Vous êtes sur la page 1sur 63

The Poverty of Neorealism

Author(s): Richard K. Ashley

Source: International Organization, Vol. 38, No. 2 (Spring, 1984), pp. 225-286
Published by: The MIT Press
Stable URL: http://www.jstor.org/stable/2706440
Accessed: 04-09-2015 08:01 UTC

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/

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 Organization.


This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism RichardK. Ashley

The theory of knowledge is a dimension of political theory because the

specificallysymbolic power to impose the principlesof the construction
of reality-in particular,social reality-is a major dimension of political
Pierre Bourdieu

It is a dangerousthing to be a Machiavelli. It is a disastrousthing to be

a Machiavelli without virtu.
Hans Morgenthau
Almost six years ago, E. P. Thompson fixed his critical sights across the
English Channel and let fly with a lengthy polemic entitled The Povertyof
Theory. Thompson's immediate target was Louis Althusser. His strategic
objective was to rebut the emergent Continentalorthodoxy that Althusser
championed: structural Marxism, a self-consciously scientific perspective
aiming to employ Marxian categories within a structuralistframeworkto
producetheoreticalknowledgeof the objective structuresof capitalistreality.
The chargesThompson hurled defy brief summary,but some key themes
can be quickly recalled. Althusser and the structuralists,Thompson con-

This articledevelops ideas from a draftpaper,"The Hegemonyof Hegemony,"and "Realist

Dialectics"(Presentedat the September 1982 meeting of the AmericanPolitical Science As-
sociation,Denver, Colo.). My thinkingon this topic has benefitedenormouslyfrom comments
and criticismsgenerouslyprovidedby GordonAdams, HaywardR. AlkerJr., AlbertBergesen,
ChristopherChase-Dunn,Richard Dagger,Felicia Harmer,Robert 0. Keohane, Stephen D.
Krasner,Ivy Lang, Dickinson McGaw, George Modelski, Craig Murphy, Robert C. North,
MarkReader,John G. Ruggie,KennethN. Waltz, and David Winters,and the editorsof 10.
The argumenthere is controversial.It is thereforeall the more noteworthythat, despite deep
differences,communicationswith diverse audiences representingallegedly incommensurable
points of view have been so intelligentand, for me at least, rewarding.To all concernedI offer
thanks-and my exoneration.
1. E. P. Thompson, The Povertyof Theoryand OtherEssays (New York: MonthlyReview
Press, 1978). See also PerryAnderson'srejoinder,ArgumentswithinEnglishMarxism(London:
New Left Books, 1980).
InternationalOrganization38, 2, Spring 1984 0020-8183/84/020225-61 $1.50
C 1984 by the MassachusettsInstituteof Technologyand the World Peace Foundation

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
226 InternationalOrganization

tended, were guilty of an egregiouslyselective, hopelessly one-sided repre-

sentation of the Marxianlegacy they claimed to carryforward.In the name
of science, Althusser had purged the legacy of its rich dialectical content
while imposing a deadeningahistoricalfinalityupon categoriesstolen from
Marx'swork.To producethis backhandedhagiography,Thompson charged,
Althusserhad superimposeda positivistunderstandingof scienceupon Marx
even as he claimed to surpass the limits of positivism. What is worse, his
structuralMarxism had to ignore the historical context of Marx's work,
subordinatethe dialectical"Young Marx"to the objectivist"MatureMarx"
of the Grundrisse,cast disrespecton old Engels, "the clown," and system-
atically forget much of the Marxist literaturesince Marx, includingLenin.
In Thompson's view, this readingof Marx produced a mechanistictheory
of capitalist society-a machine-like model comprised of self-contained,
complete entities or parts connected, activated, and synchronized by all
mannerof apparatuses.It was, Thompsoncomplained,"an orreryof errors."2
Thompson's attack was by no means a plea for fidelityto Marx'soriginal
texts. Rather,it was primarilyconcernedwith restoringa respectfor practice
in history. In Thompson's view, structuralMarxismhad abolished the role
of practicein the constitution of history, includingthe historicalmaking of
social structures.It had produced an ahistorical and depoliticized under-
standing of politics in which women and men are the objects, but not the
makers,of theircircumstances.Ultimately,it presenteda totalitarianproject,
a totalizing antihistoricalstructure,which defeats the Marxian project for
change by replicatingthe positivist tendency to universalizeand naturalize
the given order.
Repeated in the context of current Europeanand Latin American social
theory, non-Marxist as well as Marxist, Thompson's assault might today
seem anachronistic.The fortresshe attackedis alreadyin ruins. In Europe,
at least, the unquestionedintellectualparamountcyof structuralismhas seen
its day. True, Europeansocial theory remains very much indebtedto struc-
turalistthought-that set of principlesand problematicsdifferentlyreflected
in, say, Saussure'slinguistics,Durkheim'ssociology, Levi-Strauss'scultural
anthropology,or Piaget's developmental psychology. Yet today, that debt
is honored not by uncriticaladherenceto structuralistprinciplesbut by the
poststructuralistquestioningof their limits.
On this side of the Atlantic, however, the themes of Thompson's attack
are still worth recalling.For just as the dominance of structuralistthought
is waning elsewhere, North American theorists of internationaland com-
parative politics claim to be at last escapingthe limits of what Piagetcalled
"atomistic empiricism."Just as the United States' position of hegemony in
2. Also called a "planetarium,"an orreryis a mechanicaldevice used to illustratewith balls
of various sizes the relative motions and positions of the bodies in a solar system. It takes its
name from CharlesBoyle, the Earlof Orrery,for whom one was made.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 227

the world economy is called into question, North American theorists of

internationalrelationsareproudlyproclaimingtheirown belated"structuralist
turn." The proponents of this North American structuralisminclude some
of the last two generations'most distinguishedand productive theorists of
international relations and comparative politics: Kenneth Waltz, Robert
Keohane,StephenKrasner,RobertGilpin, RobertTucker,GeorgeModelski,
and CharlesKindleberger,amongmany others.The movementthey represent
is known by many names: modem realism, new realism, and structural
realism, to name a few. Let us call it "neorealism."3
Like Althusserand other proponentsof structuralMarxism,North Amer-
ican proponents of neorealism claim to carry forward a rich intellectual
traditionof long standing.The neorealisttypicallydefineshis or her heritage,
as the name implies, in the Europe-borntradition of "classicalrealism"-
the tradition associated in the United States with Morgenthau,Niebuhr,
Herz, and Kissinger. Like Althusser's structuralism,too, neorealist struc-
turalism claims to surpass its predecessorsby offeringa "truly scientific"
renderingof its subject matter-an objective, theoreticalrendering,which
breaksradicallywith its predecessors'allegedlycommonsensical,subjectivist,
atomistic, and empiricist understandings.Like Althusser's structuralism,
neorealismclaims to graspa structuraltotality that constrains,disposes, and
finally limits political practice. Like Althusser's structuralism,neorealism
has achieved consensusabout the categoriesdefiningthe dominantstructures
of the totality examined:in the case of neorealism,these categoriesrefernot
to social classes and the arenas and instruments of class struggle but to
modern states, their strugglesfor hegemony, and the instrumentsby which
and arenas in which they wage it. And like Althusser'sstructuralMarxism,
neorealism has very quickly become a dominant orthodoxy. In France of
the late 1960s and 1970s, Althusserianstructuralismprovided the pivotal
text upon which the intellectualdevelopment of a generationof radicalphi-
losophers would turn. In the United States of the 1980s, neorealismand its
structuraltheory of hegemony frames the measureddiscourse and ritual of
a generationof graduatestudents in internationalpolitics.
It is time for anotherpolemic. Settingmy sightson neorealiststructuralism,
I offer an argumentwhose main themes closely parallelThompson's attack
on structuralMarxism. I want to challenge not individual neorealists but
3. In speakingof a "neorealismmovement," it is necessaryto confrontseveralissues. First,
the name "neorealism"is not universallyrecognizedby those I am callingneorealists.Some
no doubt assume that theirwork reflectsno largermovement or trendthey themselvesdid not
consciouslyset into motion;they thus rejectthe applicationof generallabelsto theirown work.
Second, I recognizethat the scholarshere regardedas neorealisthave many seriousdifferences
and quarrelsamong themselves. Third, I stress that my treatmenthere is with respectto the
structureof an overall movement in its context and not the expressed pronouncementsor
conscious intentionsof individualscholarswhose work sometimes may, and sometimes may
not, contributeto that movement.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
228 InternationalOrganization

the neorealistmovement as a whole.4LikeThompson'scritique,my argument

has both negative and positive aspects: both its critical attack and its im-
plications for an approachthat would do better. In spirit with Thompson,
let me phrase key themes of that critiquein deliberatelyexaggeratedterms.
On the negative side, I shall contend that neorealism is itself an "orrery
of errors,"a self-enclosed,self-affirmingjoiningof statist,utilitarian,positivist,
and structuralistcommitments. Although it claims to side with the victors
in two American revolutions-the realist revolution against idealism, and
the scientific revolution against traditionalism-it in fact betrays both. It
betraysthe former'scommitment to politicalautonomy by reducingpolitical
practiceto an economic logic, and it neutersthe criticalfacultiesof the latter
by swallowing methodological rules that render science a purely technical
enterprise.From realism it learns only an interest in power, from science it
takesonly an interestin expandingthe reachof control,and fromthis selective
borrowingit creates a theoreticalperspectivethat paradesthe possibilityof
a rationalpower that need never acknowledgepower'slimits. What emerges
is a positivist structuralismthat treats the given order as the naturalorder,
limits ratherthan expands political discourse, negates or trivializesthe sig-
nificance of variety across time and place, subordinatesall practice to an
interestin control, bows to the ideal of a social power beyond responsibility,
and therebydeprives politicalinteractionof those practicalcapacitieswhich
makesociallearningand creativechangepossible.Whatemergesis an ideology
that anticipates,legitimizes, and orients a totalitarianprojectof global pro-
portions:the rationalizationof global politics.'
On the positive side, I shall suggest that theoreticalalternativesare not
exhausted by the false choice between neorealism's"progressive"structur-
alism and a "regression"to atomistic, behavioralist,or, in Waltz's terms,
"reductionist"perspectiveson internationalpolitics.This dichotomyof wholes
and parts,often invoked by neorealistorthodoxy,obscuresanothercleavage
of at least equal importance. This is a cleavage that pits early structuralist
"compliance models" of action and social reality (physicalisticmodels as
seen in earlyDurkheim,for instance)againstdialectical"competencemodels"
(as seen in poststructuralistthought over the last few decades).6Against the

