Walt - Alliance Formation - Test Theory
Walt - Alliance Formation - Test Theory
Stephen M. Walt
International Organization, Vol. 42, No. 2. (Spring, 1988), pp. 275-316.
Stable URL:
http://links.jstor.org/sici?sici=0020-8183%28198821%2942%3A2%3C275%3ATTOAFT%3E2.0.CO%3B2-1
International Organization is currently published by The MIT Press.
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/mitpress.html.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact [email protected].
http://www.jstor.org
Sun Aug 5 00:52:47 2007
When will states form alliances, and what determines their choice of allies?
This question is important for at least two reasons. First, from a policy
perspective, competing hypotheses about the causes of alignment lie at the
heart of many recurring debates in U.S. foreign policy. One fundamental
issue is whether states tend to balance strong or threatening powers by
allying against them, or whether they are more likely to "bandwagon" by
allying with the most powerful or threatening states. For example, the widespread belief that states tend to bandwagon has magnified concern for the
credibility of U.S. commitments and is still used to justify U.S. involvement
in Third World conflicts.' Likewise, the fear of "Finlandization" rests upon
the belief that U.S. allies will tilt towards the Soviet Union should the U.S.
commitment to Europe d e ~ l i n e By
. ~ contrast, others invoke the logic of
balancing to argue that U.S. allies would do more for themselves if the
United States did less.3
I would like to thank Doug Blum, Selig Harrison, Ted Hopf, Robert Jervis, Zalmay
Khalilizad, Stephen Krasner, Bruce R. Kuniholm, Deborah Welch Larson, Jack Snyder,
Stephen Van Evera, W. Howard Wriggins, and two anonymous reviewers for their comments,
and the Carnegie Endowment for International Peace for financial support.
1. As President Reagan stated in a speech advocating support for the contras: "If we cannot
defend ourselves [in Central America] . . . then we cannot expect to prevail elsewhere. . . .
[Olur credibility will collapse and our alliances will crumble. . . . If Central America were to fall,
what would the consequences be for our position in Asia and Europe and for alliances such as
NATO? . . . Which ally, which friend would trust us then?" See New York Times, 28 April 1983,
p. A12.
2. For example, General Bernard Rogers, former supreme allied commander in Europe, has
stated that "if the U.S. withdraws 100,000 troops from Europe, this won't make the Europeans
do more. . . . [it] is going to send the kind of message that will lead [the European NATO
members] to start to accommodate to the East." Quoted in Edgar Ulsamer, "The Potential
Checkmate in Europe," Air Force Magazine 69 (November 1986), pp. 55-56.
3. See, for example, Christopher Layne, "Atlanticism Without NATO," Foreign Policy 67
(Summer 1987); Earl Ravenal, "Europe Without America: The Erosion of NATO," Foreign
Affairs 63 (Summer 1985); Irving Kristol, "What's Wrong with NATO?" New York Times
Magazine, 25 September 1983; and David P. Calleo, Beyond American Hegemony: The Future
of the Western Alliance (New York: Basic Books, 1987).
International Organization 42, 2, Spring 1988
O 1988 by the World Peace Foundation and the Massachusetts Institute of Technology
international politics. At the same time, its limitations suggest that certain
modifications are in order. Because predictions about alliances are a central
part of the theory, and because alliances play an especially important role in
international politics, examining alliance formation is an obvious place to
start.8
In this spirit, several recent works have examined competing hypotheses
about alliance formation, drawing upon both traditional diplomatic history
and recent alliance commitments in the Middle ~ a s tThese
. ~ studies revealed
that states form alliances primarily to balance against other states, and that
"bandwagoning" behavior-that is, alignment with the dominant state or
coalition-was relatively rare. Contrary to structural balance-of-power theory, however, these studies also showed that states balanced against several
different kinds of threats, and not just against power alone. In this article, I
refine these hypotheses further and present additional evidence to support
these conclusions.
To do this, I examine alliance formation in Southwest Asia since World
War 11, focusing on the alliance commitments of Iran, Turkey, India, and
Pakistan. As I shall discuss at greater length, these cases are especially
useful because they provide strong tests of several especially important
propositions.
Thus, this article has two broad objectives: first, to present evidence
identifying which hypotheses on alliance formation provide the best predictions of state behavior; and second, to highlight several limitations in structural balance-of-power theory and to suggest how they can be overcome.
Part 1 outlines the different hypotheses and examines several important
issues of interpretation. Part 2 describes the postwar alliance policies of
Iran, Turkey, India, and Pakistan, and identifies why these states chose the
allies that they did. Part 3 analyzes the different hypotheses in light of these
decisions and explores the implications of these results.
8. The importance of alliances in international politics is widely acknowledged. According to
Julian Friedman, "alliances are the central feature of international political life." See his "Alliance in International Politics," in Friedman, Christopher Bladen, and Steven Rosen, eds.,
Alliance in International Politics (Boston: Allyn & Bacon, 1970). In George Modelski's view,
alliance is "one of a dozen or so key terms of international relations." See his "The Study of
Alliances: A Review," Journal of Conjict Resolution 7 (December 1963), p. 773. Ole Holsti
describes alliances as "a universal component of relations between political units," and Hans
Morgenthau states that "alliances are a necessary function of the balance-of-power operating in
a multiple state system." See Holsti, P. Terrence Hopmann, and John D. Sullivan, Unity and
Disintegration in International Alliances (New York: Wiley, 1973) p. 2, and Hans J. Morgenthau, "Alliances in Theory and Practice," in Arnold Wolfers, ed., Alliance Policy in the Cold
War (Baltimore: Johns Hopkins University Press, 1959), p. 185.
9. See Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press,
1987), and "Alliance Formation and the Balance of World Power," International Security 9
(Spring 1985).
When confronted by a significant external threat, states may either balance or bandwagon. Balancing is alignment against the threatening power to
deter it from attacking or to defeat it if it does. Bandwagoning refers to
alignment with the dominant power, either to appease it or to profit from its
victory. The question is: which tendency is more common?
The answer has far-reaching implications. If states tend to balance, then
great powers can take a relaxed view of most international events. Aggressors will face numerous opponents and sustained efforts to expand are likely
to fail. Statesmen should avoid bellicose behavior because it will encourage
others to align against them. But if bandwagoning is the dominant tendency,
then security is scarce because allies will be prone to defect. l 1 Great powers
must strive to maintain their credibility so that fear does not lead their allies
to realign. Threats and intimidation are more likely to work, and empires will
both be easier to amass and more likely to disintegrate.'*
Ironically, although balance-of-power notions pervade the scholarly literature on international politics, the United States (and to a lesser extent, the
Soviet Union as well) has often acted as if bandwagoning were the dominant
tendency. U.S. leaders have been preoccupied with preserving credibility,
fearing that setbacks in one area would lead to widespread defections elsewhere. As John F. Kennedy once put it: "If the United States were to falter,
the whole world would inevitably move towards the Communist bloc."13
10. Portions of this section are based on Walt, Origins of Alliances, chaps. 2 and 8.
1 1 . See W. Scott Thompson, "Toward a Communist International System," Orbis 20 (Winter 1977); Charles Krauthammer, "Isolationism: Left and Right," New Republic, 4 March 1985;
and Colin S. Gray, Maritime Strategy, Geopolitics, and the Defense of the West (New York:
National Strategy Information Center, 1986), pp. 38-40. None of these authors provide persuasive evidence to support their claims.
12. Empires are easier to amass because each successful conquest will encourage others to
"leap on the bandwagon," thereby increasing the dominant power's capabilities while reducing
resistance. Empires will also be fragile, however, because minor setbacks may trigger a wave of
defections.
13. Quoted in Seyom Brown, The Faces of Power: Constancy and Change in United States
Foreign Policy from Truman to Johnson (New York: Columbia University Press, 1968), p. 217.
For additional evidence on U.S. perceptions, see: Deborah Welch Larson, "The Bandwagon
Metaphor and American Foreign Policy ," paper presented at the International Studies Association convention, 25-29 March 1986; Walt, Origins ofAlliances, chaps. 1 and 2; Patrick Morgan,
"Saving Face for the Sake of Deterrence," in Robert Jervis, Richard Ned Lebow, and Janice
Gross Stein, eds., Psychology and Deterrence (Baltimore: Johns Hopkins University Press,
1985); and John Lewis Gaddis, Strategies of Containment (New York: Oxford University
Press, 1981). The Soviet belief in bandwagoning can be seen in the claim that positive shifts in
the "correlation of forces" will make additional international gains easier rather than harder,
and in the corresponding belief that the growth of Soviet power forced the United States to
accept detente in the 1970s. See William Zimmermann, Soviet Perspectives on International
Relations (Princeton, N.J.: Princeton University Press, 1969), pp. 159-64; Raymond J. Garthoff, Detente and Confrontation: American-Soviet Relations from Nixon to Reagan (Washington, D.C.: Brookings Institution, 1985), pp. 36-53; and Coit D. Blacker, "The Kremlin and
Alliance formation
279
The belief that allies will bandwagon helps justify both intervention abroad
and military spending at home to convince allies that their partner is still
reliable. Yet despite the importance of this hypothesis as a justification for
policy, it has rarely been challenged or tested.
According to structural balance-of-power theory, the uncertainties inherent in anarchy encourage balancing behavior. Bandwagoning is risky because it requires trust; one assists a dominant power in the hope that it will
remain benevolent. It is more prudent to balance, in case the strongest state
turns out to be aggressive. Joining the weaker side thus prevents the emergence of a hegemon that could threaten the independence of all. Furthermore, alignment with the weaker side enhances one's influence within the
resulting coalition because the weaker side has greater need of assistance. l4
Bandwagoning may be chosen under some conditions, however. Extremely weak states may be more inclined to bandwagon; because they have
little deterrent or defensive strength to contribute, they must seek the winning side at all costs.15 And when allies are unavailable, weak states may be
forced to bandwagon even if this is not their first choice. Bandwagoning is
also more likely when the dominant power appears to be appeasable, that is,
when its ambitions can be satisfied or deflected should the threatened power
opt for alignment with it. Because perceptions are unreliable and intentions
can change, however, bandwagoning should be relatively rare.
Balance of power or balance of threat?
In structural balance-of-power theory, balancing and bandwagoning are
defined solely by capabilities. Balancing is alignment with the weaker side,
bandwagoning means to choose the stronger. Although this conception is
elegant and parsimonious-the distribution of power (defined as aggregate
capabilities) is the only important variable-it has serious limitations as
well. In particular, states may balance by aligning with the stronger side, if a
weaker power is more dangerous for other reasons (for example, if its intentions are especially malign). By focusing solely on the distribution of capabilities, structural balance-of-power theory ignores the other factors that
statesmen will consider when making alliance choices. As a result, the theory cannot explain why balances often fail to form.
For example, the coalitions that formed against Germany and its allies in
World Wars I and I1 possessed far greater overall capabilities, and the disparity grew steadily until Germany was totally defeated.16 Although balanceof-power theory helps explain why these alliances eventually dissolved (that
is, German power had been nullified by its defeat), it cannot explain why
Germany attracted such widespread opposition, given that it was ultimately
weaker than the coalition it fought."
Similarly, if states were concerned solely with balancing power, America's predominant position after World War I1 should have led the nations of
Western Europe to align with the Soviet Union against the United States.''
