6, 1961, 4:30 p.m., FRUS, 1961–1963, Vol. 5, p. 233.
116. Westad, Brothers in Arms, p. 381.
117. Ibid.; Li and Xia, “Jockeying for Leadership,” pp. 31–33; National Intelligence Estimate, Feb-
ruary 21, 1962, FRUS, 1961–1963, Vol. 5: Soviet Union, p. 375; and National Intelligence Estimate,
May 2, 1962, FRUS, 1961–1963, Vol. 5: Soviet Union, p. 428.
with Communist China.118 We argue that this is an example of the logic of se-
lective accommodation. Arms control was an area of mutual interest, and the
Kennedy administration pressed forward with the arms control negotiations,
betting that U.S.-Soviet détente would surface the covert Sino-Soviet split.119
From 1961 to mid-1962, the Soviet Union and China suppressed their mu-
tual antagonism: Chairman of the Chinese Communist Party (CCP) Mao
Zedong needed to recover from the Great Leap Forward and Khrushchev
wanted to preserve Communist unity.120 But for Mao, conºict over leader-
ship of the Communist movement and Khrushchev’s revisionism meant that
détente could not last.121 In March, the Soviet Union raised the issue of a com-
prehensive test ban agreement with the West, but China opposed it because
the Soviets had no rights to stop socialist countries from obtaining nuclear
weapons.122 In August, the United States gave the Soviet Union two proposals:
a comprehensive test ban (including atmosphere, outer space, underwater, and
underground testing) and a partial test ban (without underground testing). Al-
though the Soviets rejected both proposals, they agreed to resume negotia-
tions.123 U.S. ofªcials persuaded the Soviet Union to agree to a partial test
ban by pointing out that it would beneªt from continuing underground
tests given the Soviet need to conduct future tests of this kind, which Soviet
Ambassador to the United States Anatoly Dobrynin found appealing.124
China condemned the Soviet Union’s agreement to negotiate as “collusion . . .
to prevent us [China] from doing atomic research,” which deepened the Sino-
Soviet conºict.125
The Cuban missile crisis in October alerted both sides to the danger of an in-
advertent nuclear war, and thus worsened the Sino-Soviet split by rekindling
the prospect of a U.S.-Soviet–backed test ban treaty.126 China denounced
Khrushchev’s retreat in Cuba and refusal to back China’s war with India,
118. Memorandum from Hughes to Rostow, March 6, 1962, FRUS, 1961–1963, Vols. 7, 8, 9: Arms
Control; National Security Policy; Foreign Economic Policy, Microªche Supplement (Washington, D.C.:
GPO, 1997), p. 938.
119. National Intelligence Estimate, January 17, 1961, FRUS, 1961–1963, Vol. 5: Soviet Union, p. 19;
and Memorandum Prepared by the Central Intelligence Agency, November 29, 1962, FRUS, 1961–
1963, Vol. 5: Soviet Union, pp. 582–587.
120. Li and Xia, “Jockeying for Leadership,” pp. 38–42.
121. Ibid.
122. Westad, Brothers in Arms, pp. 179–180.
123. Chang, Friends and Enemies, p. 234. In 1963, Khrushchev supported a limited rather than com-
prehensive test ban treaty because it did not require intrusive inspections.
124. Memorandum from the Ambassador at Large (Thompson) to Secretary of State (Rusk), Octo-
ber 10, 1962, FRUS, 1961–1963, Vol. 5: Soviet Union, p. 509.
125. Quoted in Li and Xia, “Jockeying for Leadership,” pp. 45–46; and Zubok, “Look What
Chaos,” p. 159.
126. On the Cuban missile crisis and Sino-Soviet relations, see Enrico Maria Fardella, “Mao
Zedong and the 1962 Cuban Missile Crisis,” Cold War History, Vol. 15, No. 1 (2015), pp. 73–88,
doi.org/10.1080/14682745.2014.971017; and Li and Xia, “Jockeying for Leadership.”
which further jeopardized the Sino-Soviet détente and pushed the Soviet
Union closer to the United States.127 Concurrently, Khrushchev was under
pressure to make his “peaceful coexistence” policy (including the test ban
treaty) successful in order to refute China’s assertion that he was being tricked
by the West.128 Meanwhile, the United States developed a framework for a test
ban treaty that would limit both China’s and West Germany’s nuclear pro-
grams. Targeting West Germany would constitute a signiªcant accommoda-
tion of Soviet security interests.129 Kennedy’s push for a test ban would also ªt
Khrushchev’s aspiration to improve relations with the West to counter China’s
criticism of Khrushchev.130 Senior U.S. ofªcials were doubtful about the Soviet
Union’s leverage over China despite its renewed interest in a test ban treaty.131
National Security Advisor McGeorge Bundy argued that the treaty would not
be valuable without Soviet assurance that the Chinese would agree to a test
ban.132 Secretary of State Dean Rusk advised Kennedy to seek a Soviet guaran-
tee that China would adhere to the test ban in exchange for U.S. assurance
that West Germany would, too.133 Hence, during a meeting between Kennedy
and a Dobrynin-led Soviet delegation in January 1963, Kennedy pressed the
Soviets to get China to comply.134 That month, the Central Intelligence Agency
(CIA) concluded that the Soviet Union and China would “increasingly view
each other as hostile rivals and competing powers,” even if they did not repu-
diate the military alliance.135
From early 1963, Kennedy ofªcials studied ways to exacerbate the Sino-
Soviet split via test ban negotiations. After concluding that “Khrushchev’s
primary interest in a test ban relates to Communist China and West Germany,”

