I Ken Berry 1998

Institutions, Strategic Restraint, and the Persistence of American Postwar Order
Author(s): G. John Ikenberry

Source: International Security, Vol. 23, No. 3 (Winter, 1998-1999), pp. 43-78
Published by: The MIT Press
Stable URL: http://www.jstor.org/stable/2539338
Accessed: 13-09-2015 14:24 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
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
This content downloaded from 132.174.254.127 on Sun, 13 Sep 2015 14:24:28 UTC
All use subject to JSTOR Terms and Conditions
Institutions,
StrategicG.John
Ikenberry
and the
Restraint,
Persistenceof
AmericanPostwar
Order
One of themostpuz-
zling aspects of world order afterthe Cold War is the persistenceof stable and
cooperative relations among the advanced industrial democracies. Despite
the collapse of bipolarity and dramatic shiftsin the global distributionof
power, America's relationswith Europe and Japan have remained what they
have been for decades: cooperative,stable, interdependent,and highlyinsti-
tutionalized. The Cold War is over, but the postwar order forged between
the United States and its allies remains alive and well fiftyyears after its
founding.
This is surprising.Many observershave expected dramatic shiftsin world
politicsaftertheCold War-such as the disappearance of Americanhegemony,
the returnof great power balancing,the rise of competingregionalblocs, and
the decay of multilateralism.Yet even withoutthe Soviet threatand Cold War
bipolarity,the United States along with Japan and WesternEurope have re-
affirmedtheir alliance partnerships,contained political conflicts,expanded
trade and investmentbetween them,and avoided a returnto strategicrivalry
and great power balancing.
The persistenceof the postwar Westernorder is particularlya puzzle for
neorealism.Neorealism has two clearlydefinedexplanationsfororder:balance
of power and hegemony.Neorealist theoriesof balance argue that order and
cohesion in the West are a resultof cooperationto balance against an external
threat,in thiscase the Soviet Union, and with the disappearance of the threat,
is Associate
G. JohnIkenberry ofPoliticalScienceat theUtniversity
Professor and in
ofPennsylvania
1998-99Visiting
Scholar
at theWoodrowWilsonCenter D.C. He isatuthor
in Washington, ofReasonsof
State:Oil Politicsand the Capacities of AmericanGovernment(Ithaca,N.Y.: CornellUniversity
Press,
1988),coauthor
ofTheState(Minneapolis: 1989),andcoeditor
ofMinnesota,
University ofNew Thinking
in International Press,1997).He is theauthorof the
Colo.:Westview
RelationsTheory(Bouilder,
book,AfterVictory:Institutions,StrategicRestraint,and the Rebuilding of Order after
forthcoming
Major Wars.
I wish to thank Daniel Deudney, James Fearon, Joseph M. Grieco, John A. Hall, Charles A.
Kupchan, David Lake, Keir Lieber,Charles Lipson, Michael Mastanduno, Nicholas Onuf, Duncan
Snidel, and PIPES seminarparticipantsat the Universityof Chicago.
International Vol.23,No. 3 (Winter

Secuirity, pp. 43-78
1998/99),
? 1998 by the Presidentand Fellows of Harvard College and the Massachusetts Instituteof Technology.
43
Security23:3 | 44
International
alliance and cooperation will decline.' Neorealist theoriesof hegemony con-

tend that order is a result of the concentrationof material power capabilities
in a single state,which uses its commanding position to create and maintain
order; with the decline of hegemonic power, order will decay.2In this view, it
is the preponderanceof Americanpower thatexplains order,with the United
States using its materialresourcesto co-opt,coerce,and induce European and
Asian countriesto cooperate.
But neorealist theories are inadequate to explain both the durability of
Westernorder and its importantfeatures,such as its extensiveinstitutionali-
zation and the consensual and reciprocalcharacterof relationswithinit. To be
sure,decades of balancing against Soviet power reinforcedcooperationamong
these countries,but the basic organizationof Westernorderpredated the Cold
War and survives today without it. Hegemonic theoryis more promisingas
an explanation,but the neorealistversion is incomplete.It misses the remark-
ably liberal characterof American hegemony and the importanceof interna-
tional institutions in facilitating cooperation and overcoming fears of
domination or exploitation.It was the exercise of strategicrestraint-made
good by an open polity and binding institutions-more than the direct and
instrumentalexerciseof hegemonicdominationthatensured a cooperativeand
stable postwar order. For all these reasons, it is necessary to look beyond
neorealism for an understanding of order among the advanced industrial
societies.
To understand the continued durabilityand cohesion of the advanced in-
dustrialworld, we need to turnneorealisttheoriesof order on theirhead. It is
actually the ability of the Westerndemocracies to overcome or dampen the
underlyingmanifestationsof anarchy(orderbased on balance) and domination
(orderbased on coercivehegemony)thatexplains thecharacterand persistence
of Westernorder.Neorealism misses the institutionalfoundationsof Western
political order-a logic of order in which the connectingand constraining
effectsof institutionsand democraticpolities reduce the incentivesof Western
states to engage in strategicrivalryor balance against American hegemony.
1. See KennethN. Waltz, TheoryofInternational Politics(Reading, Mass.: 1979). For extensionsand

debates, see Robert0. Keohane, ed., Neorealismand Its Critics(New York: Columbia University
Press, 1986). For an extensionof thisbasic positionto the West,see JohnJ.Mearsheimer,"Why We
Will Soon Miss the Cold War,"AtlanticMonthly,August 1990,pp. 35-50.
2. See Robert Gilpin, Warand Change in WorldPolitics(New York: Cambridge UniversityPress,
1981). See also Stephen D. Krasner,"American Policy and Global Economic Stability,"in William
P. Avery and David P. Rapkin, eds., Americain a ChangingWorldPoliticalEconomy(New York:
Longman,1982); and see the debate in David Rapkin,ed., WorldLeadership and Hegemony(Boulder,
Colo.: Lynne Rienner,1990).
Persistence
ofPostwarOrder| 45
The Westernstates are not held togetherbecause of external threatsor the

simple concentrationof power. Rather,Westernorderhas what mightbe called
"constitutionalcharacteristics"-a structureof institutionsand open polities
that constrainpower and facilitate"voice opportunities,"therebymitigating
the implicationsof power asymmetriesand reducing the opportunitiesof the
leading state to exit or dominate.
This is one of the greatpuzzles thathas eluded fullexplanation:Why would
the United States, at the height of its hegemonic power afterWorld War II,
agree to limitthatpower? It is equally a puzzle why weak and secondarystates
mightagree to become more ratherthan less entangled with such a potential
hegemon. The answer has to do with the incentivesthatpowerfulstateshave
to create a legitimate political order, and it also has to do with the way
institutions-when wielded by democracies-allow powerful states to both
lock in a favorable postwar order and overcome fears of domination and
abandonmentthatstand in the way of postwar agreementand the creationof
legitimateorder. Weak and secondary states get institutionalizedassurances
that they will not be exploited. Paradoxically,a leading or hegemonic state,
seeking to hold on to its power and make it last, has incentivesto findways
to limitthatpower and make it acceptable to otherstates.
This institutionalexplanation of Westernorder is developed in four steps.
First,I argue thatthe basic logic of order among the Westernstates was set in
place during and immediatelyafterWorld War II, and it was a logic that
addressed the basic problem of how to build a durable and mutually accept-
able orderamong a group of stateswith huge power asymmetries.To gain the
cooperation and compliance of secondary states, the United States had to
engage in strategicrestraint-toreassure weaker states thatit would not aban-
don or dominate them.Cooperative order is built around a basic bargain: the
hegemonic state obtains commitmentsfrom secondary states to participate
within the postwar order, and in returnthe hegemon places limits on the
exercise of its power. The weaker states do not fear domination or abandon-
ment-reducing the incentivesto balance-and the leading statedoes not need
to use its power assets to enforceorder and compliance.
Second, strategicrestraintis possible because of the potentialbindingeffects
of internationalinstitutions.Internationalinstitutionsdo not simply serve the
functionalpurposes of states,reducingtransactioncosts and solving collective
actionproblems,but theycan also be "sticky"-locking statesintoongoingand
predictablecourses of action. It is this lock-ineffectof institutionsthatallows
them to play a role in restrainingthe exerciseof state power. In effect,institu-
tionscreateconstraintson stateaction thatserve to reducethereturnstopower-
Security23:3 | 46
International
thatis, theyreduce the long-termimplicationsof asymmetriesof power. This

is preciselywhat constitutionsdo in domestic political orders. Limits are set
on what actors can do with momentaryadvantage. Losers realize that their
losses are limited and temporary-to accept those losses is not to risk every-
thingnor will it give the winners a permanentadvantage. Political orders-
domestic and international-differwidely in their "returns to power." The
Westernpostwar order is so stable and mutually acceptable because it has
found institutionalways to reduce the returnsto power.
Third, the Western postwar order has also been rendered acceptable to
Europe and Japan because American hegemony is built around decidedly
liberal features.The penetrated character of American hegemony,creating
transparencyand allowing access by secondary states, along with the con-
strainingeffectsof economic and securityinstitutions,has provided mecha-
nisms to increaseconfidencethatthe participatingstateswould remainwithin
the order and operate according to its rules and institutions.American hege-
mony has been renderedmore benign and acceptable because of its open and
accessible internalinstitutions.
Fourth,the Westernorder has actually become more stable over time be-
cause the rules and institutionshave become more firmlyembedded in the
wider structuresof politicsand society.This is an argumentabout the increasing
returnsto institutions,
in this case Westernsecurityand economic institutions.
Over the decades, the core institutionsof Westernorder have sunk theirroots
ever more deeply into the political and economic structuresof the states that
participate within the order. The result is that it is becoming increasingly
difficultfor "alternativeinstitutions"or "alternativeleadership" to seriously
emerge.Westernorderhas become institutionalizedand path dependent-that
is, more and more people will have to disrupt theirlives if the order is to
radicallychange. This makes wholesale change less likely.
Overall, the durabilityof Westernorderis built on two core logics. First,the
constitution-like characterof the institutionsand practicesof the order serve
to reduce the returnsto power, which lowers the risks of participationby
strongand weak states alike. This in turn makes a resort to balancing and
relativegains competitionless necessary.Second, the institutionsalso exhibit
an "increasing returns" character,which makes it increasinglydifficultfor
would-be orders and would-be hegemonic leaders to compete against and
replace the existingorder and leader. Although the Cold War reinforcedthis
order,it was not triggeredby it or ultimatelydependent on the Cold War for
its functioningand stability.
ofPostwarOrder| 47
Persistence
The implicationof this analysis is that the West is a relativelystable and

expansive political order. This is not only because the United States is an
unmatched economic and militarypower, but also because it is uniquely
capable of engagingin "strategicrestraint,"reassuringpartnersand facilitating
cooperation. Because of its distinctivelyopen domestic political system,and
because of the arrayof power-dampeninginstitutionsit has createdto manage
political conflict,the United States has been able to remain at the centerof a
large and expanding institutionalizedand legitimatepoliticalorder.Its capac-
ityto win in specificstruggleswith otherswithinthe systemmay rise and fall,
and the distributionof power can continue to evolve in America's favor or
against it,but the largerWesternorderremainsin place with littleprospectof
decline.
In the next section,I sketch the theoreticaldebate about American power
and post-WorldWar II order and the contendingpositions of neorealismand
liberalism.Following this,I develop an institutionaltheoryof orderformation
and the "constitutionalbargain" thatlies at the heart of the Westernpostwar
political order. I then examine America's postwar strategyof hegemonic
reassurance and the major institutionalcharacteristicsof this order that, in
effect,"doomed" the United States to succeed in overcomingfearsof domina-
tion and abandonment.Finally,I develop the argumentthatthe postwar order
has experiencedan increasingreturnsto institutions,which in turnreinforces
the underlyingstabilityof relationsbetween the United States and the other
advanced industrialcountries.
TheDebateaboutOrder
The debate over American grand strategyafterthe Cold War hinges on as-
sumptions about the sources and characterof Westernorder.Neorealism ad-
vances two clearly defined answers to the basic question of how order is
created among states: balance of power and hegemony.Both are ultimately
pessimisticabout the futurestabilityand coherence of economic and security
relationsbetween the United States,Europe, and Japan.3
3. For an extended critiqueof neorealistexpectationsof post-Cold War order,see Daniel Deudney