4. As discussed here, neorealism is not just an amalgam of individual scholars'traits or

opinions, nor is it the lowest common denominatoramong them. Rather,my contentionsare
with respectto neorealismas a collective movement or projectemergingin a sharedcontext,
having shared principlesof practice,and observing certain backgroundunderstandingsand
normsthat participantsmutuallyacceptas unproblematicand thatlimitand orientthe questions
raised,the answerswarranted,and the conductof discourseamongneorealists- this regardless
of the fact that the participantsmay not be conscious of (may merely take for grantedthe
universaltruthof) the normsand understandingsintegratingthem as one movement.In Waltz's
now well-knownterminology,mine is a systemic, not a reductionist,accountof the neorealist
5. The term "totalitarian"is, to say the least, provocative. As seen below, my usage is
consistent with that of Hans Morgenthau.
6. This is John O'Neill's terminology.The distinctionwill be elaboratedbelow.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 229

neorealisttendency to march triumphantlybackwardto compliancemodels

of the 19th century,I shall be suggestingthat the rudimentsof an alternative
competence model of internationalpolitics, a model more responsive to
contemporaryarguments in social theory, are already present in classical
realist scholarship.Drawingespecially upon the work of PierreBourdieu,I
shall suggest that a dialectical competence model would allow us to grasp
all that neorealism can claim to comprehend while also recovering from
classicalrealismthose insightsinto politicalpracticewhich neorealismthreat-
ens to purge. Such a model, fully developed, would reinstatethe theoretical
role of practice. It would sharpen the depiction of the currentworld crisis,
includingdilemmas of hegemonicleadership.And it would shed lighton the
role and limits of knowledge, including neorealism, in the production,ra-
tionalization,and possible transformationof the currentorder.
A critiqueof this breadthnecessarilyfindsits inspirationin severalquarters.
In addition to Thompson, I should single out two poststructuralistsources,
one Frenchand one German.The Frenchsourceof inspiration,as indicated,
is primarilyBourdieu'sdialecticalOutlineof a Theoryof Practice.7The Ger-
man source of inspiration is the critical theory of JuirgenHabermas and,
more distantly, the whole tradition of the FrankfurtSchool.8 Habermas's
theme of the "scientizationof politics" is more than faintly echoed in my
critique of neorealism. His diagnosis of a "legitimationcrisis" in advanced
capitalist society complements my discussion of the historicalconditions of
At the same time, the studied parochialismof American international
political discourse would make it too easy to deploy alien concepts from
Europeansocial theoryto outflank,pummel, and overwhelmthat discourse.
Such a strategywould be self-defeatinggiven my intentions. My arguments
here, intentionally phrased in provocative terms, are like warning shots,
meant to provoke a discussion, not destroy an alleged enemy. Thus, I feel
an obligation to present my position in "familiar"terms, that is, in a way
that makesreferenceto the collectiveexperiencesof North Americanstudents
7. Pierre Bourdieu,Outline of a Theoryof Practice, trans. by Richard Nice (Cambridge:
CambridgeUniversityPress, 1977). See also Michel Foucault,Power/Knowledge: SelectedIn-
terviewsand Other Writings,1972-1977, ed. and trans. by Colin Gordon et al. (New York:
Pantheon, 1980); Foucault,Language, Counter-Memory,Practice(Oxford:Blackwell,1977);
Foucault,TheArchaeologyof Knowledge(New York:Pantheon,1972);and Foucault,TheOrder
of Things(New York: Pantheon, 1970).
8. Jiirgen Habermas, Towardsa Rational Society, trans. by Jeremy J. Shapiro (London:
Heinemann, 1971); Habermas, Theoryand Practice,trans. by John Viertel (Boston: Beacon
Press, 1974); Habermas, Communicationand the Evolution of Society, trans. by Thomas
McCarthy(Boston:Beacon Press, 1979).
9. JuirgenHabermas,LegitimationCrisis,trans.by Thomas McCarthy(London:Heinemann,
1976). Of course, the figurescited can hardlybe said to occupy one school; in fact, there are
very sharpdifferencesamong them. Thompson, for instance,would be among the last to align
happily with Foucault, "Althusser'sformerstudent";Habermas'srationalismwould set him
apartfrom Bourdieu.On the theme of the "economization"of politics,see also HannahArendt,
The Human Condition(Chicago:Universityof ChicagoPress, 1958).

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
230 InternationalOrganization

of internationalrelations.As it turns out, this is not so hard to do. For those

experiencesare not nearly so impoverishedas the keepersof neorealistlore
would make them seem.

1. The lore of neorealism

Everygreatscholarlymovement has its own lore, its own collectivelyrecalled

creation myths, its ritualizedunderstandingsof the titanic strugglesfought
and challenges still to be overcome in establishingand maintainingits par-
amountcy. The importance of this lore must not be underestimated:to a
very considerabledegree, the solidarityof a movement depends upon the
members' abilities to recount this lore and locate their every practicein its
terms. Small wonder, therefore,that rites of passage,such as oral qualifying
examinations,put so much stresson the student'sability to offera satisfying
reconstructionof the movement's lore and to identify the ongoing struggles
that the student,in turn, will continueto wage.Two generationsago, aspiring
North American students of internationalrelationshad to show themselves
readyto continueclassicalrealism'snoblewaragainstan entrenchedAmerican
idealism. A generation ago, they had to internalizeanother lore: they had
to sing the battle hymns of behavioralscience triumphantagainsttradition-
alism. Today, thanks to the emergence of a neorealistorthodoxy, students
must preparethemselves to retell and carry forwardyet another lore.

a. The triumphof scientificrealism

The lore of neorealism might be retold in several ways, and each telling
might stress different heroes, but a central theme is likely to remain the
same. Neorealism, according to this theme, is a progressive scientific re-
demption of classical realist scholarship.It serves the interests of classical
realism under new and challengingcircumstancesand as advantagedby a
clearer grasp of objective science's demands and potentialities. As such,
neorealismis twice blessed. It is heir to and carriesforwardboth of the great
revolutions that preceded it: realism against idealism, and science against
A fuller recountingof the lore would begin by diagnosingsome lapses in
the classical realist scholarshipof, say, Morgenthau,Kissinger,and Herz. In
neorealisteyes, and for reasons considered below, these and other classical
realistswere quite correctin their emphasis on power, nationalinterest,and
the historically effective political agency of the state. Unfortunately,when
held up to modem scientificstandardsof theory,these classicalrealistscholars
seemed to fall woefully short. Four lapses in the classical heritagemight be
First, classical realist concepts, arguments,and knowledge claims might

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 231

be said to be too fuzzy, too slippery,too resistantto consistent operational

formulation, and, in application, too dependent upon the artful sensitivity
of the historicallyminded and context-sensitivescholar.Somehow, classical
realistconcepts and knowledgeclaims never quite ascend to Popper's"third
world of objective knowledge,"because classical realistshold that the truth
of these concepts and claims is to be found only throughthe situation-bound
interpretationsof the analyst or statesman.10
Second, and closely related, classical realists might be said to distinguish
insufficientlybetweensubjectiveand objectiveaspectsof internationalpolitical
life, thereby underminingthe building of theory. Such a concern is to be
found,for example,in Waltz'scomplaintsaboutMorgenthau'sand Kissinger's
understandingsof the internationalsystem.They are,forWaltz,"reductionist"
becausethey tend to accordto the "attribute"of actors'subjectiveperceptions
an importantrole in constitutingand reproducingthe "system."They thereby
deny the system a life of its own as an objective social fact to be grasped
by theory."
Third, it might be claimed that, in Gilpin'swords, classicalrealistscholar-
ship "is not well groundedin social theory."'2For all its strengths,classical
realism could be claimed to exhibit a lamentable lack of learningfrom the
insights of economics, psychology, or sociology.
The fourth lapse, however, is the most salient from the neorealist point
of view, for it marksboth a failureof realistnerve and a point of considerable
vulnerabilityin the defense of a key realist principle:the principleof "the
autonomy of the political sphere." Classical realists limited themselves to
the domain of political-militaryrelations,where balance of power could be
granted the status of a core concept. As a result, realism was naive with
respect to economic processes and relations;it left them to the power-blind
eyes of liberal interdependencethinkersand the questioningeyes of radical
theorists of dependency and imperialism.As neorealistssee it, this was not
just a matterof rivalrybetweenscholarlyparadigms.Sinceeconomicprocesses
and relationshave definitepower-politicalramificationsover the longerterm,
and since these same processes are badly describedby referenceto balance-
of-power logics, classical realism's blindness with respect to economics had
several relatedeffects:it situatedinterstatepolitics in a reactive"superstruc-
tural"pose vis-a-vis economic dynamics,renderedclassicalrealismincapable

10. Karl Popper, "On the Theory of Objective Mind," and "Epistemology without a Knowing
Subject," in Objective Knowledge (London: Oxford University Press, 1972). As Morgenthau
says again and again, the application of every universalizing formulation "must be filtered
through the concrete circumstances of time and place." Hans J. Morgenthau, Politics among
Nations: The Struggle for Power and Peace, 5th ed. (New York: Knopf, 1978), p. 8.
1 1. See Kenneth N. Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley,
1979), pp. 62-64.
12. Robert Gilpin, "Political Change and International Theory" (Paper presented at the
annual meeting of the American Political Science Association, New York, 3-6 September 198 1),
p. 3.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
232 InternationalOrganization

of graspingpolitical-economicdilemmas, and limited realism's capacity to

guide state practice amidst these dilemmas. Given all of this, and given a
period of world economic crisis that increasinglycalls into question states'
capacitiesto justifythemselvesas managersof economicdysfunctions,realism
was in dangerof failingin one of its foremost functions:as a frameworkthat
could be deployed to legitimize and orient the state.
This situationand the neorealistresponsecan be phrasedin more definite
terms. In a period of world economic crisis, wellingtransnationalistoutcries
againstthe limits of the realistvision, and evidentlypoliticizeddevelopments
that realism could not comprehend,the classicalrealisttraditionand its key
conceptssuffereda crisisof legitimacy,especiallyin the United States.Sensing
this crisis, a number of American scholars, most of whom are relatively
young and very few of whom are steeped in the classic tradition, more or
less independently undertook to respond in a distinctly American fashion;
that is, scientifically.3 They set out to develop and historically to corroborate
scientific theories that would portray or assume a fixed structureof inter-
national anarchy;'4trim away the balance-of-powerconcept's scientifically
inscrutableideologicalconnotations;reducebalanceof power'sscientificstatus
to that of a systemic propertyor a situationallogic undertakenby rational,
calculating,self-interestedstates;and, most importantly,disclose the power-
political strugglefor hegemony behind the economic dynamics that liberal
and radicalanalysts had too often falsely treatedin isolation from interstate
politics.'5More than that, they set out to construct theories that would lay
bare the structuralrelations-the causal connections between means and
ends-that give form to the dynamics of hegemonic rise and decline and in
light of which a hegemon might orient its effortsboth to secureits hegemony
and to preserve cooperative economic and ecological regimes. Political-
economic orderfollows from the concentrationof political-economicpower,
say these theories. Power begets order. Order requires power. The realist
emphasis on the role of state power had been saved.

13. As I shall indicatebelow, neorealismholds to a very definite,highlyrestrictivemodel of

social science.
14. A few neorealistsare extremely hostile to the use of the wordanarchy(e.g., as used in
Waltz'swork),even thoughthey acceptthe absenceof centralrule(Waltz'sdefinitionof anarchy)
as a hard-coreassumption.GeorgeModelskitakes "worldleadership"as his "centralconcept."
Thus, he writes, "we make it clear that we do not regardthe modern world as some sort of
anarchicalsociety. To the contrary,our analysisclarifiesthe principlesof orderand authority
that have governedthat worldfor the past half millenniumand that, whilefamiliarto historians
in each particularinstance, have not been previouslyput togetherin quite this manner and
have been generallyunfamiliarto studentsof internationalrelations.Anarchycould be in the
eye of the beholder."Modelski,"LongCyclesand the Strategyof U.S. InternationalEconomic
Policy,"in WilliamP. Avery and David P. Rapkin,eds., Americain a ChangingWorldPolitical
Economy (New York: Longman, 1982), p. 99.
15. Again, neorealistsdiffer,and the words they choose to use is one of the differences.One
mightspeakof order,anotherof stability,and still anotherof leadership.The word"hegemony"
itself is certainlyin some dispute,even though all agreethat hegemony(whateverone chooses
to call it) follows from power or the distributionof the attributesof power.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 233