128. Allen Pietrobon, “The Role of Norman Cousins and Track II Diplomacy in the Breakthrough
to the 1963 Limited Test Ban Treaty,” Journal of Cold War Studies, Vol. 18, No.1 (Winter 2016), pp. 60–
79, doi.org/10.1162/JCWS_a_00619.
129. Memorandum from Secretary of State Rusk to President Kennedy, November 27, 1962, FRUS,
1961–1963, Vol. 7, pp. 618–619.
130. Memorandum Prepared by the Central Intelligence Agency, November 29, 1962, FRUS, 1961–
1963, Vol. 5: Soviet Union, pp. 582–587.
131. Letter from the Deputy Director for Intelligence, Central Intelligence Agency (Cline) to the
Director of the Arms Control and Disarmament Agency (Foster), FRUS, 1961–1963, Vol. 7: Arms
Control and Disarmament, p. 583.
132. Memorandum from the President’s Special Assistant for National Security Affairs (Bundy) to
President Kennedy, November 8, 1962, FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament,
p. 598.
133. Memorandum from Secretary of State Rusk to President Kennedy, November 27, 1962, FRUS,
1961–1963, Vol. 7: Arms Control and Disarmament, pp. 618–619.
134. Memorandum of Conversation, January 9, 1963, FRUS, 1961–1963, Vol. 7: Arms Control and
Disarmament, pp. 628–629.
135. Central Intelligence Agency, “Sino-Soviet Relations at a New Crisis,” Freedom of Information
Act Electronic Reading Room, January 14, 1963, pp. 1–6, https://www.cia.gov/readingroom/
document/cia-rdp79t00429a000300020013-1.
they sought to capitalize on the Soviets’ renewed interests.136 The Kennedy ad-
ministration sought to restrain China’s nuclear program by isolating it.137 For
the administration, a test ban agreement would both stiºe China’s nuclear am-
bitions and subvert Sino-Soviet alignment. The U.S. Joint Chiefs of Staff were
skeptical about a partial test ban per se, because it would let the Soviet Union
continue testing below the detection threshold, and continued testing was es-
sential for “the United States [to] achieve or maintain superiority in all areas of
nuclear weapons technology.”138 But, in the words of U.S. Assistant Secretary
of Defense Paul Nitze, even if “it was difªcult to support the test ban treaty on
national security grounds alone, it could be supported on foreign policy
grounds.”139 Although Kennedy (and the Joint Chiefs of Staff) expected that
China would not comply, Nitze was willing to risk some Soviet cheating in or-
der to get its help making China adhere to the test ban.140 In a May 1963 testi-
mony before the Senate Preparedness Investigating Subcommittee, head of the
Arms Control and Disarmament Agency, William Foster, argued that a test
ban would both combine U.S. and Soviet efforts to stop China’s nuclear pro-
gram and “have a divisive effect on Sino-Soviet relations,” even if China
would not adhere.141
Despite China’s growing claims to lead the international Communist move-
ment, in April 1963 Khrushchev still hoped to preserve unity with China
under “peaceful coexistence.”142 His dilemma was whether to align Soviet pol-
icies with China or blame it for the collapse of Sino-Soviet relations.143 If
Khrushchev agreed to a test ban treaty, it would be hard to blame China for the
split.144 U.S. ofªcials believed that, besides inspection problems, the “major
factor” impeding an improvement in U.S.-Soviet relations was “Moscow’s pre-
occupation with the Chinese Communist problem.”145 Kennedy understood
that it was this dilemma that inhibited Khrushchev from advancing test
136. Memorandum from the Ambassador at Large (Thompson) to Secretary of State Rusk,
March 21, 1963, FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament, p. 658.
137. Kochavi, A Conºict Perpetuated, pp. 217–223.
138. Memorandum from the Joint Chiefs of Staff to Secretary of Defense McNamara, April 20,
1963, FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament, p. 684.
139. Seaborg, Kennedy, Khrushchev, and the Test Ban, p. 223.
140. Kochavi, A Conºict Perpetuated, p. 219; and Memorandum from the Joint Chiefs of Staff to Sec-
retary of Defense McNamara, April 29, 1963, FRUS, 1961–1963, Vol. 7: Arms Control and Disarma-
ment, pp. 689–691.
141. Chang, Friends and Enemies, p. 238.
143. Message from President Kennedy to Prime Minister Macmillan, March 28, 1963, FRUS, 1961–
1963, Vol. 7: Arms Control and Disarmament, p. 659.
144. Memorandum from the Ambassador at Large (Thompson) to Secretary of State Rusk,
March 21, 1963, FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament, pp. 657–658.
145. Telegram from the Embassy in the Soviet Union to the Department of State, March 16, 1963,
FRUS, 1961–1963, Vol. 5: Soviet Union, p. 642.
ban negotiations rather than disagreements over the number of inspections
and the status of West Germany.146 U.S. ofªcials highlighted the need to get
Khrushchev on board with a test ban by stressing that Chinese nuclear weap-
ons could “be directed westward against the USSR as well as eastward
against the US.”147 In May 1963, Bundy told Dobrynin that China’s nuclear
ambition was an issue of common interest between the United States and the
Soviet Union.148
To help Khrushchev deºect Chinese pressure, Kennedy mentioned his sup-
port of a nuclear test ban treaty during his June 1963 American University
commencement address.149 A Chinese observer in the Peking Review accused
Kennedy of “wooing the Soviet Union, opposing China and poisoning Sino-
Soviet relations.”150 After that speech, the White House stated in a memo to
the national security bureaucracy, “the President is not choosing sides, in any
ostentatious way, between Moscow and Peiping, but his speech is designed to
emphasize the positive opportunities for a more constructive and less hostile
Soviet policy.” As one of Kennedy’s test ban negotiators conjectured, that im-
plied “the president was taking sides in an unostentatious way. . . . [The]
Chinese were regarded by many as . . . the more dangerous of the Communist
giants. President Kennedy seemed to be one who held this view (italics in orig-
inal).”151 In July 1963, the CIA reported that “the Sino-Soviet rupture has been
the by-product of US policy,” and U.S.-Soviet cooperation to impede the
Chinese nuclear program through “arms and testing negotiations” was one of
the areas that the United States could work on to worsen the split.152 That
month, Khrushchev tried to salvage the relationship with China by proposing
to work with it over “peaceful coexistence” at a fraternal party conference.153
146. Wenger and Gerber, “Kennedy and the Limited Test Ban Treaty,” pp. 470–473.
147. Memorandum from the Assistant Director of Central Intelligence for National Estimates
(Kent) to the Under Secretary of State for Political Affairs (Harriman), FRUS, 1961–1963, Vol. 7:
Arms Control and Disarmament, p. 771.
148. Memorandum of Conversation between the President’s Special Assistant for National Secu-