and G. JohnIkenberry, "Realism, StructuralLiberalism,and the WesternOrder,"in Ethan Kapstein
and Michael Mastanduno, eds., UnipolarPolitics:Realismand StateStrategies
aftertheCold War(New
York: Columbia UniversityPress, forthcoming).For a sophisticatedsurvey of realistand liberal
theoriesof internationalorder,see JohnA. Hall, International
Order(Cambridge: PolityPress,1996),
chap. 1.
International
Security23:3 | 48
Balance-of-powertheoryexplains order-and the rules and institutionsthat

emerge-as the result of balancing to counterexternalor hegemonic power.4
Order is the product of the unending process of balancing and adjustment
among states under conditions of anarchy.Balancing can be pursued both
internallyand externally-through domestic mobilization and through the
formationof temporaryalliances among threatenedstates to resistand coun-
terbalancea looming or threateningconcentrationof power. Under conditions
of anarchy,alliances will come and go as temporaryexpedients, states will
guard theirautonomy,and entanglinginstitutionswill be resisted.5
A second neorealist theory,hegemonic stabilitytheory,holds that order is
created and maintained by a hegemonic state,which uses power capabilities
to organize relationsamong states.The preponderanceof power held by a state
allows it to offerincentives,both positive and negative,to the otherstates to
agree to participationwithinthehegemonicorder.Accordingto RobertGilpin,
an internationalorder is, at any particularmomentin history,the reflectionof
the underlyingdistributionof power of states withinthe system.6Over time,
that distributionof power shifts,leading to conflictsand ruptures in the
system,hegemonic war, and the eventual reorganizationof order so as to
reflectthe new distributionof power capabilities. It is the rising hegemonic
state or group of states,whose power position has been ratifiedby war, that
defines the terms of the postwar settlement-and the characterof the new
order.
The continuityand stabilityof the Westernpostwar order is a puzzle for
both varietiesof neorealism.With the end of the Cold War,balance-of-power
theoryexpects theWest,and particularlythesecurityorganizationssuch as the
North Atlantic TreatyOrganization (NATO) and the U.S.-Japan alliance, to
weaken and eventually returnto a pattern of strategicrivalry.The "semi-
sovereign" securityposture of Germanyand Japan will end, and these coun-
4. Waltz, TheoryofInternational Politics.On the balance-of-threattheory,see Stephen M. Walt,The

OriginsofAlliances(Ithaca, N.Y.: Cornell UniversityPress, 1987).
5. Three assumptionslie behind the neorealistclaim thatbalance of power is the only real solution
to the problem of order.First,all states seek security,but because otherstates can always become
threats,that security is never absolute. Second, the intentionsof other states are inherently
uncertain.A state can never be absolutely certainthat its currentallies will remain allies in the
futureand not turninto adversaries. This is because all states have the capacity to threatenand
theirintentionsare unknowable. Finally,relative capabilities are more importantthan absolute
capabilities to ensure the securityof states because securityis derived fromthe relativestrength
of a state in relationto competingstates.
6. Gilpin, Warand Changein WorldPolitics.
ofPostwarOrder| 49
Persistence
trieswill eventuallyrevertback to traditionalgreatpowers.7The Soviet threat

also served to dampen and contain economic conflictwithin the West-and
afterthe Cold War, economic competitionand conflictamong the advanced
industrialsocieties is expected to rise.8Neorealist theoriesof hegemony also
expect that the gradual decline of American power-magnified by the Cold
War-should also lead to rising conflictand institutionaldisarray.9More re-
cently,some realistshave argued thatit is actuallythe extremepreponderance
of American power, and not its decline, that will triggercounterbalancing
reactionsby Asian and European allies.10
The basic thrustof these neorealist theories is that relations among the
Western states will returnto the problems of anarchy afterthe Cold War:
economy rivalry,securitydilemmas, institutionaldecay, and balancing alli-
ances. The fact that post-Cold War relations among the Westernindustrial
countrieshave remained stable and open, and economic interdependenceand
institutionalizedcooperationhave actuallyexpanded in some areas, is a puzzle
7. KennethN. Waltz,"The EmergingStructureof InternationalPolitics,"International Security,Vol.

18, No. 2 (Fall 1993), pp. 44-79; JohnJ. Mearsheimer,"Back to the Future: Instabilityin Europe
afterthe Cold War,"International Security,Vol. 15, No. 1 (Fall 1990), pp. 5-56; Mearsheimer,"Why
We Will Soon Miss the Cold War"; Conor Cruise O'Brien, "The Future of the West," National
Interest,Vol. 30 (Winter1992/93),pp. 3-10; and BradleyA. Thayer,"Realism,the Balance of Power,
and Stabilityin the Twenty-first Century,"unpublished paper, Harvard University,1997.
8. As RobertLieber argues: "In the past, the Soviet threatand the existenceof an American-led
bloc to contain it placed limitson the degree of frictionthat could develop among the Western
partnersand Japan.The need to cooperate in the face of the Soviets meant thatdisagreementsin
economic and otherrealms were preventedfromescalatingbeyond a certainpoint because of the
perceptionthattoo bitteran intra-alliedconfrontation could only benefittheiradversaryand thus
weaken common security."Lieber,"Eagle without a Cause: Making Foreign Policy without the
Soviet Threat,"in Lieber,ed., Eagle Adrift:AmericanForeignPolicyat theEnd of theCentury(New
York:Longman, 1997), p. 10. This point is also made by KennethOye in "Beyond Postwar Order
and the New World Order," in Oye, Lieber, and Donald Rothchild,eds., Eagle in a New World:
AmericanGrandStrategyin thePost-Cold WarEra (New York:HarperCollins,1992), pp. 3-33.
9. For contrastingviews, see RobertGilpin, The PoliticalEconomyofInternational Relations(Prince-
ton,N.J.:PrincetonUniversityPress, 1987); Paul M. Kennedy,The Rise and Fall oftheGreatPowers
(New York:Random House, 1987);JosephS. Nye,Jr.,BoundtoLead:TheChangingNatureofAmerican
Power(New York:Basic Books, 1990); Henry Nau, TheMythofAmerica'sDecline:LeadingtheWorld
Economyin the1990's (New York:OxfordUniversityPress, 1990); and Susan Strange,"The Persist-
ent Mythof Lost Hegemony,"International Organization, Vol. 41, No. 4 (Autumn 1987),pp. 551-574.
10. See, for example, ChristopherLayne, "The Unipolar Illusion: Why New Great Powers Will
Arise," International Vol. 17, No. 4 (Spring 1993), pp. 5-51; Layne, "From Preponderance
Security,
to OffshoreBalancing: America's Future Grand Strategy,"InternationalSecurity,Vol. 22, No. 1
(Summer 1997), pp. 86-124; and JosefJoffe, "'Bismarck'or 'Britain'?Toward an American Grand
StrategyafterBipolarity,"InternationalSecurity,Vol. 19, No. 4 (Spring 1995), pp. 94-117. For a
discussion of these views, see Michael Mastanduno, "Preservingthe Unipolar Moment: Realist
Theories and U.S. Grand Strategyafterthe Cold War,"International Security,Vol. 21, No. 4 (Spring
1997), pp. 49-88.
International
Security23:3 | 50
thatneorealismis hard-pressedto explain. Despite sharp shiftsin the distribu-

tion of power withinthe West,the politicalorderamong the industrialdemoc-
racies has remained quite stable. Highly asymmetricalrelationsbetween the
United States and the other advanced industrialcountries-in the 1940s and
again today-or declines in thoseasymmetries-in the 1980s-have not altered
the basic stabilityand cohesion in relationsamong these countries.
Liberal theoriesprovide some promisingleads in explainingfeaturesof the
postwar Westernorder,but theytoo are incomplete.11Many of these theories
would also predictorderand stabilityin the West-but theircausal arguments
are too narrow.The key focusof liberalinstitutionaltheoryis theway in which
institutionsprovide informationto states and reduce the incentivesforcheat-
ing.12 But this misses the fundamentalfeatureof order among the advanced
industrialcountries:the structuresof relationsare now so deep and pervasive
thatthe kind of cheatingthatthese theoriesworryabout eithercannothappen,
or if it does it will not really matterbecause cooperation and the institutions
are not fragilebut profoundlyrobust.The basic problem is thatthese institu-
tionalistargumentshave not incorporatedthe structuralfeaturesof Western
order in their explanations. In particular,they miss the problems of order
associated with the great asymmetriesof power between Westernstates,the
path-dependentcharacterof postwar institutions,and the importanceof the
open and accessible characterof Americanhegemony.
In general terms, liberal theories see institutionsas having a variety of
internationalfunctionsand impacts-serving in various ways to facilitateco-
operation and alter the ways in which states identifyand pursue theirinter-
11. No single theoristrepresentsthis composite liberal orientation,but a variety of theorists

provide aspects. On the democraticpeace, see Michael Doyle, "Kant, Liberal Legacies, and Foreign
Affairs,"Philosophyand Public Affairs, Vol. 12, No. 3 (Summer 1983), pp. 205-235, 323-353. On
securitycommunities,see Karl Deutsch, PoliticalCommunity and theNorthAtlanticArea (Princeton,
N.J.: PrincetonUniversityPress, 1957). On the interrelationshipof domestic and international
politics, see James N. Rosenau, ed., LinkagePolitics:Essays on the Convergenceof National and
International Systems(New York:Free Press,1969). On functionalintegrationtheory,see Emst Haas,
BeyondtheNation-State: Functionalism and InternationalOrganization(Stanford,Calif.:StanfordUni-
versityPress, 1964). On the fragmentedand complex nature of power and interdependence,see
RobertKeohane and JosephNye, Powerand Interdependence (Boston: Little,Brown,1977). On the
modemization theoryunderpinningsof the liberaltradition,see Edward Morse, Modernization and
the Transformation of InternationalRelations(New York: Free Press, 1976); and James Rosenau,
Turbulence in WorldPolitics:A TheoryofChangeand Continuity (Princeton,N.J.:PrincetonUniversity
Press, 1991).
12. Robert 0. Keohane, AfterHegemony:Cooperationand Discord in the WorldPoliticalEconomy
(Princeton,N.J.:PrincetonUniversityPress,1984); and Lisa Martin,CoerciveCooperation: Explaining
MultilateralEconomicSanctions(Princeton,N.J.:PrincetonUniversityPress, 1992).
PersistenceofPostwarOrder| 51
ests.13 Liberal theories have also identifiedand stressed the importance of

institutionsamong states that serve as foundationalagreementsor constitu-
tional contracts-what Oran Young describes as "sets of rightsand rules that
are expected to govern theirsubsequent interactions." 14 Liberal institutional
theoriesare helpful in explaining why specificinstitutionsin the West may
persist-even after the power and interests that established them have
changed."5But therehas been less attentionto the ways thatinstitutions can be
used as strategiesto mitigatethe securitydilemma and overcome incentives
to balance. Liberal theoriesgrasp the ways in which institutionscan channel
and constrainstate actions, but they have not explored a more far-reaching
view in which leading statesuse intergovernmental institutionsto restrainthe
arbitraryexerciseof power and dampen the fearsof dominationand abandon-
ment.
The approach to institutionsI am proposing can be contrastedwith two
alternativetheories-the rationalist(or "unsticky")theoryof institutions,and
the constructivist(or "disembodied") theoryof institutions.16Rationalistthe-
ory sees institutionsas agreementsor contractsbetween actors that function
to reduce uncertainty,lower transactionscosts, and solve collective action
problems. Institutionsprovide information,enforcementmechanisms, and
otherdevices thatallow states to realize joint gains. Institutionsare explained
in termsof the problems theysolve-they are constructsthatcan be traced to
13. For a recentsurvey,see Lisa Martin,"An Institutionalist View: InternationalInstitutionsand

