Alliance Formation and The Balance of World Power - Stephen Walt (1985)
Alliance Formation and The Balance of World Power - Stephen Walt (1985)
Alliance Formation and The Balance of World Power - Stephen Walt (1985)
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
The MIT Press is collaborating with JSTOR to digitize, preserve and extend access to International Security.
http://www.jstor.org
AllianceFormationStephenM. Walt
and the Balance of
WorldPower
The
question "what
I would like to thank RobertArt, George Breslauer,Lynn Eden, Charles Glaser, Lori Gronich,
Fen Hampson, JohnMearsheimer,Kenneth Oye, Glenn Snyder, JackSnyder, Marc Trachtenberg, and Kenneth Waltz fortheirthoughtfulcommentson earlierdraftsof this article.
StephenM. Walt is an AssistantProfessorof Politicsand International
Affairsin the Departmentof
Politicsand theWoodrowWilsonSchoolat PrincetonUniversity.
1. For representativeexamples of typical scholarlyefforts,consult: Robert Rood and Patrick
McGowan, "Alliance Behavior in Balance of Power Systems," AmericanPoliticalScienceReview,
Vol. 69, No. 3 (September 1975); George T. Duncan and Randolph Siverson, "Flexibilityof
Alliance PartnerChoice in Multipolar Systems," Initernational
StudiesQuarterly,Vol. 26, No. 4
(December 1982); R.P.Y. Li and W.R. Thompson, "The StochasticProcess of Alliance Formation
Behavior," AmericanPoliticalScienceReview,Vol. 72, No. 3 (December 1978). More traditional
Spring 1985 (Vol. 9, No. 4) 0162-2889/85/040003-41
$02.50/0
Internationial
Secuirity,
C 1985 by the Presidentand Fellows of Harvard College and of the Massachusetts Instituteof Technology.
International
Security14
some of the most importanthypotheses of alliance formation,and by exploring the policy implicationsof each. The firstsection explores the competing propositions that states eitherbalance against strongor threatening
states or, alternatively,that they "bandwagon" with them. I shall also consider the sharplydifferent
foreignand defense policies thateach proposition
implies. The second section develops the contrastinghypotheses that ideological or cultural similaritiescan eitherbind states togetheror drive them
apart. The thirdsectionexamines the abilityofstatesto createallies or proxies
by militaryand economic aid, propaganda, or politicalpenetration.Finally,
the last section demonstrates how these hypotheses, taken together,can
explain the currentstructureof world power, and suggests what theyimply
forAmerican national securitypolicy.
BalancingVersusBandwagoning:
Alliancesas a Responseto Threat
Alliances are most commonlyviewed as a response to threats,yet there is
sharp disagreement as to what that response will bo. When entering an
alliance, states may eitherbalance(ally in opposition to the principalsource
of danger) or bandwagon(ally with the state that poses the major threat).2
These contrastinghypotheses depict very differentworlds, and the policies
thatfollow fromeach are equally distinct.In the simplestterms,ifbalancing
is more common than bandwagoning, then states are more secure because
aggressorswill face combined opposition. Status quo states should therefore
avoid provokingcountervailingcoalitionsby eschewing threateningforeign
and defense policies. But if bandwagoning is the dominant tendency,then
securityis scarce because aggressionis rewarded. A more belligerentforeign
AllianceFormation| 5
BEHAVIOR
The proposition that states will join alliances in order to avoid domination
by strongerpowers lies at the heart of traditionalbalance of power theory.3
Accordingto this hypothesis,statesjoin alliances to protectthemselvesfrom
states or coalitions whose superiorresources could pose a threat.States will
choose to balance fortwo main reasons.
First,states risktheirown survivaliftheyfailto curb a potentialhegemon
beforeit becomes too strong.To ally withthe dominantpower means placing
one's trustin its continued benevolence. The safer strategyis to join with
those who cannot readily dominate their allies, in order to avoid being
dominated by those who can.4 As Winston Churchill explained Britain's
traditionalalliance policy:
For four hundred years the foreignpolicy of England has been to oppose
the strongest,most aggressive, most dominatingpower on the Continent.
... it would have been easy ... and temptingto join with the strongerand
3. For impressive analyses of the classical writingson the balance of power, see: Edward V.
Gulick,Europe'sClassicalBalanceofPower(New York: W.W. Norton, 1955), Part I; F.H. Hinsley,
Powerand thePursuitofPeace: Theoryand Practicein theHistoryofRelationsbetweenStates(Cambridge:Cambridge UniversityPress, 1963), PartI; Inis L. Claude, Powerand International
Relations
(New York: Random House, 1962), Chapters 2 and 3; RobertOsgood and RobertTucker,Force,
Order,and Justice(Baltimore:JohnsHopkins UniversityPress, 1967), pp. 96-104 and passim;and
MartinWight,"The Balance of Power," in MartinWightand HerbertButterfield,
eds., Diplomatic
Investigations
(London: Allen and Unwin, 1966). For modern versions of the theory,see Waltz,
Theory
ofInternational
Politics,Chapter 6; MortonKaplan, Systemand Processin International
Politics
(New York: JohnWiley, 1957); and Hans J. Morgenthau,PoliticsAmongNations,5th ed. (New
York: AlfredA. Knopf, 1978), Part IV.
4. As Vattelwrote several centuriesago: "The surestmeans of preservingthisbalance of power
would be to bring it about that no State should be much superior to the others . . . but this
could not be realized without injustice and violence. . . [It] is simpler,easier, and more just
. . . to formalliances in order to make a stand against a very powerfulsovereign and prevent
him fromdominating." Quoted in Gulick, Europe'sClassicalBalanceofPower,pp. 61-62.
International
Securityj 6
share the fruitsof his conquest. However, we always took the hardercourse,
joined with the less strongPowers, . . . and thus defeated the Continental
militarytyrantwhoever he was....5
In the same way, Henry Kissingeradvocated rapprochement
withChina rather
than the Soviet Union because he believed that,in a triangularrelationship,
it was betterto align with the weaker side.6
Second, joining the more vulnerable side increases the new member's
influence,because the weaker side has greaterneed forassistance. Joining
the strongerside, by contrast,reduces the new member'sinfluence(because
it adds relativelyless to the coalition) and leaves it vulnerable to the whims
of its new partners. Alignmentwith the weaker side is thus the preferred
choice.7
BEHAVIOR
The beliefthat states will tend to ally withratherthan against the dominant
side is surprisinglycommon. Accordingto one scholar,
[In internationalpolitics] momentum accrues to the gainer and accelerates
his movement. The appearance of irreversibility
in his gains enfeebles one
side and stimulatesthe otherall the more. The bandwagon collectsthose on
the sidelines.9
Storm(Boston: Houghton
5. WinstonS. Churchill,TheSecondWorldWar: VolumeI, TheGathering
Mifflin,1948), pp. 207-208.