rity Affairs (Bundy) and the Soviet Ambassador (Dobrynin), May 17, 1963, FRUS, 1961–1963,
Vol. 5: Soviet Union, p. 675.
149. Pietrobon, “Norman Cousins,” p. 76. On the genesis, content, and impact of the speech, see
Seaborg, Kennedy, Khrushchev, and the Test Ban, pp. 211–218. James Goodby, a member of the advi-
sory and negotiating teams during the period, also argued that the Soviet dispute with China mo-
tivated Khrushchev to agree to a limited test ban. See James Goodby, “The Limited Test Ban
Negotiations, 1954–63: How a Negotiator Viewed the Proceedings,” International Negotiation,
Vol. 10, No. 3 (2005), pp. 381–404, doi.org/10.1163/157180605776087507.
150. Renmin Ribao Observer, “No Meddling in Sino-Soviet Differences by U.S. Imperialism,”
Peking Review, July 19, 1963, p. 9, https://www.marxists.org/subject/china/peking-review/1963/
PR1963-29.pdf.
152. Central Intelligence Agency, “Implications of the Sino-Soviet Rupture for the U.S.,” Free-
dom of Information Act Electronic Reading Room, July 18, 1963, pp. 7–9, https://www.cia.gov/
readingroom/document/cia-rdp79t00429a001300040006-6.
However, China’s June 14 article in the Peking Review criticized Soviet coopera-
tion with U.S. imperialism and proposed a complete prohibition of nuclear
weapons.154 That caught Khrushchev by surprise.155 When Soviet and Chinese
leaders met in July to discuss the future of the international Communist move-
ment, they failed to reach any agreements.156 In the words of the leader of the
Chinese delegation Deng Xiaoping, the United States “exercise[s] deception
and trickery toward other imperialist powers. How can one in such a case be-
lieve that there can be total unity between an imperialist country [and] a so-
cialist country?”157 He denounced the test ban as a Soviet attempt to bind
“China by the hands and feet through an agreement with the USA.”158 On
July 14, the Congress of the Soviet Communist Party (CPSU) responded to the
CCP June 14 article with an “Open Letter” that blamed the CCP for the Sino-
Soviet split and defended “peaceful coexistence.”159 The CPSU and the CCP
were openly deadlocked.
On July 15, the ªrst day of the test ban negotiation, Kennedy instructed the
lead negotiator, Averell Harriman, to remind Khrushchev of the Chinese nu-
clear threat and to “elicit Khrushchev’s view of means of limiting or prevent-
ing Chinese nuclear development.”160 After that meeting, Harriman reasoned
that Khrushchev’s interest in the test ban was because of his concern over
China’s challenges to the Soviet leadership of the international Communist
movement.161 Harriman explained to Kennedy that, since the Soviet Union
could not make China agree to the test ban, the Soviets would want all coun-
tries to adhere to the ban. Pressure from world opinion rather than just the
Soviets would isolate China.162 Thus, the Soviet Union would try to maintain
leadership of the Communist bloc and blame the split on China. When U.S. of-
ªcials and the Soviets agreed to sign the LTBT in late July, China openly ac-
cused the Soviet Union of “allying with the US against China” and declared it
154. CPC Central Committee’s Letter in Reply to the CSPU Central Committee’s Letter,
“A Proposal Concerning the General Line of the International Communist Movement,” Peking
Review, June 21, 1963, pp. 6–14, https://www.marxists.org/subject/china/peking-review/1963/
PR1963-25.pdf.
156. Ibid.
158. Ibid., p. 381.
159. Communist Party of the Soviet Union, “Open Letter of the Central Committee of the Com-
munist Party of the Soviet Union to All Party Organizations, to All Communists of the Soviet
Union,” Pravda, July 14, 1963, https://www.marxists.org/history/international/comintern/sino-
soviet-split/cpsu/openletter.htm.
160. Telegram from the Department of State to the Embassy in the Soviet Union, July 15, 1963,
FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament, p. 801.
161. Telegram from the Embassy in the Soviet Union to the Department of State, July 18, 1963,
FRUS, 1961–1963, Vol. 7: Arms Control and Disarmament, p. 808.
162. Ibid.
would not adhere to such a “dirty fraud.”163 The United States, Britain, and the
Soviet Union signed the LTBT on August 5, 1963. The Soviets used the occa-
sion to condemn China’s call for a summit on nuclear disarmament as a “cover
for China’s refusal to sign” the LTBT; meanwhile, China was preparing for a
complete split of the Communist movement.164
effect
Despite some military disadvantages, the United States supported the LTBT
for its political merits. President Kennedy instructed the Joint Chiefs of Staff to
consider both the political and military/technical effects of the treaty. The
president, as James Lebovic notes, “obviously thought that the political impli-
cations weighed toward ratiªcation.”165 As the U.S. Air Force Chief of Staff,
Curtis LeMay, explained in an August 1963 testimony to the Senate Foreign
Relations Committee, the key factor in the Joint Chiefs of Staff’s support for
the treaty was political: “we examined the military and the technical aspects
and came up with a net disadvantage in that ªeld. . . . Then we examined the
political gains that were possible, and we came up with a net advantage there
which we thought offset the [military] disadvantages.”166 One of the key politi-
cal beneªts was that the LTBT “would contribute to a further division of the
Sino-Soviet Bloc, and this result would be a major political achievement with
important and favorable military implications.”167 Rusk concurred, pointing
out that the LTBT “could contribute to what appears to be a split between the
Soviets and the Chinese Communists and . . . may lead to further favorable de-
velopments in our relations with the Soviets.”168
The LTBT did indeed exacerbate the Sino-Soviet split. China believed
that the Soviet Union had joined the United States’ anti-China campaign
and betrayed “the principle of proletarian internationalism” by refusing to
provide China with nuclear weapons.169 China condemned the Soviet Union
163. Central Committee of the Chinese Communist Party, “Statement of the Chinese Government
Advocating the Complete, Thorough, Total and Resolute Prohibition and Destruction of Nuclear
Weapons Proposing a Conference of the Government Heads of All Countries of the World,” Peking
Review, August 2, 1963, pp. 7–8, https://www.marxists.org/subject/china/peking-review/1963/
PR1963-31.pdf.
164. Central Intelligence Agency, “The President’s Intelligence Checklist,” Freedom of Information
Act Electronic Reading Room, August 5, 1963, https://www.cia.gov/readingroom/document/
0005996489.
165. Lebovic, Flawed Logics, p. 54.
167. Statement of the Position of the Joint Chiefs of Staff on the Three-Environment Nuclear
Test Ban Treaty, August 12, 1963, FRUS, 1961–1963, Vols. 7, 8, 9, https://history.state.gov/