State Strategies,"in T.V. Paul and JohnA. Hall, eds., International Orderand theFutureof World
Politics (New York: Cambridge UniversityPress, 1999). The liberal literatureon international
institutionsand regimes is large. See Stephen D. Krasner,ed., International Regimes(Ithaca, N.Y.:
Cornell UniversityPress, 1981); Stephan Haggard and Beth Simmons,"Theories of International
Regimes,"International Organization,Vol. 41, No. 3 (Summer1987),pp. 491-517; and VolkerRittber-
ger,ed., RegimeTheonyand International Relations(Oxford:OxfordUniversityPress, 1995).
14. Oran Young,"Political Leadership and Regime Formation:On the Development of Institutions
in InternationalSociety,"InternationalOrganization, Vol. 45, No. 3 (Summer 1991), p. 282. See also
Young, InternationalCooperation: BuildingRegimesforNaturalResourcesand theEnvironment (Ithaca,
N.Y.: Cornell UniversityPress, 1989). The concept of constitutionalcontractis discussed in James
M. Buchanan, The LimitsofLiberty(Chicago: Universityof Chicago Press, 1975), esp. chap. 5.
15. See Robert B. McCalla, "NATO's Persistenceafterthe Cold War," Internationial Organization,
Vol. 50, No. 3 (Summer 1996), pp. 445-476; Celeste Wallander and Robert 0. Keohane, "When
ThreatsDecline, Why Do Alliances Persist?An InstitutionalApproach," unpublishedms., Harvard
Universityand Duke University,1997; and JohnS. Duffield,"InternationalRegimes and Alliance
Behavior: ExplainingNATO Conventional Force Levels," International Vol. 46, No. 4
Organization,
(Autumn 1992), pp. 819-855.
16. For a good survey of these two alternativetraditionsof institutionaltheory,see Robert 0.
Keohane, "InternationalInstitutions:Two Perspectives,"International StudiesQuarterly,
Vol. 32, No.
4 (December 1988), pp. 379-396.
Security23:3 | 52
International
individuals or groups.17 Constructivisttheorysees

the actionsof self-interested
institutionsas diffuseand socially constructedworldviews that bound and
shape the strategicbehavior of individuals and states. Institutionsare over-
archingpatternsofrelationsthatdefineand reproducetheinterestsand actions
of individuals and groups. Institutionsprovide normativeand cognitivemaps
forinterpretation and action,and theyultimatelyaffecttheidentitiesand social
purposes of the actors.18 A thirdposition advanced here sees institutionsas
both constructsand constraints.19 Institutionsare the formal and informal
organizations,rules, routines,and practices that are embedded in the wider
political order and define the "landscape" in which actors operate. As such,
institutionalstructuresinfluencethe way power is distributedacross individu-
als and groups withina politicalsystem-providing advantages and resources
to some and constrainingthe options of others.20
17. The rationalistapproach argues thatinstitutionsare essentiallyfunctionalor utilitarian"solu-

tions" to problemsencounteredby rationalactors seeking to organize theirenvironmentin ways
that advance their interests.Kenneth A. Shepsle describes institutionsas "agreements about a
structureof cooperation" that reduce transactioncosts, opportunism,and other formsof "slip-
page." Shepsle, "InstitutionalEquilibriumand Equilibrium Institutions,"in HerbertF. Weisberg,
ed., PoliticalScience:The ScienceofPolitics(New York:Agathon,1986), p. 74.
18. As Alexander Wendtargues, "Constructivistsare interestedin the constructionof identityand
interestsand, as such, take a more sociological than economic approach" to theory.Wendt,
"Collective IdentityFormationand the InternationalState," AmericanPoliticalScienceReview,Vol.
88, No. 2 (June1994), pp. 384-385.
19. This "thirdview" of institutionsis dissatisfiedwith both rationalistand constructivist theory.
Rationalisttheoryproves too "thin"-it provides too much agency and not enough structure.It is
silenton what otherssee as the heartof how institutionsmatter,namely,the impactof institutional
structureson the interestsand goals of individuals and groups. Constructivisttheory,on the other
hand, is too "thick"-it does not allow enough agency,and it has problemsexplaininginstitutional
change. It rendersuniversal and deterministicwhat is really only a contingentoutcome, namely,
the impact of institutionson the way people conceive theiridentitiesand interests.
20. This theoreticalview-often called "historicalinstitutionalism"-makesseveral claims. First,
state policy and orientationsare mediated in decisive ways by political structures-such as
institutionalconfigurations of government.The structuresof a polityshape and constrainthegoals,
opportunities,and actionsof thegroups and individuals operatingwithinit.Second, to understand
how these institutionalconstraintsand opportunitiesare manifest,theymust be placed withina
historicalprocess-timing, sequencing, unintended consequences, and policy feedback matter.
Third, institutionshave path-dependentcharacteristics-institutions are established and tend to
persistuntil a later shock or upheaval introducesa new momentof opportunityforinstitutional
change. Finally,institutionalstructureshave an impact because theyfacilitateor limitthe actions
of groups and individuals-which means thatinstitutionsare never offeredas a complete expla-
nationof outcomes.The impactsof institutions, therefore, tend to be assessed as theyinteractwith
other factors,such as societal interests,culture,ideology,and new policy ideas. For surveys of
historicalinstitutionaltheory,see Peter A. Hall and RosemaryC.R. Taylor,"Political Science and
the Three New Institutionalisms," PoliticalStudies,Vol. 44, No. 5 (April 1996); and Sven Steinmo
et al., StructuringPolitics:HistoricalInstitutionalism
in Comparative Analysis(New York:Cambridge
UniversityPress, 1992).
This approach to institutionsgives attentionto the ways in which institu-

tions alter or fixinto place the distributionof power withina political order.
It offersa more "sticky" theoryof institutionsthan the rationalistaccount,
but unlike constructivism,institutionalstickinessis manifestin the practical
interactionbetween actors and formaland informalorganizations,rules, and
routines.Because of the complex causal interactionbetween actors and insti-
tutions,attentionto historicaltimingand sequencing is necessaryto appreciate
the way in which agency and structurematter.
Debates over post-Cold War order hinge on claims about the characterof
relations among the major industrial democracies. Neorealist theories trace
order to the operation of the balance of power or hegemony,and theyantici-
pate rising conflictand strategyrivalrywithin the West. Liberal theoriesare
more inclined to see continuityand inertiain the institutionsand relationsof
postwar order,even if threatsdisappear and power balances shiftsharply.It
is the sharply contrastingview of institutional"stickiness,"as RobertPowell
argues, thatdifferentiates realistand liberal institutionaltheories.21The argu-
mentadvanced here is thatinstitutionsare potentiallyeven more "sticky"than
liberal theories allow, capable under specificcircumstancesof locking states
into stable and continuous relationsthat place some limitson the exercise of
state power, therebymitigatingthe insecuritiesthatneorealismtracesto anar-
chy and shiftingpower balances.
TheConstitutional
Bargain
The most fundamentalstrategicrealityafterWorld War II was the huge dis-

parity of power among the great powers that had foughtthe war-and, in
particular,the commanding hegemonic position of the United States. British
scholar Harold Laski, writingin 1947, captured the overarchingcharacterof
American power: "Today literallyhundreds of millions of Europeans and
Asiaticsknow thatboth the quality and the rhythmof theirlives depend upon
decisions made in Washington.On the wisdom of those decisions hangs the
21. Robert Powell, "Anarchy in InternationalRelations Theory: The Neorealist and Neoliberal
Debate," InternationalOrganization,Vol. 48, No. 2 (Spring 1994), pp. 313-344. For a skepticalrealist
view of institutions,see JohnJ. Mearsheimer,"The False Promise of InternationalInstitutions,"
InternationalSecurity,Vol. 19, No. 3 (Winter1994/95), pp. 5-49. For a less skeptical view, see
Randall L. Schweller and David Priess, "A Tale of Two Realisms: Expanding the Institutions
Debate," MershonInternational StudiesReview,Vol. 41, Supplement 1 (May 1997), pp. 1-32.
International
Security23:3 | 54
fate of the next generation."22George Kennan, in a major State Department

review of Americanforeignpolicyin 1948,pointed to thisnew strategicreality:
"We have about 50% of the world's wealth but only 6.3% of its population....
Our real task in the comingperiod is to devise a patternof relationshipswhich
will permitus to maintainthispositionof disparitywithoutpositivedetriment
to our national security."23
It is the examinationof the choices and options that
the United States and the major European and Asian statesfaced afterthe war
thatallows us to see the underlyinglogic of postwar order.
In a commanding postwar position, the United States had three major
options. It could dominate-use its power to prevail in the endless distributive
struggleswith other states. It could abandon-wash its hands of Europe and
Asia and returnhome. Or it could seek to convertits favorablepostwar power
positioninto a durable orderthatcommanded the allegiance of theotherstates
within it. A legitimatepolitical order is one in which its members willingly
participateand agree with the overall orientationof the system.24To achieve
a legitimateordermeans to secure agreementamong the relevantstateson the
basic rules and principlesof political order.States abide by the order's rules
and principlesbecause theyaccept them as theirown.
To gain the willing participationof other states, the United States had to
overcome theirfearsthatAmerica mightpursue its otheroptions: domination
or abandonment.In this situation,the criticalelementto order formationwas
the abilityof the United States to engage in "strategicrestraint"-to convey to
its potential partnerscredible assurances of its commitmentsto restrainits
power and operate within the agreed-upon rules and principles of postwar
order. In the absence of these assurances, the weaker states of Europe and
Japanwould have serious incentivesto resistAmericanhegemonyand engage
in strategicrivalryand perhaps counterbalancingalliances.
A durable and legitimatepostwar order was possible preciselybecause the
United States had the ability to engage in strategicrestraint.In effect,the
United States agreed to move toward an institutionalizedand agreed-upon
politicalprocess and to limitits power-made credibleby "sticky"institutions
and open polities-in exchange forthe acquiescence and compliantparticipa-
22. Quoted in Norman Graebner,Americaas a WorldPower:A RealistAppraisalfromWilsontoReagan

(Wilmington,Del.: ScholarlyResources, 1984), p. 275.
23. George Kennan,Memorandum by the Directorof the Policy PlanningStaffto the Secretaryof
State and the Under Secretaryof State Lovett,ForeignRelationsof the UnitedStates,1948, Vol. 1
(Washington,D.C.: GovernmentPrintingOffice),February24, 1948, p. 524.
24. See David Beetham,The Legitimation ofPower(London: Macmillan, 1991).
tion of secondary states. At the heart of the Western postwar order is an

ongoing trade-off:the United States agrees to operate withinan institutional-
ized political process and, in return,its partnersagree to be willing partici-
pants.25
More specifically,the United States had an incentive to move toward a
"constitutional"settlementafterthe war-that is, to create basic institutions
and operating principles that limit what the leading state can do with its
power.26In effect,constitutionalagreementsreduce the implicationsof "win-
ning" in internationalrelationsor, to put it more directly,theyserve to reduce
thereturnstopower.This is fundamentallywhat constitutionsdo withindomes-
tic orders.They set limitson what a state thatgains disproportionatelywithin
the order can do with those gains, therebyreducing the stakes of uneven
gains.27This means that constitutionsreduce the possibilitiesthat a state can
turn short-termgains into a long-termpower advantage.28Taken together,
constitutionalagreementsset limits on what actors can do with momentary
advantages. To lose is not to lose everything-the stakes are limited; to lose
today does not necessarilydiminishthe possibilityof winning tomorrow.29
The role of constitutionallimits on power can be seen within domestic
constitutionalpolities. When a party or leader wins an election and takes
controlof the government,thereare fundamentaland strictlydefined limits
on the scope of the power thatcan be exercised.A newly elected leader cannot
use the militaryto oppress or punish his rivals, or use the taxing and law-
enforcementpowers of governmentto harm or destroythe opposition party
or parties.The constitutionsets limitson the use of power-and thisserves to
reduce the implicationsof winning and losing withinthe political system.To
lose is not to lose all, and winningis at best a temporaryadvantage. As a result,
both parties can agree to stay withinthe systemand play by the rules.
25. Anotherrecentcase of strategicrestraintis Germanywithinthe European Union: here also is

a powerful state that has agreed to operate within an institutionalizedpolitical process, and in
return,its partnersagree to be willing participants.Neorealists missed this development, and
expected Germanyto be a predatorypower afterthe Cold War.
26. See Alec Stone, "What Is a Supranational Constitution?An Essay in InternationalRelations
Theory,"ReviewofPolitics,Vol. 56, No. 3 (Summer 1994), pp. 441-474.
27. See discussion in Jon Elster and Rune Slagstad, eds., Constitutionalisin
and Democracy(New
York:Cambridge UniversityPress, 1988).
28. See Adam Przeworski,Democracyand the Market(New York: Cambridge UniversityPress,
1991), p. 36.
29. For a discussion of how formalinstitutionsprovide ways forpowerfulstatesto convey credible
restraintto weaker states,see Kenneth W. Abbottand Duncan Snidal, "Why States Act through
Formal InternationalOrganizations,"JournalofConflictResolution,Vol. 42, No. 1 (February1998),
pp. 3-32.
Security23:3 | 56
International
Limits on power are never as clear-cut,absolute, or guaranteed in relations

between states. The underlyingstructure,even in highly complex and inte-
grated orders,is still anarchic. But where institutionscan be established that
provide some measure of mutuallybinding constraintson states,and where
the polities of the participatingstates are open democracies, the conditions
exist fora settlementwith constitutionalcharacteristics.
STRATEGIC RESTRAINT AND POWER CONSERVATION