According to neorealist lore, this rescue of realist power politics was by

no means a paltry act. It was, if anything, heroic. For it depended, above
all, upon one bold move: a move of cunning and daring against stiff odds
and in opposition to the mass of sedimentedsocial-scientifichabits. In order
to bringscience to bear in saving and extendingrealism,neorealistshad first
to escape the limits of logical atomism, then prevailingamong "scientific"
approachesto the study of internationalrelations.To do this, they adopted
a critical stance with respect to "reductionist"arguments,arguments that
would reduce "systems" to the interactionsamong distinct parts. In their
place, neorealistserectedwhat have come to be called "systemic,""holistic,"
or "structuralist"arguments.
For the neorealistrescue of realist power politics, this structuralistmove
was decisive. By appeal to objective structures,which are said to dispose
and limit practicesamong states (most especially, the anarchicstructureof
the modern states system), neorealistsseemed to cut throughthe subjectivist
veils and dark metaphysicsof classical realist thought. Dispensingwith the
normatively laden metaphysics of fallen man, they seemed to root realist
power politics, including concepts of power and national interest, securely
in the scientificallydefensible terrain of objective necessity. Thus rooted,
realist power politics could be scientificallydefended againstmodernistand
radical critics. Without necessarily denying such tendencies as economic
interdependenceor uneven development,neorealistscould arguethat power-
political structureswould refractand limit the effects of these tendencies in
ways securingthe structuresthemselves.
Such is the stuff of legends. Even in neorealistlore, to be sure, this rev-
olutionarystructuralistturn is only part of neorealism'sstory. The graduate
student going throughneorealistrites of passagewould have to graspa good
deal more. As will become clear later, the aspiringstudent would also have
to come to grips with neorealistperspectiveson internationalcollaboration
and the role of regimes, on the role and limits of ideology, and on the
dynamics of hegemonic succession and "system change."Most of all, he or
she would have to demonstrate an ability to interpret state practices in
neorealistterms, which is to say as calculating,"economically"rationalbe-
haviors under constraints. Still, it is the structuralistturn that is decisive,
the sine qua non of neorealism'striumph. Let us take a closer look at this
vaunted structuralistaspect.

b. The structuralistpromise
As John Ruggie has been among the first to point out, the promise of
neorealism, like the promise of Immanuel Wallerstein'sworld systems per-
spective, is in very large measure attributableto its structuralistaspect.'6
16. John G. Ruggie, "Continuity and Transformation in the World Polity: Toward a Neorealist
Synthesis," World Politics 35 (January 1983), especially pp. 261-64.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
234 InternationalOrganization

Ruggie is right. There are indeed certain isomorphismsbetween aspects of

neorealist argumentand elements of structuralistargument(as seen in the
work of, say, Saussure,Durkheim, Levi-Strauss,and Althusser).Noting the
isomorphisms,one can let neorealismbaskin the reflectedgloryof yesteryear's
structuralisttriumphs in fields such as linguistics,sociology, anthropology,
and philosophy. One can say that structuralism'ssuccesses in other fields
suggest neorealism'spromise for the study of internationalrelations.'7
At the risk of oversimplification,it is possible to abstract a number of
more or less continuous "elements" of structuralistthought. Five of these
elements-overlapping aspects,really-are especiallyimportantfor my pres-
ent purposes.They suggestsome of the parallelsbetweenneorealistargument
and structuralistargumentin general.
1. Wherever it has emerged, structuralistargument has taken form in
reactionagainstphenomenologicalknowledgeand speculative,evolutionary
thought.'8Structuralistthought breaksradicallywith the formerbecause of
phenomenology'sdebt to a conscious subjectivitythat, in structuralisteyes,
is always suspect. It poses precisely the question that phenomenological
knowledge excludes: how is this familiar apprehensionof the given order,
and hence the community itself, possible? Structuralismalso breaks with
speculative, evolutionary thought, regardingit as nothing more than the
"other side" of phenomenology.Evolutionarythought too often fails to see
that what pretends to promise change is but an expression of continuity in
the deeper order of things.
2. Structuralistargumentaims to constructthe objectiverelations(linguistic,
economic, political, or social) that structurepracticeand representationsof
practice,includingprimaryknowledgeof the familiarworld.'I Human con-
duct, includinghuman beings' own understandings,is interpretedas surface
practicegeneratedby a deeper, independentlyexisting logic or structure.In
strivingto comprehendthis deeper logic, structuralismbreakswith individ-
ualist perspectiveson social subjectivity,as in the Cartesiancogito. In the
same stroke, it attempts to transcendthe subject/objectdualism. For struc-
turalism, to simplify, social consciousness is not "transparentto itself." It
is generatedby a deep social intersubjectivity-linguisticrules,for example-
which is itself regardedas the objectivestructureof society. In Paul Ricoeur's
words, "Structuralismis predicatedon a Kantian rather than a Freudian

17. See Jonathan Culler, Structuralist Poetics (London: Routledge & Kegan Paul, 1975); Edith
Kurzweil, The Age of Structuralism (New York: Columbia University Press, 1980); Anthony
Giddens, Central Problems in Social Theory (Berkeley: University of California Press, 1979),
especially chap. 1; Paul Ricoeur, "Structure and Hermeneutics," and "Structure, Word, Event,"
in Conflict of Interpretations: Essays in Hermeneutics (Evanston, Ill.: Northwestern University
Press, 1974); and Miriam Glucksman, Structuralist Analysis in Contemporary Social Thought
(London: Routledge & Kegan Paul, 1974).
18. Giddens, Central Problems, p. 9.
19. Bourdieu, Outline, p. 3.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 235

unconscious, on structuralimperativesthat constitute the logical geography

of mind."20
3. Thus, structuralismshifts toward the interpretationof practicefrom a
social, totalizing,or "systemic"point of view. Ferdinandde Saussure'sdis-
tinction between speech and language (parole et langue) is paradigmatic;
what concerned him was not speech per se but the logical conditions of its
intelligibility(an inherentlysocial or "systemic"concern).2'What concerns
structuralistsin generalis not practiceper se but the logical conditions that
account for the significanceand significationof practicewithin a community
(again, a social or "systemic"relation).Saussurelocated his logical precon-
ditionsfor the intelligibilityof speechin language:speechbecomesthe product
of language.Structuralistsin generallocate their explanationsin deep social
structures:practicebecomes the productof structure.For Saussure,language
contained possible speech, that is, speech that will be understoodwithin the
languagecommunity. For structuralistsmore generally,structureis a system
of constitutive rules "which do not regulatebehavior so much as create the
possibility of particularforms of behavior."
4. Consistentwith its totalizinginclinations,structuralismpresupposesnot
only the priorityof structureover practice but also the "absolute predom-
inance of the whole over the parts."22Structuralistsemphasizethe "system"
not only in contrast to but also as constitutiveof the elements that compose
it. The overall structureexists autonomously, independent of the parts or
actors, and the identities of the constituent elements are attributednot to
intrinsic qualities or contents of the elements themselves but to the differ-
entiationamong them suppliedor determinedby the overall structure.Thus,
the units have no identity independent of the structuralwhole. Saussure's
position is again exemplary:"In language,"he wrote, "there are only dif-
ferences.... [A] differencegenerallyimplies positive terms between which
the differenceis set up; but in languagethere are only differenceswithout
positive terms."23
5. In their treatmentsof time and change, structuralistargumentstend to
presupposean absolutedistinctionbetweensynchronic(static)and diachronic
(dynamic)viewpoints, and they tend to accentuatethe one-way dependence
of diachrony (dynamics) upon synchrony (statics).24Change, for the struc-
turalist, is always to be grasped in the context of a model of structure-an
elaborated model whose elements are taken to be fixed and immutable in
the face of the changes it conditions and limits.
Cursorythough it is, this listing suggests some obvious correspondences

20. Ricoeur,"Structureand Hermeneutics,"p. 79.

21. Bourdieu,Outline,pp. 23-24.
22. BertellOllman,Alienation:Marx'sConceptionof Man in CapitalistSociety,2d ed. (Cam-
bridge:CambridgeUniversity Press, 1976), p. 266.
23. Quoted in Giddens, CentralProblems,p. 12.
24. Ibid., p. 46.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
236 InternationalOrganization

between neorealistargumentand some of the fundamentalsof structuralism.

Consider the first "element": neorealism's criticism of classical realism's
subjectivisttendencies(the tendenciesof Morgenthauand Kissinger,among
others, to adopt the posture of an ethnomethodologistof a diplomaticcom-
munity) closely parallelsstructuralism'sreaction against phenomenological
knowledge. The neorealist reaction to the writingsof transnationalistsand
modernists similarly parallelsstructuralism'sattitude with respect to spec-
ulative, evolutionary thought. The shallow analysis behind such writings,
neorealiststend to feel, mistakesthe ephemeralfor the eternaland too eagerly
seizes upon epiphenomenalchange as evidence of system change.
The second,third,and fourth"elements"are equallysuggestiveof parallels.
It might be argued,for example, that the centralimportanceof Waltz'swell-
known work lies in its attempt to realize these "elements"for the study of
internationalpolitics. Waltz's argumentagainst "attributetheories"and on
behalfof "systemic"theoriesmightseem to locate the properobjectof theory
not in "parts,"and not in externalrelationsamongthem, but in independently
existing objective "wholes," which, as orderingand orientingpropertiesof
a system, constitute parts and generaterelationsamong them. His argument
clearly adopts a totalizing stance in that he focuses not on explainingthe
variety of foreign-policybehavior per se (such behavior remains indeter-
minate) but on uncovering the objective structuresthat determine the sig-
nificanceof practicewithin the context of an overallsystem. And while Waltz
allows that there may be considerablevariety among "actors,"only those
forms of differentiationsignificantwithin the overall structure,namely dis-
tributionsof capabilities,are of concern to his theory.
Finally, the fifth "element" of structuralistargument,having to do with
time and change,finds expressionin neorealism:RobertGilpin'srecent War
and Change in WorldPolitics offers one example, George Modelski's im-
portant "long-cycle" argument another.25Indeed, the preoccupationwith
cycles of hegemonic rise and decline would seem near-perfectlyto illustrate
the structuralisttendency to emphasize synchrony over diachrony. As in
structuralistthought, dynamics of change are of concern to neorealistspri-
marily insofar as their structuraldeterminantscan be theoreticallygrasped.
In view of these isomorphisms,it is easy to see why neorealismmight be
viewed as a "welcome antidote" to the "prevailingsuperficiality"of much
international relations discourse. If nothing else, neorealists, like Waller-
steinians, have illustratedthat scientificinternationalrelationsdiscoursecan
entertainstructuralistarguments,can transcendempiricistfixations,and can
in principleescape the limits of logicalatomism. At least, researchprograms
now purportto try. In turn, the field is encouragedto recognizethat reality

25. Robert Gilpin, War and Change in WorldPolitics (New York: CambridgeUniversity
Press, 1981); George Modelski, "The Long-Cycleof Global Politics and the Nation-State,"
ComparativeStudies in Society and History20 (April 1978), pp. 214-35.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 237

is not all "on the surface,"that it has, or might have, depth levels, that an
adequate social or political analysis cannot be reduced to a concatenation
of commonsense appearances,and that one can look for a unity behind and
generatingevident differences.Herein is the neorealistpromise.
If neorealism is to bathe in the glow of structuralistaccomplishments,
however, it must also be preparedto suffercriticisms as to structuralism's
limits. Above all, such critiquesstress the troublingconsequencesof struc-
turalism'stendency to "put at a distance, to objectify,to separateout from
the personal equation of the investigatorthe structureof an institution, a
myth, a rite."26In trying to avoid "the shop-girl'sweb of subjectivity"or
"the swampsof experience,"to use Levi-Strauss'swords,structuralistsadopt
a posture that denies the role of practicein the making and possible trans-
formationof social order. In part, of course, such critiquesare animatedby
revulsionat structuralism's"scandalousanti-humanism."27 But in part,also,
they are animated by a concern for the disastrousconsequencesfor political
theory and the possibly dangerousconsequences for political practice. An
adequate critique of neorealism must develop these themes.