historicaldocuments/frus1961-63v07-09mSupp/d218.
168. Memorandum for the Record, August 13, 1963, FRUS, 1961–1963, Vol. 7: Arms Control and
Disarmament, p. 879.
for embracing “the U.S. imperialists in joyous abandon”; the Soviets, in
turn, denounced China’s “fabrications” and “Trotskyite foreign policy.”170
On August 15, 1963, China declared that it would not join the LTBT, and
the Soviets, in return, threatened to expel China from the international
Communist movement if it continued to attack the test ban treaty.171 By 1964,
China considered the Soviet Union one of its two main enemies, and in March
1966, it broke interparty ties.172
According to Gordon Chang, the post-LTBT split occurred exactly how the
United States wanted it to.173 By cementing Soviet and Chinese differences
over their respective policies toward the United States, the leadership of the
Communist movement, and China’s nuclear program, the LTBT propelled and
rendered irreconcilable the Sino-Soviet split. The then deputy director of the
CCP Central International Liaison Department, Wu Xiuquan, noted that China
perceived the Soviet’s pressure to join the LTBT as an attempt to deny
China its nuclear capability.174 Across the aisle, a retired Soviet military ofªcer,
Victor Gobarev, argued that “one of the principal causes of the Split was
China’s insistence on pursing their nuclear weapons program at any costs.”175
One of those costs was arguably the Sino-Soviet alliance. As Lorenz Lüthi
summarizes, though, “the ground for the Sino-Soviet rift” was already pre-
pared, the LTBT “helped it to burst into the open.”176
SALT and the U.S.-China Rapprochement
The Soviet Union attempted to use SALT I from 1969 to 1972 to dampen
U.S.-China rapprochement. Studies of Soviet arms control policy in the early
détente period tend to highlight direct-effect incentives: to gain U.S. recogni-
tion of strategic parity; to reduce the risks of a nuclear war that could not be
won; and to protect Soviet military advantages while blunting those of the
United States.177 These studies suggest that the China factor did not exert sig-
170. William E. Grifªth, “The Aftermath of the Test Ban Treaty,” The Sino-Soviet Rift (Cambridge,
Mass.: Massachusetts Institute of Technology Press, 1964), pp. 162–164.
171. Ibid., p. 173; and “Statement by the Spokesman of the Chinese Government: A Comment on
the Soviet Government’s Statement of August 3,” Peking Review, August 16, 1963, pp. 7–8, https://
www.massline.org/PekingReview/PR1963/PR1963-33.pdf.
172. Li and Xia, “Jockeying for Leadership,” p. 59.
173. Chang, Friends and Enemies, p. 249.
174. Nicholas Khoo, “Breaking the Ring of Encirclement: The Sino-Soviet Rift and Chinese Policy
toward Vietnam, 1964–1968,” Journal of Cold War Studies, Vol. 12, No. 1 (Winter 2010), pp. 3–42, esp.
pp. 9–14, https://www.jstor.org/stable/26923059.
175. Ibid., p. 13.
176. Lüthi, Sino-Soviet Split, p. 272.
177. Raymond L. Garthoff, Détente and Confrontation: American-Soviet Relations from Nixon to Rea-
gan (Washington, D.C.: Brookings Institution Press, 1985), pp. 53–68; Philip J. Farley, “Strategic
Arms Control, 1967–1987,” in Alexander L. George, Philip J. Farley, and Alexander Dallin, eds.,
niªcant effects on Soviet and U.S. maneuvers during SALT I negotiations.178
Some analysts, however, suggest that SALT might have weakened alignment
within NATO by creating a false sense of security or igniting fears of super-
power collusion.179 In contrast, our framework shows both how SALT’s wedge
effects extended to the U.S.-China-Soviet Union strategic triangle and how, in
particular, the Soviets used the peace offensive mechanism to stave off U.S.-
China collusion.
motive
As relations with China deteriorated in the late 1960s, the Soviet Union was
afraid of Beijing improving its ties with the United States. After the CCP broke
off ties in 1966, the 1969 border clash reinforced Sino-Soviet antagonism.180 The
Soviet Union had by then identiªed China as a primary threat; moreover,
U.S. signals during the 1969 clash “could only be construed as opposition to a
Soviet attack on China [and] resonate[d] with expressed apprehension over
a coincidence in Sino-American anti-Soviet interests that might draw Beijing
and Washington together.”181 By embracing SALT, the Soviet Union had two
major goals: (1) to concentrate on the threat coming from China after the 1969
border conºict, and (2) to prevent an anti-Soviet U.S.-China alignment.182
The strategic triangle schema that animated President Richard Nixon’s and
National Security Advisor Henry Kissinger’s “grand design” of pursuing both
détente with the Soviet Union and rapprochement with China presaged these
two Soviet motivations.183 More precisely, they were an expected response to
the leverage that the United States gained from holding a “China card.” The
Soviet Union’s impulse to divide the United States and China—by accom-
modating the United States through strategic arms control—was partly a
U.S.-Soviet Security Cooperation: Achievements, Failures, Lessons (New York: Oxford University Press,
1988), pp. 225–227; Condoleezza Rice, “SALT and the Search for a Security Regime,” in George,
Farley, and Dallin, eds., U.S.-Soviet Security Cooperation, pp. 294–295; Coit D. Blacker, “Learning in
the Nuclear Age: Soviet Strategic Arms Control Policy, 1969–1989,” in George W. Breslauer and
Philip E. Tetlock, eds., Learning in U.S. and Soviet Foreign Policy (Boulder, Colo.: Westview, 1991),
p. 434; and Fen Osler Hampson, “SALT I: Interim Agreement and ABM Treaty,” in Carnesale and
Haass, eds., Superpower Arms Control, pp. 68–72, 78–84.
178. James Cameron, The Double Game: The Demise of America’s First Missile Defense System and the
Rise of Strategic Arms Limitation (New York: Oxford University Press, 2018), pp. 152–154.
179. Lynn-Jones, “Lulling and Stimulating Effects,” pp. 225–229; Seay, “What Are the Soviets’ Ob-
jectives,” pp. 62–64, 96–97; and Luttwak, “SALT and the Meaning of Strategy,” pp. 26–27.
180. Y. Kuisong, “The Sino-Soviet Border Clash of 1969: From Zhenbao Island to Sino-American
Rapprochement,” Cold War History, Vol. 1, No. 1 (2000), pp. 21–52, doi.org/10.1080/713999906.
181. Allen Whiting, “Soviet Policy toward China, 1969–1988,” in Breslauer and Tetlock, eds.,
Learning in U.S. and Soviet Foreign Policy, p. 520.
182. On the connection between détente and the Sino-Soviet conºict, see M.E. Sarotte, Dealing with
the Devil: East Germany, Détente, and Ostpolitik, 1969–1973 (Chapel Hill: University of North
Carolina Press, 2001). Thanks to Kraig Larkin for the suggestion.
183. Caldwell, American-Soviet Relations, pp. 90–91.
consequence of the United States becoming a “pivot” between the two com-
munist powers.184
peace offensive mechanism