Why would a newly hegemonicstatewant to restrictitselfby agreeingto limits
on the use of hegemonic power? The basic answer is that a constitutional
settlementconserveshegemonicpower,fortwo reasons. First,ifthehegemonic
state calculates that its overwhelming postwar power advantages are only
momentary,an institutionalizedordermight"lock in" favorablearrangements
that continue beyond the zenith of its power. In effect,the creation of basic
orderinginstitutionsare a formof hegemonic investmentin the future.The
hegemonic state gives up some freedomon the use of its power in exchange
fora durable and predictableorder thatsafeguardsits interestsin the future.
This investmentmotive rests on several assumptions. The hegemonic state
must be convinced that its power position will ultimatelydecline-that it is
currentlyexperiencinga momentarywindfallin relativepower capabilities.If
this is the state's strategicsituation,it should want to use its momentary
positionto get what it wants. On theotherhand, ifthenew hegemon calculates
that its power position will remain preponderantinto the foreseeablefuture,
the incentiveto conserve its power will disappear. Also, the hegemon must be
convinced that the institutionsit creates will persist beyond its own power
capabilities-that is, it must calculate that these institutionshave some inde-
pendent orderingcapacity.30If institutionssimply are reflectionsof the distri-
bution of power, the appeal of an institutionalsettlementwill obviously
decline. But if institutionsare potentially"sticky,"powerful states that are
farsightedenough to anticipatetheirrelativedecline can attemptto institution-
30. The argument that internationalregimes and institutions,once created, can have an inde-
pendent orderingimpact on states comes in several versions. The weak version of this claim is
the modifiedstructuralrealistpositionthatsees lags in the shiftsof regimesas power and interests
change. See Stephen D. Krasner, "Structural Causes and Regime Consequences: Regimes as
Regimes(Ithaca,N.Y.: Cornell UniversityPress,
InterveningVariables,"in Krasner,ed., International
1983), pp. 1-12. The strongerversion entails assumptions about path dependency and increasing
returns.For a survey,see WalterW. Powell and Paul J.DiMaggio, eds., TheNew Institutionalism in
OrganizationalAnalysis(Chicago: Universityof Chicago Press, 1991), Introduction,pp. 1-38.
alize favorable patternsof cooperation with other states that persist even as
power balances shift.
The second reason why a hegemon mightwant to reach agreementon basic
institutions,even ifit means giving up some autonomyand short-term advan-
tage, is that it can reduce the "enforcementcosts" of maintainingorder.The
constantuse of power capabilities to punish and reward secondary statesand
resolve conflictsis costly.It is far more effectiveover the long termto shape
the interestsand orientationsof other states ratherthan directlyshape their
actions throughcoercion and inducements.31 A constitutionalsettlementre-
duces the necessityof the costlyexpenditureof resourcesby the leading state
on bargaining,monitoring,and enforcement.
It remains a question why weaker states might not just resistany institu-
tional settlementafter the war and wait until they are strongerand can
negotiatea more favorable settlement.Several factorsmightmake this a less
attractiveoption. First,without an institutionalagreement,the weaker states
will lose more than theywould under a settlement,where the hegemonicstate
agrees to forgosome immediategains in exchange forwilling participationof
secondarystates.Withoutan institutionalsettlement,bargainingwill be based
simply on power capacities, and the hegemonic state will have the clear
advantage. The option of losing more now to gain more later is not attractive
fora weak statethatis strugglingto rebuildafterwar. Its choices will be biased
in favorof gains today ratherthan gains tomorrow.The hegemon,on the other
hand, will be more willing to trade offgains today for gains tomorrow.The
differencein the two time horizons is crucial to understandingwhy a consti-
tutionalsettlementis possible.
A second reason why weaker statesmightopt fortheinstitutionalagreement
is that-if the hegemon is able to crediblydemonstratestrategicrestraint-it
does buy them some protectionagainst the threatof dominationor abandon-
ment. As realist theorywould note, a central concern of weak or secondary
states is whether they will be dominated by the more powerful state. In an
internationalorder that has credible restraintson power, the possibilityof
indiscriminateand ruthless domination is mitigated.Justas important,the
possibilityof abandonmentis also lessened. If the hegemonicstateis rendered
31. The argumentthathegemons will want to promotenormativeconsensus among states so as

to reduce the necessityof coercive management of the order is presented in G. JohnIkenberry
and Charles A. Kupchan, "Socialization and Hegemonic Order,"Initerniationial
Organiizationi,Vol. 44,
No. 3 (Summer 1990), pp. 283-315.
International
Security23:3 | 58
more predictable,the secondarystatesdo not need to spend as many resources

on "risk premiums,"which would otherwisebe needed to prepare foreither
domination or abandonment. In such a situation,the asymmetriesin power
are renderedmore tolerableforweaker states.
Importantly, institutionalagreementis possible because of the different
time
horizons that the hegemonic and secondary states are using to calculate their
interests.The leading state agrees to forgo some of the gains that it could
achieve if it took full advantage of its superior power position, doing so to
conserve power resourcesand invest in futurereturns.The weaker states get
more returnson theirpower in the early periods,but in agreeingto be locked
into a set of postwar institutions,they give up the opportunityto take full
potentialadvantage of risingrelativepower capacities in later periods. These
alternativecalculations are summarized in Figures 1 and 2. The leading state
trades short-term gains forlong-termgains, takingadvantage of the opportu-
nityto lay down a set of institutionsthat will ensure a favorable order well
into the future.Gains in the later periods are greaterthan what that state's
power capacities alone, withoutthe institutionalagreement,would otherwise
yield. Weaker and secondary statesgive up some later opportunitiesto gain a
more favorable returnon theirrisingrelativepower, but in returntheyget a
betterpostwar deal in the early postwar period. The option of losing more
Figure1. TimeHorizonsand the Returnon Power Assets: LeadingState.
Relativepower
capabilities(P)
High
Returnon
power assets
(P)
Low
T(1) T(2) T(3)
Figure2. TimeHorizonsand the Returnon Power Assets: SecondaryState.
Relativepower
capabilities(P)
High
(P)
Returnon
powerassets
Low
T(1) T(2) T(3)
now so as to gain more later is not an attractiveoption fora weak state that
is strugglingto rebuild afterwar. But beyond this,the weaker states also get
an institutionalagreementthatprovides some protectionsagainst the threatof
domination or abandonment-if the leading state is able to crediblydemon-
stratestrategicrestraint.
Taken together,the Westernpostwar order involves a bargain: the leading
state gets a predictable and durable order based on agreed-upon rules and
institutions-itsecures the acquiescence in this order of weaker states,which
in turnallows it to conserve its power. In return,the leading state agrees to
limitson its own actions-to operate according to the same rules and institu-
tions as lesser states-and to open itselfup to a politicalprocess in which the
weaker states can activelypress theirinterestsupon the more powerfulstate.
The hegemonicor leading stateagrees to forgosome gains in theearlypostwar
period in exchange forrules and institutionsthatallow it to have stable returns
later,while weaker states are given favorable returnsup frontand limits on
the exerciseof power.
STRATEGIES OF RESTRAINT: BONDING, BINDING, AND VOICE OPPORTUNITIES

The American postwar hegemonic order could take on constitutionalcharac-
teristicsbecause of the way institutions,created by and operated between
Security23:3 | 60
International
democraticstates,could be wielded to facilitatestrategicrestraint.Institutions

can shape and limit the way power can be used in the system, thereby
renderingasymmetricpostwar power relationsless potentiallyexploitiveand
commitmentmore certain.The returnsto power are reduced.32Where institu-
tions createrestraintson power,weaker and secondarystateshave less fearof
abandonment, or domination is lessened. Institutionscan also dampen the
effectsof the securitydilemma and reduce the incentives for weaker and
secondary states to balance against the newly powerful state. By creatinga
mutuallyconstrainingenvironment,institutionsallow states to convey assur-
ances to each otherand mitigatethe dynamics of anarchy.33
The hegemonic state has a disproportionaterole in creatingconfidencein
postwar order: it has the most capacity to break out of its commitmentsand
take advantage of its position to dominate or abandon the weaker and secon-
dary states.34As a result,in its effortsto draw other states into the postwar
order,the leading state will have strongincentivesto find ways to reassure
these other states, to demonstrate that it is a responsible and predictable
wielder of power and thatthe exerciseof power is, at least to some acceptable
degree,circumscribed.To achieve thisgoal, the leading statecan pursue strate-
gies thatinvolve bonding,binding,and institutionalizedvoice opportunities.
Bonding means to certifystate power: to make it open and predictable.A
powerful leading state with a governmentaldecisionmaking process-and
wider politicalsystem-that is open and transparent,and thatoperates accord-
32. Obviously,neorealistsargue that such guarantees are never sufficientenough to go forward

with far-reachinginstitutionalizedcooperation. Under conditions of anarchy,neorealistsassert,
stateswill be reluctantto seek even mutuallyadvantageous agreementsifitleaves themvulnerable
to cheatingand/or the relativegains of others.In a self-helpsystemsuch as anarchy,states face
huge obstacles to institutionalizedcooperation because the interdependenceand differentiation
thatcome with it are manifestwithinan anarchyas vulnerability. See JosephM. Grieco,"Anarchy
and the Limitsof Cooperation: A Realist Critique of the Newest Liberal Institutionalism,"
Interna-
tionalOrganization, Vol. 42, No. 3 (Summer 1988), pp. 485-507.
33. For a discussion of the general problem of credible commitmentand its importanceto insti-
tutionaldevelopment and the rule of law, see BarryWeingastand Douglass C. North,"Constitu-
tions and Commitment: The Evolution of Institutions Governing Public Choice in
Seventeenth-CenturyEngland," Journalof Economnic History,Vol. 44, No. 4 (December 1989),
pp. 803-832.
34. As PeterCowhey argues, otherstates"will not become fullycommittedto workingwithinthe
multilateralorder unless they believe the dominant powers intend to stay within it." Cowhey,
"Elect Locally-Order Globally,"in JohnGerard Ruggie,ed., Multilateralismn:TheTheoryand Practice
ofan Institutionlal
Formn(New York:Columbia UniversityPress, 1993),p. 158. Secondary statesmay
have feweroptions than the leading state,but theirwillingnessto participatewithinthe order-
that is, to engage in voluntary compliance-will hinge on the ability of the leading state to
demonstrateits reliability, commitment,and willingnessof forgothe arbitraryexerciseof power.
ing to predictableinstitutionalrules and procedures,can reassure weaker and

secondarystates thatthe exerciseof power will not be arbitraryor exploitive.
JonElsterargues that"bonding" is analogous to theefforts ofa firmattempting
to attractoutside shareholders: "It must incur 'costs of bonding,' such as
conservativeprinciplesof accountancy,in order to attractcapital."35Potential
shareholders are more likely to lend their capital to a firmwhen the firm
operates according to established and reasonable methods of record keeping
and accountability.Similarly,when a state is open and transparentto outside
states,it reduces the surprisesand allows otherstatesto monitorthe domestic
decisionmaking that attends the exercise of power. The implication of this
argumentis thatdemocraticstateshave an advantage in the process of bond-
ing. Democraticstateshave a more ready capacityto incurthe costs of bonding
because those costs will be relativelylow. Democratic states already have the
decentralized and permeable institutionsthat provide secondary states with
information, access, and ultimatelyreassurance.
The leading state can go beyond internalopenness to establishformalinsti-
tutionallinks with other states, limitingstate autonomy and allowing other
statesto have institutionalized"voice opportunities"in the decisionmakingof
the leading state.These institutionalbinding strategieshave been explored by
JosephGrieco and Daniel Deudney. Grieco argues that weaker states within
the European Union (EU) have had an incentiveto create institutionallinks
with strongerstates so as to have a "voice" in how the strongstates exercise
their power, therebypreventingdomination of the weaker by the stronger
states. Weak states are likely to find institutionalizedcollaboration with
strongerstates attractiveif it provides mechanisms to influencethe policy of
the strongerstates. "States . . . are likely to assign great significanceto the
enjoymentof such effectivevoice opportunitiesin a cooperativearrangement,
forit may determinewhetherstates can obtain redress if theyare concerned
about such mattersas the compliance of strongerpartnerswith theircommit-
mentsin the arrangement,or imbalances in the division of otherwisemutually
positive gains thatmay be produced by theirjointeffort."36 Put differently,
the
institutionalizationofrelationsbetween weak and strongstates,when it creates
35. Elster,"Introduction,"in Elsterand Slagstad, Constitutionalism