Instead, they chose to balance the Soviet Union by aligning with the United
States, because the former's proximity, its impressive military power, and
its apparently aggressive aims made it appear more threatening. As a result,
the U.S. alliance network dwarfs the Soviet Union and its allies on most
measures of national power, and this imbalance of power has remained
remarkably constant for nearly four decades.19 At a minimum, therefore,
structural balance-of-power theory often leads to inaccurate prediction^.^'
These deficiencies can be overcome by recognizing that states seek allies
to balance threats, and that power is merely one element in their calculations-albeit an important one. In addition to its overall capabilities, the
degree to which a state threatens others is also affected by its geographic
16. In World War I, the alliance of Great Britain, France, and Russia controlled 27.9% of
world industrial production, while Germany and Austria together controlled only 19.2%. With
Russia out of the war but with the United States joining Britain and France, the percentage
opposing the Dual Alliance reached 51.7%, an advantage of more than 2 to 1. In World War 11,
the defense expenditures of the United States, Britain, and Russia exceeded those of Germany
by roughly 4.5 to 1. Even allowing for German control of Europe and the war against Japan, the
Grand Alliance possessed an enormous advantage in capabilities. Thus, the coalitions that
waged the two most important wars of this century cannot be explained by focusing solely on
the distribution of capabilities. For these and other statistics on relative power in the two world
wars, see Paul M. Kennedy, "The First World War and the International Power System,"
International Security 9 (Summer 1984); and Kennedy, The Rise and Fall of British Naval
Mastery (London: Allen & Unwin, 1983), pp. 309-15.
17. The explanation, of course, is that the Allies were united by the common awareness that
although Germany was ultimately less powerful, its aggressive actions made it more dangerous.
18. In 1950, the United States produced approximately 40% of gross world product; the
Soviet Union managed only 13.5%. U.S. naval and air power were far superior, and the United
States had a clear advantage in deliverable atomic weaponry.
19. The United States and its allies lead the Soviet Union and its allies by more than 3 to 1 in
GNP and nearly 2 to 1 in population. The two coalitions are roughly equal in number of men
under arms, and the Western alliance spends roughly 15% more each year on defense. Although the U.S. share has declined, the overall disparity has changed little since 1950. Indeed,
the most significant shift-China's realignment with the West-has added even more to the
current imbalance of power. For the evidence upon which this assessment is based, see Walt,
Origins of Alliances, Appendix 2.
20. As A. F. K. Organski puts it, "If we look at the whole sweep of international history for
the past 150 years, we find that balances of power are the exception, not the rule." See his
World Politics (New York: Knopf, 1968), p. 293.
Alliance formation
281
Balance-of-power theory
Imbalances of power
cause
)
Alliances
against the
most powerful state
An imbalance of power occurs when the strongest state or coalition in the system possesses
significantly greater power than the second strongest. Power is the product of many
components, including population, economic and military capability, technological capacity,
political cohesion, and so on.
Balance-of-threat theory
Imbalances of threat
causc
An imbalance of threat occurs when the most threatening state or coalition is significantly
more dangerous than the second most threatening state or coalition. The degree to which a
state threatens others is the product of its aggregate power, geographic proximity, offensive
capability, and the aggressiveness of its intentions.
FIGURE 1. Balance-ofpower versus balance-ofthreat theory
Source. Adapted from Stephen M. Walt, The Origins of Alliances (Ithaca, N . Y . : Cornell University Press, 1987). Used by permission.
proximity, offensive capabilities, and perceived intentions. Other things being equal, states that are nearby are more dangerous than those that are far
away. States with large offensive capabilities-defined as the capacity to
threaten the sovereignty of other states-pose a greater threat than states
whose capabilities are more suitable for defense. Lastly, states with aggressive intentions are more threatening than those who seek only to preserve
the status quo. If balancing behavior is the norm, therefore, an increase in
any of these factors-power, proximity, offensive capabilities, or aggressive
intentions-should encourage other states to ally against the most threatening power.
'Balance-of-threat" theory should be viewed as a rejinement of balanceof-power theory. Where balance-of-power theory predicts that states ally in
response to imbalances of power, balance-of-threat theory predicts that
states seek allies when there is an imbalance of threat (that is, when one
state or coalition is especially dangerous). The two theories are equally
parsimonious, but balance-of-threat theory is more general and abstract. As
shown in Figure 1, the main concept informing balance-of-power theory is
the distribution of capabilities, which is based on population, economic
capacity, military power, political cohesion, etc. The central concept of
balance-of-threat theory is the distribution of threats, which consists of capabilities, proximity, offensive power, and intentions. Thus, both theories
are based on a single general concept incorporating a number of components, although balance-of-threat theory subsumes balance-of-power theory
by incorporating capabilities, geography, and intentions. As we shall see, it
Problems of interpretation
Before turning to the historical record, several issues of interpretation
should be clarified. First, balancing and bandwagoning are ideal types, and
actual behavior will only approximate either model. States that choose to
bandwagon will not leave themselves completely vulnerable, and they may
offer only modest support to the dominant power. States that choose to
balance may also seek cordial relations with their opponents, while simultaneously obtaining protection against them. Accordingly, testing these hypotheses requires careful historical interpretation.
A second issue concerns the definition of bandwagoning itself. Whenever
one state seeks better relations with an opponent, its allies may view this as
a form of bandwagoning. If they believe that this behavior is caused by a
decline in their own position, they will fear that additional defections are
likely. Yet an ally's decision to seek better relations with the adversary may
reflect its own belief that the adversary is no longer as dangerous. If the
adversary has become less aggressive or if one's own capabilities have
grown, for example, then a dCtente may be desirable. Far from reflecting
fear or vulnerability, a successful dCtente may actually reveal that earlier
efforts to balance have worked quite well.
Several criteria can help distinguish between bandwagoning and dCtente.
Bandwagoning involves unequal exchange; the vulnerable state makes
asymmetrical concessions to the dominant power and accepts a subordinate
role. DCtente, by contrast, involves roughly equal concessions in which both
sides benefit. Bandwagoning is an accommodation to pressure (either latent
or manifest) while dCtente implies a mutual recognition of legitimate interests. Most important of all, bandwagoning suggests a willingness to support
or tolerate illegitimate actions by the dominant ally. By contrast, dCtente
carries no such implication; indeed, a detente will probably collapse if an
opponent seeks to exploit it.22
Third, when examining the historical record, we should focus not only on
what states did, but also on what they preferred to do. As noted, states may
be forced to bandwagon if they are unable to obtain adequate support. If the
United States returned to an isolationist policy, for example, some of its
21. Balance-of-threat theory can also be conceived in systemic terms, because its predictions
are based on the distribution of threats rather than the absolute "threat potential" of a single
state. Thus, if two states are equally powerful but one seems more aggressive, the latter would
be more threatening and more likely to provoke others to align in response. Measuring the
"balance of threats" would be difficult, but so is measuring the balance of power.
22. An obvious example is the breakdown of the Soviet-American detente after its heyday in
the early 1970s, amid mutual accusations that the other side had sought to exploit the relationship.
allies might choose to realign with the Soviet Union. But this is not the real
issue. The real question is the relative propensity for states to balance or
bandwagon: how much support is needed to convince them to balance a
threat? If bandwagoning is the dominant tendency, then patrons must work
very hard to prevent defections. But if balancing is the preferred choice,
then even modest allied support should convince others to oppose an aggressor.
Fourth, allies often exaggerate their propensity to bandwagon, in order to
extract greater support from their patrons. During the War of Attrition with
Israel, for example, President Gamal Abdel Nasser of Egypt convinced the
Soviet Union to increase its military aid by threatening to resign in favor of a
pro-U.S. president.23Because the propensity for states to bandwagon will be
exaggerated if we look only at what elites say and ignore what they do,
careful interpretation is necessary.
Finally, efforts to test these competing hypotheses can be undermined by
circular reasoning and excessive reliance on post hoc analysis. The former
problem is straightforward and easily avoided: we cannot use the existence
of an alliance as evidence that a threat existed, and then use that same
alliance as "evidence" that external threats cause states to seek an ally.
Instead, independent evidence of an external threat (for example, increases
in military power, bellicose statements or actions, and so on) should be
compared with each state's response (that is, did it balance or bandwagon?).
More importantly, because both balance-of-power and balance-of-threat
theory are extremely general and abstract, it may be difficult to reject either
one on the basis of a few cases. After all, one can always salvage a theory by
introducing ad hoc explanations to account for persistent anomalies. This
problem is compounded by the fact that alliance choices are based on subjective assessments: it is the actors' perceptions of threat that count, not the
analyst's "objective" evaluation. And because geographic location and intentions affect the level of threat that a state poses to others, different states
will view potential allies and adversaries in different ways, unless one state
is so strong and aggressive that the threat it presents is overwhelmingly
clear.
The danger, of course, is that anomalies can always be explained away by
arguing that the actors failed to follow the theory's predictions because they
misperceived the "true" situation. Fortunately, this pitfall is not a serious
problem here, because balance-of-threat theory explicitly acknowledges the
importance of perceptions (of threat). Assuming we have accurate indepen23. See Mohamed Heikal, The Road to Ramadan (New York: Quadrangle, 1975), pp. 82-90.
And in the early years of the cold war, the French executive director of the World Bank
remarked that "the Communists are rendering [Europe] a great service. Because we have a
'Communist danger,' the Americans are making a tremendous effort to help us. We must keep
up this indispensable Communist scare." Quoted in Walter Isaacson and Evan Thomas, The
Wise Men: Six Friends and the World They Made (New York: Simon & Schuster, 1986), p. 426.
dent information on perceptions, we can test whether balancing or bandwagoning was preferred by asking whether policymakers chose to align with
or against the states they perceived as the most powerful or threatening. We
can also explore which sources of threat were most important in their calculations. The problem of misperception can also be reduced by examining
alliance formation over an extended period, because any significant gaps
between perceptions and reality are likely to be clarified over time.24 Thus,
the danger of salvaging the theory through post hoc assessment is hardly
insurmountable.
Because conclusive evidence may be scarce, however, we should examine a large number of cases and remain alert for events that are consistent
with one theory and inconsistent with its rivals. We should also seek strong
tests of likely hypotheses, rather than relying solely upon cases in which the
theory's predictions are overwhelmingly likely to hold.
For all of these reasons, examining alliance formation in Southwest Asia is
especially valuable. Because earlier studies have focused on other areas,
these cases add new evidence from a region that is of increasing strategic
interest, yet rarely examined by theorists of international politics. Even
more important, the recent history of this region provides for a strong test of
the hypotheses on balancing and bandwagoning. Each of these states has
faced a number of internal and external challenges. All are close to the Soviet
Union and far from the United States, which increases their vulnerability to
Soviet pressure. Because U.S. involvement in this region was minimal prior
to World War 11, the United States had no legacy of commitment to bolster
its credibility. With respect to the Soviet Union, therefore, these states were
likely candidates for bandwagoning. If they chose to balance instead, that
would be strong evidence that balancing is the preferred response. Similarly,
if these states sought allies primarily to counter the most powerful state in
the system, or if their commitments shifted with changes in the distribution
of capabilities, that would tend to support balance-of-power theory. But if
they sought allies in order to oppose other regional powers, or if changes in
intentions were more important than shifts in relative power, then we should
conclude that it is the balance of threats, not the balance of power alone, that
plays the principal role in determining alliance choices. Finally, we can also
consider the impact of ideology: have these states tried to align with states
whose domestic systems resemble their own, or are they relatively indifferent to such considerations when seeking external support? With these predictions in mind, let us now turn to the historical record.
24. Thus, the widespread belief that Great Britain was still a superpower was corrected
rather quickly after World War 11, and the sluggish response to the danger posed by Nazi
Germany vanished once events dispelled any confusion about Hitler's ultimate aims.
Tudeh party, several rebellious tribes, and the two dissident republics.
Qavam's efforts to appease these groups led U.S. officials to fear that he was
delivering Iran into communist hands." By October, both the Shah and the
prime minister had become concerned as well, and, with American encouragement, they abandoned their conciliatory a p p r ~ a c h . 'The
~ Tudeh ministers were ousted and the Iranian army reoccupied the rebellious republics in
December. The parliament ultimately rejected the oil agreement in 1947,
leaving the Soviets empty-handed.
Iran's behavior during the crisis is revealing. Despite its vulnerability to
Soviet pressure, Iran made only minor concessions to its more powerful
neighbor. Qavam displayed great resolution in his dealings with Stalin, and
his domestic compromises were mostly tactical. Although the Iranians did
not wish to antagonize the Soviet Union, they clearly preferred to balance it
through an alignment with the United States. Indeed, the Shah had tried to
obtain a U.S. guarantee as early as 1943, and Iranian officials made repeated
requests for American support during 1945.~'
Yet the United States did not make a clear commitment to Iran until
October 1946. Despite the Shah's requests, the United States did not adhere
to the 1942 Anglo-Russian-Iranian treaty guaranteeing Iranian sovereignty.