6. Henry A. Kissinger, WhiteHouse Years (Boston: Little,Brown, 1979), p. 178.
7. In the words of Kenneth Waltz: "Secondary states, if they are free to choose, flockto the
weaker side; forit is the strongerside that threatensthem. On the weaker side, they are both
more appreciated and safer, provided, of course, that the coalition they join achieves enough
defensiveor deterrentstrengthto dissuade adversaries fromattacking."See his TheoryofInternationalPolitics,p. 127.
8. This theme is explored in Ludwig Dehio, The PrecariousBalance(New York: Vintage, 1965);
Hinsley, Powerand thePuirsuitofPeace; and Gulick, Europe'sClassicalBalanceofPower.
9. W. Scott Thompson, "The Communist InternationalSystem," Orbis,Vol. 20, No. 4 (Winter
1977), p. 843.
AllianceFormation| 7
Scholars are not alone in this conception. For example, the German Admiral Alfred von Tirpitz's famous "risk theory" implied such a view. By
building a great battle fleet,Tirpitzargued, Germany could forceEngland
intoneutralityor alliance withitby posing a threatto England's vitalmaritime
supremacy.10More recently,American officialshave repeatedly embraced
the bandwagoning hypothesisin justifyingAmericanforeignpolicy commitments. John F. Kennedy claimed that, "if the United States were to falter,
the whole world ... would inevitablybegin to move toward the Communist
bloc.""1Although the rapprochement
with China showed his own willingness
to balance, Henry Kissinger also revealed his beliefthat most states tend to
bandwagon by suggesting that "if leaders around the world
. . .assume that the U.S. lacked eitherthe forcesor the will . . they will
accommodate themselvesto the dominanttrend."'12
And Ronald Reagan has
endorsed the same beliefs in his claim that "if we cannot defend ourselves
[in CentralAmerica] . . . then we cannot expect to prevail elsewhere . .. our
credibilitywill collapse and our alliances will crumble."13
Statements like these reveal a common theme: states are attracted to
strength.The more powerful you are and the more clearlythis is demonstrated,the more likelyothersare to ally with you. By contrast,a decline in
relativeposition will lead one's allies to opt forneutralityat best or to defect
to the other side at worst.
What is the logic behind the bandwagoning hypothesis? Two distinct
motives can be identified.First,bandwagoning may be adopted as a formof
International
Security| 8
appeasement. By aligning with the threateningstate or coalition, the bandwagoner may hope to avoid an attackon himselfby divertingit elsewhere.
Second, a state may align with the dominant side in war in order to share
the spoils of victory.Mussolini's declarationof war on France and Russia's
entryinto the war againstJapanin 1945 illustratethistypeof bandwagoning,
as do Italian and Rumanian alliance choices in World War 1.14 By joining
what they believed was the strongerside, each hoped to make territorial
gains at the end of the fighting.
Stalin's decision to ally with Hitler in 1939 illustratesbothmotives nicely.
The Nazi-Soviet Pact led to the dismembermentof Poland and may have
deflectedHitler's ambitionswestward. Stalinwas thus able to gain both time
and territory
by bandwagoning with Hitler.15In general,however, these two
motives forbandwagoning are quite different.In the first,bandwagoning is
chosen fordefensive
reasons, as a means of maintainingindependence in the
face of a potential threat.In the second, a bandwagoning state chooses the
leading side foroffensive
reasons, in order to acquire territory.
Regardless of
the specificmotive, however, bandwagoning behavior stands in sharp contrastto the predictionsof balance of power theory.The two hypotheses thus
offermutuallyexclusive explanations forhow states will make theiralliance
choices.
DIFFERENT
SOURCES
OF THREAT
14. See Denis Mack Smith,Mussolinii(New York: AlfredA. Knopf, 1982), pp. 234-235, 246-250;
and Coexistence(New York: Praeger, 1972), pp. 394-398; and A.J.P.
Adam B. Ulam, Expansioni
Taylor, TheFirstWorldWar (New York: Perigee Books, 1980), pp. 88-90, 153.
pp. 276-277; Isaac Deutscher, Stalin:A PoliticalBiography
15. See Ulam, Expansionand Coexistenice,
(London: Pelican Books, 1966), pp. 437-443; and JoachimFest, Hitler(New York: Vintage, 1974),
pp. 583-584, 592-593.
16. The preeminentexample of balance of power theoryfocusingexclusivelyon the distribution
Politics,Chapter 6. For examples of theoristswho
of capabilitiesis Waltz, TheoryofIniternationial
acknowledge that other factorscan be important,see Gulick, Europe'sClassicalBalanceofPower,
pp. 25, 45-47, 60-62.
AllianceFormationI 9
threatening
power. For example, states may balanceby allying with other
strongstates, if a weaker power is more dangerous forotherreasons. Thus
the coalitions that defeated Germany in World Wars I and II were vastly
superior in total resources, but united by their common recognitionthat
German expansionism posed the greater danger.17Because balancing and
bandwagoning are more accuratelyviewed as a response to threats,it is
importantto consider all the factorsthat will affectthe level of threatthat
states may pose. I shall thereforediscuss the impact of: 1) aggregatepower;
2) proximity;3) offensivecapability;and 4) offensiveintentions.
AGGREGATE POWER. The greatera state's total resources (i.e., population,
industrialand militarycapability,technologicalprowess, etc.), the greatera
potential threat it can pose to others. Recognizing this, Walter Lippmann
and George Kennan defined the aim of American grand strategyto be preventingany single statefromcontrollingthe combinedresourcesof industrial
Eurasia, and theyadvocated U.S. interventionon whicheverside was weaker
when thisprospectemerged.18Similarly,Lord Grey,BritishForeignSecretary
in 1914, justifiedBritishinterventionagainst the Dual Alliance by saying:
To stand aside would mean the domination of Germany;the subordination
of France and Russia; the isolation of Britain,. . . and ultimatelyGermany
would wield the whole power of the continent.19
In the same way, Castlereagh's aim to createa "just distributionof the forces
in Europe" reveals his own concern forthe distributionof aggregatepower,
as does Bismarck's dictum that "in a system of five great powers, the goal
17. In World War I, the alliance of Great Britain,France, and Russia controlled27.9 percentof
world industrialproduction,while Germanyand Austria togethercontrolledonly 19.2 percent.