awedDemocracy,p. 15.
36. JosephM. Grieco, "State Interestsand InstitutionalRule Trajectories:A Neorealist Interpreta-
tion of the MaastrichtTreatyand European Economic and MonetaryUnion," SecurityStudies,Vol.
5, No. 3 (Spring 1996), p. 288. See also Grieco, "Understanding the Problem of International
Cooperation:The Limitsof Neoliberalismand the Futureof RealistTheory,"in David A. Baldwin,
ed., NeorealismanidNeoliberalism:The Contemporary Debate (New York:Columbia UniversityPress,
International
Security23:3 | 62
voice opportunitiesfor the weaker states,can be a solution forthese weaker

states thatwant to work with but not be dominated by strongerstates.
Deudney also describes the dynamic of binding,but emphasizes its other
feature:it is a practiceof establishinginstitutionallinksbetween the units that
reduce theirautonomyvis-'a-visone another.37 In agreeingto be institutionally
connected,statesmutuallyconstraineach otherand therebymitigatethe prob-
lems of anarchythatlead to securitydilemmas and power balancing. Accord-
ing to Deudney, binding practicesare particularlyavailable to and desired by
democraticpolities thatwant to resistthe state-strengthening and centralizing
consequences of balance-of-powerorders. Binding restrictsthe range of free-
dom of states-whether weak or strong-and when states bind to each other,
theyjointlyreduce the role and consequences of power in theirrelationship.
Each of these strategies involves the institutionalizationof state power.
Asymmetriesof power do not disappear, but institutions-democraticinstitu-
tions and intergovernmental institutions-channel and circumscribethe way
that state power is exercised. Institutionsmake the exercise of power more
predictableand less arbitraryand indiscriminate,up to some point. When a
newly hegemonic state seeks to create a mutually acceptable order,doing so
to preserveand extendthe returnsto its power into the future,institutionscan
be an attractivetool: theylock other states into the order,and theyallow the
leading state to reassure and co-opt other states by limitingthe returnsto
power.
PostwarHegemonicReassurance
By design,inadvertence,and structuralcircumstance,the United States was in

a positionafterWorldWar II to mute the implicationsof unprecedentedpower
asymmetriesbetween itselfand its postwar partners,alleviatingfearsof domi-
nation and abandonment. These strategiesof reassurance were manifestin
1993), pp. 331-334; and Grieco, "The MaastrichtTreaty,Economic and MonetaryUnion, and the
Neo-Realist Research Programme,"Reviezvof International Stuldies,Vol. 21, No. 1 (January1995),
pp. 21-40. The classic formulationof this logic is Albert0. Hirschman,Exit,Voice,and Loyalty-
ResponsestoDecliniein Firmns,
Organizations,and States(Cambridge,Mass.: Harvard UniversityPress,
1970).
37. Daniel Deudney,"The PhiladelphianSystem:Sovereignty, ArmsControl,and Balance of Power
in the AmericanStates-Union,"International Organization,Vol. 49, No. 2 (Spring 1995),pp. 191-228;
and Deudney, "Binding Sovereigns: Authorities,Structures,and Geopolitics in Philadelphian
Systems," in Thomas Bierstekerand Cynthia Weber, eds., State Sovereignty as Social Construtct
(Cambridge: Cambridge UniversityPress, 1996), esp. pp. 213-216.
three ways: in America's "reluctant"orientationtoward hegemonic domina-

tion; in the open and penetratedcharacterof the American polity;and in the
ways postwar economic and securityinstitutionsshaped and circumscribed
the exercise of American power. Together,these strategiesand circumstances
served to limitthe returnsto power in the postwar order,creatingthe condi-
tions forits legitimacyand durability.
RELUCTANT HEGEMONY
Behind specificinstitutionalstrategiesof reassurance,the United States exhib-
ited a general orientationtoward postwar order and the exercise of coercive
hegemonicpower thatserved to reassureotherstates.The United Statesmight
be seen as a "reluctanthegemon" in many respects,seekingagreementamong
the Westernstateson a mutuallyacceptable order,even ifthismeant extensive
compromise,and pushing for an institutionalizedorder that would require
littledirect "management" by the United States or the active exercise of its
hegemonicpower. It is revealingthatthe initialand most forcefully presented
American view on postwar order was the State Department'sproposal for a
postwar systemof freetrade. This proposal did not only reflectan American
convictionabout the virtuesof open markets,but it also was a vision of order
that would require very littledirectAmerican involvementor management.
The system would be largely self-regulating,leaving the United States to
operate withoutthe burdens of directand ongoing supervision.
This view on postwar trade reflecteda more general American orientation
as thewar came to an end. The new hegemonwanted a world orderthatwould
advance Americaninterests,but it was not eager to activelyorganize and run
thatorder.In this sense, the United States was a reluctantsuperpower.38This
generalcharacteristicwas not lost on the Europeans, and it matteredas Amer-
ica's potentialpartnerscontemplatedwhetherand how to cooperate with the
United States. To the extentthatthe United States could convey the sense that
it did not seek to dominate the Europeans, it gave greater credibilityto
America's proposals for an institutionalizedpostwar order.It provided some
reassurancethatthe United States would operate withinlimitsand not use its
overwhelmingpower position simply to dominate.
A HistoryofAmerica'sGlobalEcoiionoic
38. See RichardHolt, The ReluctantSuperpozwer: Reoch(New
York: Kodansha International,1995); see also Geir Lundestad, "An Empire by Invitation?The
United States and WesternEurope, 1945-1952,"JourinalofPeace Research,Vol. 23, No. 3 (September
1986).
Security23:3 | 64
International
More generally,the overall patternof Americanpostwar policies reflecteda

self-consciouseffortby administrationofficialsto infuse the postwar system
with a sense of legitimacyand reciprocal consent. When American officials
began to organize Marshall Plan aid for Europe, for example, there was a
strongdesire to have the Europeans embrace Americanaid and plans as their
own-thus enhancing the legitimacyof the overall postwar settlement.At a
May 1947 meeting, George Kennan argued that it was importantto have
"European acknowledgementof responsibilityand parentage in the plan to
preventthe certainattemptsof powerful elements to place the entireburden
on the United States and to discreditit and us by blaming the United States
forall failures."Similarly,State DepartmentofficialCharles Bohlen argued that
United States policy should not be seen as an attempt"to force'the American
way' on Europe."39The United Stateswanted to createan orderthatconformed
to its liberal democraticprinciples,but this could be done only if other gov-
ernmentsembraced such a systemas theirown.
This orientationwas also reflectedin the compromisesthattheUnited States
made in accommodatingEuropean views about the postwar world economy.
The Britishand the continentalEuropeans, worried about postwar depression
and the protectionof their fragile economies, were not eager to embrace
America's starkproposals foran open world tradingsystem,favoringinstead
a more regulated and compensatorysystem.40The United States did attempt
to use its material resources to pressure and induce Britain and the other
European countriesto abandon bilateraland regional preferentialagreements
and accept the principlesof a postwar economy organized around a nondis-
criminatorysystemof trade and payments.4'The United States knew it held a
commanding position and sought to use its power to give the postwar order
a distinctiveshape. But it also prized agreementover deadlock, and it ulti-
39. Quoted in "Summary of Discussion on Problemsof Relief,Rehabilitation,and Reconstruction

of Europe," May 29, 1947,Foreigni RelationsoftheUnitedStates,1947,Vol. 3, p. 235.
40. The strongestclaims about Americanand European differences over postwarpoliticaleconomy
are made by Fred Block, The Originsof InternatiolnalEcoionmic Disorder(Berkeley:Universityof
CaliforniaPress, 1977), pp. 70-122.
41. The 1946 Britishloan deal was perhaps the most overteffortby the Trumanadministrationto
tie Americanpostwar aid to specificpolicy concessionsby allied governments.This was the failed
Anglo-AmericanFinancial Agreement,which obliged the Britishto make sterlingconvertiblein
exchange for American assistance. See Richard Gardner,Sterling-Dollar Diplomacy,2d ed. (New
York: Columbia UniversityPress, 1980); and AlfredE. Eckes, Jr.,A Search forSolvency: Bretton
Woods and the InternationalMonetarySystem,1944-71 (Austin: Universityof Texas Press, 1971).
matelymoved a great distance away fromits originalproposals in settingup

the various postwar economic institutions.42
OPEN HEGEMONY
A second way thatthe United States projectedreassurancewas structural-its
own liberal democratic polity.The open and decentralized characterof the
American political systemprovided opportunitiesforother states to observe
decisionmakingand exercise their"voice" in the operation of American post-
war foreignpolicy,therebyreassuringthese states thattheirinterestscould be
activelyadvanced and thatprocesses of conflictresolutionwould exist.In this
sense, the American postwar order was a "liberal hegemony,"an extended
systemthatblurreddomestic and internationalpolitics as it created an elabo-
rate transnationaland transgovernmentalpolitical system with the United
States at its center.43
There are actually several ways in which America's open hegemony has
reinforcedthe credibilityof the United States' commitmentto operate within
an institutionalizedpolitical order.The firstis simply the transparencyof the
system,which reduces surprisesand allays worriesby partnersthattheUnited
Statesmightmake abruptchanges in policy.This transparencycomes fromthe
fact that policymakingin a large, decentralized democracy involves many
players and an extended and relativelyvisible politicalprocess. However, it is
not only that it is an open and decentralized system; it is also one with
competingpolitical parties and an independent press-features that serve to
expose the underlyingintegrityand viabilityof major policy commitments.44
The open and competitiveprocess may produce mixed and ambiguous policies
at times,but thetransparencyof theprocess at least allows otherstatesto make
more accurate calculations about the likely direction of American foreign
policy,which lowers levels of uncertaintyand provides a measure of reassur-
ance, which-everything else being equal-creates greater opportunitiesto
cooperate.
42. See JohnGerard Ruggie, WinningthePeace: Amer-ica and WorldOrderin theNew Era (New York:
Columbia UniversityPress, 1996), chap. 5.
43. This section and the nextbuild on Daniel Deudney and G. JohnIkenberry,"The Sources and
Characterof Liberal InternationalOrder,"ReviezvofInternational Studies,forthcoming.
44. This poilntis made in James Fearon, "Domestic Political Audiences and the Escalation of
InternationalDisputes," AmericanPoliticalScienceReviezv,
Vol. 88, No. 3 (September1994), pp. 577-
592.
Security23:3 | 66
International
Anotherway in which the open hegemonic order provides reassurances to

partnersis that the American systeminvites (or at least creates opportunities
for) the participationof outsiders.The fragmentedand penetratedAmerican
systemallows and invitesthe proliferationof a vast networkof transnational
and transgovernmentalrelations with Europe, Japan, and other parts of the
industrialworld. Diffuseand dense networksof governmental,corporate,and
private associations tie the systemtogether.The United States is the primary
site for the pulling and hauling of trans-Atlanticand trans-Pacificpolitics.
Europeans and Japanesedo not have elected officialsin Washington-but they
do have representatives.45 Although this access to the American political pro-
cess is not fullyreciprocatedabroad, the openness and extensivedecentraliza-
tion of the Americanliberal systemassures otherstatesthattheyhave routine
access to the decisionmakingprocesses of the United States.
The implicationof open hegemonyis thatthe United States is not as able to
use its commanding power position to gain disproportionatelyin relations
with Japan and Europe-or at least it diminishes the leverage that would
otherwiseexist.For example, thereis littleevidence thatthe United States has
been able to bring more pressure to bear on Japanese importpolicy,even as
its relative power capacities have seeminglyincreased in the 1990s with the
end of theCold War and the slump in theJapaneseeconomy.Beginningin July
1993 with the signingof a "framework"agreement,the Clintonadministration
launched a series of effortsto pin Japan to numericalimporttargets,includ-
ing threatenedsanctions to boost American automobile imports.But despite
repeated effortsin February 1994 and the spring of 1995, Prime Minister
Morihiro Hosakawa, helped by protests from the European Union, was
largely able to resist American pressure.46The advent of the World Trade
Organization (WTO) also has provided additional ways forJapan to narrow
45. For the transnationalpoliticalprocess channeled throughthe Atlanticsecurityinstitutiolns,