In his talks with Qavam in March 1946, the U.S. ambassador refused to offer
tangible support because he "could not be sure that the U.S. Government or
the U.N.O. [United Nations Organization] could save Iran [if the Soviets
atta~ked]."~'
In August, Qavam's requests for American help received only
the "suggestion of American support . . . less than a forthright promise,"
and by September, the Iranians had concluded that concrete aid was unlikely.32This conclusion was not surprising: when the Iranians requested a
28. Qavam tried unsuccessfully to negotiate with the Azerbaijani and Kurdish republics in
June, and he invited three members of the Tudeh Party to join the cabinet in August. This was
probably just a tactical concession, as Qavam also formed his own "Democratic Party of Iran"
at this time in order to oppose the Tudeh in the upcoming elections.
29. According to the U.S. ambassador, Qavam began contemplating a "sharp change in
policy" in September because "he realized that [the] policy of conciliation towards Azerbaijan
had not yielded favorable results." See FRUS 1946, vol. 7, p. 518.
30. In 1943, the Shah told American officials that he would "like to see American interest in
Iran continue and grow," and he also stated that an American guarantee "would be most
helpful in counteracting the ever-growing Soviet threat." See Ramazani, Iran's Foreign Policy,
pp. 57-58, 254. In November and December of 1945, Iranian officials in the United States
inquired further about American intentions, and warned that "the confidence of small countries" would be eroded if the United States did not help Iran. See Kuniholm, Origins of the Cold
War, pp. 279, 284-85.
31. Ambassador Murray also noted that "we cannot provide Iran with an insurance policy
against all dangers." See FRUS 1946, vol. 7, pp. 374-75; and Kuniholm, Origins of the Cold
War, p. 327. The Iranians were troubled by America's rapid demobilization after the war, and
by the fact that the United States and Great Britain had withdrawn their troops from Iran while
Soviet forces remained.
32. According to Bruce R. Kuniholm, "Qavam and the Shah asked [Ambassador] George
Allen repeatedly what either U.S. or U.N. assistance meant in concrete terms. On 13 August,
on 24 August, in early September, and on 29 September, Allen received pointed inquiries. But
Alliance formation
287
$250 million credit, they were informed that $10 million was a more realistic
possibility .33
It is true that George V. Allen, who became U.S. ambassador to Iran in
May 1946, worked hard to bolster Qavam's resolve.34Yet it is also striking
how small the U.S. commitment was and how long the Iranians waited for
it.35 First, the United States did not pledge its support until after Qavam
.~~
expelled the Tudeh representatives from the cabinet on 18 O ~ t o b e rAllen
himself believed that U.S. backing for Iran had been paltry, and he was
convinced that Qavam preferred to resist Soviet pre~sure.~'
As he wrote
several weeks later: "many people wanted the [ouster of the Tudeh], not
least of whom was Qavam
Finally, Qavam and the Shah chose to
subdue the rebellious province of Azerbaijan (thereby risking direct Soviet
intervention), despite their awareness that direct U.S. assistance was unlikely.39They courted Soviet displeasure again several months later by rehis cautious advice, together with Henry Wallace's speech of 12 September, were taken by the
Iranians as evidence that Iran could expect little in the way of concrete assistance from the
United States." See his Origins of the Cold War, p. 384 (emphasis added). At the beginning of
October, Undersecretary of State Dean Acheson told the U.S. ambassador that "responsibility
for maintaining Iran's independence and integrity rests primarily with the government of Iran."
See FRUS 1946, vol. 7, p. 527. See also Pfau, "Containment in Iran," pp. 362-63.
33. See FRUS 1946, vol. 7, pp. 519-20.
34. A number of accounts conclude that Allen's behind-the-scenes activities were crucial in
persuading Qavam and the Shah to resist Soviet and Tudeh pressure in October. Allen pressed
Qavam to take action against the Tudeh on 11 October, and repeated his warnings to the Shah
three days later. According to the Shah's version of this episode, Qavam was reportedly fearful
of the Soviet reaction but the Shah ordered him to oust the Tudeh from his cabinet and restore
the government's authority. See Pfau, "Containment in Iran," pp. 365-68; Paterson, SovietAmerican Confrontation, pp. 177-83; and Kuniholm, Origins of the Cold War, especially pp.
388-91.
35. Having ordered Qavam to oust the Tudeh on 17 October, the Shah made yet another
request for American support the following day. The United States finally announced on 28
October that it would sell Iran several million dollars of military equipment, and Ambassador
Allen told an Iranian newspaper that he thought sending troops into Azerbaijan was a "perfectly
normal and proper decision." Encouraged by an off-the-record statement by Acheson in
November and by Allen's continued cooperation, Qavam and the Shah proceeded with their
plans to restore authority in Azerbaijan. See Kuniholm, Origins of the Cold War, pp. 393-94.
36. Qavam had begun "contemplating a sharp change of policy" in late September, convinced that his policy of accommodating the Tudeh had not worked. The limited nature of U.S.
support is revealed by Allen's request on 30 September that he "would welcome instructions
which would justify [a] more favorable response to Qavam and also to [the] Shah. . . ." FRUS
1946, vol. 7, pp. 519-20.
37. As Allen told Secretary of State James F. Byrnes: "We must keep in mind [that Qavam's]
appeals to the U.S. for help have been answered by advice to depend on United Nations for
security. Our advice has been best we could give, but Qavam's problems are immediate. . . ."
FRUS 1946, vol. 7, p. 542. Significantly, Allen made this observation after Qavam had taken the
decisive step of ousting the Tudeh from the Cabinet.
38. Quoted in Kuniholm, Origins of the Cold War, p. 389, note 18 (emphasis added).
39. Discussing his plans to oust the rebel government in Azerbaijan, Qavam told Allen on 23
November that "he realized troops could probably not be sent to aid Iran but he felt Iran must
bring to [the U.N. Security] Council's attention a situation which threatened world peace. . . .
He hoped members of Council would at least show their approval of [the] Iranian Government's
efforts to maintain its sovereignty." FRUS 1946, vol. 7, p. 548.
neging on the oil agreement. The United States supported each of these
actions, but still refused to provide Tehran with significant economic or
military aid.40
In sum, the crisis of 1946 suggests that it takes relatively little effort to
persuade states to balance. Despite their own vulnerability, repeated Soviet
threats, and very limited American support, Iran's leaders clearly preferred
to balance rather than bandwagon. Qavam and the Shah were challenged by
the Tudeh party, by autonomy movements in Azerbaijan and Kurdistan, by
several rebellious tribes, and by British efforts to preserve their access to
Iranian oil. Alignment with the United States helped meet these interrelated
internal and external threats. Indeed, using a less threatening third power to
counter British and Russian interference was a familiar Iranian t a ~ t i c . ~ '
Until the mid-1950s, Iran's efforts to obtain a greater U.S. commitment
met with little success. Despite the Shah's repeated pleas, the principal link
was a small U.S. military training mission, and the United States provided a
mere $16 million worth of military credits (and no economic aid) from 1946
to 1951.42Taking Iran for granted was easy: although he was perennially
dissatisfied with the level of American aid, the Shah made no attempt to
obtain support from the Soviet
The "globalization" of containment in the 1950s breathed new life into the
U.S.-Iranian relationship. The United States and Great Britain staged a
successful covert operation to restore the Shah to the throne in 1953, after
Prime Minister Mohammed Mossadegh's efforts to nationalize the AngloIranian Oil Company led to prolonged internal turmoil and American fears of
a Communist takeover.44 Subsequent efforts to establish an anti-Soviet al40. In April 1947, Loy Henderson, director of Near East and African affairs, told a group of
Iranian representatives that "[ilt is desirable that Iran should keep the best possible relations
with the Soviet Union . . . [economic or military] grants can be given only to those countries
under direct threat and danger." FRUS 1947, vol. 5, p. 905. Similarly, Allen told the Shah that
"occasions arise when small states are called upon to defend their essential rights regardless of
assurances . . . defending their essential interests even in the absence of airtight guarantees of
help." FRUS 1947, vol. 5, p. 922.
41. This theme is elaborated in R. K. Ramazani, The Foreign Policy of Iran, 1500-1941: A
Developing Nation in World Affairs (Charlottesville: University of Virginia Press, 1966), pp.
277-300.
42. Ramazani argues that Iran's desire for American assistance was primarily political; the
Shah sought economic ties with the United States to deepen the U.S. political commitment to
Iran. See his Iran's Foreign Policy, pp. 163-64.
43. According to George McGhee, assistant secretary of state for Near Eastern affairs, the
Shah told President Truman in 1949 that Iran had "always fought Russia with her bare hands,"
and that he would do so with or without American help. This suggests that American credibility
had little to do with Iran's interest in balancing against the Soviet Union. See his Envoy to the
Middle World (New York: Harper & Row, 1983), p. 69 and passim. See also Barry Rubin,
Paved with Good Intentions: The American Experience and Iran (New York: Oxford University Press, 1980), pp. 36-41, 49-51.
44. Although he had little sympathy for the Soviet Union, Mossadegh balanced AngloAmerican pressure by seeking support from the Tudeh Party prior to his ouster. For accounts of
this episode, see: Richard J. Barnet, Intervention and Revolution (New York: New American
Alliance formation
289
liance in the Near East brought an enthusiastic response from Tehran, and
Iran joined the so-called Baghdad Pact despite U. S. reservations .45 Iran
began receiving substantial U.S. assistance, the Shah endorsed the anticommunist "Eisenhower Doctrine," and relations grew even closer following the Iraqi revolution in July 1958. The Shah rejected Soviet pressure for
Iranian neutrality, U.S. military aid increased even further, and the two
countries signed a bilateral security agreement in March 1 9 5 9 . ~ ~
These steps made Iran a central member in the web of anti-Communist
alliances established by the United States. For the United States, Iran was
another link in the chain of containment intended to balance the threat of
Soviet expansion. For Iran, U.S. support served as a shield against pressure
from either the Soviet Union or hostile neighbors like Iraq, and also bolstered the Shah's position vis-a-vis his various internal opponents.47
In the 1960s, however, Iran began a significant dCtente with the Soviet
Union, under the guise of the Shah's "Independent National Policy." In
response to Soviet requests, Iran announced that it would not permit foreign
missile bases on its territory. This decision paved the way for a series of
agreements on trade, transit, and economic assistance. The Shah visited
Moscow in June 1965, and agreed to purchase roughly $100 million worth of
Soviet arms in 1967.~'
This shift was the result of several factors. First, the Shah resented American pressure for internal reforms and for limits on Iranian military expendiLibrary, 1968), pp. 226-29; Kermit Roosevelt, Countercoup: The Struggle for Control of Iran
(New York: McGraw-Hill, 1979); Rubin, Paved with Good Intentions, chap. 3 ; and Ramazani,
Iran's Foreign Policy, pp. 247-50.
45. The Baghdad pact was a treaty network linking Turkey, Pakistan, Iraq, Iran, and Great
Britain, with the United States as an observer. Secretary of State John Foster Dulles was
opposed to Iran's membership (believing it would jeopardize the Shah's internal position) but
the Shah staged a diplomatic fait accompli by announcing his desire to join. See Shahram
Chubin and Sepehr Zabih, The Foreign Relations of Iran: A Developing State in a Zone of
Great-Power Conjict (Berkeley: University of California Press, 1974), pp. 90-92; and Rubin,
Paved with Good Intentions, p. 97. Iran's motives were revealed by Premier Fazlollah Zahidi,
who stated in 1954 that "we have witnessed how aggressors have wantonly occupied neutral
countries in defiance of international law . . . Therefore, it is certain . . . that a government can
preserve itself only if it has the power of resistance against an aggressor." New York Times, 28
July 1954.