The Soviet Union was a common enemy of both the United States and China
in the late 1960s and it worried, with good reason, about a U.S.-China conver-
gence against it.185 During the Lyndon B. Johnson administration, the Soviet
Union and the United States contemplated negotiations to address the huge in-
crease in both sides’ nuclear arsenals. However, the Soviet 1968 invasion of
Czechoslovakia impeded the plan. Talks began only after Nixon came into
ofªce.186 In August 1969, as the Sino-Soviet border conºict was ongoing, a U.S.
National Intelligence Estimate (NIE) predicted that the Soviet Union would
seek to improve relations with the West to “contain” China.187 When the Soviet
Union began the SALT negotiations in November 1969, it did so with the
Chinese threat in mind. The Soviet wedge motive was consistent throughout
the three-year negotiations over strategic arms limitations. Although both the
Soviet and U.S. ofªcials did not discuss China explicitly during the negotia-
tions, the chief U.S. negotiator at SALT, Gerard Smith, concluded that China
was “a constant presence” and a key reason for the Soviet Union’s participa-
tion in SALT.188 Similarly, when the talks began, Dobrynin depicted them as “a
barometer of our [Soviet] relations with the United States.”189
In 1969, the Soviet Union still believed that it was in an advantageous posi-
tion in the strategic triangle.190 Taiwan, the Vietnam War, and China’s repre-
sentation at the United Nations would impede U.S.-China rapprochement.191
That year, the Soviet Union initiated the ªrst of several attempts to settle its
184. See Jervis, System Effects, pp. 181–191, esp. p. 182; and Gaddis, Strategies of Containment,
pp. 295–298.
185. Because Soviet primary documents on SALT negotiations remain scarce—and no Soviet ne-
gotiators, except for Anatoly Dobrynin, produced a major memoir on the process—our account of
this aspect of Soviet arms control diplomacy draws upon contemporary assessments from
U.S. policy and intelligence sources, and secondary works by historians and arms control and So-
viet policy experts. On the lack of Soviet SALT accounts, see John Maurer, “An Era of Negotia-
tions: SALT in the Nixon Administration, 1969–1972,” Ph.D. dissertation, Georgetown University,
2017, pp. 13–14.
186. Bohlen, “Rise and Fall of Arms Control,” pp. 9–12.
187. National Intelligence Estimate, August 12, 1969, FRUS, 1969–1976: Soviet Union, January 1969–
October 1970, Vol. 12 (Washington, D.C.: GPO, 2006), p. 225.
188. Gerard C. Smith, Doubletalk: The Story of the First Strategic Arms Limitation Talks (Garden City,
N.Y.: Doubleday, 1980), pp. 34–35.
189. Anatoly Dobrynin, In Conªdence: Moscow’s Ambassador to America’s Six Cold War Presidents
(Seattle: University of Washington Press, 1995), p. 202.
190. Chi Su, “U.S.-China Relations: Soviet Views and Policies,” Far Eastern Survey, Vol. 23, No. 5
(1954), pp. 557–561, 571, doi.org/10.2307/2644464.
191. Ibid, p. 561.
border dispute with China via a nonaggression treaty and a new border agree-
ment.192 Meanwhile, Nixon stalled strategic arms talks to decide how to link
them with Vietnam and Middle East priorities. The Soviet Union was prepared
to wait Nixon out193 because it opposed arms control linkage.194
Because they disagreed about the scope of an agreement, SALT negotiators
believed that neither side felt “any overriding need” to reach a deal, and that
the ªrst one would be modest.195 In June 1970, however, the United States de-
tected signs that the Soviets were using SALT negotiations to not only pursue
arms control but also isolate China by facilitating U.S.-Soviet cooperation
against it.196 We argue that the Soviet Union used a peace offensive logic to
reassure the United States of its cooperative intentions, which the Soviets
would turn against China in an effort to dampen a potential U.S.-China bal-
ancing coalition. This logic was embodied in a proposal by the Soviet Union’s
lead negotiator, Vladimir Semenov, to lessen the likelihood that a provoca-
tive nuclear attack of a third country would cause an inadvertent U.S.-Soviet
nuclear war.197
Raymond Garthoff, the lead CIA analyst supporting the U.S. SALT negotia-
tors, described the Soviet proposal as a highly political one “that rais[ed] deli-
cate and serious questions of condominium that concerned American relations
with its allies and, above all, China.”198 Indeed, the abstract scenario of such an
attack tapped into the U.S. strategic community’s longstanding, thinly veiled,
anti-China fear. Nuclear strategy and arms control specialists called it the
problem of “catalytic war,” and as the 1960s progressed, the specter of a nu-
clear-capable China as provocateur was their central concern.199 The Soviets’
provocative attack proposal thus envisioned a kind of bilateral crisis manage-
ment hotline mechanism that leading arms control theorists had championed.
But it also called for U.S.-Soviet “joint action to prevent or retaliate against pro-
vocative attack” by a third party.200 For a potential attack, the Soviet Union and
the United States would exchange information about their respective prepara-
192. Vitaly Kozyrev, “Soviet Policy toward the United States and China, 1969–1979” in William C.
Kirby, Robert S. Ross, and Gong Li, eds., Normalization of U.S.-China Relations: An International His-
tory (Cambridge, Mass.: Harvard University Press, 2005), pp. 261–262.
193. Dobrynin, In Conªdence, pp. 206–207.
194. Maurer, “An Era of Negotiations,” pp. 68–75; and Hampson, “SALT I,” p. 68.
195. Smith, Doubletalk, pp. 114–115.
196. Maurer, “An Era of Negotiations,” p. 183.
197. Garthoff, Détente and Confrontation, pp. 175–181.
198. Ibid., p. 175.
199. Arthur Lee Burns, The Rationale of Catalytic War (Princeton, N.J.: Princeton University Center
of International Studies, 1959); and Henry S. Rowen, “Catalytic Nuclear War,” in Allison,
Carnesale, and Nye, eds., Hawks, Doves, and Owls, pp. 148–166.
200. Garthoff, Détente and Confrontation, p. 176.
tions, use necessary measures to stop it, consult to prevent “an automatic start
of nuclear war” if nuclear weapons were used, and pledge to retaliate against
the provocateur.201
A U.S. State Department internal document on the political aspect of the
proposal concluded that “China is the primary object of the proposed agree-
ment, even though Semenov does not say it outright. The Soviet Union would
evidently like to do what it can to reduce any chances of improved US-Chinese
relations.”202 Semenov was more concerned about the overall political aspect
of the proposal than its speciªc details, because the Soviet Union often wor-
ried China would stir “troubles between the USSR and the US.”203 Helmut
Sonnenfeldt, a counselor to Kissinger at the National Security Council (NSC),
called the proposal “tantamount to an alliance against China” and possibly
“the most important Soviet political initiative in years.”204 Because the ar-
rangement would give the Soviet Union the justiªcation to attack China, he
advised “it might be best to kill the anti-China concept from the start.”205
Smith, too, speculated that the Soviet insistence on the proposal reºected “a
high-level” political decision.206 Kissinger explained to Dobrynin that “politi-
cal cooperation implying a major change in international alignments and
clearly aimed against third countries was out of the question.”207 Conse-
quently, the Nixon administration rejected the Soviet proposal at that time.208
Throughout the end of 1970 and early 1971, negotiations between the Soviet
Union and the United States remained stalled over antiballistic missiles, the
scope of offensive weapons, and an agreement on Berlin’s status. In April 1971,
U.S. leaders believed, however, that the incentive to limit U.S.-China rap-
prochement would pull the Soviet Union forward. Nixon speciªcally said,
“without China, they [the Soviet Union] never would have agreed to the
SALT.”209 That proved true in May, when the United States and the Soviet
201. Smith, Doubletalk, p. 139.