see
Thomas Risse-Kappen,Cooperation amongDemocracies:The EuropeanInfluence on U.S. ForeignPolicy
(Princeton,N.J.:PrincetonUniversityPress, 1995). On the U.S.-Japaneseside, see PeterJ.Katzen-
stein and Yutaka Tsujinaka,"'Bullying,''Buying,'and 'Binding':U.S.-JapaneseTransnationalRela-
tions and Domestic Structures,"in Risse-Kappen, ed., BringingTransnationalRelationsBack In:
Non-StateActors,DomesticStructures, and Interncational
Institutions
(Cambridge: Cambridge Univer-
sityPress, 1995), pp. 79-111.
46. See Andrew Pollack, "U.S. Appears to RetreatfromSettingTargets to Increase Japan's Im-
ports," New YorkTimes,July 10, 1993; David Sanger, "Hosakawa's Move Foils U.S. Strategy,"
InternationalHeraldTribune,April 11, 1994; Nancy Dunne and Michito Nakamato, "Wiser U.S. to
Meet Chastened Japan,"FinancialTimes,May 18, 1994; and Reginald Dale, "JapanGains the Edge
in Trade War,"International HeraldTribune,September 13, 1994.
Persistence
ofPostwarOrder| 67
trade disputes to specificissues, bringinternationalproceduresof review into

play, and diminish the capacity of the United States to bring its hegemonic
power to bear. In the recentKodak case, forexample, the WTO basically ruled
against the United States in its effortto push Japan toward greateropenness.
Afterthis ruling,the United States quietly lowered its voice and sat down.47
The open and penetratedcharacterof the United States serves to fragmentand
narrow policy disputes, creates a more level playing field for European and
Japaneseinterests,and reduces the implicationsof hegemonicpower asymme-
tries.
BINDING INSTITUTIONS
A finalway in which reassurancehas been conveyed is throughtheinstitutions
themselves,which provide "lock in" and "binding" constraintson the United
States and its partners,therebymitigatingfears of domination or abandon-
ment.The Westerncountrieshave made systematiceffortsto anchor theirjoint
commitmentsin principled and binding institutionalmechanisms. Govern-
mentsmightordinarilyseek to preservetheiroptions,to cooperate with other
statesbut to leave open the option of disengaging.What the United Statesand
the otherWesternstates did afterthe war was exactlythe opposite: theybuilt
long-termeconomic,political,and securitycommitmentsthatwere difficultto
retract.They "locked in" theircommitmentsand relationships,to the extent
thatthis can be done by sovereign states.
The logic of institutionalbinding is best seen in securityalliances. Alliances
have oftenbeen formednot simply or even primarilyto aggregate power so
as to balance against externalthreats,but ratherto allow alliance partnersto
restraineach other and manage joint relations. Alliances have traditionally
been seen as temporaryexpedients that bring states togetherin pledges of
mutual assistance in the face of a common threat,a commitmentspecifiedin
the casus foederisarticleof the treaty.But as Paul Schroeder and othershave
47. As Clyde V. Prestowitz,Jr.writes: "In the past, the United States would have attemptedto
negotiate a bilateral settlementwith the potential imposition of sanctions lurkingin the back-
ground as an incentiveto reach an agreement.Under the new WTO rules, however,all disputes
are supposed to be submittedto the WTO forcompulsoryarbitrationand unilateralimpositionof
trade sanctions is illegal....But if America then attempts to solve the problem unilaterallyby
imposing sanctions,Japan could have the U.S. sanctions declared illegal by the WTO. Knowing
this,Japaneseofficialshave refusedeven to meet with U.S. negotiators-and have effectively told
the United States to buzz off."Prestowitz,"The New Asian Equation," Washinigton Post,April 14,
1996.
International
Security23:3 | 68
noted, alliances have also been created as pacta de contrahendo-pactsof re-

straint.48 They have served as mechanisms for states to manage and restrain
their partners within the alliance. "Frequently the desire to exercise such
controlover an ally's policy,"Schroederargues, "was the main reason thatone
power, or both, entered into the alliance."49Alliances create binding treaties
thatallow states to keep a hand in the securitypolicy of theirpartners.When
alliance treatiesare pactade contrahendo,potentialrivals tie themselvesto each
other-alleviating suspicions,reducinguncertainties,and creatinginstitutional
mechanismsforeach to influencethe policies of the other.50
The practiceof mutual constraintmakes sense only if internationalinstitu-
tions or regimes can have an independent orderingimpact on the actions of
states.The assumption is thatinstitutionsare sticky-that theycan take on a
lifeand logic of theirown, shaping and constrainingeven the statesthatcreate
them.When states employ institutionalbinding as a strategy,theyare essen-
tiallyagreeing to mutuallyconstrainthemselves.In effect,institutionsspecify
what it is that states are expected to do, and theymake it difficultand costly
forstatesto do otherwise.51 In thissense, institutionalbindingis like marriage:
two individuals realize thattheirrelationshipwill eventuallygenerateconflict
and discord, so they bind themselvesin a legal framework,making it more
difficultto dissolve the relationshipwhen those inevitablemomentsarrive.In
the case of internationalinstitutions,examples of bindingmechanismsinclude
treaties,interlockingorganizations,joint managementresponsibilities,agreed-
upon standards and principles of relations,and so forth.These mechanisms
raise the "costs of exit" and create "voice opportunities,"therebyproviding
mechanismsto mitigateor resolve the conflict.
48. See Paul W. Schroeder,"Alliances, 1815-1945:Weapons of Power and Tools of Management,"

in Klaus Knorr,ed., HistoricalDimensionsofNationalSecurityProblems(Lawrence: UniversityPress
of Kansas, 1975),pp. 227-262. As Schroedernotes,the internalconstraintfunctionof alliances was
earlier observed by George Liska. See Liska, Nations in Alliance:The Limitsof Interdepetndence
(Baltimore,Md.: JohnsHopkins UniversityPress, 1962), p. 116; ImperialAmnerica: The International
Politicsof Primnacy(Baltimore,Md.: JohnsHopkins UniversityPress, 1967); pp. 9-11, 20-21; and
AlliancesanidtheThirdWorld(Baltimore,Md.: JohnsHopkins UniversityPress, 1968), pp. 24-35.
49. Schroeder,"Alliances, 1815-1945," p. 230.
50. For a recentextension of this argument,see Patricia A. Weitsman,"Intimate Enemies: The
Politics of Peacetime Alliances," SecurityStudies,Vol. 7, No. 1 (Autumn 1997), pp. 156-192.
51. This view accords with our general view of what institutionsare and do. As Lorenzo Ornaghi
argues: "The role of institutionsin politics is to give the rules of the game, in that,by reducing
the uncertainand unforeseeablecharacterof interpersonalrelations,insurance is mutually pro-
vided." Ornaghi, "Economic Structureand Political Institutions:A Theoretical Framework,"in
Mauro Baranzini and RobertoScazzieri, eds., The Econiomnic Thieory of Structureand Chanige(Cam-
bridge: Cambridge UniversityPress, 1990), p. 27.
ofPostwarOrder| 69
Persistence
The BrettonWoods economic and monetary accords exhibit the logic of

institutionallock-in. These were the firstaccords to establish a permanent
internationalinstitutionaland legal frameworkto ensure economic coopera-
tion between states. They were constructedas elaborate systemsof rules and
obligations with quasi-judicial procedures for adjudicating disputes.52The
dollar-gold standard collapsed in 1971, but the broader institutionsof joint
managementremain. In effect,the Westerngovernmentscreated an array of
functionallyorganized transnationalpolitical systems.Moreover, the demo-
craticcharacterof the United Statesand the otherWesterncountriesfacilitated
the constructionof these dense interstateconnections. The permeabilityof
domesticinstitutionsprovided congenial grounds forreciprocaland pluralistic
"pulling and hauling" across the advanced industrialworld.
It was here thatthe Cold War's securityalliances provided additional insti-
tutionalbindingopportunities.The old sayingthatNATO was createdto "keep
the Russians out, the Germans down, and the Americans in" is a statement
about the importanceof the alliance structuresforlocking in long-termcom-
mitmentsand expectations.The U.S.-Japansecurityalliance also had a similar
"dual containment"character.These institutionsnot only served as alliances
in the ordinarysense of organized effortsto balance against externalthreats,
but they also provided mechanisms and venues to build political relations,
conduct business, and regulateconflict.
The constitutionalfeatures of the Western order have been particularly
importantforGermany and Japan. Both countrieswere reintegratedinto the
advanced industrialworld as "semisovereign"powers: thatis, theyaccepted
unprecedented constitutionallimits on their military capacity and inde-
pendence.53 As such, they became unusually dependent on the array of
Western regional and multilateraleconomic and security institutions.The
Westernpolitical order in which they were embedded was integralto their
stabilityand functioning.The ChristianDemocrat WaltherLeisler Kiep argued
in 1972 that "the German-Americanalliance . . . is not merelyone aspect of
modern German history,but a decisive element as a result of its preeminent
place in our politics.In effect,it provides a second constitutionforour coun-
52. See Harold James,International MonetaryCooperationsince BrettonWoods(New York: Oxford

UniversityPress, 1995).
53. On the notionof semisovereignty, see PeterJ.Katzenstein,Policyand Politicsin WestGermany:
The Grozvthofa Semi-Sovereign State(Philadelphia: Temple UniversityPress,1987). For a discussion
of Japanese semisovereigntyand the postwar peace constitution,see Masaru Tamamoto,"Reflec-
tions on Japan's Postwar State," Daedalus,Vol. 125, No. 2 (Spring 1995), pp. 1-22.
International
Security23:3 | 70
try."54 This logic of Germany'sinvolvementin NATO and the EU was reaffir-

med recentlyby theGermanpoliticalleader VoigtKarsten:"We wanted to bind
Germanyinto a structurethatpracticallyobliges Germanyto take the interests
of its neighbors into consideration.We wanted to give our neighborsassur-
ances thatwe won't do what we don't intendto do."55Westerneconomic and
securityinstitutionsprovide Germany and Japan with a political bulwark of
stabilitythatfartranscendstheirmore immediate and practicalpurposes.
The recentrevision of the U.S.-Japansecuritytreatyin May 1996 is another
indicationthatboth countriessee virtuesin maintaininga tightsecurityrela-
tionshipregardlessof the end of the Cold War or the rise and fall of specific
securitythreatsin the region.56Even though the threatsin the region have
become less tangibleor immediate,the alliance has been reaffirmedand coop-
erationand jointplanninghave expanded. Partof thereason is thatthealliance
is still seen by many Japanese and American officialsas a way to render the
bilateralrelationshipmore stable by bindingeach to the other.57The expansion
of NATO is also, at least in part,drivenby the binding aspects of the alliance.
This view of NATO as "architecture"that would stabilize relations within
Europe and across the Atlantic,ratherthan primarilyan alliance to counter
externalthreats,was firstsignaled in Secretaryof State JamesBaker's famous
speech in the aftermathof thefallof theBerlinWall in 1989.58Some supporters
of NATO expansion see it as an insurance policy against the possibilityof a
54. Quoted in Thomas A. Schwartz, "The United States and Germany after 1945: Alliances,
TransnationalRelations,and the Legacy of the Cold War," DiplomaticHistony,Vol. 19, No. 4 (Fall
1995), p. 555.
55. Quoted in Jan Perlez, "Larger NATO Seen as Lid on Germany,"Internationcal Herald Tribunie,
December 8, 1997.
56. PresidentBill Clinton and Prime MinisterRyutaroHashimoto signed a JointDeclaration on
Securityon April 17, 1996, which was a revision of the 1978 Guidelines forU.S.-JapanDefense
Cooperation. The agreement declared that the U.S.-Japan securitytreatyof 1960 "remains the
cornerstone"of theirpolicies, thattheircombined forcesin Japan would engage in policy coordi-
nation fordealing with regionalcrises,and on a reciprocalbasis provide equipment and supplies.
Overall, the Japanese made a commitmentto actually move toward closer securityrelationswith
the United States.
57. Peter J. Katzensteinand Yutaka Tsujinaka argue that "the securityrelationshipbetween the
United States and Japan is best described by 'binding,'with the United States doing most of the
'advising' and Japanmost of the 'accepting.'By and large since the mid-1970sdefensecooperation
has increasedsmoothlyand apparentlyto the satisfactionof both militaries.Since thatcooperation
involved primarilygovernmentsand sub-units of governmentsimplementingpolicy, 'binding'
resultsprimarilyfromtransgovernmental relations."Katzensteinand Tsujinaka,"'Bullying,''Buy-
ing,'and 'Binding,"'inRisse-Kappen,Bringin1gTrau1sn1ation1al
Relations BackIn, p. 80. See also Richard
Finn, "Japan's Search for a Global Role," in WarrenS. Hunsberger,Japan'sQuest: The Searchfor
InternationalRole,Recognition,
and Respect(New York:M.E. Sharpe, 1977), pp. 113-130.
58. See Michael Smith and Stephen Woolcock, The UnitedStatesand theEuropeanl Communilityin a
Transformed World(London: Pinter/RoyalInstituteof InternationalAffairs,1993), p. 1.
ofPostwarOrder| 71
Persistence
futureresurgentand revisionistRussia.59But others,particularlyin theClinton