With Russia out of the war but the United States joining Britainand France, the percentage
opposing the Dual Alliance reached 51.7 percent,an advantage of more than 2 to 1. In World
War II, the defense expenditures of the U.S., Great Britain,and the Soviet Union exceeded
those of Germanyby roughly4.5 to 1. Even allowing forGermany's controlof Europe and the
need to fightJapan,the Grand Alliance possessed an enormous advantage in latentcapabilities.
Thus balancing against powerwas not the sole explanation for these alliances. For these and
otherstatisticson the relativepower in these two wars, see: Paul M. Kennedy, "The FirstWorld
Security,Vol. 9, No. 1 (Summer 1984),
War and the InternationalPower System," International
pp. 7-40; and TheRise and Fall ofBritishNaval Mastery(London: Macmillan, 1983), pp. 309-315.
(New York:
ofContainment
18. For a summaryof these ideas, see JohnLewis Gaddis, Strategies
OxfordUniversityPress, 1982), pp. 25-88. Kennan's own thoughtsare found in his Realitiesof
AmericanForeignPolicy(New York: New American Library,1951), p. 10. Lippmann's stillcompellinganalysis is found in his The Cold War: A Stuidyof U.S. ForeignPolicy(New York: Harper
Brothers,1947).
19. Quoted in Bernadotte Schmitt,The Comingof theWar ini1914 (New York: Howard Fertig,
1968),Vol. 2, p. 115.
International
SecurityI 10
20. Castlereagh's policy is described in Harold Nicolson, The Congressof Vienna (New York:
Harcourt,Brace, Jovanovich,1946), pp. 205-206. Bismarck'sstatementis quoted in William L.
2nd ed. (New York: Random House, 1950), p. 197.
Langer, EuropeanAlliancesand Alignments,
21. See Harvey Starrand Benjamin A. Most, "The Substance and Study of Borders in InternaStudiesQuarterly,
Vol. 20 (1976). For a discussion of the
tional Relations Research," International
relationshipbetween power and distance, see Kenneth A. Boulding, Conflictand Defense:A
GeneralTheory(New York: Harper Torchbooks, 1962), pp. 229-230, 245-247. For an interesting
Vol. 46, No. 2
practicalcritique,see AlbertWohlstetter,"Illusions of Distance," ForeignAffairs,
(Fall 1968).
Antagonism(London: Allen and
22. Quoted in Paul M. Kennedy, The Rise of theAnglo-German
Unwin, 1980), p. 421.
23. Kautilya's analysis ran as follows: "The king who is situated anywhereimmediatelyon the
is termedthe enemy.The kingwho is likewise situated
circumference
of the conqueror's territory
close to the enemy, but separated fromthe conqueror only by the enemy is termed the friend
(of the conqueror).... In frontof the conquerorand close to the enemy, therehappened to be
situated kings such as the conqueror's friend,next to him the enemy's friend,and next to the
last the conqueror's friend'sfriend,and next, the enemy's friend'sfriend."See "Arthasastra"
(Science of Politics), in Paul A. Seabury, ed., BalanceofPower(San Francisco: Chandler, 1965),
p. 8.
AllianceFormationI 11
and the "LittleEntente" in the 1930s; the Soviet Union and Vietnam against
China and Cambodia in the 1970s; the U.S.S.R. and India against the U.S.
and Pakistan presently; and the tacit alignment between Iran and Syria
against Iraq and its various Arab supporters.When a threatfromproximate
power leads to bandwagoning, by contrast,the familiarphenomenon of a
"sphere of influence" is created. Small states borderinga great power may
be so vulnerable that they choose to bandwagon ratherthan balance, especiallyif theirpowerfulneighborhas demonstratedits abilityto compel obedience. Thus Finland, whose name has become synonymouswith bandwagoning, chose to do so only afterlosing two major wars against the Soviet
Union withina five-yearperiod.
OFFENSIVE POWER. All else being equal, states with large offensivecapabilitiesare more likelyto provoke an alliance than those who are eithermilitarily
weak or capable only of defending.24Once again, the effectsof this factor
vary. On the one hand, the immediatethreatthatsuch capabilitiespose may
lead states to balance by allyingwith others.25Tirpitz's"riskstrategy"backfiredforpreciselythis reason. England viewed the German battle fleetas a
potentoffensivethreat,and redoubled its own naval effortswhile reinforcing
its ties with France and Russia.26On the otherhand, when offensivepower
permits rapid conquest, vulnerable states may see littlehope in resisting.
Balancing may seem unwise because one's allies may not be able to provide
assistance quicklyenough. This is anotherreason why "spheres of influence"
may form:states borderingthose with large offensivecapabilities(and who
are farfrompotentialallies) may be forcedto bandwagon because balancing
alliances are simplynot viable.27
24. The best discussions of the implicationsof offenseand defense are: RobertJervis,"CooperationUnder the SecurityDilemma," WorldPolitics,Vol. 30, No. 2 (January1978); Stephen W.
Van Evera, "Causes of War" (Ph.D. dissertation,Universityof California,Berkeley,1984); and
System(New York: Wiley, 1977).
and Defensein theInternational
George H. Quester, Offense
pp. 3-5; Raymond J. Sontag, EuropeanDiplo25. See Langer, EuropeanAlliancesand Alignments,
maticHistory,1871-1932 (New York: Appleton-CenturyCrofts,1933), pp. 4-5; Jervis,"Cooperand Defensein theInternational
ation Under the SecurityDilemma," p. 189; and Quester, Offense
System,pp. 105-106.
26. As Imanuel Geiss notes: "Finding an agreementwith Britainalong German lines withouta
substantialnaval agreementthus amounted to squaring the circle."See his GermanForeignPolicy
(London: Routledge and Kegan Paul, 1977), p. 131. See also Kennedy, The Rise of theAngloGermanAntagonism,
pp. 416-423.
27. Thus alliance formationbecomes more freneticwhen the offenseis believed to have the
advantage: great powers will balance more vigorously while weak states seek protectionby
bandwagoning more frequently.A world of tightalliances and few neutral states is the likely
result.
International
Security| 12
INTENTIONS.
It cannot for a moment be questioned that the mere existence and healthy
activityof a powerfulGermanyis an undoubted blessing forall.... So long,
and moralleadership
then,as Germanycompetes
foran intellectual
of theworldin
relianceon its own naturaladvantagesand energiesEngland cannot but admire.