2. The structure of neorealist structuralism: an orrery of errors

I am, however, a step or two ahead of myself. I have so far spoken only of
the neorealistlore, includingthe structuralistpromise neorealismoften pur-
ports to bear. I have tried to assay that promise by drawingout parallels
between neorealistargumentand the now classic positions of structuralism.
Still, such comparisonsare more than a triflemisleading.For there is at once
more and less to neorealism than might be inferredfrom its isomorphisms
with structuralistargument.There is more to neorealism in that it exhibits
three furthercommitments: statist, utilitarian,and positivist. There is less
to neorealism in that, thanks to the prioritygiven to these commitments,
neorealist "structuralism"takes a shallow, physicalisticform-a form that
exacerbatesthe dangerswhile negatingthe promise of structuralism.
Within neorealism, I suggest, structuralism,statism, utilitarianism,and
positivism are bound together in machine-like, self-enclosing unity. This
machine-likejoining of commitments appearsas if designedto defy criticism
or to drawall oppositioninto its own self-centeredarc. Hereinis neorealism's
answer to Althusser's "orrery"-an orreryof errors. Far from questioning
commonsenseappearances,the neorealistorreryhypostatizesthem. Far from
expandinginternationalpoliticaldiscourse,the neorealistorreryexcludes all
standpointsthat would expose the limits of the given orderof things. Before
26. Ricoeur,as quoted in Paul Rabinowand William M. Sullivan,eds., InterpretiveSocial
Science:A Reader(Berkeley:Universityof CaliforniaPress, 1979), pp. 10-1 1.
27. Giddens, Central Problems, p. 38.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
238 InternationalOrganization

returningto the matter of neorealist "structuralism,"let me take up each

of the other elements of this orrery-neorealism's statist, utilitarian,and
positivist commitments-in turn.

a. Statism
Neorealism is bound to the state. Neorealist theory is "state centric" or
"statist," as Krasnerhas labeled the position.28It offers a "state-as-actor"
model of the world. So long as one proposes to be understoodamong neo-
realists, one must work within this model. At a minimum, this means that
for purposesof theory,one must view the state as an entity capableof having
certainobjectivesor interestsand of decidingamongand deployingalternative
means in their service. Thus, for purposesof theory,the state must be treated
as an unproblematicunity:an entity whose existence,boundaries,identifying
structures, constituencies, legitimations, interests, and capacities to make
self-regardingdecisionscan be treatedas given, independentof transnational
class and human interests,and undisputed(except,perhaps,by other states).
In all of these respects, the state is regardedas the stuff of theorists' un-
examined assumptions-a matter upon which theorists will consensually
agree, and not as a problematicrelationwhose consensualacceptanceneeds
explanation.The propositionthat the state might be essentiallyproblematic
or contested is excluded from neorealisttheory. Indeed, neorealisttheory is
preparedto acknowledgeproblems of the state only to the extent that the
state itself, within the frameworkof its own legitimations,might be prepared
to recognizeproblems and mobilize resourcestoward their solution.
True, individual neorealists sometimes allow that the theoretical com-
mitment to the state-as-actorconstructinvolves a distortionof sorts. Waltz,
for instance,writesthat he "canfreelyadmit that statesarein fact not unitary,
purposive actors.""9Gilpin can acknowledgethat, "strictlyspeaking,states,
as such, have no interests, or what economists call 'utility functions,' nor
do bureaucracies,interest groups, or so-called transnationalactors, for that
matter." He can even go on to say that "the state may be conceived as a
coalition of coalitions whose objectives and interestsresult from the powers
and bargainingamong the several coalitions comprisingthe largersociety
and political elite."30And Keohane, as coauthor of Power and Interdepen-
dence, can certainly recognize that the conditions of "complex interdepen-
dence," includingthe fact of transnationaland transgovernmentalrelations,
fall well short of the "realist"assumption that states are "coherent units"
with sharp boundariesseparatingthem from their externalenvironments.3'
28. See Stephen D. Krasner, Defending the National Interest: Raw Materials Investments
and U.S. ForeignPolicy (Princeton:PrincetonUniversityPress, 1978).
29. Waltz, Theory of International Politics, p. 91.
30. Gilpin, War and Change, p. 18.
31. Robert 0. Keohane and Joseph S. Nye, Power and Interdependence: World Politics in
Transition(Boston:Little, Brown, 1977), especiallychap. 2.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 239

The issue, however, is the theoreticaldiscourse of neorealismas a move-

ment, not the protectiveclausesthat individualneorealistsdeploy to preempt
or deflectcriticismsof that discourse'slimits. Once one entersthis theoretical
discourse among neorealists, the state-as-actormodel needs no defense. It
stands without challenge. Like Waltz, one simply assumes that states have
the status of unitaryactors.32 Or, like Gilpin, one refuses to be deterredby
the mountainousinconsistenciesbetween the state as a coalitionof coalitions
(presumablyin opposition to the losing coalitions againstwhich the winning
coalition is formed) and the state as a provider of public goods, protector
of citizens' welfare, and solver of the free-riderproblem in the name of
winners and losers alike. Knowing that the "objectivesand foreign policies
of statesare determinedprimarilyby the interestsof theirdominantmembers
or rulingcoalitions,"3one nonethelesssimplyjoins the victorsin proclaiming
the state a singularactor with a unified set of objectives in the name of the
collectivegood. This proclamationis the startingpointof theoreticaldiscourse,
one of the unexamined assumptions from which theoretical discourse
In short, the state-as-actor assumption is a metaphysical commitment
priorto science and exempted from scientificcriticism.Despite neorealism's
much ballyhooedemphasis on the role of hardfalsifyingtests as the measure
of theoreticalprogress,neorealismimmunizes its statist commitments from
any form of falsification.Excluded, for instance, is the historicallytestable
hypothesis that the state-as-actorconstructmight be not a first-ordergiven
of internationalpolitical life but part of a historicaljustificatoryframework
by whichdominantcoalitionslegitimizeand secureconsentfortheirprecarious
conditions of rule.
Two implicationsof this "state-centricity,"itself an ontological principle
of neorealisttheorizing,deserve emphasis. The first is obvious. As a frame-
work for the interpretationof internationalpolitics, neorealisttheory cannot
accordrecognitionto-it cannot even comprehend-those globalcollectivist
concepts that are irreducibleto logical combinations of state-bounded re-
lations.In other words,globalcollectivistconcepts-concepts of transnational
classrelations,say, or the interestsof humankind-can be grantedan objective
status only to the extent that they can be interpretedas aggregations of
relations and interests having logically and historically prior roots within
state-boundedsocieties. Much as the "individual"is a prism throughwhich
methodologicalindividualistscomprehend collectivist concepts as aggrega-
tions of individualwants,needs,beliefs,and actions,so also does the neorealist
refractall global collectivist concepts through the prism of the state.34Im-
32. Waltz, Theoryof InternationalPolitics, p. 91.
33. Gilpin, Warand Change,p. 19.
34. Popper understandsmethodologicalindividualismas the principlethat "all social phe-
nomena, and especiallythe functioningof all social institutions,should always be understood
as resultingfrom the decisions, actions, attitudes,etc. of human individuals.... [W]e should
neverbe satisfiedby an explanationin termsof so-called'collectives.'" KarlPopper,The Open
Society and Its Enemies, vol. 2 (London:Routledge, 1966), p. 98. Takingstates as the living
individualsof internationallife, neorealiststatism is understandablein analogousterms.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
240 InternationalOrganization

portantly, this means that neorealist theory implicitly takes a side amidst
contending political interests. Whatever the personal commitments of in-
dividualneorealistsmightbe, neorealisttheoryallieswith, accordsrecognition
to, and gives expression to those class and sectoral interests (the apexes of
Waltz's domestic hierarchiesor Gilpin's victorious coalitions of coalitions)
that are actuallyor potentiallycongruentwith state interestsand legitimations.
It implicitlyopposes and denies recognitionto those class and humaninterests
which cannot be reducedto concatenationsof state interestsor transnational
coalitions of domestic interests.
The second implicationtakes longerto spell out, for it relatesto neorealist
"structuralism"-the neorealist position with respect to structuresof the
internationalsystem. Reflecting on the fourth element of structuralistar-
gument presented above, one might expect the neorealist to accord to the
structureof the internationalsystem an identity independentof the parts or
units (states-as-actorsin this case);the identitiesof the unitswouldbe supplied
via differentiation.The neorealistorrerydisappointsthese expectations,how-
ever. For the neorealist, the state is ontologically prior to the international
system. The system's structureis produced by definingstates as individual
unities and then by noting propertiesthat emerge when several such unities
are brought into mutual reference. For the neorealist, it is impossible to
describe internationalstructureswithout first fashioning a concept of the
The proper analogy, as Waltz points out, is classical economic theory-
microtheory,not macrotheory.As Waltz puts it, "International-political sys-
tems, like economic markets, are formed by the coaction of self-regarding
units." They "are individualistin origin, spontaneouslygenerated,and un-
intended."35Other neorealists would agree. Gilpin, for example, follows
economists Robert Mundell and Alexander Swoboda in defining a system
as "an aggregationof diverse entities united by regularinteractionaccording
to a form of control."36He then names states as "the principalentities or
actors," and he asserts that control over or governanceof the international
system is a function of three factors, all of which are understood to have
their logicaland historicalroots in the capabilities,interests,and interactions
of states: the distributionof power among states, the hierarchyof prestige
among states, and rightsand rulesthat have their "primaryfoundation... in
the powerand interestsof the dominantgroupsor statesin a social system."37
For Gilpin, as for other neorealists, the structureof internationalpolitics,
far from being an autonomous and absolute whole that expresses itself in
the constitution of acting units, is an emergent property produced by the
joining of units having a prior existence.
35. Waltz, Theoryof InternationalPolitics, p. 91.
36. Robert A. Mundelland AlexanderK. Swoboda, eds., MonetaryProblemsin the Inter-
national Economy (Chicago:University of Chicago Press, 1969), p. 343; Gilpin, War and
Change,p. 26.
37. Gilpin, Warand Change,p. 25.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 241

Ruggie's recent review of Waltz's Theoryof InternationalPolitics brings

this point home by diagnosinga lapse in Waltz's "structuralism."Informed
by structuralistliteratures,Ruggieconsidersthe three analyticalcomponents
(or "depth levels") of Waltz's political structure-organizational principle,
differentiationof units, and concentrationor diffusion of capabilities-and
pinpoints what he takes to be a problem:
... [A] dimension of change is missing from Waltz's model. It is miss-
ing because he drops the second analyticalcomponent of political struc-
ture, differentiationof units, when discussinginternationalsystems.
And he drops this component as a result of giving an infelicitousinter-
pretationto the sociologicalterm "differentiation,"taking it to mean
that which denotes differencesratherthan that which denotes
The alleged problem, in other words, is that Waltz has misunderstoodthe
structuralistposition on identityand difference(the fourthelement presented
above). Ruggie moves to put it right by restoringthe second "depth level"
of politicalstructure,now as principlesof differentiationthat tell us "on what
basis" acting units are individuated.Specifically,he contends that there are
contrastingmedieval and modem variantsof the second depth level of struc-
ture:a "heteronomous"institutionalframeworkfor the medieval versus the
modern institutionalframeworkof "sovereignty."Ruggie'sargumentis im-
portant. From a genuine structuralistpoint of view, it is indispensable.
Ruggie introduces his argument as a contributionto a "neo-realistsyn-
thesis," it is true, and he couches it in an extremelygenerousinterpretation
of Waltz'stheory.By posingand tryingto repairthe problemof differentiation
in Waltz'stheory,however,Ruggieindirectlyissueswhatis so farthe strongest
critiqueof the structuralistpretensionsin Waltz's neorealism.By posing the
problem of differentiationfrom a structuraliststandpoint,Ruggieinvites us
to wonder why neorealists, most especially Waltz, had not considered the
problem before. The answeris simple: neorealismis statist before it is struc-
turalist.From a neorealistpoint of view, Ruggie'sargumentis simply super-
fluous because it treats as problematic, and hence in need of a structural
accounting,what neorealistsinsist on treatingas unproblematic-the identity
of the state.
In neorealist eyes, there is nothing "infelicitous"about Waltz's interpre-
tation of differentiation.When Waltz takes differentiationto refer to spec-
ification of the "functions performed by differentiatedunits," he is giving
the only interpretationpossible from a neorealist standpoint.39There is no
need to decide the basis upon which units are individuated, because the
essential individualityof states is alreadytaken for granted.It is embedded
in a definition of sovereignty that neorealists accord to states independent