202. “SALT: The Politics of ‘Provocative Attack,’” July 31, 1970, folder “SALT Talks (Vienna),”
Vol. 12, July–Sept. 70, to “SALT Talks (Helsinki),” Vol. 13, Oct. 70–Dec. 70, box 879, National Secu-
rity Council (NSC) Files SALT, Nixon Presidential Materials Staff, Richard Nixon Presidential
Library and Museum (hereafter Nixon Library). The authors thank Andrew Bowen for sharing
Nixon Library documents.
203. “Semenov’s Proposal,” folder “SALT Talks (Vienna),” Vol. 12, July–Sept. 70, to “SALT Talks
(Helsinki),” Vol. 13, Oct.–Dec. 70, box 879, NSC Files SALT, Nixon Library.
204. Memorandum from Helmut Sonnenfeldt of the National Security Council Staff to the Presi-
dent’s Assistant for National Security Affairs (Kissinger), October 16, 1970, FRUS, 1969–1976,
Vol. 32: SALT I, 1969–1972 (Washington, D.C.: GPO, 2010), p. 349.
205. Ibid.
206. Smith, Doubletalk, p. 142.
207. Henry A. Kissinger, White House Years (Boston: Little Brown, 2000), pp. 454–455.
208. Still, the proposal would later become the 1973 Agreement on the Prevention of Nuclear War.
209. Conversation among President Nixon, the President’s Assistant for National Security Affairs
(Kissinger), and the Assistant to the President (Haldeman), April 23, 1971, FRUS, 1969–1976,
Vol. 32: SALT I, 1969–1972, p. 460.
Union broke the deadlock over the linkage between defensive and offensive
arms. One of the things that motivated the Soviets to compromise was concern
about signals of warming U.S.-China relations that appeared the prior month,
most notably through Ping-Pong Diplomacy.210 The Soviets wanted to reach an
agreement to avert consolidation of U.S.-China relations against the Soviet
Union.211 The peace offensive logic thus guided the Soviet accommodation.
In the months that followed, Kissinger’s secret visit to China in July 1971
alarmed the Soviet Union and propelled it to improve relations with the
United States.212 In August 1971, Dobrynin probed Kissinger about whether
the delay in announcing Nixon’s visit to the Soviet Union in 1972 was an “anti-
Soviet” maneuver, a suggestion that Kissinger denied.213 Still, the NSC con-
cluded that the Soviet Union’s fear of isolation and its wish to isolate China
contributed to improved relations between the United States and the Soviet
Union in late 1971—including the 1971 Four Power Agreement on Berlin,
Soviet Foreign Minister Gromyko’s visit to the United States to discuss arms
control, and the announcement of the 1972 Moscow summit.214 Sonnenfeldt
agreed that the Soviets relaxed their stance on ABM negotiations in August
and September 1971 in reaction to the announcement of Nixon’s impending
visit to China.215
When Nixon visited China in February 1972, the Soviets downplayed
Vietnam as an obstacle to reaching arms control agreements.216 An April 1972
NIE saw the Soviet Union fearing a U.S.-China combination and eager to foster
better terms with the United States and progress in strategic arms talks before
the Moscow summit.217 The NSC anticipated that the Soviets might use the
summit to probe U.S. ofªcials about their relations with the Chinese and warn
them against closer ties to China.218 Another NSC report noted that Leonid
210. On Ping-Pong Diplomacy, see Robert S. Ross, Negotiating Cooperation: The United States and
China 1969–1989 (Stanford, Calif.: Stanford University Press, 1995), p. 35.
211. Richard C. Thornton and William H. Lewis, “Arms Control and Heavy Missiles,” Naval War
College Review, Vol. 37, No. 1 (January/February1984), p. 96, https://digital-commons.usnwc.edu/
nwc-review/vol37/iss1/9; and Su, “U.S.-China Relations,” p. 572.
212. Su, “U.S.-China Relations,” pp. 562, 574; and Smith, Doubletalk, pp. 254, 339.
213. Memorandum of Conversation, August 17, 1971, Soviet-American Relations: The Détente Years,
1969–1972 (Washington, D.C.: GPO, 2007), p. 432.
214. Paper Prepared by the National Security Council Staff, undated, FRUS, 1969–1976: Soviet
Union, October 1971–May 1972, Vol. 14 (Washington, D.C.: GPO, 2006), pp. 908–915.
215. Memorandum from Helmut Sonnenfeldt of the National Security Council Staff to the Presi-
dent’s Assistant for National Security Affairs (Kissinger), undated, FRUS, 1969–1976, Vol. 14:
Soviet Union, October 1971–May 1972, pp. 173–176.
216. Kozyrev, “Soviet Policy,” p. 264. The Soviet Union worried that North Vietnam’s Easter Of-
fensive against South Vietnam might hurt a U.S.-Soviet summit. Consequently, the politburo de-
cided to ignore North Vietnam to approach United States. Dobrynin, In Conªdence, pp. 248–249.
217. National Intelligence Estimate, April 20, 1972, FRUS, 1969–1976, Vol. 14: Soviet Union, October
1971–May 1972, pp. 468–474.
218. Paper Prepared by the National Security Council Staff, undated, FRUS, 1969–1976, Vol. 14:
Soviet Union, October 1971–May 1972, pp. 931–933.
Brezhnev, general secretary of the CPSU, saw the summit as an “oppor-
tunity to demonstrate a closer, more substantively-grounded relationship
with the US than Peking can command,” and wanted to prevent an anti-Soviet
U.S.-China collusion by signing bilateral agreements, including arms control,
at the summit.219
The Moscow summit’s results conªrmed those expectations. In May 1972,
Nixon and Brezhnev inked the SALT and ABM treaties.220 According to Soviet
specialist Jonathan Haslam, Brezhnev “saw SALT as a demonstrable success in
blocking the route to an [U.S.-China] anti-Soviet alliance.”221 In a May 1972
presentation to the Central Committee of the CPSU, Brezhnev emphasized the
importance of constant dialog and improvement in U.S.-Soviet relations to
prevent the U.S.-China convergence, a theme that would be repeated in the
same venue the next year.222 A Soviet brieªng for the East German government
in June explained that “the results of the talks, our agreements with the USA,
objectively create a serious obstacle with which to disrupt [China’s] foreign
policy toward a Sino-American rapprochement on an anti-Soviet basis and
lessen the likelihood of such an outcome.”223 The NSC mirrored the Soviet
Union’s assessment.224
effect
The Soviet Union’s attempt to use SALT to dampen an emerging U.S.-China
alignment partly worked. Despite the rapprochement after Nixon’s February
1972 visit to China, the two sides still had fundamental differences over their
policies toward the Soviet Union. China sought a “horizontal line” strategy
that entailed an anti-Soviet alliance stretching from Japan to Europe to the
United States.225 The United States, in contrast, promoted both détente and
U.S.-China rapprochement.226 China was skeptical about a U.S.-Soviet détente
because it knew that the Soviet Union was using détente to isolate China.
Thus, China warned the United States against appeasing the Soviets to “push
the ill waters of the Soviet Union . . . eastward.”227 Kissinger responded to
219. Ibid., p. 913.