. . . [So] longas Germany'sactiondoesnotoverstepthelineoflegitimate
protection
and goodwill, and even
ofexistingrightsit can alwayscountupon thesympathy
the moral support of England. . . . It would be of real advantage if the
determinationnot to bar Germany's legitimateand peacefulexpansionwere
made as patent and pronounced as authoritatively
as possible, provided that
care was taken at the same time to make it quite clear that thisbenevolent
attitudewillgive way to determined
opposition
at thefirstsign of Britishor allied
interestsbeing adversely affected.30
28. For a discussion of Libya's internationalposition, see Claudia Wright,"Libya and the West:
Affairs(London), Vol. 58, No. 1 (Winter1981-82),
Headlong Into Confrontation?,"International
pp. 13-41.
29. See Craig, Germany:1866-1945, pp. 101, 242-247, and Chapter 10; Geiss, GermanForeign
Antagonism,
Chapters 14 and 20.
Policy,pp. 66-68; and Kennedy, The Rise oftheAnglo-German
30. "Memorandum by Sir Eyre Crowe on the PresentState of BritishRelations withFrance and
Germany,January1, 1907," in G.P. Gooch and Harold Temperley,eds., BritishDocumentson
theOriginsoftheWar, 1898-1914 (London: BritishForeign Office,1928), Volume 3, pp. 403 and
passimn
(emphasis added). See also G.W. Monger, The End ofIsolation:BritishForeignPolicy19001907 (London: Thomas Nelson, 1963), pp. 313-315.
AllianceFormation| 13
OF BALANCING
AND BANDWAGONING
International
Security| 14
AllianceFormation| 15
IS MORE
COMMON
THAN
BANDWAGONING
Which of these two worlds most resembles reality?Which hypothesis describes the dominant tendencyin internationalpolitics?Although statesmen
frequentlyjustifytheir actions by invoking the bandwagoning hypothesis,
historyprovides littleevidence for this assertion. On the contrary,balance
of power theoristsfromRanke forwardhave persistentlyand persuasively
shown that states facingan externalthreatoverwhelminglypreferto balance
against the threatratherthan bandwagon with it. This is primarilybecause
an alignmentthat preserves most of a state's freedomof action is preferable
to accepting subordinationunder a potential hegemon. Because intentions
can change and perceptions are unreliable, it is safer to balance against
potentialthreatsthan to hope that strongstates will remain benevolent.
The overwhelmingtendencyforstates to balance ratherthan bandwagon
defeated the hegemonic aspirations of Spain under Philip II, France under
Louis XIV and Napoleon, and Germanyunder WilhelmII and Hitler.Where
the bandwagoning hypothesispredictsthatthese potentialhegemons should
have attractedmore and more supportas theyexpanded, the actual response
of the powers that they threatened was precisely the opposite. The more
clearlyany one state sought to dominatethe rest,the morereliablythe others
combined to counterthe threat.35
34. The French attemptto contain Germany afterWorld War I was undermined both by the
LocarnoTreaty(which guaranteed the Frenchborderwith Germanybut failedto provide similar
guarantees for its allies) and by the French adoption of a defensive militarydoctrine,which
made it impossible for it come to the aid of its allies. See TelfordTaylor, Munich: The Priceof
Peace (New York: Vintage, 1980), pp. 111-112; and Richard D. Challener, The FrenchTheoryof
theNationin Arms(New York: Columbia UniversityPress, 1969), pp. 264-265. For the effectsof
Soviet pressure on Turkey, see: George Lenczowski, The Middle East in WorldAffairs(Ithaca:
CornellUniversityPress, 1980), pp. 134-138; and Bruce R. Kuniholm,TheOriginsoftheCold War
in theNear East (Princeton:PrincetonUniversityPress, 1980), pp. 355-378.
35. See JackS. Levy, "Theories ofGeneral War," unpublished ms., 1984. (An extensivelyrevised
versionof thispaper will be published in WorldPolitics,April 1985.)
Intermational
Security| 16
36. For evidence and analysis on this point, see Stephen M. Walt, "The Origins of Alliances"
(Ph.D. dissertation,Universityof California,Berkeley,1983), especially Chapter 6.
37. See Mahnaz Zehra Ispahani, "Alone Together:Regional SecurityArrangementsin Southern
Security,Vol. 8, No. 4 (Spring 1984), pp. 152-175.
Africaand the Arabian Gulf," International
AllianceFormation| 17
especially weak states will be more likelyto bandwagon, both because they
are more vulnerable to pressure and because the capabilitiestheycan add to
eitherside are unlikelyto make much difference.Because they can do little
to affectthe outcome, they are more likely to opt for the winning side.38
Thus King Leopold of Belgium and Urho Kekkonen of Finland justifiedtheir
own alliance policies with referenceto the special vulnerabilitiesof small
statesborderingupon greatpowers.39A furtherdeductionis thatweak states
may balance against other weak states, but may be relativelymore likelyto
bandwagon when confrontedby a great power.
Second, weak states are more likelyto bandwagon when allies are simply
unavailable. Even weak states may be persuaded to balance when they are
confidentof allied support; in its absence, however, accommodation with
the threateningpower may be the only viable alternative.Thus a further
prerequisiteforeffectivebalancing behavioris an active systemof diplomatic
communication,permittingpotentialallies to recognizetheirshared interests
and coordinatetheirresponses.40If weak states see no possibilityof external
assistance, accommodation throughalignmentwith the threateningpower
may be chosen as a last resort.Thus the firstShah of Iran took the British
withdrawal fromKandahar in 1881 as a signal to bandwagon with Russia.
As he told the Britishrepresentative,all he had received fromBritainwas
"good advice and honeyed words-nothing else.i"41Finland's foreignpolicy
suggests the same lesson. Finland's bandwagoning alliance with the Soviet
Union afterWorld War II was encouraged by the factthatFinland's balancing
38. See Rothstein,Alliancesand SmallPowers,p. 11.
39. As King Leopold explained Belgian neutralityafterWorld War I, "an alliance, even ifpurely
defensive,does not lead to the goal [of security],forno matterhow promptthe help of an ally
mightbe, it would not come until afterthe invader's attackwhich will be overwhelming...."
Quoted in Rothstein,Alliancesand Small Powers,pp. 111-112. Kekkonen of Finland argued for
accommodationwith the U.S.S.R. by saying: "A small state cannot stand foreverarmed to the
teeth. . . the firstto be overrunby the enemy, and devoid of politicalimportanceto lend any
significanceto its word when decisions over war and peace are being taken ....." See Urho
Kekkonen,A President'sView,trans. GregoryCoogan (London: Heinemann, 1982), pp. 42-43.