38. Ruggie,"Continuityand Transformation,"pp. 273-74, emphasisin original.

39. Waltz, Theory of International Politics, p. 93.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
242 InternationalOrganization

of the system. For Waltz, "To say that a state is sovereign means that it
decides for itself how it will cope with its internaland external problems."
For Gilpin, "The state is sovereign in that it must answer to no higher
authorityin the internationalsphere."40Whetherit be one state in the lone
isolation of universal dominion or many interacting,the definition is the
Ruggie'scritiqueof Waltz has a familiarring. His position vis-a-vis Waltz
is not unlike the critiqueof "utilitarianindividualism"in the work of Durk-
heim, upon whom Ruggie draws. "The clincherin Durkheim'sargument,"
writes John O'Neill, "is his demonstrationthat modern individualismso far
from creating industrial society presupposes its differentiationof the so-
ciopsychicspace which createsthe concepts of personalityand autonomy.""4
The clincherin Ruggie'sargumentis his attempt to show that the sovereign
state, so far from creatingmodern internationalsociety, presupposesinter-
national society's production of the sociopolitical space within which sov-
ereigntycould flourishas the modem conceptof internationalpoliticalidentity
and liberty.

b. Utilitarianism
The aptness of the analogy is no accident. For if neorealism'sfirst com-
mitment is to statism, its second commitment is to a utilitarianperspective
on action, social order, and institutionalchange. By utilitarianism,I do not
mean the moral philosophy often associated with Bentham and Mill-a
philosophy that holds, for example, that the proper measure of the moral
worth of acts and policies is to be found in the value of their consequences.
My usageof the termis broader,much morein the sociologicalsenseemployed
by Durkheim, Polanyi, Parsons, and, more recently, Brian Barry, Charles
Camic, and MichaelHechter.i2As these people have made clear,sociological
and utilitarianpositions stand sharply opposed. As Camic argues, modem
sociology emerged as the critiqueof utilitarianism.43 Still, the utilitarianpo-
sition has refused to die. Indeed, the utilitarianperspective-first outlined
by Hobbesand Mandeville,evolvingthroughthe classicalpoliticaleconomists,
and findingmore recentexpressionin the writingsof von Mises and Hayek-
has "been making steady inroads into the territorythat sociology had tra-

40. Ibid., p. 96; Gilpin, Warand Change,p. 17.

41. John O'Neill, "The HobbesianProblemin Marxand Parsons,"in O'Neill, Sociologyas
a Skin Trade(New York: Harper& Row, 1972), pp. 195-96.
42. See BrianBarry,Sociologists,Economists,andDemocracy(Chicago:Universityof Chicago
Press, 1970);Talcott Parsons,The Structureof Social Action(New York:McGraw-Hill,1937);
Karl Polanyi, The Great Transformation(1944; rpt. Boston: Beacon Press, 1957), and The
Livelihoodof Man (New York:Academic Press, 1977); CharlesCamic, "The UtilitariansRe-
visited,"AmericanJournalof Sociology85, 3 (1979), pp. 516-50; and MichaelHechter,"Karl
Polanyi'sSocial Theory:A Critique,"Politics & Society 10, 4 (1981), pp. 399-429.
43. Camic, "UtilitariansRevisited."

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 243

ditionally staked out as its own." Today it finds expression in the form of
microeconomictheoriesof politics,gametheory,exchangetheory,and rational
choice theory. Today, Hechtercan arguethat, "if currentsocial science can
boast anything remotely resembling a paradigm, then utilitarianismis its
leading candidate."44Neorealism shares in the "paradigm."
Broadlyconstrued, utilitarianismis characterizedby its individualistand
rationalistpremises. Its individualism stipulates the theoreticalprimacy of
individualactors ratherthan of social collectives. The individualactingunit
is taken to be essentiallyprivate. It exists priorto and independentof larger
social institutionsand is understoodas the autonomousgeneratorof its own
ends. Social realityis understoodas made up of many such individualactors,
inhabitinga world characterizedby scarcity-a world in which not all goals
can be equally realized and, hence, choices have to be made. Utilitarian
rationalismdefinesrationalityin means-endsor instrumentalterms:efficient
action in the service of establishedends whose value or truth is properlythe
province of the individual actor and cannot be held to account in public
terms. Economic rationalityis the archetype,the ideal form. What Weber
called "substantiverationality"or Habermascalled "practicalreason"(both
of which can pass judgment on ends as well as means) are excluded from
the utilitariannotion of rationality.Indeed, insofaras substantiverationality
and practicalreason presupposenormativestructurestranscendingand irre-
ducible to individual wants and needs, the utilitarianwould hold them to
be scientificallyindefensible metaphysicalnotions.
Upon these premises,utilitariansfound theirtheoriesof action, interaction,
order, and change. Utilitarian theories of action hold that actors behave
rationally,in the narrow instrumentalistsense. Actors strive to serve their
intrinsic(biologicallyor psychologicallyproduced)desiresor ends in the most
efficientmeans possible.Socialinteractionis interpretable,by directextension,
as instrumentalcoaction or exchange among individual actors, each party
regardedas an external object or instrument in the eyes of the rationally
acting other. Utilitarian theories also hold that, at base, social order is a
derivative relation. It derives entirely from equilibria (dynamic or static,
stable or unstable) in the instrumentalrelations and mutual expectations
among rational egoistic individuals. Social institutions are taken to be the
consequence of the regularizationof mutual expectations.As for its theory
of institutionalchange, utilitarianismproposes that changes occur sponta-
neously, as a consequence of relative changes in the competing demands
and capabilities of individual actors. Social order being a consequence of
instrumentalrelations among individual actors, changes in actors' interests
and means give rise to demands for change and, among other things, new
It is important to add that such modes of action, interaction,order, and

44. Hechter, "Karl Polanyi's Social Theory," p. 399.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
244 InternationalOrganization

changearedeemedintrinsicallyobjective,in need of neithernormativedefense

nor historicalaccounting.Their realizationin practice,while not always to
be observedhistorically,is takento be an essential,objective,and progressive
tendency of history. It follows that, for the utilitarian,modes of action fol-
lowingthe logicof economicrationalityare inherentlyobjective.The existence
of an economy whose actors obey this form of rationalityis interpretedas
a realizationof universal and objective truths existing independent of any
social-normativebasis. Hence, for the utilitarian,the market presents itself
as an idealmodel of rational,objectiveaction,interaction,order,and change-
a frameworkfor the interpretationof political as well as economic life.
Neorealism approachesthe internationalsystem from a utilitarianpoint
of view. The major difference,of course, stems from the neorealist'sstatism.
For the neorealist, states are the rational individual actors whose interests
and calculatingactions and coactions give form and moment to the inter-
nationalsystem. Such a positioncould easilyprovokelengthycriticalanalysis.
For presentpurposes,I shall confine myself to a brief,two-step commentary.
The first step is simply to note that the utilitarianmodel is indeed the
effectivemodel of internationalpolitics in neorealistresearchprograms.This
is not to say that neorealists systematicallyexclude insights or hypotheses
from other points of view. Among neorealism'snoteworthytraitsis an unex-
celled eclecticism:many neorealistswill use an argument,a clause, a phrase
from almost any source if it suits their purposes.45The point, rather,is that
utilitarianpremisestogetherwith statistcommitmentsestablishthe anchoring
"purposes"that all these borrowingsserve. To use Imre Lakatos'sfamiliar
terminology,utilitarianstatism is the "hardcore"of the neorealist"scientific
research programme."Around this hard core, neorealistsdevelop a "pro-
tective belt" of "auxiliary hypotheses" derived from many sources.46
This claim, whichgoes to the orientingstructureor "grammar"of neorealist
practice, cannot be demonstratedin a few pages. Two examples will have
to suffice.The first is the neorealisttreatmentof power. In neorealism,there
is no concept of social power behind or constitutive of states and their
interests. Rather, power is generally regardedin terms of capabilitiesthat
are said to be distributed,possessed, and potentiallyused among states-as-
actors. They are said to exist independent of the actors' knowing or will.
They are regardedas finallycollapsible,in principle,into a unique, objective
measure of a singular systemic distribution(as if there were one uniquely
true point of view from which the distributioncould be measured).Waltz
puts it this way: "To be politically pertinent, power has to be defined in

45. What are we to make of a structuralism,for example, that deploys both Adam Smith
and Emile Durkheim for its authoritieswithout once stopping to consider the contrarieties
between the two?
46. Imre Lakatos,"Falsificationand the Methodologyof ScientificResearchProgrammes,"
in Lakatosand Alan Musgrave,eds., Criticismand the Growthof Knowledge(London:Cambridge
University Press, 1970).

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 245

terms of the distributionof capabilities[among agents or actors];the extent

of one's power cannot be inferred from the results one may or may not
get.... [A]n agent is powerfulto the extent that he affectsothers more than
they affecthim."47Gilpin's understandingis not dissimilar.Power,he writes,
"refers simply to the military, economic, and technologicalcapabilitiesof
states."As he is quickto add, "Thisdefinitionobviouslyleaves out important
and intangible elements that affect outcomes of political actions, such as
public morale,qualitiesof leadership,and situationalfactors.It also excludes
what E. H. Carr called 'power over opinion.' These psychologicaland fre-
quently incalculableaspects of power and internationalrelations are more
closely associated with the concept of prestige.... "48 Such understandings
of power are rooted in a utilitarianunderstandingof internationalsociety:
an understandingin which (a) there exists no form of sociality,no intersub-
jective consensual basis, priorto or constitutive of individualactors or their
private ends, and hence (b) the essential determinants of actors' relative
effects on one another will be found in the capabilities they respectively
control. Only within such a conception could one believe, as Waltz believes,
that "power provides the means of maintaining one's autonomy." Only
within such a frameworkis one inclined to join Gilpin in reducingmatters
of morale, leadership,and power over opinion to "psychological"factors.
The second is the neorealist conception of internationalorder. For the
neorealist, there are no rules, norms, mutual expectations, or principlesof
practice prior to or independent of actors, their essential ends, and their
capabilities.In the last analysis,if not immediately,the evolution of all rules
follows from the regularizationand breakdownof mutual expectations in
accordancewith the vectoringof power and interestamong states-as-actors.
It follows that for the neorealist, a world of a multiplicityof actors having
relatively equal power is a formula for chaos. The potentiality for order
increases as the hierarchicalconcentration of power steepens. For Waltz,
who is concernedlest the envisionedconcentrationreduceto a singledominant
state, therebyoverturningthe fundamentalorganizationalprincipleof inter-
national politics, the optimal concentration is with two states. For other
neorealists, who somehow manage to ignore Waltz's concerns while citing
his "structuralist"authority,the condition of maximal order is a hierarchy
centeringpower within the graspof a singularhegemon, a state, in Keohane
and Nye's words, that is "powerfulenough to maintain the essential rules
governing interstaterelations, and willing to do so."49 Even in the analysis
47. Waltz, Theoryof InternationalPolitics, p. 192.
48. Gilpin, Warand Change, pp. 13-14.
49. Keohane and Nye, Power and Interdependence,chap. 3. See also Robert 0. Keohane,
"The Theory of HegemonicStabilityand Changesin InternationalRegimes, 1967-1977," in
Ole R. Holsti, RandolphM. Siverson,and AlexanderGeorge,eds., Changein the International
System (Boulder,Colo.: Westview Press, 1980); and Keohane, "HegemonicLeadershipand
U.S. Foreign Economic Policy in the 'Long Decade' of the 1950s," in Avery and Rapkin,
Americain a Changing World.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
246 InternationalOrganization