46. The Soviets offered Iran a SO-year non-aggression treaty and substantial foreign aid if
Iran would formally adopt a neutral position. Although the Shah was concerned by the U.S.
refusal to sign a formal treaty (the 1959 accord was an executive agreement), he declined the
Soviet offer. Indeed, these negotiations may well have been a ploy to attract greater American
support. See R. K. Ramazani, The United States and Iran: The Patterns of Injuence (New
York: Praeger, 1982), pp. 38-39; Iran's Foreign Policy, pp. 280-85,295-99; Rubin, Paved with
Good Intentions, pp. 101-2; and Alvin Z. Rubinstein, Soviet Policy Toward Turkey, Afghanistan, and Iran: The Dynamics of InJluence (New York: Praeger, 1982), pp. 68-69.
47. For example, U.S. aid enabled the Shah to balance Iran's persistent budgetary deficits
throughout the 1960s. See Rubin, Paved with Good Intentions, chap. 4.
48. See Chubin and Zabih, Foreign Relations of Iran, p. 77 and passim, and Sepehr Zabih,
"Iran's International Posture: De Facto Non-Alignment Within a Pro-Western Alliance," Middle East Journal (Summer 1970), p. 313.
boom." Security ties with the United States expanded even more (primarily
through arms sales), despite the Shah's complaints about its failure to support Pakistan in 1965 and 1971. In particular, his desire to increase Iran's
military capabilities fit in well with the Nixon Doctrine, which called for the
United States to rely less on its own troops and more on regional allies.54
As a result, Nixon endorsed the Shah's imperial ambitions during his visit
to Tehran in 1972." Iran now enjoyed unrestricted access to U.S. weaponry,
and the Shah began taking direct action against the radical forces that he saw
as the main threat to his rule. Iranian troops intervened to defeat the Dhofar
rebellion in Oman, and the Shah collaborated with Israel and the CIA in
supporting the Kurdish insurgency in Iraq. Iran occupied several disputed
islands in the Persian Gulf, and the Shah began a campaign to establish a
collective security organization in the region. Following Pakistan's dismemberment in its 1971 war with India, Iran also increased its support for Islamabad to discourage separatist movements that might endanger both countries.56
Predictably, relations with Moscow were guarded. The Soviets were understandably opposed to the Shah's willingness to defend U.S. interests, his
hostility towards Iraq (which signed a Treaty of Friendship with the Soviet
Union in 1972), and his recognition of Communist China in 1971. Although
economic exchanges continued, Soviet-Iranian relations grew more distant
after 1973. This development reveals the Shah's preferences clearly; although he still sought cordial relations with the Soviet Union, his overriding
interest lay in a continued alignment with the United States.
The revolution that overthrew the Shah in 1979 brought the U.S.-Iranian
alignment to an abrupt end. Iran's new rulers were understandably suspicious of the Shah's principal patron, and this attitude was intensified by an
ideological antipathy towards the West in general and the United States in
particular.57Among other things, these events reveal that balance-of-threat
theory cannot explain everything; Iran's realignment was due to a domestic
upheaval rather than a change in the distribution of external threats. Because Khomeini and his followers defined Iran's interests in a radically
different way-for both ideological and historical reasons-they abandoned
54. On the Nixon Doctrine and its impact on Iran, see Robert F. Litwak, Detente and the
Nixon Doctrine (Cambridge: Cambridge University Press, 1984), pp. 139-43 and passim.
55. Iranian arms imports averaged $146.5 million per year for the period 1965-70. By contrast, arms imports for the period 1971-77 were almost eight times greater, averaging $1.15
billion per year. See U.S. Arms Control and Disarmament Agency, World Military Expenditures and Arms Transfers (Washington, D . C . : GPO, various years); and Leslie Pryor, "Arms
and the Shah," Foreign Policy 31 (Summer 1978).
56. On these events, see R. K. Ramazani, "Iran's Search for Regional Cooperation," Middle
East Journal 30 (Spring 1976); Ramazani, "Emerging Regional Relations," pp. 1051-54, 106062; Halliday, Iran, pp. 270-77; and Tad Szulc, The Illusion of Peace: Foreign Policy in the
Nixon Years (New York: Viking, 1978), pp. 582-87.
57. On Khomeini's ideology, and especially his view of the United States and Soviet Union
as two "Satans," see R. K. Ramazani, Revolutionary Iran: Challenge and Response in the
Middle East (Baltimore: Johns Hopkins University Press, 1986), pp. 21-22.
the Shah's traditional allies (for example, the United States and Israel) in
favor of a policy of non-alignment in association with Syria, Libya, and
various Shi'ite radical groups in Lebanon and elsewhere.58
Although domestic factors have largely determined Iran's international
behavior since the revolution, its tendency to balance threats remained.
Among other things, the revolution did not lead Iran to move towards the
Soviet Union; indeed, Soviet-Iranian relations have probably been worse
under Khomeini than they were under the
Iran has forged close ties
with Syria to balance Iraq (a traditional enemy for both states), and the
prolonged Gulf war has forced the Islamic Republic to seek assistance from
a diverse array of otherwise unlikely supporter^.^^ It is also worth noting that
Iran's bellicosity has encouraged balancing behavior by its various neighbors, as the formation of the Gulf Cooperation Council, the tacit alignment
between Saudi Arabia and Iraq, and the growing naval commitment by the
United States, Britain, and France all reveal.
Summary
Iran's behavior throughout the cold war supports the balancing hypothesis. Despite lukewarm American support, Iran chose to resist Soviet
pressure after World War 11. The Shah sought an active American commitment during the 1950s, but shifted to a more independent posture as the
Soviet threat declined and Iran's own capabilities improved. At the same
time, he balanced against emerging regional threats by forging cooperative
ties with Israel, Saudi Arabia, Pakistan, and Oman. In the 1970s, Iran took a
more active role in the Gulf, relying on its own capabilities and indirect
support from the United States. Although the fall of the Shah altered Iran's
aims and interests considerably, it did not alter its tendency to balance
threats. Iran's alliance policy, in short, provides little evidence of bandwagoning .
Turkey
tween Nazi Germany and its traditional Russian rival. Turkey swung towards the Allies as Germany's defeat approached to ensure Western support
against any future threat from the Soviet ~ n i o n . Soviet
~'
pressure on Turkey
increased steadily, and the Soviets denounced the Soviet-Turkish NonAggression Treaty in March 1945. In June, they informed the Turks that
renewal of the treaty was contingent upon revision of the Montreux Agreement governing the Turkish Straits and upon territorial concessions in eastern Turkey. Stalin pressed unsuccessfully for territorial adjustments at the
Potsdam Conference in July, and the Soviets began to publicize their "historical" claims to additional portions of the Black Sea coastline.62
Turkey refused to budge. Although both the United States and Great
Britain were initially willing to discuss the Soviet demands, the Turks rejected them outright and began an active campaign for American support.63
Stalin's continued threats and the simultaneous crisis in Iran encouraged a
favorable U.S. response: President Truman sent the battleship Missouri to
Turkey in April 1946, and additional Soviet pressure in August led Truman
to decide to "resist with all means at our disposal . . . any Soviet aggression
against Turkey."@' Although the United States rejected several requests for
economic and military aid, Turkey received credits worth between $25 and
$50 million in November 1946. Finally, in March 1947, the Truman Doctrine
publically committed the United States to defend Turkey against Soviet
e~pansion.~'
Still fearful of the Soviets, Turkey sought a more substantial Western
~ o r n m i t m e n tAfter
. ~ ~ a lengthy campaign, Turkey was admitted to the North
Atlantic Treaty Organization (NATO) in 1952.67American assistance av61. Although this might appear to be an example of bandwagoning, it was in fact an effort to
balance against the Soviet Union by obtaining British or American support. As the danger from
Germany receded, the traditional threat from Russia dominated Turkish calculations. See
Ferenc Vali, Bridge Across the Bosporus: The Foreign Policy of Turkey (Baltimore: Johns
Hopkins University Press, 1971),pp. 31-33; Kuniholm, Origins of the Cold War, pp. 21, 51-55,
68-72, and passim.
62. See Kuniholm, Origins of the Cold War, p. 287.
63. At the Potsdam Conference in July, the United States and Britain agreed that a revision
of the Montreux Convention was in order, because it "failed to meet present-day conditions."
See Rubinstein, Soviet Policy, p. 10; Kuniholm, Origins of the Cold War, pp. 260-70.
64. Truman made his decision in response to a Soviet note proposing joint Soviet-Turkish
control of the Turkish Straits, which would have permitted Soviet military bases in the area.
The Turks insisted they would "resist by force Soviet efforts to secure bases in Turkish
temtory even if Turkey had to fight alone." It should be noted that Turkish resolve was doubted
by some U.S. officials. See FRUS 1946, vol. 7 , p. 841.
65. See George S. Hams, Troubled Alliance: Turkish-American Problems in Historical
Perspective, 1945-1971 (Washington, D.C.: American Enterprise Institute, 1972), pp. 17-29.
66. According to Vali, "it was the Soviet threat, more menacing in its modem Stalinist form
than the Tsarist pressures experienced in the past, that compelled Ankara to seek close political
and military ties with the West." See his Bridge Across the Bosporus, p. 35.
67. Turkey's initial request for NATO membership was rejected by Britain and France on the
grounds that Turkey was not a European country. Turkey's strong anti-Soviet position and its
eraged nearly $100 million per year from 1950 to 1960, the United States
enjoyed access to Turkish bases, and Turkey participated actively in
NATO's military planning. Turkey agreed to join the Baghdad Pact in February 1955, and concern over Soviet involvement in Syria and Egypt led them
to support an ill-conceived Western campaign against Syria in 1 9 5 7 . After
~~
the Iraqi Revolution, the Turks agreed to join the Central Treaty Organization (CENT0)-the successor to the Baghdad Pact-and a bilateral defense
agreement with the United States was signed at this time. Although disturbed by America's failure to become a full member of CENTO, Turkey's
anti-Soviet position remained intact through 1 9 6 0 . ~ ~
As with Iran, however, Turkey's unquestioning alignment with the West
eroded in the 1960s. Relations with the Soviet Union, in turn, improved
dramatically. Several factors were at work. Kennedy's decision to withdraw
U.S. missiles from Turkey following the Cuban Missile Crisis alarmed the
Turks, primarily because they were not consulted in advance.70 Even more
important, the growing struggle with Greece over Cyprus led to a serious rift
between Washington and Ankara. The Turks were especially incensed by
President Johnson's warning against the use of U.S. arms in the event of
hostilities with Greece. Indeed, Johnson hinted that NATO might not support Turkey if its actions on Cyprus led to Soviet intervention." Lastly, the
growing burden of the Vietnam War led to a significant reduction in U.S. aid,
despite Turkey's continued need for assistance.
Feeling betrayed, the Turks sought to increase their diplomatic options. A
Turkish delegation attended the Non-Aligned Conference in Cairo in October 1964, signaling a desire to expand their relations with the Third World. In
addition, Turkey and the Soviet Union began what Alvin Rubinstein has
termed a "wary dktente." Soviet Foreign Minister Andrei Gromyko and
Turkish Prime Minister Suat Urgiiplii exchanged visits in 1964, Soviet President Nikolai Podgorny came to Ankara the following year, and Alexei Kosygin followed suit in 1966. The principal avenue of cooperation was
active participation in the Korean War overcame this resistance. See Bruce R. Kuniholm,
"Turkey and NATO: Past, Present, and Future," Orbis 27 (Summer 1983); Rubinstein, Soviet
Policy, u. 12; H a m s , Troubled Alliance, pp.
.. 35-44; and Vali, Bridge Across the Bosporus, pp.
36-37, i16-18.
68. For details on the Svrian affair. see: Alexander L. George and Richard Smoke. Deferrence in American ~ o r e i i nPolicy: Theory and Practice ( ~ e wYork: Columbia University
Press, 1974), pp. 332-33; Patrick Seale, The Struggle for Syria:A Study ofArab Politics, 19451958 (London: Oxford University Press, 1965), pp. 291-96; and Wilbur Clare Eveland, Ropes of
Sand: America's Failure in the Middle East (New York: Norton, 1980), chaps. 19 and 23.