40. One reason forRome's durable hegemonyin the ancientworld was the factthatits various
opponents lacked the diplomaticmeans to coordinateopposition against Rome effectively.See
oftheRomanEmpire(Baltimore:JohnsHopkins University
Edward N. Luttwak,TheGrandStrategy
Press, 1976), pp. 192, 199-200. When a workable diplomatic system was established in the
Renaissance,prospects forEuropean hegemony declined drastically.On this point, see Gulick,
Europe'sClassicalBalanceofPower,p. 16; Hedley Bull, TheAnarchicalSociety(New York: Columbia
UniversityPress, 1977), p. 106 and Chapter 7; GarrettMattingly,RenaissanceDiplomacy(Boston:
Houghton Mifflin,1971), Chapters 13-16; and Harold Nicolson, Diplomacy(London: Oxford
UniversityPress, 1963), Chapter 1.
(New York: Macmillan, 1967), p. 85.
41. Quoted in C.J. Lowe, TheReluctantImperialists
International
Security118
AllianceFormationI 19
to this hypothesis,the more similartwo or more states are, the more likely
theyare to ally. Althoughmost scholarsbelieve thatthisis at best a secondary
explanationforalliances,44the beliefthatideological affinities
are crucialoften
appears in the rhetoricof statesmen seeking to justifyalignmentwith one
side or opposition to another.Thus Samora Machel ofMozambique explained
his close relationship with the U.S.S.R. by describing the two states as
"natural socialist allies."45Lord Palmerstonof Britain,despite his assertion
thatEngland had "no permanentfriends. . . only permanentinterests,"also
believed in the naturalaffinity
of democracies.In a statementthatalso reveals
a beliefthat weak states will bandwagon, Palmerstonsaid:
Our policy ought now to be to forma Westernconfederacyof freestates as
a counterpoise to the Eastern League of arbitrarygovernments.England,
France, Spain, and Portugal . . . will forma political and moral power in
Europe....
We shall be on the advance, they on the decline, and all the
smaller planets in Europe will have -a natural tendencyto gravitatetowards
our system.46
More recently,JohnFosterDulles justifiedAmericansupportforChiang Kaishek and SynghmanRhee by proclaimingthatthese leaders "were Christian
In the same spirit,Ronald
gentlemen. . . who have sufferedfortheirfaith."47
Reagan has praised the factthat the U.S. and its allies have "rediscovered
theirdemocraticvalues," values that"unite us in a stewardshipof peace and
And throughoutthe Cold War, Amerfreedomwith our allies and friends."48
ican opposition to leftistmovementsin the Third World has been based on
hegemonic power has increased. Obviously, this tendencyis more pronounced in the Western
Hemispherethan in EasternEurope, because geographymakes iteasier forthe Soviets to enforce
theircontrol.
44. For scholarly discussions that question the importanceof ideology in alliance formation,
see: Edwin Fedder, "The Concept of Alliance," International
StudiesQuarterly,Vol. 12 (1968), p.
86; Morgenthau,PoliticsAmongNations,pp. 183-184; ErnstB. Haas and Allen Whiting,Dynamics
ofInternational
Relations(New York:McGraw-Hill,1956), pp. 167-168;RobertE. Osgood, Alliances
andAmericanForeignPolicy(Baltimore:JohnsHopkins UniversityPress, 1968), p. 20; and Harold
Guetzkow, "Isolation and Collaboration:A PartialTheory of InternationalRelations," Journalof
Conflict
Resolution,Vol. 1, No. 1 (1957), p. 158.
45. Quoted in Committeeon InternationalRelations,"The Soviet Union and the Third World,"
pp. 46-48. Under pressure fromSouth Africa,Machel has recentlymoderated his pro-Soviet
stance.
46. Quoted in Charles K. Webster, The ForeignPolicyofPalmerston(London: G. Bell and Sons,
1951),Vol. 1, p. 390.
47. Quoted in Townsend Hoopes, The Devil and JohnFosterDulles (Boston: Little,Brown, 1973),
pp. 77-78.
48. "State of the Union Message," TheNew YorkTimes,January26, 1983.
International
Security| 20
AllianceFormation| 21
OF A FEATHER
FLYING
APART:
DIVISIVE
IDEOLOGIES
International
Security| 22
rivalriesbetweenCommunist states have been among the world's most virulent quarrels. The "natural" cohesion of the movementwas guaranteed only
in Eastern Europe, and there only by force.
The historyof Pan-Arabism provides an even more strikingillustration.
Despite the many attempts to translate the Arab world's common ethnic
characterand ideological vision intoworkablepoliticalcohesion, the ideology
of Pan-Arabism has led to repeated rivalries.58And the more serious the
commitmentto unity,the more intense the conflict.Thus the bitterestrivalries in the Arab world took place between Nasser (the leading Pan-Arab
figure) and the transnational,explicitlyPan-Arab Ba'ath party. And the
Ba'ath Partyitselfeventuallysplitinto rivalSyrianand Iraqi factionsin 1966,
a schism that persiststo this day.59
The explanation for these rivalrieslies in the contradictorypremises of
Pan-Arab ideology. Although support forArab unitywas an importantcomponent of regime legitimacyafter1955,60 implementation
of the ideal threatened the existence of each separate regime. If unitywere ever achieved, all
elites save the one that emerged on top would be eliminated. The various
attemptsat formalunion thus quicklybecame strugglesforpower, in which
the ideology was used to justifyextrememeasures against rivals.61As one
memberof the Ba'ath explained: "the rupture[of the United Arab Republic]
viewoftheunion."62 Afterthe
. .. was caused by a certainEgyptianhegemonic
split, Nasser himselfseemed to recognize the fundamentalcontradiction:
58. For analyses of the various riftsbetween the Arabs, see: Malcolm Kerr, TheArabCold War:
Gamal'AbdelNasserand His Rivals(London: OxfordUniversityPress, 1971); Nadav Safran,"Arab
Politics:Peace and War," Orbis,Vol. 18, No. 2 (Summer 1974); and FromWarto War(New York:
Dial Press, 1969), Chapter 2; and Walt, "The Origins of Alliances," Chapter 7.
59. For discussions of the split within the Ba'ath Party, see: JohnF. Devlin, The Ba'ath Party
(Stanford,Calif.: Hoover InstitutePress, 1968), pp. 313-315; and ItamarRabinovich,SyriaUnder
theBa'ath, 1963-1966: TheArmy-Party
Symbiosis(New York: Halsted Press, 1974), pp. 207-208.
60. For a discussion of the role of Pan-Arabismin the legitimacyformulaof Arab regimes, see
Michael Hudson, Arab Politics:The Searchfor Legitimacy(New Haven: Yale UniversityPress,
1977), Chapter 2 and p. 242.