of internationalregimes, this emphasis persists. As Krasnerputs it, "The

most common proposition[amongneorealists]is that hegemonicdistributions
of power lead to stable, open economic regimes because it is in the interest
of a hegemonic state to pursue such a policy and because the hegemon has
the resourcesto provide the collective goods needed to make such a system
functioneffectively."'50In short,neorealismregardsinternationalorderentirely
as a derivative relation. Deriving from the rational coactions of individual
actors, order is taken to be finallydependentupon their respectiveinterests
and relative means of influencingone another.5'
The second step in this two-step commentary is to consider some of the
objectionswith which neorealism,as an instanceof utilitarianthought,must
contend. Three establishedcriticismsof utilitarianthought, all centeringon
the utilitarianconception of order, deserve mention. As will be seen, the
objections suggest a contradictionin neorealistthought, one that threatens
to fracturethe statist pillars of neorealistinternationalpolitical theory.
The three objections can be briefly summarized.The first objection has
its roots in sociology. It is found in Talcott Parsons's diagnosis (informed
by Durkheimand Weber)of the so-calledHobbesianproblem:in the absence
of a frameworkof norms consensually accepted by its members, it might
be possible momentarily to establish an orderly social aggregate(a "social
contract," for example) among instrumentallyrational individuals. Except
under conditions of absolute stasis, however, it cannot be maintained.The
second objection to the utilitarianconception of order is developed within
the utilitarianframeworkitself. This is MancurOlson's critique.52As aptly
summarizedby Hechter:
Rational self-interestedactors will not join large organizationsto pursue
collective goods when they can reap the benefit of other people's activ-
50. Stephen D. Krasner,"Regimes and the Limits of Realism: Regimes as Autonomous
Variables,"InternationalOrganization36 (Spring 1982), p. 499. Krasnerin this paperdem-
onstratesthat he is among the most open-mindedand criticism-consciousof neorealists.He
explores the limits of neorealism;in fact, he goes rightto the brinkof underminingits statist
props altogether.Exploringvarious relationshipsbetween regimes,state interests,politicalca-
pabilities,and state practices,he comes close to raisingthe possibilitythat regimes(principles,
norms, and proceduresthat have some autonomyfrom the vectoringof state behaviors)might
be constitutiveof states and their interests.
51. I am carefulin my wordinghere, because neorealists,like most utilitarianthinkers,are
slipperyabout the position they in fact take regardingrationalaction and the productionof
order. In a recent review of MancurOlson's The Rise and Decline of Nations (New Haven:
Yale UniversityPress, 1982), BrianBarrymakes a similarpoint. He notes that Olson could be
offeringa "monocausalexplanation,"a primus inter pares explanation,or an explanationin
terms of a factor that is not always the most importantbut that will always emerge on top
when other factorsare not too strong(which is not sayingmuch). Barrysays that he is "not at
all clear what position MancurOlson himself wants to take." Barry,"Some QuestionsAbout
Explanation,"InternationalStudies Quarterly27 (March 1983). Consideringthe same three
possibilitiesin neorealistexplanationsof order, I am not at all sure what position neorealists
mean to take.
52. See MancurOlson, TheLogicof CollectiveAction(Cambridge:HarvardUniversityPress,

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 247

ity to pursue those ends. This means that the rationalactor in the utili-
tarian model will always be a free rider whenever given the
opportunity.Thus, accordingto utilitarianbehavioral premises, social
organizationis unlikely to arise even among those individualswho have
a strong personal interest in reapingthe benefits that such organization
The thirdobjection,and no doubtthe most important,is Marx's.Anticipating
the broad outlines of both Parsons's and Olson's arguments, Marx went
beyond them to try to draw out what utilitariansmust presupposeif they
are to hold to their "contractarian"(i.e., instrumentalistor exchange-based)
understandingsof order in society. Marx arguedconvincinglythat the myth
of the contract, put into practice, depends upon a dominant class's ability
to externalizethe costs of keepingpromisesonto a class that lacksthe freedom
to contract;the Hobbesian "state of war" is thus held in check throughone-
sided power in a "class war."54Utilitarianorder thus presupposesclass re-
lations (and associated political, legal, and institutionalrelations),which its
consciousindividualistpremisesprohibitit from confronting,comprehending,
or explaining.
How do neorealistsdeal with these objections?The answer,quite simply,
is that they finesse them. In a bold stroke, neorealism embraces these ob-
jections as articles of faith. Turning problems of utilitariananalysis into
virtues,neorealismredefinesthe Hobbesianproblemof orderas an "ordering
principle"of internationalpolitics. Strugglesfor power among states become
the normal processof orderlychangeand succession.The free-riderproblem
among states becomes a global "sociological"legitimation for hegemonic
states, whose private interestsdefine the public "good" and whose prepon-
derant capabilitiessee to it that more "good"gets done. As for the Marxian
critique, it is accepted, albeit with a twist. It is accepted not as global class
analysis per se but in the idea that order among the great powers, the great
states, is ever dependent on the perpetuationof a hierarchyof domination
among great and small states. Inequality,Waltz says, has its virtues. Order
is among them.55
One has to have some grudgingadmirationfor theoristswho would make
such a move. They must have enormous courage,and not just because such
positions expose neorealiststo a lot of self-righteousmoralizing.Neorealists
must be courageousbecause their attempt to finesse objectionsto utilitarian
accounts of order involves a bluff of sorts. It counts on our failureto notice
that, at a certain moment in making their move, neorealistsare suspended
in thin idealist air.
That moment comes when, conceding objections to utilitarianaccounts,
53. Hechter, "Karl Polanyi's Social Theory," p. 403, note 6.
54. O'Neill, "The Hobbesian Problem."
55. Waltz, Theory of International Politics, pp. 131-32.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
248 InternationalOrganization

the neorealist embraces them to describe internationalorder among states

at the "level" of the internationalsystem. The rhetoricalforce of this conces-
sion, ironically,is to divert the critic of utilitarianconceptions of order into
momentarycomplicitywith the neorealist'sown statism,a statismthat would
collapse on its face if the critic were to raise the same objectionsat the level
of the state. That is to say, the neorealistcounts on our being so awestruck
by the Hobbesianand free-riderdilemmas we confrontat the "international
level" that we shall join in neglectingthe same dilemmas at the level of the
state.The neorealistcountson our failureto noticethatthe objectionsaccepted
at the level of the internationalsystem can equally well be turned against
the metaphysicalprop upon which dependsthe reificationof an international
politicalsystem analyticallydistinguishablefrom domestic and transnational
relations:the conception of the state-as-actor.
The neorealistmove is, in short, a sleight of hand. For despite its statism,
neorealismcan produceno theory of the state capableof satisfyingthe state-
as-actor premises of its internationalpolitical theory. On the contrary,by
adoptinga utilitariantheoryof action,order,and change,neorealistsimplicitly
give the lie to their idkefixe, the ideal of the state-as-actorupon which their
distinction among "levels" and their whole theory of internationalpolitics

c. Positivism
I am being unfair. To suggest, as I have, that neorealists play a trick of
sorts is to imply some kind of intentional duping of an innocent audience.
This is surely wrong. It is wrong because neorealistsare as much victims as
perpetrators.And it is wrong because, in truth, the bedazzled audience is
far from innocent. We alreadyshare complicity in the illusion. Neither neo-
realists nor we, the fawning audience, can imagine seeing the world in any
other way.
Why should this be so? Why, for example, is it so difficultto see that the
utilitarianperspectiveneorealistsembraceat the "internationallevel" under-
mines the state-as-actornotion upon which their whole theoreticaledifice,
includingthe distinction between levels, depends?The history of utilitarian
thought is, after all, largelythe story of philosophicaloppositionto the "per-
sonalist" concept of state required by neorealism's internationalpolitical
theory. In part, surely, this refusal to see is due to the blinding light of the
halo surroundingthe state in neorealistthought.But in part,too, this blindness
is due to the third commitment of the neorealist orrery.Neorealist theory
is theory of, by, and for positivists. It secures instantaneousrecognition,I
want to suggest, because it merely projectsonto the plane of explicit theory
certain metatheoreticalcommitments that have long been implicit in the
habits of positivist method. It tells us what, hidden in our method, we have
known all along.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 249

Bornin struggle,"positivism"is of coursea disputedterm. ManyAmerican

political scientists are unaware of its rich currents of meaning in recent
European,Latin American,and North Americansociology, philosophy,and
anthropology.Many trivializeand thus evade the term by misequatingpos-
itivism with "mindless number crunching,"brute empiricism, inductivist
logic, or narrow logical positivism of the Vienna Circle. And the term has
suffered at the hands of a number of silly or naive radicals who, having
encountered Lenin's indictment, use the term as a synonym for regime-
supportingscholarshipor bourgeois social science. Many of these radicals
are positivists themselves.56
At the very minimum, positivism means two complementarythings. In
its most generalmeaning,positivism refersto the so-called"receivedmodel"
of naturalscience.57At the same time, and aproposthe subject-object,man-
nature dualisms implicit in this "received model," one can follow Michel
Foucault in distinguishingpositivist from eschatologicaldiscourse. For es-
chatologicaldiscourse (evident in phenomenology,ethnomethodology,and
some hermeneuticalsciences) the objective truth of the discourselies within
and is producedby the discourse itself. By contrast,for positivist discourse,
with its naturalisticbias, the truth of discourse lies in the external object.58
In general, positivist discourseholds to four expectations.The firstis that
scientificknowledgeaims to graspa realitythat exists in accordwith certain
fixed structuralor causal relationswhich are independentof human subjec-
tivity (hence their objectivity) and internallyharmoniousor contradiction-
free (as if authored from a single point of view). The second is that science
seeks to formulate technically useful knowledge, knowledge that enhances
humancapacitiesto make predictions,orientefficientaction,and exertcontrol