220. Kozyrev, “Soviet Policy,” p. 264.
221. Haslam, Russia’s Cold War, p. 263.
222. Kozyrev, “Soviet Policy,” p. 264.
223. Quoted in Haslam, Russia’s Cold War, p. 263, original source in fn. 376.
224. Paper Prepared by the National Security Council Staff, undated, FRUS, 1969–1976, Vol. 32:
SALT I, 1969–1972, p. 817. The report added, “the demonstration of a Superpower relationship ex-
ploitable against China, was an underlying Soviet motive in the past negotiations, and is an incen-
tive for keeping the dialogue alive in the future [italics in original].”
225. Wang Zhongchun, “The Soviet Factor in Sino-American Normalization, 1969–1979” in Kirby,
Ross, and Li, eds., Normalization of U.S.-China Relations, pp. 156–157.
226. Ibid.
227. Evelyn Goh, “Nixon, Kissinger, and the ‘Soviet Card’ in the U.S. Opening to China, 1971–
China that improving U.S.-Soviet relations were aimed at reassuring the
Soviets of the U.S. intention not to encircle them, and that China should wait
for diplomatic normalization while the United States tempered Soviet ambi-
tions.228 We contend that the Soviet pursuit of strategic arms control with the
United States from 1969 to 1972 weakened alignment between China and
the United States and helped to delay the normalization of U.S.-China diplo-
matic ties to 1978. Ultimately, the wedge motive’s importance was revealed by
its diminution: U.S.-China rapprochement slowed down in 1973–74, and the
Soviet Union’s efforts to negotiate SALT II also decreased because it was less
concerned about U.S.-China collusion.229
Conclusion
Although states have used strategic arms control to promote stability, cut
costs, or secure a military edge, they have also exploited it to divide oppo-
nents. We have argued that this calculus was a key component of states’ deci-
sions to negotiate the Washington Naval Treaty, the LTBT, and the SALT I. The
recurrence of the wedge strategy logic, across the pre- and post-nuclear eras
and different adversarial alignment trajectories, shows the reach of the theory.
The wedge logic remains relevant today. First, we show how the United
States can employ the extension of the New START and future arms control
agreements to drive a wedge between China and Russia. Second, we con-
tend that the United States must be cautious about arms control deals with
North Korea.
In today’s context of the United States-Russia-China triangle, the United
States increasingly worries about the growing strategic partnership between
Russia and China.230 Such concern is reasonable.231 And it has inspired some
analysts to call for the United States to look for ways to cooperate with
Moscow that will distance Russia from China and allow the United States to
better concentrate on China.232 One of the most important ways for the United
1974,” Diplomatic History, Vol. 29, No. 3 (June 2005), pp. 475–488, doi.org/10.1111/j.1467-7709.2005
.00500.x.
228. Ibid., pp. 490–491.
229. Thomas W. Wolfe, “The SALT Process and Détente,” The SALT Experience: Its Impact on U.S.
and Soviet Strategic Policy and Decisionmaking (Santa Monica, Calif.: RAND, 1975), pp. 204–205.
230. Daniel R. Coats, Worldwide Threat Assessment of the U.S. Intelligence Community, U.S. Senate
Select Committee on Intelligence (Washington, D.C.: Ofªce of the Director of National Intelligence,
January 29, 2019), https://www.dni.gov/ªles/ODNI/documents/2019-ATA-SFR—-SSCI.pdf.
231. In October 2020, Putin speculated that a Russia-China military alliance would be possible
and such cooperation could help the two countries regularly hold ground and naval drills and ex-
change military technologies and practices. “Military Alliance with China to Tie Russia’s Hands,
Scare Off Partners, Says Expert,” TASS Russian News Agency, October 30, 2020, https://tass.com/
defense/1218485.
232. Robert D. Blackwill, Implementing Grand Strategy Toward China: Twenty-Two U.S. Policy Pre-
States to cooperate with Russia is via strategic arms control. When the
United States and Russia extended New START in February 2021, U.S.
President Joseph R. Biden, Jr. conveyed to Russian President Vladimir Putin
that the United States wants to work with Russia on matters concerning strate-
gic stability and arms control.233 Negotiating a bridge from New START to a
successor framework of U.S.-Russia strategic arms control could have the di-
rect effect of not only restraining an arms race but also, as the peace offensive
logic suggests, improving the United States’ ties with Russia and dampening
Russia-China convergence by unwinding their common threat perceptions, es-
pecially if the new pact is coupled with other initiatives to conciliate Russia.234
The future focus of U.S.-Russia strategic arms control should remain bilateral,
and it should be coupled with other measures to reduce tensions and build
new areas of cooperation between the United States and Russia.235 Moreover,
better U.S.-Russia ties can help the United States build trust with Russia’s tra-
ditional partners like India and Vietnam.
If the United States can use strategic arms control to divide adversaries, so
can others. North Korea has long opposed the continuation of the U.S.-South
Korea alliance, and one of the ways for North Korea to split the alliance could
be an arms control deal with the United States that limits North Korea’s devel-
opment of an intercontinental ballistic missiles (ICBM) capability in exchange
for reducing U.S. security commitments to South Korea via the speciªc linkage
logic. This reduction could include withdrawing U.S. troops from South Korea,
rolling back U.S. extended deterrence, or halting joint military exercises.236
Such a deal may be hard for the United States to turn down given North
Korea’s unimpeded development of nuclear and missile capabilities and the
scriptions, Council Special Report No. 85 (New York: Council on Foreign Relations, January
2020), pp. 23, 36, https://cdn.cfr.org/sites/default/ªles/report_pdf/CSR85_Blackwill_China.pdf;
Charles A. Kupchan, “The Right Way to Split China and Russia: Washington Should Help Moscow
Leave a Bad Marriage,” Foreign Affairs, August 4, 2021, https://www.foreignaffairs.com/articles/
united-states/2021-08-04/right-way-split-china-and-russia; and Timothy W. Crawford, “How to
Distance Russia from China,” Washington Quarterly, Vol. 44, No. 2 (Fall 2021).
233. The White House, “Readout of President Joseph R. Biden, Jr. Call with President Vladimir
Putin of Russia,” January 26, 2021, https://www.whitehouse.gov/brieªng-room/statements-
releases/2021/01/26/readout-of-president-joseph-r-biden-jr-call-with-president-vladimir-putin-
of-russia/; and Kremlin, “Telephone Conversation with U.S. President Joseph Biden,” January 26,
2021, http://en.kremlin.ru/events/president/news/copy/64936.
234. Sarah Bidgood, “U.S.-Russia Relations and the Future of Arms Control: How the Compre-
hensive Nuclear-Test-Ban Treaty Could Restore Engagement on Nuclear Issues,” Nonproliferation
Review, Vol. 25, Nos. 3–4 (2018), pp. 307–318, doi.org/10.1080/10736700.2018.1512203.
235. Crawford, “How to Distance Russia from China;” and Kupchan, “The Right Way to Split
Russia and China.”
236. Terence Roehrig, Japan, South Korea, and the United States Nuclear Umbrella: Deterrence after the
Cold War (New York: Columbia University Press, 2017), pp. 13–37.
ability of those ICBMs to target the U.S. East Coast, which is the essence of alli-
ance decoupling.237
Strategic arms control is, among other things, a tool of power politics and
one of many ways for states to increase their net security. At the level of grand
strategy, the logic of arms control wedge strategies thus points, somewhat
heretically, to an important possibility: Although it may be optimal for states
to gain advantage in both direct effects (i.e., increased stability, limited war,
and reduced expenses) and indirect effects of strategic arms control, they may
achieve large gains in security and relative power from the indirect political ef-
fects alone, despite some disadvantageous arms limits in the treaty terms. If
the indirect beneªts are expected to be more rewarding than direct ones, a state
may rationally accept unbalanced or otherwise imperfect terms to improve the
chances of a successful deal that allows it to drive a wedge between its adver-
saries. What appears to be a negotiating failure (i.e., the agreement terms
are technically disadvantageous) may be a victory for the state’s larger strate-
gic goals. The conventional focus on wins and losses in bilateral bargain-
ing over the formal terms and direct effects of strategic arms control will
miss, or underrate, this outside game, and misconstrue the nature and sig-
niªcance of them.
237. Nicholas L. Miller and Vipin Narang, “North Korea Deªed the Theoretical Odds: What Can
We Learn from Its Successful Nuclearization?” Texas National Security Review, Vol. 1, No. 2 (Febru-
ary 2018), pp. 65–67, doi.org/10.15781/T2M32NT02. On alliance decoupling, see Khang X. Vu,
“Driving the Enemies Apart: How Nuclear-Armed States Decouple Adversarial Alliances,” mas-
ter’s thesis, Dartmouth College, 2019.