61. As the EgyptianNational Charterstated: "Egyptis bound to spread its mission and put the
principlesupon which it rests at the disposal of all the Arabs, disregardingthe worn-outnotion
that in doing so it is interfering
in otherpeople's affairs."Quoted in Adeed Dawisha, Egyptin
the Arab World(London: Macmillan, 1976), p. 35. "Spreading its mission" involved military
interventionin the Yemeni civilwar, assassination attemptsagainst otherArab leaders, support
forNasserist groups in other countries,and continuous propaganda over Radio Cairo.
62. Quoted in RobertStephens, Nasser:A PoliticalBiography
(London: Allen Lane/ThePenguin
Press, 1971), p. 343. Emphasis added.
AllianceFormation| 23
International
Security124
OF IDEOLOGICAL
SOLIDARITY
Is ideological solidarityan importantcause of alliances? Under what conditionsdoes it play a greateror lesser role?These questions are difficult
because
ideology is but one factoramong many. Nonetheless, several conclusions
can be stated with confidence.
First, states are more likely to follow theirideological preferenceswhen
theyare already fairlysecure. When faced by great danger, one takes whatever allies one can get. Winston Churchillcaptured this in his famous statementthat"ifHitlerinvaded Hell, I should at least make a favorablereference
to the Devil in the House of Commons," a sentimentthat FranklinD. Roosevelt shared.68These reactions may be compared with earlier Britishand
American policies. In the 1920s, Germany's weakness made it possible for
Britain,France, and the United States to treatthe Soviet Union with disdain,
a revulsionbased largelyon ideology and echoed by the Soviets. Only when
Nazi Germanybegan to pose a significantthreatdid these ideological preferences lose theirpower.69In otherwords, securityconsiderationstake precedence over ideological preferences,and ideologicallybased alliances are
unlikelyto survive when more pragmaticinterestsintrude.
Several interestingimplicationsfollow fromthis conclusion. In particular,
those factorswhich tend to make states more secure should increase the
importanceof ideological considerationsin alliance choices. IfKennethWaltz
is correctthatbipolar worlds are the most stable, then the impactof ideology
may increase because all states are more secure. Not only will the bipolar
rivalryencourage both superpowers to support thirdparties freely(giving
thirdparties the option to choose the ideologically most compatible side),
but the caution that bipolarityimposes on superpower conduct may permit
most other states to follow ideological preferencesratherthan securityre-
AllianceFormation| 25
International
Security 26
AllianceFormationj 27
TheInstruments
ofAllianceFormation:"Bribery"
and Penetration
States seeking allies will employ specificpolicy instrumentsto attractothers
to theirside. The use of such instruments(or the interpretationplaced on
theiruse by others) rests upon implicithypotheses about the relativeeffectiveness of such tactics. The most importantof these hypotheses concerns
the instrumentsof "bribery"and penetration.What are these hypotheses,
and how seriously should we take them?
"INTERNATIONAL
BRIBERY:
FOREIGN
FORMATION
Security| 28
International
AllianceFormation| 29
International
Security130
85. For an analysis of the historicalrecord,see Walt, "The Originsof Alliances," Chapter 8. See
also Robert0. Keohane, "The Big Influenceof Small Allies," ForeignPolicy,Number 2 (Spring
1971), pp. 161-182.
BritishForeignAid in the Wars with France
86. See Robert Sherwig, Guineas and Guinpozwder:
(Cambridge: Harvard UniversityPress, 1969), pp. 311-313, 350-355.
AllianceFormationj 31
Althoughpenetrationhas receivedrelativelylittleattentionin recentscholarly research,87examples are easy to find. The Turkishdecision to ally with
Germany in World War I was due in part to the influence of Liman von
Sanders, a German officercommanding the TurkishArmy in Constantinople.88During the war itself,Britainconducted an effectivepropaganda campaign in the United States,which played a significantrole in the U.S. decision
to enter.89During the 1950s, the "China Lobby" exerteda substantialinfluence over American policy in the Far East by manipulatingpublic opinion
and influentialAmerican officials.90
Finally,the belief that penetrationmay
be an effectivetool of alliance-buildinghas inspired the politicalindoctrination programs that accompanied American militarytrainingforvarious developing countries, not to mention American concern over similar Soviet
educational and militaryassistance programs.91
The circumstancesin which penetrationwill have a significanteffect,however, are limited. First,it is more likelyto succeed in open societies, where
influentialelites are more accessible to foreignideas. Second, if the effortto
penetrateis viewed by the targetstateas subversiveor illegitimate,it is likely
to reactby moving away fromthe state seeking to enhance its influenceand
the penetrationwill thereforebe counterproductive.This possibilityimplies
that penetrationwill be safest (i.e., have the best chance of success) when
there are already strong incentives for the two states to align, so that the
activitiesto encourage the alliance via penetrationwill not appear as dangerous. Of course, this implies that penetrationis at best servingto supplementthe incentivesforalignmentthat already exist.
International
Security132
In addition, to inferthat penetrationplays a strongcausal role in contemporary internationalalliances is to reverse the likelycausal relationshipbetween the decision to ally and the developmentofextensivecontactsbetween
two states. As with foreign aid, a large militaryor educational training
programis one effect
of good politicalrelations,but itis rarelyan independent
cause.92
AllianceFormation| 33
alignment. In short, penetrationmay preserve or enhance an existingalliance, but it rarelycreates one by itself.
Finally, the cases where penetrationhas had a strongindependent effect
reinforcethese conclusions. Penetration will be most effectivewhen the
centralauthorityof the targetstate is extremelyweak. In such circumstances,
penetrationmay provide a foreignpower with de factocontrol.For example,
the Warsaw Pact presence in South Yemen has enabled the U.S.S.R. to
protectthe alliance by ensuring that pro-Soviet factionsremain in power.
This was demonstratedin 1978, when Cuban and East German forces supported a violent coup ousting a moderate factionled by Rubay Ali.95Similar
events seem to have occurredin Afghanistanas well. Of course, states that
are this vulnerable to outside interferenceare rarelyimportantplayers on
the internationalstage. As with foreignaid, therefore,penetrationis by itself
most useful for acquiring allies that don't really mattervery much, or in
cementingties between states whose interestsare already highlycompatible.
and theBalanceofWorldPower
Conclusion:AllianceFormation
The analysis above may be summarizedas follows.First,statesformalliances
to balance against threats rather than bandwagon with them. Threats, in
turn, are the product of several differentsources. Second, ideology is a
weaker cause of alliance formation,and ideological movements that strive
fortightcentralauthorityare more likelyto lead to conflictthan cooperation.
Third,the instrumentsof "bribery"and penetrationare by themselvesweak
determinantsof alignment;they make existingalliances more effective,but
rarelycreate them in the absence of common interests.