56. I hold that all social science aspiring to theory has a positivist aspect in the sense given
below. This is true of Hegel, Marx, Lenin, Bourdieu, Foucault, Morgenthau, Alker, and me.
Following Bourdieu, even dialectical knowledge contains the objectivistic, the positivistic. As
I use the term here, however, a movement is "positivist" if it appears to be a one-dimensional
positivism. The issue is not the purging of positivism-the positivist moment is an inescapable
moment of all inquiry-but the realization of a more adequate "two-dimensional" or dialectical
perspective by bringing the positivist moment into unceasing critical tension with the practical
moment such that each side ever problematizes the other. Valuable readings on the subject of
positivism and its limits include Gerard Radnitzky, Contemporary Schools of Metascience, 3d
enl. ed. (Chicago: Regnery, 1973); Richard Bernstein, The Restructuring of Political and Social
Theory (New York: Harcourt Brace Jovanovich, 1976); the first chapter of Michael J. Shapiro,
Language and Political Understanding (New Haven: Yale University Press, 1981); Hayward
R. Alker Jr., "Logic, Dialectics, Politics: Some Recent Controversies," in Alker, ed., Dialectical
Logics for the Political Sciences, vol. 7 of the Poznan Studies in the Philosophy of the Sciences
and the Humanities (Amsterdam: Rodopi, 1982); and Theodore W. Adorno et al., The Positivism
Dispute in German Sociology, trans. by G. Adey and D. Frisby (New York: Harper Torchbooks,
In my present discussion, I am especially concerned with that strain, still predominant in
Anglo-American sociology, anchored in Weberian solutions to the problem of human subjectivity
and meaning in a naturalistic social science.
57. Giddens, CentralProblems,p. 257.
58. Foucault, The Order of Things.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
250 InternationalOrganization

in the serviceof given human values. The thirdis that sought-afterknowledge

is value-neutral.The fourth, consistent with the first three, holds that the
truth of claims and concepts is to be tested by their correspondenceto a
field of external experience as read via (problematic)instrumentsor inter-
pretative rules."9
When one turns to positive social science, at least one other expectation
needs to be added to the list. This is the expectation that "the phenomena
of human subjectivity... do not offerany particularbarriersto the treatment
of social conduct as an 'object'on a par with objects in the naturalworld."60
Obviously,this is a most troublesomeexpectation.Makinggood on it requires
overcoming a double problem inherent in human subjectivity.On the one
side, human subjectivity raises a problem from the perspective of social
actors: the problem of meaningful,value-laden social action. On the other
side, there is a problem from the analyst's point of view: the analyst'sown
norms, values, and understandingspotentiallynegate the analyst'sability to
detach himself or herself from the social world, to treat it, on a par with
nature,as an external,objective, "dumb generality."Positivist social science
has had to "solve" this double problem.
As it turns out, the "solutions" are worth a few moments of our time.
For it is in these solutions that we encounter the social-theoreticalcom-
mitments embedded within dominant conceptionsof social science itself. In
particular,I have in mind positivist solutions to the problem of human
subjectivityanchoredin an unquestionedcommitment to the objective, his-
torical force of instrumentalor technicalrationality.Let me brieflydescribe
this commitmentand then considerits role in "solutions"to the dual problem
of subjectivity in positivist social science. As I shall indicate, the result is a
metatheoreticaloutlook implicit in positivist method, which circumscribes
scientificcriticism and limits the rangeof theories about society that can be
scientificallyentertained.As I shall also suggest,these limits establishamong
positivists an uncriticalreceptivity to neorealists'conceptions of the inter-
national system.
Again, the commitment in question is a commitment to the essential
objectivity of technical rationality. According to this (typically unspoken)
commitment, which also appearsat the centerof utilitarianthought, means-
ends rationalityis inherentlyobjective, value-neutral,void of normative or
substantive content. Technical rationalityis said to inhabit the domain of
the "is" ratherthan the domain of the "ought,"and hence its truth requires
no normative defense. Indeed, as exemplified by Max Weber's resignation
to the world historical "rationalization"of all modes of life, technical ra-
tionalityis takento be a necessaryprogressiveforcein history.Rationalization
59. Anthony Giddens, ed., Positivismand Sociology (London:Heinemann, 1974), chap. 1.
Comparewith the list in Alker, "Logic,Dialectics,Politics."
60. Giddens, Positivismand Sociology, p. 4.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 251

involves the breakingdown of traditionallimitsand the progressiveabsorption

of all institutions of life within a mode of thought that aims to reduce all
aspects of human action to matters of purposive-rationalaction-efficiency
in the service of pregiven ends. For Weber, this tendency was inexorable,
its outcome inevitable: the "iron cage" of a totally bureaucratizedlife.6'
Science, committed to the objectivity of technical reason, is on the side of
this necessary historicaltendency. It is at the leading edge.
Immediatelyone can see that this commitment replicatesin a novel way
the classicaljustificationof positivist science as a critical,even revolutionary
force, a force that demystifiesall forms of romanticism,dispenseswith atav-
istic myth, and establishes the "end of ideology." What may be harder to
see, especially for positivists, is that this commitment ties positivism to an
ideology of its own. It endorses a metahistoricalfaith in scientific-technical
progressthat positivist science itself cannot question. Insofar as the com-
mitment affords "solutions" to the dual problem of human subjectivity,it
justifies itself in its own technicalterms, enrichingthe theoreticalcontent of
positivist method qua politicalideology.Having mentionedWeber'sposition
as exemplary, it is appropriateto consider the role of this commitment in
Weber'sown (now conventional) solutions to the two sides of the problem.
In Weber,the first side of the problem,the side concernedwith the mean-
ingful characterof social action, could be reducedto this: how can there be
a naturalisticsocial science, one that producesobjective knowledgecapable
of calculatingand predictingsocial outcomes, given that human action is
necessarily"subjective"in character?Weberconfrontedthis problemin the
specificcontext of the German historicalschool.62Authors like Roscherand
Knies had concluded that, given the subjective quality of human action,
human action is not calculableor predictablein the same way that one might
calculateor predictevents in the naturalworld.In this sense, they concluded,
human action has an "irrational"quality.63In Weber'sview, this conflation
of "subjectivism"and "irrationalism"presented a serious obstacle to the
reconciliationof naturalismand sociologicaland historicalmethod. He thus
set out the classic synthesisto which much of modernpositivistsocial science
is indebted.
Premisedon the inherentobjectivityof technicalrationality,the synthesis
was this:if we abstractand regardas objectivelygiven an agent'ssubstantively

61. See Max Weber,From Max Weber.Essays in Sociology,ed. and trans.by H. Gerthand
C. Wright Mills (New York: Oxford University Press, 1972), pp. 44-45. See also Herbert
Marcuse,"Industrializationand Capitalismin the Workof Max Weber,"in Negations:Essays
in Critical Theory(Boston:Beacon Press, 1968);JiurgenHabermas,"Technologyand Science
as Ideology,"in Towardsa Rational Society;and Anthony Giddens,Politicsand Sociologyin
the Thoughtof Max Weber(London:Macmillan, 1972).
62. Giddens, Positivismand Sociology, p. 5.
63. Max Weber,"Roscherund Knies und das Irrationalitatsproblem," in Wissenschaftslehre
(Tilbingen:J. C. B. Mohr),pp. 127-37, and translatedas "Subjectivismand Determinism,"in
Giddens, Positivismand Sociology,pp. 23-31.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
252 InternationalOrganization

empty logic of technical reason, then in interpretingthe agent's action we

can assume that, from this objective standpoint, society will appearto the
individual agent as a subjectless set of external constraints,a meaningless
second nature. We shall then be able to say that meaning enters society
primarily through the autonomously generated ends of individual acting
agents: meaningful action is merely motivatedaction. With that, we have
our objective, naturalisticsocial science. For with knowledgeof an agent's
pregivenends and "meaningless"social constraints,"meaningful"and "ra-
tional" subjective relations become calculable,predictable,and susceptible
to causal accounts.64
For most North Americantheoristsof internationaland comparativepol-
itics, Weber's solution is a "methodologicalprinciple"whose obviousness
precludes any need for justification. Yet as recollection of the Weberian
moment makes clear, the methodologicalprincipleimplicit in this solution
restrictsus to a particularconceptionof society. We may call this conception
an actor model. Upon commencingany analysisof a social system, the habit-
born principle predisposes the positivist to identify the irreducibleactors
whose rationaldecisions will mediate the entryof meaninginto social reality.
Thanks to this "principle,"the committed positivist knows almost "instinc-
tively" that all explanations of social action must ultimately come to rest
with the interpretationof some frozen set of actors, their values and their
ends. All analysis comes to rest with actors who are capable of exercising
technical rationality;whose ends, values, and boundaries separatingone
anotherare taken to be given and independentof communicationand inter-
action among the several; who accordingly must appear to one another,
individuallyand in aggregates,as externalconstraints;and who must relate
to one another, in the last analysis, in strictly instrumentalterms.65
Weber'ssolution to the second side of the problem of human subjectivity
is equallyimportant.The problem, seen from the second side, is the possible
confoundingof scientific detachment and objectivity owing to the fact that
the social scientist's own norms, values, and understandingsimplicate him
or her in the social world examined. As Weberrecognized,even one's cate-
gories of analysis and the meaningsone attaches to them depend upon nor-
mative commitments that bind one to the social world. All knowledgehas
its socially rooted presuppositions.
Weber's solution to this second side of the problem is also anchoredin a
commitment to the essentialobjectivityof technicalrationality.The solution
64. See ibid.
65. I am sayingnot that the predispositiontowardactormodels reflectsconsciousconformity
to a norm, but that social scientistsdo not conceive of the principlebecause it is so faithfully
observed that, in general,social scientistscannot conceive of thinkingabout the world in any
other ways. The principleat once exhaustsand limits the span of active social reasoning.My
thinkingregardingthe irresistibletug of "actor models" is largelysparkedby a conversation
with Robert North, althoughI do not know that he would agreewith my characterizationof
this predispositionas methodologicallyrooted.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 253

involves radicalizingthe separationbetweenthe processby whichthe validity

of scientific concepts and knowledge claims may be scientificallydecided
and the process by which scientists take interest in, generate, or come to
recognize as meaningfultheir concepts and knowledge claims. In Weber's
view, socialscientificdiscoursewould centeron the formerprocess-a process
whose objectivity would be assuredbecause it could and should be monop-
olized by the logic of technical rationality. It would concentrateon issues
decidable within technical rationality'sown inherentlyobjective terms.66
Thus, while individual scientists' norms, values, and socially established
understandingsmay help decide the directionin which the scientificbeacon
will cast its light, science as an enterprise cannot pass judgment on the truth
of values, ethics,ends, or understandings,includingthose at workin scientists'
choices of what to study. Scientificdiscourse cannot criticallyexamine the
meaningstructuresat workin and accountingfor scientists'mutualrecognition
of the concepts they deploy. Scientificdiscourse can speak decisively only
to the efficiency of means. In sum, science as an enterprise preserves its
objectivity by excluding from its terrain all questions that cannot be for-
mulatedand solved withinthe allegedlyobjectivelogicof technicalrationality.
This solution, like the first, is now widely taken for granted as one of
science'sdelimitingfeatures.Like the first,too, it buttressesthe commitment
of positivist science to an actor model. It does so primarilyby limiting the
range of scientificcriticism. In particular,it excludes discussion of forms of
social consensus that might themselves be value-laden, that might be his-
torically contingent and susceptible to change, and that might nonetheless
coordinatehuman practicesand distributionsof resourcesin ways that pro-
duce and accordrecognitionto the consensuallyrecognizedactors(including
their boundariesand ends) which positivists take as the irreducibleelements
of analyses.
Taken together, then, the two solutions establish a methodologicalpre-
disposition that is anything but neutral with respect to social orderingpos-
sibilities. On the contrary,they implicate and profoundlylimit the rangeof
possibilitiesthat theorycan contemplateif it is to findacceptanceas objective,
scientifictheory.Even beforethe firstself-consciouslytheoreticalwordpasses
anyone's lips, a theoreticalpictureworth a thousandwords is alreadyetched
in the minds of positivist speakersand hearers.Born of long practicecon-
formingto the solutionsjust described,this picture,a kind of scheme, orders
and limits expectationsabout what explicit theoreticaldiscoursecan do and
say. In particular,it commits scientific discourse to an "actor model" of
social reality-a model within which science itself is incapableof questioning
the historical constitutionof social actors, cannot question their ends, but
can only advise them as to the efficiencyof means.
Here in this theory-masked-as-methodwe find a partial explanation of