69. In particular, the military government that seized power in 1960 assured Turkey's allies
that it would maintain Turkey's existing commitments. On these events, see Vali, Bridge
Across the Bosporus, pp. 126-28; and H a m s , Troubled Alliance, pp. 63-65.
70. See Vali, Bridge Across the Bosporus, p. 129; and Rubinstein, Soviet Policy, p. 132.
71. For the text of Johnson's letter, see Middle East Journal 20 (Summer 1966), pp. 386-93.
See also H a m s , Troubled Alliance, chap. 5, especially pp. 112-15; Vali, Bridge Across the
Bosporus, pp. 130-31; and Richard C. Campany, Turkey and the United States: The Arms
Embargo Period (New York: Praeger, 1986), pp. 32-33.
Since World War 11, Turkey's security from Soviet pressure has been
based on a formal alliance with Europe and the United States. Once security
against the Soviets was achieved, however, cordial relations with Moscow
have been clearly in Turkey's interest. The Soviet Union has become a
valuable economic partner, it no longer poses an imminent threat, and
76. On these points, see Bulent Ecevit, "Turkey's Security Policies," Survival 20 (September-October 1978), pp. 203-8; and Boll, "Turkey's New National Security Concept," p. 619;
and Rubinstein, Soviet Policy, pp. 38-42.
77. See Boll, "Turkey's New National Security Concept," pp. 615-16.
78. See U.S. Congress, Subcommittee on Europe and the Middle East, 96th Congress, 2d
Session, United States-Turkey Defense and Economic Cooperation Agreement, 1980 (Washington, D.C.: GPO, 1980), pp. 35-69.
Soviet-Turkish ties make it less likely that NATO will take the Turks for
granted. Thus, dCtente with the Soviet Union has reflected both NATO's
success in providing for Turkey's basic security and its inability to support
all of Turkey's objectives (for example, its policy on Cyprus). After thirtyfive years, Turkey remains fully integrated in NATO defense planning (for
example, nearly five hundred U.S. nuclear weapons are deployed on Turkish
All things considered, Turkish foreign policy offers substantial
support for the balancing hypothesis.
India
Indian foreign policy is based on an explicit policy of non-alignment. Its
size and stature have enabled it to avoid formal attachments, and the legacy
of colonial rule has reinforced suspicion of Western security schemes. Despite its public posture, however, India has departed from strict nonalignment whenever serious threats emerged. Given the nature of these
threats and India's geographic position, this has usually meant a close relationship with the Soviet union."
Pakistan has been India's most important rival. The Indo-Pakistani conflict arises from the traditional enmity between Hindus and Moslems on the
subcontinent, compounded by the specific disputes that emerged during
independence and partition.s' Given that approximately 500,000 people died
in communal violence after independence, triggering a migration of roughly
79. See William M. Arkin and Richard W. Fieldhouse, Nuclear Battlejields: Global Links in
the Arms Race (Cambridge: Ballinger, 1984), pp. 232-33. For summaries of Turkey's role in
NATO and the Turks' continued interest in close ties with the West, see Kuniholm, "Turkey
and NATO," pp. 436-45; and House Committee on Foreign Affairs, Turkey's Problems and
Prospects: Implications for U . S . Interests (Washington, D.C. : GPO, 1980) pp. 14-28, 49.
80. Non-alignment was the brainchild of Prime Minister Jawarharlal Nehru, the dominant
figure in Indian foreign policymaking and a leading spokesman for the Non-Aligned Movement.
An excellent guide to his thinking is J. Nehru, India's Foreign Policy: Selected Speeches,
September 1946-April 1961 (New Delhi: Ministry of Information and Broadcasting, 1961). See
also Michael Brecher's very flattering Nehru: A Political Biography (London: Oxford University Press, 1959), especially chap. 19; and Taya Zinkin, "Indian Foreign Policy: An Interpretation of Attitudes," World Politics 7 (January 1955). On India's policy towards the other states in
South Asia, see Leo Rose, "India and Its Neighbors: Regional Foreign and Security Policies,"
in Lawrence Ziring, ed., The Subcontinent in World Politics: India, Its Neighbors, and the
Great Powers (New York: Praeger, 1982).
81. The most significant problem was the fate of Kashmir, a predominantly Moslem province
that India occupied after independence, triggering a brief war between the two newly independent states. Its importance was heightened because: 1) Kashmir was strategically valuable to
both sides; 2) each side's claims reflected competing formulas for legitimacy (e.g., India is a
secular state, Pakistan is formally Moslem); and 3) both sides believed that the other had acted
illegally in pressing its claims. For various accounts of the dispute, see: Michael Brecher, The
Struggle for Kashmir (London: Oxford University Press, 1952); Russell Brines, The IndoPakistani Conjlict (London: Pall Mall, 1968), chaps. 4 and 5; and Lord Birdwood, Two Nations
and Kashmir (London: Hale, 1956).
ten million people, it is hardly surprising that the conflict has dominated both
countries' security concerns since then.82
China was a potential threat as well, but Prime Minister Jawaharlal Nehru
actively sought good relations with India's large northern neighbor. He supported China's entry into the United Nations and helped mediate between
China and the West during the Korean War. When China's occupation of
Tibet in 1950 threatened the Indian border, Nehru made only minor protests
and negotiated a conciliatory agreement on the issue. This policy was temporarily successful: Nehru and Zhou Enlai announced in 1954 that SinoIndian relations would henceforth be guided by the principles of "panch
sheel" (peaceful coexistence) .83
The decision to accommodate China was based on several factors. Nehru
placed great value on "Asian solidarity," and saw both China and India as
successful opponents of imperialist domination. As a result, he assumed that
Sino-Indian differences could be overcome easily. Nehru also saw SinoIndian cooperation as a way for both countries to protect themselves against
great power interference. Finally, he believed that a confrontation with
China would require sacrifices India could not afford, given the existing
conflict with Pakistan and India's pressing domestic needs.84
India's attitude towards both superpowers was more reserved. In contrast
to Turkey and Iran, India did not support America's campaign to "contain"
communism. As a result, early contacts between Nehru and the United
States were d i ~ a p p o i n t i n gNehru
. ~ ~ was at least partly sympathetic to socialist ideas (though he occasionally imprisoned Communists in India), and he
did not view the Soviet Union as a significant threat. India's refusal to sign
the Japanese peace treaty or to condemn China's intervention in the Korean
War also undermined its image in the United
Relations with the
Soviet Union were limited, however, because Stalin believed that the former
82. See William J. Barnds, India, Pakistan, and the Great Powers (New York: Praeger,
1972), pp. 32-33. This rivalry was exacerbated by disputes over water rights, financial compensation, and border demarcation.
83. According to Nehru, the principles of "panch sheel" were: 1) mutual respect for each
other's territorial integrity and sovereignty; 2) mutual non-aggression; 3) mutual noninterference in each other's internal affairs; 4) equality and mutual benefit; and 5) peaceful
coexistence. See Nehru, India's Foreign Policy, p. 99.
84. On Indian attitudes towards China, see Vidya Prakah Dutt, "India and China: Betrayal,
Humiliation, Reappraisal," in A. M. Halpern, ed., Policies Towards China: Views from Six
Continents (New York: McGraw-Hill, 1965), pp. 202-9; Nehru, India's Foreign Policy, pp.
303-12; Brecher, Nehru, pp. 588-92; and Zinkin, "Indian Foreign Policy," pp. 201-5.
85. For accounts of Nehru's largely unsuccessful visit to the United States in 1949, see
Brecher, Nehru, pp. 419-20; and McGhee, Envoy to the Middle World, pp. 46-52.
86. India supported the U.N. Resolution condemning North Korea's aggression in 1950, but
viewed the Chinese intervention later in the year as justified. See Barnds, India, Pakistan, and
the Great Powers, p. 61; G. W. Choudhury, India, Pakistan, Bangladesh and the Great Powers
(New York: Free Press, 1975), pp. 15,74-77; S. M. Burke, Mainsprings ofIndian and Pakistani
Foreign Policy (Minneapolis: University of Minnesota Press, 1974), pp. 127-29.
colonies were still pawns of the imperialist powers.87 This situation presented little difficulty for India: so long as relations with China were good
and Pakistan did not become too powerful, India had little need for allies.88
The first hint that India would seek Soviet assistance followed Pakistan's
decision to join the Baghdad Pact and Southeast Asia Treaty Organization
(SEATO) in 1955. Convinced that Stalin's successors were more moderate,
Nehru countered by inviting Premier Nikita Khrushchev and Prime Minister
Nikolai Bulganin to India in 1955. The Soviets backed India's position on
Kashmir, and India received a $112 million loan at this time.89 In return,
Nehru paid a well-publicized visit to the Soviet Union later in the year.
Although India continued to receive economic aid from the West and remained a member of the British Commonwealth, it had demonstrated its
diplomatic options.
The policy of non-alignment weakened further once relations with China
deteriorated. Although Chinese and Indian interests clashed on a number of
issues, the principal quarrel concerned several disputed border areas.90The
Tibetan revolt of 1959 accelerated the decline: Indian public opinion was
aroused while the Chinese suspected India of supporting the rebels. Talks
between Nehru and Zhou in 1960 failed to resolve their differences, and
public opinion forced Nehru to take an increasingly provocative p o ~ i t i o n . ~ '
In response, India increased its security efforts and sought greater foreign
support. Defense spending doubled from 1959 to 1962, and India consolidated its influence in the buffer states of Nepal, Bhutan, and ~ i k k i m . ~ ~
India agreed to purchase Soviet arms for the first time in 1959, and negotiations to purchase advanced fighter aircraft began in 1 9 6 1 . ~
~
India's
quarrel
87. On Soviet relations with India after independence, see: Robert H. Donaldson, Soviet
Policy Towards India: Ideology and Strategy (Cambridge: Harvard University Press, 1974),
chap. 3, especially pp. 70-76; and Choudhury, India, Pakistan, Bangladesh, pp. 7-11, 15-16.
88. On this point, see Onkar Manvah, "National Security and Military Policy in India," in
Ziring, ed., The Subcontinent in World Politics; and Lome J . Kavic, India's Quest for Security,
1947-65 (Berkeley: University of California Press, 1967), pp. 40-41.
89. See Barnds, India, Pakistan, and the Great Powers, p. 118.
90. There were hints of Sino-Indian rivalry at the 1955 Conference of the Non-Aligned
Movement at Bandung. Nehru viewed himself as the leader of the movement, but was upstaged
at the meeting by Chinese Premier Zhou Enlai. See George McTuman Kahin, The AsianAfrican Conference: Bandung, Indonesia, April 1955 (Ithaca, N.Y.: Cornell University Press,
1956), pp. 14-16, 23-30; and Mohammed Habib Sidky, "Chinese World Strategy and South
Asia: The China Factor in Indo-Pakistani Relations," Asian Survey 16 (October 19761, pp. 96667.
-. .
91. For accounts of the Sino-Indian border dispute, see Kavic, India's Quest for Security,
chap. 10; Richard Ned Lebow, Between Peace and War: The Nature of International Crisis
(Baltimore: Johns Hopkins University Press, 1981), pp. 184-92,216-22; Klaus H. Pringsheim,
"China, India, and Their Himalayan Border (1961-63)," Asian Survey 3 (October 1963); and
Barnds, India, Pakistan, and the Great Powers, pp. 141-55.
92. See Barnds, India, Pakistan, and the Great Powers, p. 152.
93. See Ian Graham, "The Indo-Soviet MIG Deal and Its International Repercussions,"
Asian Survey 4 (May 1964), p. 823. Signs of the growing Sino-Soviet split were the fact that
China had not yet received these advanced systems and the Soviet decision to adopt a neutral
position in the border dispute between democratic India and communist China.
with China was especially welcome news in Washington, and the Kennedy
administration began a major effort to improve relations with New Delhi
through a combination of diplomacy and increased economic aid.94
These trends were reinforced when China inflicted a brief but decisive
defeat on the Indian army in October 1 9 6 2 . ~
~
Although
India was the principal aggressor, the United States, Great Britain and the Soviet Union all sent
military equipment to India during the war, and they continued to provide
military and economic aid to India in the months that followed.96 India's
ability to balance China was enhanced by the superpower competition and
by the fact that both were now hostile to Beijing.