These propositionstell us a greatdeal about America's global position and
theoptimalstrategiesformaintainingit. To demonstratethis,I shall conclude
by consideringthe followingquestion: what explains the currentbalance of
world power between the Soviet and American alliance systems? In other
words, why have thirdparties aligned with one or the other superpower? I
make two claims. First,in contrastto the prevailingAmerican wisdom, the
presentbalance greatlyfavors the United States and its allies. Second, this
favorableimbalanceof power can be explained by the central propositions
95. See J.B. Kelly, Arabia, the Gulf,and the West (New York: Basic Books, 1980), pp. 470-473;
Laurie A. Mylroie,The SovietPresencein thePeople'sDemocratic
RepublicofYemen:InternalVulnerabilitiesand RegionalChallenges,Research Note N-2052-AF(Santa Monica, Calif.: Rand Corporation, 1983); and Middle East Contemporary
Survey,1977-78 (London: Holmes and Meier, 1979),
pp. 655-660,662-666.
International
Security134
(IM)BALANCE
OF WORLD
POWER
GNP
Size of Armed
Forces
Defense $
U.S.R. + Allies
?
Allies?PR
U.S.SR
+ Allies + PRC
U.S.
2.25
3.26
0.99
1.17
1.81
3.30
1.32
1.30
4.08
3.52
1.49
1.32
Coalitions
AllianceFormation| 35
that reflectlatent power. Significantly,the worst case for the U.S.S.R.China allied with the U.S. and India neutral-is probably the most likely
case as well. The Soviets face a ratio of more than 3:1 in population and
GNP, to say nothingof theirtechnologicaldisadvantages.97The gap is smaller
in termsof mobilizedpower, because the Soviets and theirallies have sought
to compensate fortheirrelativeweakness by devotinga largerpercentageof
GNP to defense. Despite this disproportionateeffort,the Soviet alliance
system still trails the U.S. and its allies by a considerable margin on these
items as well.
At firstglance, thisresultwould seem to contradictthe assertionthatstates
choose alliance partnersin order to balance against the strongest.Focusing
solely on aggregate power would lead us to expect more states to ally with
the Soviet Union, in orderto preventthe United Statesfromusing its superior
overall resources in harmfulways. Judgingfromthe preponderance of aggregate power favoring the West, many states appear to have "bandwagoned" rather than balanced by aligning with the U.S. This is even more
strikingwhen one remembersthat the United States was overwhelmingly
the world's most powerfulcountryin the immediatepostwar period, yetwas
able to bringmost of the otherindustrialpowers into alignmentwith rather
than against it.98
This apparent anomaly can be resolved by consideringthe centralpropositions developed above. In particular,we should recall that states balance
against threats,of which aggregatepower is only one component. By consideringthe effectsof each separate source ofthreat,as well as the less important
but still significanteffectsof ideology, a persuasive account of the current
structureof world power can be obtained. Let us consider each element in
turn.
First, not to belabor the obvious, the Soviet-American rivalryis itself
determinedby the tendency for states to balance against aggregate power.
97. Accordingto the U.S. Undersecretaryof Defense forResearch and Engineering,in 1982 the
U.S. led the U.S.S.R. in 15 out of 23 basic technologyareas, with 6 even, and with the U.S.S.R.
ahead in 2. In deployed militarysystems,U.S. technologywas superiorin 12, even in 11, and
inferiorin 6. See Statementby the Hon. Richard E. Delauer, U.S. House of Representatives,
"The FY1983 Department of Defense Program for Research, Development, and Acquisition,"
97thCongress, 2nd session (Washington,D.C.: U.S. GovernmentPrintingOffice,1982), pp. II21 and 11-22.
98. In 1950, the United States produced approximately40 percent of Gross World Product,
while the Soviet Union managed only 13.5 percent. American naval and air power were far
superior,and the U.S. had a clear advantage in deliverableatomic weaponry.
International
Security136
99. Were Canada and Mexico to ally against the U.S., theywould trailin GNP by seven to one
and in population by almost 2.5 to one.
100. This analysis stands many familiarnotions of geopolitics on their heads. For example,
Halford Mackinder suggested thatRussia gained great advantages fromits geographic position
at the center of the world "Heartland." The implicationsof alliance theoryare that while this
mayprovide some militaryadvantages, it also greatlyincreases the numberof potentialenemies
the centrallyplaced power will face. For analyses of Mackinder's ideas, see RobertE. Harkavy,
ofAccessDiplomacy(New York: PergaforOverseasBases: The Geopolitics
GreatPowerCompetition
AllianceFormation| 37
Third, the Soviet response to this situation is both predictable and selfdefeating.Faced with an encirclingcoalition of fargreateroverall resources,
the Soviets have responded by devoting considerable effortsto defense. For
this and other reasons, theirforceposture and militarydoctrineemphasize
offensivecapabilities and operations. In the event of war, the Soviet Union
aims to fightan offensivebattle of conquest ratherthan a defensive battle
on its borders, regardless of how the war actually begins. This offensive
capability,however, merely reinforcesthe cohesion of the alliance that already opposes them, because it increases the level of threat that Soviet
neighborsperceive.101
Fourth,the finalsource of threat-perceived intentions-works against the
U.S.S.R. as well. Actions like the invasion of Afghanistan,periodic intervenorganizationsand revolutionary
tionsin EasternEurope, supportforterrorist
movements abroad, all reinforceglobal opposition to the Soviet Union. Although these actions may attractthe support of radical forces around the
globe, they have also increased the already strongtendencyforthe world's
wealthiestand most stable regimes to ally togetherformutual defense.102
Significantly,
these factorshave been largelyreversed in the Third World,
which explains why the Soviets have done betterthere while the U.S. has
done worse. The Soviet capacity for global power projectionhas been and
remains distinctlyinferiorto that of the United States, and the U.S.S.R. has
generallyadopted a much more favorableattitudetowards the aspirationsof
the non-alignedmovementand revolutionaryforcesthroughoutthe world.103
mon, 1982), Chapter 6; and Paul M. Kennedy, "Mahan Versus Mackinder:Two Views on Naval
Strategy,"in his Strategyand Diplomacy:CollectedEssays(London: Allen and Unwin, 1983).
101. Accordingto the IISS, the U.S.S.R. now leads the world in totaldefense spending. On the
offensivenature of Soviet militarydoctrine,see: Benjamin Lambeth, How to ThinkAboutSoviet
MilitaryDoctrine,Rand Paper P-5939 (Santa Monica, Calif.: Rand Corporation, 1978); and Jack
L. Snyder,"Civil-MilitaryRelations and the Cult of the Offensive,1914 and 1984," International
Security,
Vol. 9, No. 1 (Summer 1984), pp. 108-146.