Arms Control

Uploaded by

Document Information

Original Description:

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

Arms Control

Uploaded by

Copyright:

Available Formats

Arms Control as Wedge Strategy

Arms Control as Timothy W. Crawford

Existing Theories of Strategic Arms Control

the standard model of strategic arms control

the skeptic’s model of strategic arms control

Applying Wedge Strategy Theory to Arms Control

29. Coe and Vaynman, “Why Arms Control Is So Rare.”

Policymakers may employ a form of statecraft to obtain an indirect effect,

42. Jervis, System Effects, pp. 31–32.

a way to accommodate rather than coerce. Selective accommodation “singles

pp. 498–531, doi.org/10.1080/09636412.2013.816121; and Daniel Treisman, “Rational Appease-

speciªc linkage mechanism

52. Julian R. Friedman, “Alliance in International Politics,” in Julian R. Friedman, Christopher

catalyze conºict mechanism

peace offensive mechanism

Case Study Method and Research Design

Washington Naval Conference and the Anglo-Japanese Alliance

In December 1921, the United States convened the Washington Naval

of the Anglo-Japanese Alliance. As Aukland Geddes, the British Ambassador

Paciªc” to which a deal on “limitation of armaments in [the] Paciªc might

speciªc linkage mechanism

deal with the Anglo-Japanese Alliance at the Washington Naval Conference.90

LTBT and the Sino-Soviet Split

101. Klein, “Whitehall,” p. 468.

catalyze conºict mechanism

127. Li and Xia, “Jockeying for Leadership,” pp. 49–50.

ban negotiations rather than disagreements over the number of inspections

for embracing “the U.S. imperialists in joyous abandon”; the Soviets, in

SALT and the U.S.-China Rapprochement

niªcant effects on Soviet and U.S. maneuvers during SALT I negotiations.178

peace offensive mechanism

201. Smith, Doubletalk, p. 139.

Brezhnev, general secretary of the CPSU, saw the summit as an “oppor-

219. Ibid., p. 913.

China that improving U.S.-Soviet relations were aimed at reassuring the

You might also like