The Indian army's poor showing in 1962 triggered a major campaign to
enhance its military capabilities, a decision reinforced by China's growing
alignment with Pakistan. The Soviet Union was the most willing supplier,
and Soviet-Indian military ties began to expand rapidly after 1962.~'
The effort paid off in the two subsequent wars with Pakistan. In the first,
Pakistan sought to exploit India's weakness and China's support by "leaning
on ~ n d i a . "Border
~~
incidents rose steadily and Pakistan began infiltrating
troops into Kashmir in August 1965. Following several minor skirmishes,
India attacked West Pakistan in force on 5 September. After several weeks
of fighting (in which the Indian forces generally held the upper hand), a
ceasefire was arranged on 22 S e ~ t e m b e r . ~ ~
94. On the Kennedy administration's efforts to improve U.S. relations in the Third World,
see: Gaddis, Strategies of Containment, pp. 223-25; and Brown, The Faces of Power, pp. 198204.
95. Border clashes between India and China had grown increasingly serious as relations
deteriorated. Convinced that China would not attack in force, Nehru had ordered the Indian
Army to outflank the Chinese troops in the Aksai Chin (the main disputed area). Despite
repeated Chinese warnings, India continued its provocations until the Chinese decided to take
direct action to halt them. See Neville Maxwell, India's China War (Garden City, N.J.: Doubleday, 1972); and Allen S . Whiting, The Chinese Calculus of Deterrence: India and Indochina
(Ann Arbor: University of Michigan Press, 1975).
96. India received $92 million worth of arms from the United States between 1962 and 1965,
and $130 million from the Soviet Union between 1962 and May 1964. See S. Nihal Singh, "Why
India Goes to Moscow for Arms," Asian Survey 24 (July 1984). See also P. R. Chari, "IndoSoviet Military Cooperation: A Review," Asian Survey 19 (March 1979), pp. 236-37.
Significantly, the Indians begged the United States to intervene during the fighting, but China's
abrupt withdrawal rendered the request irrelevant. See Kavic, India's Quest for Security,
p. 282.
97. See Chari, "Indo-Soviet Military Cooperation," pp. 232-35; Kavic, India's Quest for
Security, chap. 11. Significantly, India has sought to avoid excessive dependence by diversifying suppliers and by negotiating agreements for coproduction and technology transfer. On this
point, see Raju G. C. Thomas, Indian Security Policy (Princeton, N.J.: Princeton University
Press, 1986), pp. 246-64.
98. The phrase was coined by President Mohammed Ayub Khan of Pakistan. Pakistan was
emboldened by Chinese support, and also felt that India's military buildup (made possible by
Soviet and American aid) would reduce their long-term prospects for regaining Kashmir. Accordingly, they sought a resolution of this issue sooner rather than later.
99. For an account of the Indo-Pakistani war of 1965, see Brines, Indo-Pakistani Conjict,
chaps. 12-14.
actions were primarily reactive rather than aggressive.lW Having demonstrated its preeminence, India gradually moved back towards nonalignment. The threat from Pakistan had been greatly reduced-a fact
signified by the Simla Accord of 197215-and China's rapprochement with
the United States was partially neutralized by India's successful nuclear test
in 1974. The Indo-Soviet treaty remained in force, and the Soviet Union
continued to be a major trading partner and India's principal source of arms.
After 1971, however, India began to diversify in both areas to avoid excessive reliance on Moscow.106Although Indo-Soviet relations survived Gandhi's electoral defeat in 1977 and the invasion of Afghanistan in 1979,
disagreements on a number of issues highlighted India's continued freedom
of action. lo'
That freedom was also revealed by the recovery of India's relations with
the United States, which had reached an all-time low during the 1971 war.
Although impeded by India's nuclear test, the U.S. rapprochement with
China, and the resumption of U.S. arms sales to Pakistan, a tangible improvement was apparent by the end of the decade. Economic aid returned to
pre-1971 levels, and President Carter staged a successful visit to India in
1978. Although Gandhi (who regained power in 1979) was critical of American aid to Pakistan after the invasion of Afghanistan, India's opposition was
relatively mild by historical standards. Indeed, Gandhi's visit to Washington
in 1982 revealed a growing awareness that American involvement in South
Asian security might be a useful hedge against further Soviet encroachments.
Finally, India's enhanced position after 1971 permitted it to seek better
relations with Pakistan and China, although they proceeded cautiously on
104. The flow of refugees from East Pakistan created a serious burden for India, and the
creation of Bangladesh reduced the potential threat from a Pakistani-U.S.-Chinese alliance.
Soviet support was useful for detemng any attempt by China to come to Pakistan's aid during
the crisis.
105. The Simla Accord was negotiated at a meeting between Gandhi and Pakistani Prime
Minister Bhutto in June and July of 1972. It provided for repatriation of prisoners of war, and
both sides pledged to settle differences through negotiation, not to interfere in each other's
internal affairs, to refrain from the use of force, and to respect the 17 December 1971 cease-fire
line. They also committed themselves to additional meetings in the future. For the text and an
analysis, see Keesing's Contemporary Archives, 19-26 August 1972, pp. 25432-33.
106. This trend should not be exaggerated. The Soviet share of Indian foreign trade remained
fairly constant throughout the 1970s (approximately 12%) although it had dropped to 8.4% by
1982. See International Monetary Fund, Direction of Trade Yearbook (Washington, D.C.:
International Monetary Fund, 1980, 1983). The Soviet share of Indian military imports fluctuated, growing from 79% in the period 1965-1974 to 82% in the period 1976-1980, and then
declining to 72% in the period 1979-1983. See ACDA, World Military Expenditures and Arms
Transfers (various years). The Indians continued to insist in coproduction arrangements for
advanced technology, and they avoided exclusive reliance on Moscow by purchasing the
Anglo-French Jaguar and the French Mirage 2000 advanced combat aircraft, as well as a
number of other systems. See Chari, "Indo-Soviet Military Cooperation"; and Glynn L. Wood
and Daniel Vaagenes, "Indian Defense Policy: A New Phase?" Asian Survey 24 (July 1984).
107. India continued to reject Soviet proposals for a collective security system in Asia, and it
has consistently sought to reduce both superpowers' presence in the Indian Ocean.
both fronts. Gandhi inaugurated a modest detente with China in 1976, and
Sino-Indian relations slowly revived on several fronts.lo8 Relations with
Pakistan remain erratic, and continue to be strained by India's concerns
over Pakistan's nuclear program and the effects of U.S. military aid.'09 In
general, however, India's willingness to pursue more cooperative relations
reflects the fact that threats from its traditional adversaries have declined.
Summary
India's foreign policy also suggests that states prefer to balance rather
than bandwagon. A partial exception was the attempt to appease China in
the 1950s, and this failure was clearly instructive. Although India's size and
status made non-alignment feasible (that is, it could always count on at least
one of the superpowers to come to its aid),''' India's leaders have abandoned
strict non-alignment whenever serious threats emerged. Indeed, the tacit
alignment between India and the Soviet Union is precisely what balance-ofthreat theory would predict, given that geography and history have placed
Pakistan and China at odds with both.
"'
Pakistan
After independence, Pakistan abandoned non-alignment in favor of a proWestern policy. Indeed, several early overtures to the Soviet Union may
have been intended primarily to attract greater American attention. 'I2 Unlike
108. On the Sino-Indian detente, see Keesing's Contemporary Archives, 23 July 1976, pp.
27846; and Robert C. Horn, Soviet-Indian Relations: Issues and Injuence (New York: Praeger, 1982), pp. 123-24, 158-59.
109. Evidence of a fragile but distinct improvement in Indo-Pakistani relations takes several
forms, including: 1) Pakistan's willingness to take vigorous measures following the hijacking of
India Air Lines jets in 1982 and 1984; 2) General Zia's visit to India in November 1982; 3) the
December 1985 agreement not to attack each other's nuclear facilities; and 4) the establishment
of an organization for South Asian Regional Cooperation (SARC), which began in 1981. On
these events, see: Jyotirmoy Banerjee, "Hot and Cold Diplomacy in Indo-Pakistani Relations,"
Asian Survey 23 (March 1983); Keesing's Contemporary Archives, March 1986, p. 34240; the
symposium entitled "South Asian Regional Cooperation: Four Views and a Comparative
Perspective," Asian Survey 25 (April 1985); and Thomas, Indian Security Policy, pp. 30-44,
295-97.
110. Nehru recognized that great power rivalries would enhance India's security at an early
stage. As he wrote in 1931: "No country will tolerate the idea of another acquiring the commanding position [in India] which England enjoyed for so long. If any power was covetous
enough to make the attempt, all the others would combine to trounce the intruder. This mutual
rivalry would itself be the surest guarantee against an attack on India." Quoted in Onkar
Marwah, "National Security and Military Policy in India," p. 73.
111. See Thomas, Indian Security Policy, pp. 16-17.
112. Convinced that the West was taking Pakistan for granted, Prime Minister Liaquat Ali
Khan agreed to visit Moscow in 1950, to counter Nehru's planned visit to the United States.
This gesture convinced the United States to pay more attention to Pakistan, and Liaquat
canceled the trip to Moscow and accepted an invitation to Washington. The Pakistani Ambassador to the United States called this "a masterpiece of strategy . . . we were taken much for
India, Pakistan supported the Japanese peace treaty and opposed China's
intervention in Korea and its admission to the United Nations. Even more
important, Pakistani and American officials began exploring the possibility
of military cooperation as early as 1951. By 1954, Pakistan had agreed to join
the Baghdad Pact, and it later joined SEAT0 as well.l13
For the United States, the alliance with Pakistan was part of the campaign
to "contain" the Soviet Union. For Pakistan, the principal motive was the
need to balance India, although domestic factors and pressure from Afghanistan played an important role as well.l14 The United States refused to
support Pakistan's efforts to regain Kashmir, but did provide an average of
$120 million in economic and military aid from 1953 to 1958.'l5 Thus the
Pakistani-U.S. alliance served compatible but not identical interests.
Pakistan's position as America's "most allied ally" declined dramatically
in the 1960s. The Pakistanis were alarmed by the Kennedy administration's
efforts to court India, especially when the United States shipped arms to
India during the Sino-Indian war of 1962. Pakistan's response was predictable; they balanced by moving closer to China. Border talks began in October 1962, an agreement was announced in March 1963, and Sino-Pakistani
cooperation expanded steadily over the next two years.ll6 In addition to
economic assistance, China supported Pakistan's claims to Kashmir while
Pakistan backed China's efforts to join the United Nations. 'I7
granted [in Washington] as good boys . . . we were treated as a country that did not seriously
matter . . . [but] overnight Pakistan began to receive the serious notice and consideration of the
U.S. government." Quoted in Choudhury, India, Pakistan, Bangladesh, p. 12.
113. On the origins of the Pakistani alliance with the United States, see Selig S. Hamson,
"India, Pakistan, and the United States: Case History of a Mistake," The New Republic, 10
August 1959; Choudhury, India, Pakistan, Bangladesh, pp. 80-81; and S. M. Burke, Pakistan's
Foreign Policy: An Historical Analysis (London: Oxford University Press, 1973), pp. 126-35
and chap. 8.
114. According to Mohammed Ayub Khan, former commander-in-chief of the Pakistani
army and president of Pakistan: "The crux of the problem from the very beginning was the
Indian attitude of hostility between us; we had to look for allies to secure our position." See his
Friends Not Masters: A Political Autobiography (London: Oxford University Press, 1967), p.
154. See also Choudhury, India, Pakistan, Bangladesh, p. 86 and passim; and George J. Lerski,
"The Pakistan-American Alliance: A Reevaluation of the Past Decade," Asian Survey 8 (May
1968), pp. 405-6. The Pakistani government also faced serious internal challenges from disaffected ethnic groups, especially a rebel movement among the Pushtuns that received intermittent support from Afghanistan. See Khurshid Hasan, "Pakistan-Afghan Relations," Asian
Survey 2 (July 1962), especially pp. 14-16; Barnds, India, Pakistan, and the Great Powers, pp.