administration,see the virtues of expansion more in termsof the stabilizing,
integrating,and binding effectsthatcome fromNATO as an institution.60
Overall, U.S. hegemonyis reluctant,open, and highlyinstitutionalized.All
these characteristicshave helped to facilitatea ratherstable and durable po-
liticalorder.AmericanstrategicrestraintafterWorldWar II leftthe Europeans
more worried about abandonment than domination,and theyactivelysought
American institutionalizedcommitmentsto Europe. The American polity's
transparencyand permeabilityfosteredan "extended" political order-reach-
ing outward to the otherindustrialdemocracies-with most of its roads lead-
ing to Washington.Transnationaland transgovernmental relationsprovide the
channels. Multiple layers of economic,political,and securityinstitutionsbind
these countriestogetherin ways thatreinforcethe credibilityof theirmutual
commitments.The United States remains the centerof the system,but other
states are highlyintegratedinto it, and its legitimacydiminishesthe need for
the exerciseof coercive power by the United States or forbalancing responses
fromsecondary states.
IncreasingReturnsto PostwarInstitutions
The bargains struckand institutionscreated at the early momentsof postwar

orderbuilding have not simplypersistedforfifty years,but theyhave actually
become more deeply rooted in the wider structuresof politics and societyof
the countriesthatparticipatewithinthe order.That is, more people and more
of theiractivitiesare linked to the institutionsand operationsof the American
liberal hegemonic order. A wider array of individuals and groups, in more
countriesand more realms of activity,have a stake-or a vested interest-in
thecontinuationof the system.The costs of disruptionor change in thissystem
have steadily grown over the decades. Together,this means that "competing
orders" or "alternative institutions"are at a disadvantage. The system is
increasinglyhard to replace.
The reason institutionshave a lock-in effectis primarilybecause of the
phenomenon of increasingreturns.61 There are several aspects to increasing
59. See Zbigniew Brzezinski,"NATO-Expand or Die?" NezwYorkTimes,December 28, 1994.

60. See Michael Cox, U.S. ForeignPolicyaftertheCold War:Superpower withouta Mission? (London:
Royal Instituteof InternationalAffairs),pp. 79-83.
61. Both rationalchoice and sociological theoriesof institutionsoffertheoriesof institutionalpath
dependency-both emphasizing the phenomenon of increasingreturns.See Paul Pierson,"Path
Dependence and the Study of Politics,"unpublished paper, Harvard University,1996.
International
Security23:3 | 72
returnsto institutions.First,thereare large initialstart-upcosts to creatingnew

institutions.Even when alternativeinstitutionsmight be more efficientor
accord more closely with the interestsof powerful states,the gains fromthe
new institutionsmust be overwhelminglygreaterbefore they overcome the
sunk costs of the existinginstitutions.62 Moreover, there tend to be learning
effectsthatare achieved in the operationof the existinginstitutionthatgive it
advantages over a start-upinstitution.Finally,institutionstend to create rela-
tions and commitmentswith otheractors and institutionsthatserve to embed
the institutionand raise the costs of change. Taken together,as Douglass North
concludes, "the interdependentweb of an institutionalmatrixproduces mas-
sive increasingreturns."63
When institutionsmanifestincreasingreturns,it becomes very difficultfor
potential replacementinstitutionsto compete and succeed. The logic is seen
most clearlyin regard to competingtechnologies.The historyof the videocas-
setterecorderis the classic example, where two formats,VHS and Beta, com-
peted for standardization.The two formatswere introduced roughly at the
same time and initiallyhad equal market share, but soon the VHS format,
throughluck and circumstancesunrelated to efficiency, expanded its market
share. Increasing returnson early gains tilted the competitiontoward VHS,
allowing it to accumulate enough advantages to take over the market.64Even
if Beta was ultimatelya superior technology,a very small marketadvantage
by VHS at an early and criticalmoment allowed it to lower its production
costs, and the accumulation of connecting technologies and products that
require compatibilitymade it increasinglyhard for the losing technologyto
compete.The costsof switchingto theothertechnologyrise as productioncosts
are lowered, learningeffectsaccumulate, and the technologyis embedded in
a wider systemof compatible and interdependenttechnologies.65
62. On sunk costs, see ArthurL. Stinchcombe,Conistrutcting Social Theories(New York: Harcourt,
Brace and World,1968), pp. 108-118.
63. Douglass C. North, Institutions, InstitutionalChange,and EconomicPerformance (New York:
Cambridge UniversityPress, 1990),p. 95. For discussions of path dependency argtuments and their
implications,see Stephen D. Krasner,"Approaches to the State: Conceptions and Historical Dy-
namics," ComparativePolitics,Vol. 16 (January1984); and Paul Pierson, "When EffectBecomes
Cause: PolicyFeedback and PoliticalChange," WorldPolitics,Vol. 45, No. 4 (July1993),pp. 595-628.
64. See W. BrianArthur,"Positive Feedbacks in the Economy,"Scientific Ame1rican(February1990),
pp. 92-99. Reprintedin Arthur,IncreasingReturnsand Path Dependencein theEconiomy (Ann Arbor:
Universityof Michigan Press, 1995), pp. 1-12.
65. W. Brian Arthur,"Competing Technologies, Increasing Returns,and Lock-In by Historical
Small Events," EconomicJournial,
March 1989, pp. 116-131.
Persistence
ofPostwarOrder| 73
American postwar hegemonic order has exhibitedthis phenomenon of in-

creasingreturnsto its institutions.At the early momentsafter1945,when the
imperial,bilateral,and regionalalternativesto America'spostwar agenda were
most imminent,the United States was able to use its unusual and momentary
advantages to tiltthe systemin the directionit desired. The pathway to the
presentliberal hegemonic order began at a very narrowpassage where really
only Britain and the United States-and a few top officials-could shape
decisively the basic orientationof the world political economy.But once the
institutions,such as those erected at BrettonWoods and the General Agree-
ment on Tariffsand Trade, were established,it became increasinglyhard for
competingvisions of postwar orderto have any viability.America'sgreatburst
of institutionbuilding afterWorld War II fitsa general patternof international
continuityand change: crisisor war opens up a momentof fluxand opportu-
nity,choices are made, and interstaterelationsget fixedor settledfora while.66
The notion of increasingreturnsto institutionsmeans that once a moment
of institutionalselection comes and goes, the cost of large-scale institutional
change rises dramatically-even ifpotentialinstitutions, when compared with
existingones, are more efficientand desirable.67In termsof American hege-
mony,this means that,shortof a major war or a global economic collapse, it
is verydifficultto envisage the typeof historicalbreakpointneeded to replace
the existingorder.This is true even if a new would-be hegemon or coalition
of states had an interestin and agenda foran alternativeset of global institu-
tions-which theydo not.68
While the increasingreturnsto institutionscan serve to perpetuateinstitu-
tions of many sorts,Americanhegemonicinstitutionshave characteristics that
particularly lend themselves to increasingreturns. First,the set of principles
thatinfusethese institutions-particularlyprinciplesof multilateralism, open-
ness, and reciprocity-are ones that command agreement because of their
seemingfairnessand legitimacy.Organized around principlesthatare easy for
66. See Peter J. Katzenstein,"InternationalRelations Theory and the Analysis of Change," in

Ernst-OttoCzempiel and James N. Rosenau, eds., Global Changesand Theoretical Challenges:Ap-
proachesto WorldPoliticsforthe1990s (Lexington,Mass.: LexingtonBooks, 1989), pp. 291-304.
67. This notion of breakpoint or critical juncture is not developed in the increasing returns
literature,but it is implicitin the argument,and it is very importantforunderstandingthe path
dependency of Americanhegemony.
68. Major or greatpower war is a uniquely powerfulagent of change in world politicsbecause it
tends to destroy and discredit old institutionsand force the emergence of a new leading or
hegemonicstate.RobertGilpin discusses the possibilitythatwith the rise of nuclearweapons, this
sort of patternof global change may end, therebyleaving in place the existinghegemonicorder.
See Gilpin, Epilogue, Warand Changein WorldPolitics,pp. 231-244.
International
Security23:3 | 74
states to accept, regardless of theirspecificinternationalpower position, the

institutionalpatternis more robustand easy to expand. Moreover,the princi-
pled basis of hegemonic order also makes it more durable. This is JohnRug-
gie's argument about the multilateralorganization of postwar international
institutions:"All other thingsbeing equal, an arrangementbased on general-
ized organizingprinciplesshould be more elastic than one based on particu-
laristicinterestsand situationalexigencies."69Potentialalternativeinstitutional
orders are at an added disadvantage because the principles of the current
institutionalorderare adaptable, expandable, and easily accepted as legitimate.
Second, the open and permeable characterof American hegemonic institu-
tions also serves to facilitateincreasingreturns.One of the most important
aspects of increasingreturnsis thatonce a particularinstitutionis established,
other institutionsand relationstend to grow up around it and become inter-
connected and mutually dependent. A good analogy is computer software,
where a softwareprovider like Microsoft,aftergaining an initial marketad-
vantage, encourages the proliferationof softwareapplications and programs
based on Microsoft'soperatinglanguage. This in turnleads to a huge complex
of providers and users who are heavily dependent on the Microsoftformat.
The result is an expanding marketcommunityof individuals and firmswith
an increasinglydense set of commitmentsto Microsoft-commitmentsthatare
notbased on loyaltybut on thegrowingrealitythatchangingto anotherformat
would be more costly,even if it were more efficient.
The open and penetratedcharacterof American hegemonyencourages this
sort of proliferationof connectinggroups and institutions.A dense set of
transnationaland transgovernmentalchannels are woven into the trilateral
regions of the advanced industrialworld. A sort of layer cake of intergovern-
mental institutionsextend outward fromthe United States across the Atlantic
and Pacific.70Global multilateraleconomic institutions,such as the Interna-
tional MonetaryFund (IMF) and WTO, are connected to more circumscribed
governance institutions,such as the Group of Seven (G-7) and the Group of
Ten. Privategroups, such as the TrilateralCommission and hundreds of busi-
ness trade associations,are also connectedin one way or anotherto individual
69. JohnGerard Ruggie, "Multilateralism:The Anatomyof an Institution,"in Ruggie, ed., Multi-

lateralismMatters:The Theonyand Praxis ofan Institutional
Form(New York: Columbia University
Press, 1993), pp. 32-33.
70. See Cheryl Shanks, Harold K. Jacobson,and Jeffrey H. Kaplan, "Inertia and Change in the
Constellationof InternationalGovernmentalOrganizations,1981-1992,"International Organization,
Vol. 50, No. 4 (Autunn 1996), pp. 593-628.
Persistence
ofPostwarOrder| 75
governmentsand theirjointmanagementinstitutions.The steady rise of trade

and investmentacross the advanced industrialworld has made these countries
more interdependent,which in turn has expanded the constituencywithin
these countriesfora perpetuationof an open, multilateralsystem.71
Not only have more and more governmentsand groups become connected
to the core institutionsof the Westernorder; still more are seeking to join.
Almost every countryin the world, including China, has now indicated a
desire to join the WTO, and the line formembershipin NATO stretchesall the
way to Moscow. In the recentAsian currencycrisis,even countrieswith little
affinityfor the IMF and its operating methods have had littlechoice but to
negotiatewithit over the termsof loans and economic stabilization.Russia has
joined the annual G-7 summit,turningit into the Summitof the Eight,and the
eventual inclusionof China is quite likely.In the meantime,the G-7 process in
the 1990s has generated an expanding array of ministerialand intergovern-
mentalbodies in a wide varietyof functionalareas, tacklingproblems such as
organized crime, energy,terrorism,the environment,aid to Ukraine, and
global finance.72Together,relationsamong the advanced industrialcountries
since the end of the Cold War are characterizedby an increasinglydense
latticeworkof intergovernmentalinstitutionsand routinized organizational
relationshipsthatare servingto draw more governmentsand more functional
parts of these governmentsinto the extended postwar Westernpoliticalorder.
This means that great shiftsin the basic organizationof the Westernorder
are increasinglycostlyto a wideningarrayofindividuals and groups thatmake
up the order.More and more people have a stake in the system,even if they
have no particularloyalty or affinityfor the United States and even if they
mightreally prefera different order.As the decades have worn on, the oper-
atinginstitutionsof the postwar Westernorder have expanded and deepened.
More and more people would have theirlives disruptedif the systemwere to
radically change-which is another way of saying that the constituencyfor
71. This is the dynamicthatRobertKeohane and JosephNye seek to capturein their"international