102. Recent examples of balancing behavior by the West include: the rapprochement
with China
in the 1970s; the modernizationof Norwegian coastal and air defenses and the pre-positioning
of equipmentfora U.S. marine battalionin Norway itself;the NATO decision and deployment
of intermediaterange nuclear missiles in Europe; the 1976 agreementforan annual 3 percent
real increase in alliance spending; and continued discussions for coordinated action in areas
outside of NATO. These responses habituallyfall short of American desires, a phenomenon
stillbest explained by the theoryof collectivegoods. On the latterpoint, see Mancur Olson and
RichardZeckhauser's classic "An Economic TheoryofAlliances," ReviewofEconomics
and Statistics
(1966); and RobertJ.Art,"FixingTransatlanticBridges,"ForeignPolicy,Number 46 (Spring 1982),
pp. 69-70.
103. See Andrew Marshall, "Sources of Soviet Power: The MilitaryPotentialin the 1980s," in
ChristophBertram,ed., Prospectsof SovietPowerin the1980s (Hamden, Conn.: Archon Books,
1980),pp. 65-66; Stephen S. Kaplan, DiplomacyofPower:SovietArmedForcesas a PoliticalInstrument
(Washington,D.C.: BrookingsInstitution,1981), Chapter 5.
International
Security138
AllianceFormation 39
Security140
International
AllianceFormation| 41
ignore the possibilityof weaning clientsaway fromMoscow throughappropriate political incentives. The U.S. is also likely to provide its own allies
with too much, in the mistaken view that this will cement theirallegiance
and enhance American control.
The arguments advanced in this paper do not mean that the American
alliance network is indestructible,that isolationism is preferable,or that
What they do mean
Western defense capabilities could not be improved.110
is that Americans could hardly ask formuch more. The principalcauses of
alliance formationwork to America's advantage, and isolate the Soviet Union
fromvirtuallyall of the world's strategicallysignificantstates. If Americans
recognize this fact, the task of formulatingappropriate national security
policies should be greatlysimplified.Even more important,the policies that
emerge will reinforce-ratherthan undermine-the considerableadvantages
the United States already enjoys.
110. For analyses of the militaryproblems facing the West along with various solutions, see:
BarryR. Posen and Stephen W. Van Evera, "Reagan AdministrationDefense Policy: Departure
from Containment," International Security,Vol. 8, No. 1 (Summer 1983), pp. 3-45; Carnegie
Endowment for InternationalPeace, Challengesfor U.S. NationialSecurity:AssessingtheBalanice:
Forces,PartII (Washington,D.C.: Carnegie Endowment, 1982);
and Conlventional
DefenseSpeniding
WilliamW. Kaufmann, "Non-nuclearDeterrence,"in JohnSteinbrunerand Leon V. Sigal, eds.,
NATO aindtheNo-First-UseQuestion(Washington,D.C.: BrookingsInstitution,
AllianceSeculrity:
Debate (Baltimore:JohnsHopkins UniversityPress,
Reformn
1984); Asa Clark et al., The Defenise
Deterrencein Europe:
Conzventional
1984); Report of the European SecurityStudy, Strengthening
Proposalsfor the 1980s (New York: St. Martin's, 1983); and Barry R. Posen, "Measuring the
European Conventional Balance: Coping with Complexityin Threat Assessment," International
Security,Vol. 9, No. 3 (Winter1984-85), pp. 47-88.
International
Security142
Appendix 1. The Balance of World Power: The Soviet Alliance Network (All Data 1982)
Population
(millions)
GNP
($ m)
# in Armed Forces
(0000s)
Defense $
($ m)
270.0
1715000
4400
257000
8.9
37451
175
3761
Czechoslovakia
15.4
147065
213
7634
East Germany
16.7
175522
233
10236
Hungary
10.7
69111
112
3108
Poland
36.2
186823
429
13494
Rumania
22.5
104827
237
4793
Afghanistan
14.2
2900
30
165
Country
U.S.S.R.
Bulgaria
Angola
7.4
7634
47
90
Cuba
9.8
21971
230
1109
Ethiopia
30.6
4635
250
455
Iraq
14.0
25177
450
11689
North Korea
3500
18.7
16200
710
Laos
3.6
320
57
30
Libya
3.3
23986
55
2800
Mongolia
1.7
1100
36
113
12.7
4786
20
163
2.7
2515
75
169
Mozambique
Nicaragua
9.4
17583
290
2528
55.8
7750
1000
1000
P.D.R. Yemen
2.0
956
23
120
Finland
4.8
51232
36
897
571.1
2624544
9108
324854
12.4
19.3
33.1
39.7
715.1
178141
1120
6223
1286.2
2802685
27.8
20.6
Syria
Vietnam
TOTAL
% of World Total
India
TOTAL WITH INDIA
% of World Total
(with India)
10228
37.0
331077
40.5
Source: U.S. Arms Control and Disarmament Agency, WorldMilitaryExpenditures and Arms
Transfers, 1972-1982 (Washington: April 1984).
AllianceFormation| 43
GNP
($ m)
232.1
9.9
Canada
China (Taiwan)
Country
United States
Belgium
Denmark
Egypt
# in Armed Forces
(1000s)
Defense
($ m)
3071400
2108
196345
101721
85
3507
24.6
276706
82
6139
18.5
50583
504
3660
5.1
61192
31
1575
44.6
29375
447
2395
France
54.4
607429
485
25612
West Germany
61.6
720485
480
24351
9.8
40209
186
2782
Italy
56.3
374162
391
9778
Japan
Greece
118.4
1226951
241
12159
South Korea
40.7
69539
600
4783
Netherlands
14.3
146305
108
4755
4.1
58521
37
1823
Spain
38.0
197175
353
4123
Turkey
48.1
65252
638
3375
United Kingdom
56.0
531606
322
27368
Norway
Thailand
49.7
38695
241
1562
Australia
15.1
180057
73
4415
Philippines
51.8
41830
155
1033
New Zealand
Brazil
Saudi Arabia
3.1
25817
13
543
128.3
295648
460
1850
10.1
16118
55
24754
Somalia
6.1
1985
54
160
Israel
3.9
22917
180
5838
Jordan
3.3
4615
65
984
Mexico
73.8
243677
145
1261
Pakistan
92.1
33598
478
2033
Portugal
10.0
24522
68
900
1283.8
8558090
9085
379863
27.7
62.8
33.0
46.5
P.R. China
1044.8
698000
4490
49500
2328.6
9256090
13575
429363
50.4
67.9
49.3
52.5
TOTAL
% of World Total
% of World Total