77-79. U.S. military aid also helped Pakistan's ruling elite maintain its dominant position over
various rival ethnic groups; alignment with the United States thus helped them balance both
external and internal threats. See Selig H. Hanison, The Widening Gulf: Asian Nationalism
and American Policy (New York: Free Press, 1979) pp. 274-75.
115. See U.S. Agency for International Development, U . S . Overseas Loans and Grants and
Aid to International Organizations (Washington, D.C.: GPO, various years).
116. China issued a $64 million interest-free loan in 1964, the first in a burgeoning series of
economic and technical agreements. See Ya'acov Vertzberger, "The Political Economy of
Sino-Pakistani Relations: Trade and Aid 1963-1982," Asian Survey 23 (May 1983), p. 645 and
passim.
117. On the origins of the Sino-Pakistani relationship, see: Anwar H. Syed, China and
These trends accelerated after Pakistan's defeat in 1965. The arms embargo revealed that the U.S. alliance was worthless against India, and President Mohammed Ayub Khan adopted a policy of "bilateralism" aimed at
expanding Pakistan's relations with China and the Soviet union.' l8 Pakistan
began receiving arms from China in 1966, while the Soviets pledged $112
million in economic aid. Trade with the Soviets began to expand as well, and
the Soviets sent several small arms shipments in 1967. Ayub also announced
that Pakistan would not renew the U.S. lease for its intelligence facility at
Peshawar, in an effort to persuade the Soviets to reduce their support for
India.
These measures were an obvious response to what Pakistan's leaders saw
as inadequate American support, but they do not qualify as bandwagoning
behavior. Pakistan did not shift to a pro-Soviet or pro-Indian position; indeed, Ayub rejected Soviet requests to use Pakistani facilities and opposed
the Soviet plan for a collective security system in Asia. And when it became
clear the Soviet Union would not reduce its support for India, SovietPakistani relations cooled quickly. In short, "bilateralism" was not an accommodation to Soviet or Indian power, it was an attempt to balance the
main threat-India. 'l9
As noted earlier, Pakistan's relations with the United States revived during the Nixon administration. The arms embargo was lifted and Nixon told
President Yahya Khan (who succeeded Ayub in 1969), that "nobody has
occupied the White House who is friendlier to Pakistan than me."lZ0 Yet
neither the United States nor China did much to prevent Pakistan's dismemberment in 1971. Nixon's "tilt" towards Pakistan had little effect, and China
carefully limited its commitment to West Pakistan only. lZ1
Gravely weakened by the 1971 debacle, Pakistan's response was twofold.
First, the new president, Zulfikar Ali Bhutto, began an active effort to apPakistan: Diplomacy of an Entente Cordiale (Amherst: University of Massachusetts Press,
1974), especially chap. 4; Ya'acov Vertzberger, "The Enduring Entente: Sino-Pakistani Relations 1960-1980," The Washington Papers, no. 95 (New York: Praeger, 1983) pp 15-24; Khalid
B. Sayeed, "Pakistan and China: The Scope and Limits of Convergent Policies," in Halpern,
Policies Towards China, pp. 236-48; J. D. Armstrong, Revolutionary Diplomacy: Chinese
Foreign Policy and the United Front Doctrine (Berkeley: University of California Press, 19771,
pp. 157-66; and 0 . N. Mehrotra, "Sino-Pak Relations: A Review," China Report 7, nos. 5-6
(1976).
118. As Ayub put it: "If we could not establish normal relations with all our three big
neighbors, the best thing was to have an understanding with two of them . . . it was on this basis
that I set out to normalize our relations with the PRC and the Soviet Union." See his Friends
Not Masters, p. 118 and passim. For a summary of "bilateralism," see Burke, Pakistan's
Foreign Policy, pp. 359-60.
119. See Barnds, India, Pakistan, and the Great Powers, pp. 212-16; G. W. Choudhury,
"Pakistan and the Communist World," PaciJic Community 6 (October 1974), p. 114 and passim;
and especially Burke, Pakistan's Foreign Policy, chap. 15.
120. Quoted in Choudhury, India, Pakistan, Bangladesh, p. 142.
121. See Choudhury, India, Pakistan, and Bangladesh, pp. 210-14; Jackson, South Asian
Crisis, pp. 94-96; and Vertzberger, Enduring Entente, chap. 4.
pease India, beginning with the Simla Accord in April 1972. A gradual dCtente began to take shape, although lingering suspicions made extensive
cooperation unlikely. Bhutto also sought cordial relations with the Soviet
Union, and his visit to Moscow in March 1972 helped overcome the hostility
that had arisen between the two countries the previous year. Finally, Bhutto
extended diplomatic recognition to Bangladesh in 1973, both to obtain the
release of West Pakistani prisoners and to begin a process of reconciliation
between the two ~ 0 u n t r i e s . l ~ ~
Second, these acts of accommodation were accompanied by an energetic
search for additional allied support. China increased its military aid substantially and continued to assist Pakistan's nuclear ~ r 0 g r a m . lBhutto
also
~~
began to seek support from other Muslim countries. Pakistan hosted a major
Islamic conference in 1974, and financial assistance from Iran, Saudi Arabia,
and others helped re-equip the Pakistani armed forces. Thus, the desire to
balance continued even as dktente with India began.124
Relations with the United States, however, remained turbulent for most of
the decade. The United States resumed its economic aid program after the
war and lifted the ban on military sales in 1975. Yet Pakistan's efforts to
obtain direct U.S. military aid failed because of U.S. opposition to Pakistan's nuclear program, its concerns about human rights under both Bhutto
and his successor, General Zia ul-Haq, and the Carter administration's campaign to improve relations with India.12' As one might expect, however, the
invasion of Afghanistan ended the impasse. After years of "benign neglect,"
Pakistan's security became a vital concern once more. The Carter administration reaffirmed the U.S. commitment to Pakistan, offered a $400 million
military aid package, and began to enhance the U.S. military posture in the
region.lz6Yet Pakistan did not rush to welcome American assistance, and its
apparent reluctance to stand up to the Soviet Union was revealed by Zia's
statement that "you can't live in the sea and create enmity with whales . . .
you have to be friendly with them."127 Zia dismissed the U.S. aid offer as
122. See Shirin Tahir-Kheli, "The Foreign Policy of 'New' Pakistan," Orbis 20 (Fall 1976),
especially pp. 736-43.
123. See Vertzberger, Enduring Entente, pp. 59-62; and "Political Economy of SinoPakistani Relations," pp. 647-49; Tahir-Kheli, "Foreign Policy of New Pakistan," pp. 734-36;
William J. Barnds, "China's Relations with Pakistan," China Quarterly 63 (September 1975);
and Mehrotra, "Sino-Pak Relations," pp. 68, 70-71.
124. See M. G. Weinbaum and Gautam Sen, "Pakistan Enters the Middle East," Orbis 22
(Fall 1978), pp. 601-3.
125. See Shirin Tahir-Kheli, The United States and Pakistan: The Evolution of an Influence
Relationship (New York: Praeger, 1982), pp. 54-62, 72-76; and Thomas B. Thornton, "Between the Stools? U.S. Policy Towards Pakistan During the Carter Administration," Asian
Survey 22 (October 1982).
126. The administration established a Raoid Deolovment Joint Task Force (RDJTF) for use
in the area and increased its military access arrangements in the region. See Thomas L.
McNaugher, Arms and Oil: U . S . Military Strategy and the Persian Gulf(Washington, D.C.:
Brookings Institution, 1985), pp. 13-17 and passim.
127. Quoted in Alvin Z. Rubinstein, "Soviet Policy Towards South and Southwest Asia:
A
Summary
The Pakistani case is especially revealing. Pakistan is much weaker than
its principal adversaries, it has faced numerous internal problems, and U.S.
and Chinese support has been erratic at best. Yet Pakistan has made only
Strategic and Political Aspects," in Alvin Z. Rubinstein, ed., The Great Game: Rivalry in the
Persian Gulf and South Asia (New York: Praeger, 1983), p. 88. On Pakistan's options and
actions during this period, see W. Howard Wriggins, "Pakistan's Search for a Foreign Policy
After the Invasion of Afghanistan," PaciJic Affairs 57 (Summer 1984).
128. See Robert G . Wirsing and James M. Roherty, "The United States and Pakistan,"
International Affairs 58 (Autumn 1982); Tahir-Kheli, U . S . and Pakistan, pp. 97-104.
129. See Mehrunnisa Ali, "Soviet-Pakistani Ties Since the Afghanistan Crisis," Asian Survey 23 (September 1983), pp. 1027-28.
130. See Wirsing and Doherty, "U.S. and Pakistan"; and Zubeida Mustafa, "Pakistan-U.S.
Relations: The Latest Phase," The World Today 36 (December 1980).
131. See W. Howard Wriggins, "The Balancing Process in Pakistan's Foreign Policy," in
Lawrence Ziring et al., eds., Pakistan: The Long View (Durham, N.C.: Duke University Press,
1977).
minor concessions to its principal opponents, and only after major defeats.
In particular, Pakistan has not adopted a pro-Indian or pro-Soviet policy,
despite its obvious vulnerability to both. Thus, this case also provides powerful support for the balancing hypothesis.
with Turkey and Iran improved dramatically after 1960, neither country
abandoned its pro-Western alignment or seemed reluctant to oppose Soviet
expansion, as their reactions to the invasion of Afghanistan suggest.'33
The key issue is whether these efforts at accommodation-however limited-were a response to increases in Soviet power or the erosion of American credibility. If so, then the bandwagoning hypothesis should be given
greater weight. Although the evidence is mixed, a careful analysis suggests
that this is not the case. On the one hand, the erosion of America's commitments in South Asia and the improvement of relations with the Soviet Union
coincided with events that revealed the limits of U.S. support. For Turkey,
it was the removal of U.S. missiles and the growing dispute over Cyprus.
For Pakistan, it was America's courtship of India and the arms embargo in
1965, reinforced by the weak support that the United States provided in
1971. For Iran, it was U.S. opposition to the Shah's domestic programs and
regional ambitions, as well as its inconsistent commitment to CENTO. Thus
there is some evidence that doubts about American credibility encouraged
detente.
On the other hand, such concerns did not lead these states to realign.
Instead, they tried to balance through new defensive arrangements. Pakistan
moved closer to China and sought new allies in the Moslem world, and its
detente with Moscow in the 1960s was due to its rivalry with India rather
than fear of the Soviet Union. Iran's detente with the Soviet Union coincided with new alliances with several regional powers (for example, Israel,
Saudi Arabia, and Oman) and a growing relationship with the United States.
Turkey joined the Non-Aligned Movement and welcomed Soviet support
over Cyprus, but remained an active member of NATO. Even if American
backing no longer seemed as reliable, this perception did not cause significant defections.
Second, these developments coincided with Soviet efforts to improve its
bellicose image. Although Soviet power was growing, a combination of economic aid, cultural exchanges, and adroit diplomacy helped reduce the
threat that its neighbors perceived. Moreover, these states were all more
capable than they had been in the early 1950s. All save India enjoyed formal
security ties with the West, and the territorial status quo had acquired considerable legitimacy since 1945. Detente with the Soviet Union was less
dangerous, and the United States' own efforts to reduce tensions with Moscow made it a permissible step for others as well. Far from being evidence of
bandwagoning tendencies, the emerging detente between the Soviet Union
133. Obviously, the Iranian revolution ended Iran's close relationship with the United States,
but this was primarily because of domestic political events and ideological differences. Moreover, the Islamic Republic of Iran is probably more suspicious of the Soviet Union than the
Shah was.
and these states in fact supports the balancing hypothesis. As the level of
threat declined, the need to balance declined as ~ e 1 1 . l ~ ~
Third, the erosion of America's earlier position was caused more by the
tensions that are part of every alliance than to declining U.S. credibility. The
United States did not support Pakistan's ambitions in Kashmir, Turkey's
policies on Cyprus, or all of the Shah's regional pretensions. But this did not
mean that its commitment to defend them against Soviet aggression had
waned. Once the Soviet threat no longer loomed as large, the United States
and its allies had less reason to overlook these bilateral disputes. Moreover,
U.S. support had helped its allies gain the strength necessary to pursue more
independent policies. Thus, a decline in alliance cohesion was virtually inevitable. 135 Rather than providing evidence of Soviet intimidation, these events
suggest that the effort to balance the Soviet Union had worked quite well.