organization" model. "Internationalorganization in the broad sense of networks,norms, and
institutionsincludes the norms associated with specificinternationalregimes,but it is a broader
categorythan regime,because it also includes patternsof elite networksand (if relevant)formal
institutions."They go on to argue that "the internationalorganizationmodel assumes that a set
of networks,norms, and institutions,once established, will be difficulteither to eradicate or
drasticallyto rearrange.Even governmentswith superiorcapabilities-overall or withinthe issue
area-will find it hard to work theirwill when it conflictswith established patternsof behavior
withinexistingnetworksand institutions."See Keohane and Nye, Powerand Interdependence, p. 55.
72. See Peter I. Hajnal, "From G7 to G8: Evolution, Role, and Documentation of a Unique
Institution,"unpublished paper, Centre forInternationalStudies, Universityof Toronto,1998.
International
Security23:3 | 76
preservingthe postwar politicalorder among the major industrialcountriesis

greaterthan ever before.It is in this sense thatthe Americanpostwar order is
stable and growing.
The dominance of the United States has sparked complaintsand resistance
in various quarters of Europe and Asia-but it has not triggeredthe type of
counterhegemonicbalancing or competitiveconflictthat might otherwisebe
expected. Some argue thatcomplaintsabout America's abuse of its command-
ing power positionhave grown in recentyears.73Unwillingnessto pay United
Nations dues, the Helms-BurtonAct, which inhibits trade with Cuba, and
resistanceto commitmentsto cut greenhouse gases-these and otherfailures
are thegristof European and Asian complaintsabout Americanpredominance.
But complaintsabout theAmerican"arroganceofpower" have been a constant
minor theme of postwar Westernorder. Episodes include the "invasion" of
American companies into Europe in the 1950s, the dispute over Suez in 1953,
the "Nixon shocks" in 1971 over the surprise closure of the gold window,
America'sfailureto decontroloil prices duringthe 1970s energycrisis,and the
Euro-missilescontroversyof the early 1980s. Seen in postwar perspective,it is
difficultto argue that the level of conflicthas risen. Today,as in the past, the
differencestend to be negotiatedand resolved withinintergovernmental chan-
nels-even while the Europeans, Americans, and Japanese agree to expand
their cooperation in new areas, such as internationallaw enforcement,the
environment,and nonproliferation. More important,despite complaintsabout
the American abuse of its hegemonic position, there are no serious political
movementsin Europe or Japan that call for a radical break with the existing
Westernorder organized around American power and institutions.It is the
stabilityof the order,in spite of policy strugglesand complaints,thatis more
remarkablethan any changes in the characterof the strugglesor complaints.
Conclusion
The twentiethcenturymay be ending, but the American centuryis in full

swing. The characterof American power is as interestingand remarkableas
the factof its existence.American domination or hegemony is very unusual,
73. A reporterforthe Washington Postrecentlysummarized thisview: "The chorusof dismay with

America's overwhelmingpower has grown louder lately as the United States findsitselfincreas-
ingly accused of bullyingthe rest of the world. Indeed, the United States is discoveringthat its
behavior has come under sharpestscrutinyfromfriendlynationsthatno longer feelpreventedby
Cold War loyalties from expressing their disagreementswith Washington."William Drozdiak,
"Even Allies Resent U.S. Dominance," Washington Post,November 4, 1997.
ofPostwarOrder| 77
Persistence
and the larger Western political order that surrounds it is unique as well.
Fundamentally,Americanhegemonyis reluctant,open, and highlyinstitution-
alized-or in a word,liberal.This is what makes it acceptable to othercountries
thatmightotherwisebe expected to balance against hegemonic power, and it
is also what makes it so stable and expansive.
Even with the end of the Cold War and the shiftingglobal distributionof
power, the relationsbetween the United States and the otherindustrialcoun-
triesof Europe and Asia remainremarkablystable and cooperative.This article
offerstwo majorreasons why Americanhegemonyhas endured and facilitated
cooperation and integrationamong the major industrialcountriesratherthan
triggeredbalancing and estrangement.Both reasons underscore the impor-
tance of the liberal featuresof Americanhegemonyand the institutionalfoun-
dations of Westernpolitical order.
First,theUnited Statesmoved veryquicklyafterWorldWar II to ensure that
relations among the liberal democracies would take place within an institu-
tionalized political process. In effect,the United States offeredthe othercoun-
tries a bargain: if the United States would agree to operate within mutually
acceptable institutions, therebymutingtheimplicationsof power asymmetries,
the othercountrieswould agree to be willingparticipantsas well. The United
States got the acquiescence of the other Westernstates,and they in turngot
the reassurance that the United States would neitherdominate nor abandon
them.
The stabilityof this bargain comes fromits underlyinglogic: the postwar
hegemonic order is infused with institutionsand practices that reduce the
returnsto power. This means thatthe implicationsof winning and losing are
minimized and contained. A state could "lose" in intra-Western relationsand
yet not worrythatthe winnerwould be able to use those winningsto perma-
nentlydominate. This is a centralcharacteristicof domestic liberal constitu-
tional orders. Parties that win elections must operate within well-defined
limits.They cannot use theirpowers of incumbencyto undermineor destroy
the opposition partyor parties.They can press the advantage of officeto the
limitsof the law, but thereare limitsand laws. This reassures the losing party;
it can accept its loss and prepare for the next election. The featuresof the
postwar order-and, importantly,the open and penetrated characterof the
Americanpolityitself-has mechanismsto provide the same sortof assurances
to America's European and Asian partners.
Second, the institutionsof American hegemony also have a durabilitythat
comes from the phenomenon of increasing returns.The overall system-
organized around principlesof openness,reciprocity, and multilateralism-has
Security23:3 | 78
International
become increasinglyconnectedto the wider and deeper institutionsof politics

and society within the advanced industrialworld. As the embeddedness of
these institutionshas grown,it has become increasinglydifficultforpotential
rival states to introducea competingset of principlesand institutions.Ameri-
can hegemonyhas become highlyinstitutionalizedand path dependent. Short
of large-scalewar or a global economic crisis,the American hegemonic order
appears to be immune to would-be hegemonic challengers.Even if a large
coalition of states had intereststhat favored an alternativetype of order,the
benefitsof change would have to be radicallyhigherthan those thatflowfrom
the presentsystemto justifychange. But thereis no potentialhegemonicstate
(or coalitionof states) and no set of rival principlesand organizationseven on
the horizon. The world of the 1940s contained farmore rival systems,ideolo-
gies, and intereststhan the world of the 1990s. The phenomenon of increasing
returnsis really a type of positive feedback loop. If initial institutionsare
established successfully,where the United States and its partners have
confidencein theircredibilityand functioning,thisallows these statesto make
choices thatserve to strengthenthe binding characterof these institutions.
The postwar Westernorder fitsthis basic logic. Its open and decentralized
characterinvites participationand creates assurances of steady commitment.
Its institutionalizedcharacteralso provides mechanisms for the resolutionof
conflictsand creates assurances of continuity.Moreover,like a marriage,the
interconnectionsand institutionsof the partnershiphave spread and deep-
ened. Withinthis open and institutionalizedorder,the fortuneof particular
states will continue to rise and fall. The United States itself,while remaining
at the centerof the order,also continues to experience gains and losses. But
the mix of winningand losing across the systemis distributedwidely enough
to mitigatethe interestthatparticularstates mighthave in replacing it. In an
order where the returnsto power are low and the returnsto institutionsare
high, stabilitywill be an inevitablefeature.

I Ken Berry 1998

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

I Ken Berry 1998

Hochgeladen von

Copyright:

Verfügbare Formate

Institutions, Strategic Restraint, and the Persistence of American Postwar Order

Author(s): G. John Ikenberry

International Vol.23,No. 3 (Winter

alliance and cooperation will decline.' Neorealist theoriesof hegemony con-

1. See KennethN. Waltz, TheoryofInternational Politics(Reading, Mass.: 1979). For extensionsand

The Westernstates are not held togetherbecause of external threatsor the

thatis, theyreduce the long-termimplicationsof asymmetriesof power. This

The implicationof this analysis is that the West is a relativelystable and

3. For an extended critiqueof neorealistexpectationsof post-Cold War order,see Daniel Deudney

Balance-of-powertheoryexplains order-and the rules and institutionsthat

4. Waltz, TheoryofInternational Politics.On the balance-of-threattheory,see Stephen M. Walt,The

trieswill eventuallyrevertback to traditionalgreatpowers.7The Soviet threat

7. KennethN. Waltz,"The EmergingStructureof InternationalPolitics,"International Security,Vol.

thatneorealismis hard-pressedto explain. Despite sharp shiftsin the distribu-

11. No single theoristrepresentsthis composite liberal orientation,but a variety of theorists

ests.13 Liberal theories have also identifiedand stressed the importance of

13. For a recentsurvey,see Lisa Martin,"An Institutionalist View: InternationalInstitutionsand

individuals or groups.17 Constructivisttheorysees

17. The rationalistapproach argues thatinstitutionsare essentiallyfunctionalor utilitarian"solu-

This approach to institutionsgives attentionto the ways in which institu-

The most fundamentalstrategicrealityafterWorld War II was the huge dis-

fate of the next generation."22George Kennan, in a major State Department

22. Quoted in Norman Graebner,Americaas a WorldPower:A RealistAppraisalfromWilsontoReagan

tion of secondary states. At the heart of the Western postwar order is an

25. Anotherrecentcase of strategicrestraintis Germanywithinthe European Union: here also is

Limits on power are never as clear-cut,absolute, or guaranteed in relations

STRATEGIC RESTRAINT AND POWER CONSERVATION

31. The argumentthathegemons will want to promotenormativeconsensus among states so as

more predictable,the secondarystatesdo not need to spend as many resources

Figure1. TimeHorizonsand the Returnon Power Assets: LeadingState.

Figure2. TimeHorizonsand the Returnon Power Assets: SecondaryState.

STRATEGIES OF RESTRAINT: BONDING, BINDING, AND VOICE OPPORTUNITIES

democraticstates,could be wielded to facilitatestrategicrestraint.Institutions

32. Obviously,neorealistsargue that such guarantees are never sufficientenough to go forward

ing to predictableinstitutionalrules and procedures,can reassure weaker and

35. Elster,"Introduction,"in Elsterand Slagstad, Constitutionalism

voice opportunitiesfor the weaker states,can be a solution forthese weaker

By design,inadvertence,and structuralcircumstance,the United States was in

three ways: in America's "reluctant"orientationtoward hegemonic domina-

More generally,the overall patternof Americanpostwar policies reflecteda

39. Quoted in "Summary of Discussion on Problemsof Relief,Rehabilitation,and Reconstruction

matelymoved a great distance away fromits originalproposals in settingup

Anotherway in which the open hegemonic order provides reassurances to

45. For the transnationalpoliticalprocess channeled throughthe Atlanticsecurityinstitutiolns,

trade disputes to specificissues, bringinternationalproceduresof review into

noted, alliances have also been created as pacta de contrahendo-pactsof re-

48. See Paul W. Schroeder,"Alliances, 1815-1945:Weapons of Power and Tools of Management,"

The BrettonWoods economic and monetary accords exhibit the logic of

52. See Harold James,International MonetaryCooperationsince BrettonWoods(New York: Oxford

try."54 This logic of Germany'sinvolvementin NATO and the EU was reaffir-

futureresurgentand revisionistRussia.59But others,particularlyin theClinton

The bargains struckand institutionscreated at the early momentsof postwar

59. See Zbigniew Brzezinski,"NATO-Expand or Die?" NezwYorkTimes,December 28, 1994.

returnsto institutions.First,thereare large initialstart-upcosts to creatingnew

American postwar hegemonic order has exhibitedthis phenomenon of in-

66. See Peter J. Katzenstein,"InternationalRelations Theory and the Analysis of Change," in

states to accept, regardless of theirspecificinternationalpower position, the

69. JohnGerard Ruggie, "Multilateralism:The Anatomyof an Institution,"in Ruggie, ed., Multi-

governmentsand theirjointmanagementinstitutions.The steady rise of trade

71. This is the dynamicthatRobertKeohane and JosephNye seek to capturein their"international

preservingthe postwar politicalorder among the major industrialcountriesis

The twentiethcenturymay be ending, but the American centuryis in full

73. A reporterforthe Washington Postrecentlysummarized thisview: "The chorusof dismay with