Finally, as I suggested at the beginning of this article, client states are
likely to exaggerate their propensity to bandwagon, in order to persuade
their patrons to provide more support. The states examined here have
clearly recognized the value of this tactic. Iranian officials encouraged U.S.
support in 1946 by arguing that a Soviet victory there would undermine
international confidence in the United States and the United Nations. The
Shah claimed that America's failure to aid Pakistan in 1965 and 1971 gravely
weakened the credibility of American promises (despite the fact that Pakistan bore greater responsibility for both wars.) General Zia of Pakistan used
the same arguments when seeking a greater U.S. commitment in 1980, implying that he would be forced to succumb to Soviet pressure if U.S. aid
were not increased. Despite these threats, however, the states in question
ultimately chose to balance. Support from the United States may have contributed to these decisions, but we should not forget that client states have
powerful incentives to overstate their propensity to defect.
In short, there is little evidence that declining American power has led
others to bandwagon. The extent of accommodation was minimal, and other
factors were far more important. Alliance leaders cannot be indifferent to
their allies' interests, but they should recognize that decisions to bandwagon
are made with great reluctance. Accordingly, relatively little effort is required to prevent them from occurring.
134. It is worth noting that the invasion of Afghanistan reversed these trends, because it
revived suspicions about possible Soviet aggression.
135. This suggests that the superpowers face a fundamental tradeoff when choosing and
supporting alliance partners. The stronger or more important an ally, the more influence it can
wield within the alliance. Although powerful allies are of greater value, they are also more
difficult to control and will demand more in exchange for their allegiance. This dilemma is
compounded because establishing the alliance commits the superpower's prestige and
strengthens the client's resources. Thus, the patron's control declines further, because the
client has less need of support and the patron has an interest in preserving its investment. Thus,
the gradual erosion of Soviet and American influence is not evidence of bandwagoning; it is an
inevitable by-product of the balancing process.
make each the other's most significant rival, which explains why neither
superpower has been able to enjoy good relations with both India and Pakistan for very long.
In addition, by focusing on power alone, balance-of-power theory overlooks the fact that domestic threats may provide an important motive for
alignment. Although external threats were more important, domestic concerns also encouraged Pakistan and Iran to seek U.S. support. As Steven
David has suggested, regime stability and personal survival rank high on the
agendas of most Third World leaders.13' Balance-of-threat theory can accommodate this possibility fairly easily-states seek allies to balance both
internal or external threats, whichever is most imminent-but balance-ofpower theory cannot.
Finally, these cases provide some evidence that offensive capabilities
encourage balancing. In particular, the arms competition between India and
Pakistan (in both the nuclear and conventional realms) reveals their concern
for each other's offensive capabilities, especially when combined with
mutual suspicions and resentment^.'^^
In short, although aggregate capabilities are hardly irrelevant, they are not
the only (or even the most important) factor. Examining the different components of threat (power, proximity, offensive capability, and perceived
intentions) provides a more compelling account of how and why states seek
alliance partners. Of course, it will not always be clear which threats are
most dangerous or which allies will be most valuable. And, as the Iranian
revolution reveals, domestic upheavals may be more important than external factors in some cases. Despite these limitations, balance-of-threat theory
provides a more powerful explanation of alliance choices than structural
balance-of-power theory.
These distinctions are not merely academic. The failure to appreciate that
allies will weigh threats differently and follow a different diplomatic agenda
has been a recurring problem for the United States. Preoccupied with Soviet
power, American policymakers have too readily assumed that regional allies
share its own anti-Soviet concerns. To seek a "strategic consensus" against
the Soviet Union, as John Foster Dulles and Alexander Haig sought to do,
ignores the fact that regional threats are often of greater importance to
138. See Steven David, "The Superpowers and Regime Realignment in the Third World,"
paper presented at the annual meeting of the American Political Science Association, Chicago,
3-6 September 1987.
139. To mention a few examples: 1) India acquired the MIG-21 to counter Pakistan's purchase of the American F-86; 2) India began a major military buildup after its defeat by China in
1962, which encouraged the Sino-Pakistani military aid relationship; 3) Pakistan's nuclear
program was a clear response to India's successful effort to acquire a nuclear capability; and 4)
the Pakistani decision to acquire the F-16 encouraged India's successful effort to obtain the
Mirage 2000 and MIG-29 from the Soviet Union in 1986. For a general discussion, see Robert
Wirsing, "The Arms Race in South Asia: Implications for the U.S.," Asian Survey 25 (March
1985).
was the dominant tendency among the European great powers and among
the regional powers in the contemporary Middle East. 142 TOnote a few more
examples: the Sino-American rapprochement, the formation of the Gulf
Cooperation Council following the Iranian revolution, and the emergence of
Association of Southeast Asian Nations (ASEAN) after the U.S. withdrawal
from Vietnam all reveal the same tendency for states to seek allies to oppose
external threats.
Is this tendency universal? Apparently not. For example, balancing behavior has been relatively rare in the Western hemisphere. The United
States is far more powerful than its neighbors, it is much closer to them than
the Soviet Union is, and it has intervened in these countries on numerous
occasions. If states do tend to balance against threats, we should expect
them to align with each other and with the Soviet Union to oppose the
United States. In fact, however, these states have for the most part chosen
to bandwagon with the United States rather than balance against it.143Does
this anomaly undermine balance-of-threat theory significantly?
The answer is no. Indeed, this apparent anomaly provides additional support for the theory. As argued in the first part of this article, balancing
behavior is less likely: 1) when the states in question are too weak to alter
the balance through their own actions, 2) when effective allies are unavailable, and 3) when the threatening power is believed to be appeasable. All
three conditions apply to the Western hemisphere. Because these states are
far weaker than the United States, joining forces together would not afford
them much protection. Nor are effective allies readily available. Although
Germany flirted with several Latin American states during World Wars I and
11, the European great powers were generally too preoccupied with each
other to compete with the United States in the Western hemisphere. And as
the Sandinistas are now discovering, the Soviets are reluctant to support
additional Cubas in the American " b a ~ k y a r d . " ' ~Finally,
for most Latin
~
American elites, American intentions are not unalterably aggressive. In the
past, direct U.S. intervention has occurred primarily to prevent or reverse
alleged Communist takeovers.14' For most Latin American elites, therefore,
142. On the European great power system, see Ludwig Dehio, The Precarious Balance (New
York: Vintage, 1965) and F. H. Hinsley, Power and the Pursuit of Peace: Theory and Practice
in the History of Relations between States (Cambridge: Cambridge University Press, 1963). On
balancing behavior in the Middle East, see Walt, Origins of Alliances, chap. 5 .
143. The obvious exceptions are Cuba and Nicaragua, which have balanced against the
United States by aligning with each other and the Soviet Union.
144. See Stephen Kinzer, "For Nicaragua, Soviet Frugality Starts to Pinch," New York
Times, 20 August 1987, pp. A l , A12; and Joel Brinkley, "U.S. Says Kremlin Proposed to Cut
Aid to Nicaragua," New York Times, 4 February 1988, pp. A l , A12.
145. On this point, see Stephen D. Krasner, Defending the National Interest: Raw Materials
Investments and U . S . Foreign Policy (Princeton, N.J.: Princeton University Press, 1978).
Histories of U.S. intervention include: Cole Blasier, The Hovering Giant: The U.S. and Revolutionary Change in Latin America (Pittsburgh: University of Pittsburgh Press, 1976); and Walter
Lafeber, Inevitable Revolutions: The United States in Central America (New York: Norton,
1982).
the United States is relatively easy to appease; if they do not "go Communist," the United States is unlikely to cause them much trouble. And because radical domestic change would endanger their own positions, these
elites have been more than willing to meet this condition. Indeed, American
support has helped a number of Latin American oligarchies resist domestic
change. 146 Accordingly, balancing behavior has occurred only after an internal revolution has ousted the traditional ruling elite (such as in Cuba and
Nicaragua) and created an imminent threat of American interference. 14' In
short, although the Latin American case appears to present an exception to
balance-of-threat theory, it does not damage it irreparably. Indeed, it is an
exception that is largely consistent with the theory. Because the conditions
for balancing are lacking, bandwagoning has been more common.
In addition to confirming the merits of balance-of-threat theory, the cases
examined in this article also help us clarify the concepts of balancing and
bandwagoning. As ideal types, balancing and bandwagoning are mutually
exclusive, but reality is not always this neat. In particular, balancing against
a potential threat does not require unremitting hostility towards it. Indeed,
the optimal strategy for most states would be to forge a reliable alliance with
the least threatening great power, while maintaining reasonably good relations with the other. That is exactly what Turkey, Pakistan, and Iran have
tried to do, and India's policy of "non-aligned alignment" reflects a similar
awareness.14' It is a fundamental error to view such efforts at accommodation as bandwagoning, especially when basic security arrangements are
maintained. On the contrary, they reflect an astute effort to maximize security by aligning with one side while maintaining cordial relations with the
other.
The prevalence of balancing behavior also suggests that much of postwar
U.S. foreign policy has been based on a fundamental misunderstanding of
international politics. Because states tend to balance rather than bandwagon, the United States should worry less about its allies defecting and
worry more about how it squanders resources and provokes needless opposition in unnecessary efforts to reinforce its credibility.
Contrary to several recent proposals, however, the tendency for states to
balance does not mean that the United States can simply return to isolation146. The most obvious examples are Guatemala, El Salvador, and Nicaragua under the Somozas.
147. In addition to Cuba and Nicaragua, it is worth noting the Arbenz regime in Guatemala
sought to balance the American effort to overthrow it by purchasing weapons from the Soviet
bloc.
148. As Turkish Prime Minister Bulent Ecevit remarked in 1978: "Certainly a country always
needs armaments and armies to insure her security. But I believe . . . that establishing an
atmosphere of mutual confidence in our relations with the neighboring countries is at least as
protective as and sometimes more protective, than armaments." See Ecevit, "Turkey's Security Policies," p. 205.
ism, confident that its allies will quickly take up the ~ 1 a c k .Bandwagoning
l~~
does occur when vulnerable states are isolated, and a total U.S. withdrawal
could lead some of its weaker allies to adopt a neutralist policy. Moreover,
although our major allies would probably balance (for example, by increasing their own defense efforts), the real question is whether they would do
enough to compensate fully for the loss of U.S. support. Deterrence would
be weakened in any event because a Soviet challenge would be more likely if
they were convinced that the United States would not become involved.
Finally, the U.S. presence helps protect its allies from each other, enabling
them to cooperate more readily in both economic and military realms.
For all of these reasons, the tendency for states to prefer balancing does
not mean that the United States can simply "free ride" on the efforts of
others. lS0
A final caveat should be kept in mind. We do not know how the states
examined in this study would have reacted had the superpowers behaved
differently. Had the Soviet Union adopted a more cooperative policy in the
early cold war, and had the United States returned to an isolationist policy, a
number of U.S. allies might have chosen the Soviet Union instead. But this
is only speculation. What these cases reveal is that the likelihood of bandwagoning is greatly exaggerated. Although it may occur from time to time,
this study reinforces the conclusions of earlier analyses: relatively little
effort is required to persuade even vulnerable states to balance. For those
who rely upon allied support to contain their adversaries, this is reassuring
news.
149. See the references cited in note 3 .
150. For a more extensive analysis of this issue, see Stephen M. Walt, "Two Cheers for
Containment: U.S. Grand Strategy in the Post-Reagan Era," Cato Journal 8 (forthcoming,
Summer 1988).