66. See Shapiro, Language and Political Understanding, pp. 5-6.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
254 InternationalOrganization

the ease with which neorealistsare able to delude themselves as well as us,
their admiringaudience. Despite the contradictionbetween neorealists'util-
itarianconception of politics and their statist commitments, neorealistsare
able to perpetuatethe state-as-actorillusion in their conceptionof the inter-
national system. They are able to do so because,as positivists, we are meth-
odologicallypredisposedto look for preciselythe kindof model they "reveal."
Without an actor model, we somehow sense, we shall lack any scientific
point of entry into a meaningful understandingof the internationalsystem;
the systemwill appearto us, we worry,as a meaninglessswirlof "disembodied
forces." They are furtherable to do so because, as positivists, we join them
in excluding from the realm of proper scientific discourse precisely those
modes of criticism that would allow us to unmask the move for what it is.
At the very moment we begin to question this state-as-actorconception,we
are given to feel that we have stumbled beyond the legitimategrounds of
science, into the realm of personal ethics, values, loyalties, or ends. We are
given to feel that our complaints have no scientific standing. And so, as
scientists, we swallow our questions. We adopt the postureof Waltz's utter
detachment, Gilpin's fatalism, Krasner'swonderment, or Keohane's We-
berian resignationwith respect to the powers that be. We might not like it,
we say, but this is the world that is. As scientists, we think we cannot say

d. Structuralism
There is more to the story of neorealism'ssuccess than this, however. As
noted earlier,the decisive moment in neorealism'striumphwas its celebrated
structuralistturn. As also noted, this structuralistturn would appearto hold
out a promise for a deepening of internationalpolitical discourse. Now,
having examined the other three aspects of the neorealist orrery, we can
returnat last to neorealiststructuralismand consideronce againits attractions.
We can listen as it explodes the one-time limits of internationalpolitical
discourse. We can look to see how it penetratesbeneath commonsense ap-
pearancesof the given order. We can sift throughthe argumentsto find the
many ways in which this structuralismtranscendsthe confines of utilitar-
ianism, statism, and positivism-perhaps enrichingthem by disclosingtheir
deeper historicalsignificance.We can listen, look, and sift some more. And
what do we find? Disappointment,primarily.
The reason is now beginningto become clear:neorealistsslide all too easily
between two concepts of the whole, one structuralistin the sense described
earlierand one atomist and physicalist.The structuralistposits the possibility
of a structuralwhole-a deep social subjectivity-having an autonomous
existence independent of, prior to, and constitutive of the elements. From
a structuralistpoint of view, a structuralwhole cannotbe describedby starting
with the partsas abstract,alreadydefinedentities,takingnote of theirexternal

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 255

joining, and describingemergentpropertiesamong them. The standpointof

the structuralwhole affordsthe only objective perspective.By contrast,the
atomist conception describes the whole precisely in terms of the external
joinings of the elements, including emergent properties produced by the
joinings and potentially limiting furthermovement or relationsamong the
elements. Clearly,in this conception,the whole has no existenceindependent
of the parts taken together. But it may be possible that, from the point of
view of any one part (a point of view that remains legitimate within an
atomistic perspective),the whole may exist independentof that part or its
possible movements. From this standpoint,the standpointof the singlepart,
the whole is an external physical relation-a "second nature" to be dealt
with, in the last analysis, only physically or instrumentally.It cannot be
otherwise, for no prior intersubjectiveunity joins part and whole.67
Neorealism has managed to conflate these two concepts of the whole.
Considerthe one position, the misnamed "sociologicalposition," that many
neorealiststake to be exemplary:Waltz's position. As noted earlier,Waltz
understands"internationalstructure"not as a deep, internal relation prior
to and constitutive of social actors but as an externaljoining of states-as-
actorswho have preciselythe boundaries,ends, and self-understandingsthat
theorists accord to them on the basis of unexamined common sense. In
turn-and here is the coup-Waltz grants this structurea life of its own
independent of the parts, the states-as-actors;and he shows in countless
ways how this structurelimits and disposes action on the part of states such
that, on balance, the structureis reproducedand actors are drawninto con-
formity with its requisites. But how is the independence of this structural
whole established?It is not establishedindependentof the partstakentogether,
for it is never anythingmore than the logicalconsequenceof the partstaken
together. Nor is it established by anchoring it in any deep intersubjective
structureof the state-systemic whole. Indeed, Waltz systematicallypurges
from the realist legacy all hints that subjective relations might be, in his
terminology, "systemic"; true to Waltz's atomism, all subjective relations
are interpretedas psychologicalrelations,and propositionsthat referto them
are thus banished as "reductionist."
Rather, Waltz establishesthe independenceof the structuredwhole from
the idealized point of view of the lone, isolated state-as-actor,which cannot
alone alter the whole and cannot rely on others to aid it in bringingabout
change in the whole's deepest structures.We are encouragedto glimpse and
authenticate the independence of this structure,in other words, from the
standpoint of a frozen abstraction:the point of view of the single state-as-
actor, or the points of view of any numberof states-as-actors,one at a time.
These, though, are precisely the states-as-actors(or, more correctly,this is
the same fixed,abstractstate-as-actorcategory)with whichthe theoristbegan.

67. See Ollman,Alienation,Appendix 2.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
256 InternationalOrganization

The autonomy of the neorealistwhole is establishedpreciselyfrom the hy-

postatized point of view of the idealized parts whose appearancesas inde-
pendententitiesprovidedthe startingpoint of the analysis,the basic material,
the props without which the whole physicalstructurecould never have been
erected.From startto finish,we never escape or penetratethese appearances.
From start to finish, Waltz's is an atomistic conception of the international
At the same time, once neorealistsdo arrive at their physicalisticnotion
of structure, they do attribute to it some of the qualities of structure in
structuralistthought.Neorealistsdo tend to grantto the internationalpolitical
system "absolutepredominanceover the parts."In neorealism,as in struc-
turalism,diachronyis subordinatedto synchrony,and changeis interpretable
solely within the fixed logic of the system. And neorealists,like structuralists,
do tend to regardthe structurethat they describe in the singular.Thus, as
noted earlier,there are definite isomorphismsbetween aspects of neorealist
thought and structuralistprinciples.
This, however, is no compliment. For what it means is that neorealism
gives us the worst of two worlds. In neorealism we have atomism's super-
ficialitycombined with structuralism'sclosure such that, once we are drawn
into the neorealistcircle,we are condemnedto circulateentirelyat the surface
level of appearances. And what an idealist circle it is! What we have in
neorealism's so-called structuralismis the commonsense idealism of the
powerful, projected onto the whole in a way that at once necessitates and
forgives that power. It is the statist idealism developed from the point of
view of the one state (or, more properly,the dominant coalition) that can
afford the illusion that it is a finished state-as-actorbecause, for a time, it
is positioned such that the whole world pays the price of its illusions.
With apologies to E. P. Thompson, I would suggestthat there is a certain
"snake-like"quality to neorealiststructuralism.The head of the snake is an
unreflectivestate-as-actor,which knows itself only to rely on itself and which
will not recognize its own limits or dependence upon the world beyond its
skin. It slithersaroundhissing"self-help"and projectingits own unreflectivity
onto the world. Finding its own unreflectivenessclearly reflectedin others,
it gets its own tail into its mouth, and the system is thus defined. Asked to
describe the system so defined, the snake says that it reproducesitself, and
it swallows more of its tail. What, though, of the values or norms of this
system?The values and norms, the snake answers,are those that reflectthe
power and interests of the powerful and interested. What, then, of power?
The snake-or what is left of it, for it is now a wrigglingknot- has an answer
for this, too. Power is rooted in those capabilitieswhich provide a basis for
the state-as-actor'sautonomy. And what of autonomy?In a final gulp, the
snake answers. Autonomy is the state-as-actor'sprivilege of not having to

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 257

reflectbecause the whole world bends to its unreflectedprojectionsof itself.

"Plop! The snake has disappearedinto total theoreticalvacuity."68
As Thompson says of another structuralism:"It is, of course, a highly
conservativevacuity;whatis governswhatis whose firstfunctionis to preserve
the integrity of is-ness; what dominates has the functional imperative of
preservingits own dominance."Thompson'swordsare apt. Neorealiststruc-
turalismlends itself wonderfullywell to becoming an apologiafor the status
quo, an excuse for domination, and "an invective against 'utopian' and
'maladjusted'heretics"who would question the givenness of the dominant
In The Poverty of Historicism, Karl Popper concerned himself with the
totalitarianimplicationsof certain progressivistversions of structuralismto
which he gave the label "historicism."70What we find in neorealist struc-
turalism is a historicismof stasis. It is a historicismthat freezes the political
institutions of the current world order while at the same time rendering
absolutethe autonomyof technicalrationalityas the organonof socialprogress
to which all aspects of this order, includingstates-as-actors,must bow. It is
a historicismthat almost perfectlymirrorsHans Morgenthau'sunderstanding
of the "totalitarianstate of mind."7'
... Whereas perfectionismcreates an abstractideal to which it tries to
elevate political life through force or exhortationor reform, totalitarian-
ism, that is, the totalitarianstate of mind, identifies the ideal with the
facts of political life. What is, is good because it is, and power is to the
totalitariannot only a fact of social life with which one must come to
terms but also the ultimate standardfor judging human affairsand the
ideal source of all human values. He says "Yes" to his lust for power,
and he recognizesno transcendentstandard,no spiritualconcept which

68. Thompson, Povertyof Theory,p. 77.

69. Ibid., pp. 77, 73.
70. Popper, The Povertyof Historicism(New York: Harper& Row, 1961).
71. So dangerousis the term that I must once again hasten to stress that I am addressing
the logic of the neorealistmovement as expressedin its theoriesand not the consciouslyheld
values, intentions,or ideals of individualneorealists.I readilystipulatethat Krasner;Gilpin,
Keohane,Waltz,and other neorealistsare not championsof totalitarianismin theirconsciously
held personalvalues. I readilystipulate,too, that some neorealists,like Gilpin in his Warand
Change,can moralizeat length in their professionalwritingsand do expresspluralisticvalues
in their moralizing.The problem is-and this is my charge-that neorealistdiscoursegrants
absolutelyno scientificstandingto moral norms. At best, the moralizingof neorealistscholars
is recognizedas a proclamationof personalcommitments,belief,or faithon the partof individuals,
and not as an argumentwhose truthcontentis decidablewithinscientificdiscourseor groundable
within theory. The resultis a scientifictheorythat says no to neorealists'expressedvalues and
yes to totalitarianexpectations-hence the aura of quiet despairing(but not theoreticallyde-
scribableirony)surroundingsome neorealistarguments.Sadly,many neorealistsinterprettheir
own resignationto sucha situationas a kindof scientifictough-mindedness,a formof "realism,"
when in fact theirsituationis largelyattributableto unquestioningacceptanceof a moralsystem:
the moral norms of economic reasonand positivist science.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
258 InternationalOrganization

might tame and restrainthe lust for power by confrontingit with an

ideal alien and hostile to political domination.72
Of course, neorealism'stotalitarianimplicationsare only partlyto be dis-
covered in its celebrationof power before order. They are also present in
neorealism's silences, in those aspects of history neorealism denies, omits,
or represses.As Aldous Huxley reminds us, the greatesttriumphsof total-
itarian propagandahave been accomplished "not by doing something, but
by refrainingfrom doing. Great is the truth,but still greater,from a practical
point of view, is silence about truth."73Neorealist structuralismis silent
about four dimensions of history. I will call these the "four p's": process,
practice, power, and politics.
First, neorealist structuralismdenies history as process. Like other static
structuralisms,neorealist theory has two characteristics.One is a "fixity of
theoreticalcategories"such that each is a categoryof stasis even when it is
set in motion among other moving parts. The other characteristicis that all
movement is confined within a closed field whose limits are defined by the
pregiven structure.Thompson very clearly articulatesthe consequences of
such a conception:"[H]istoryas process,as open ended indeterminateeven-
tuation-but not for that reason devoid of rational logic or of determining
pressures-in which categories are defined in particularcontexts but are
continuously undergoinghistoricalredefinition,and whose structureis not
pre-givenbut protean,continuallychangingin form and in articulation-all
of this . .. must be denied."74
Second, neorealismjoins all modes of historicismin denyingthe historical
significanceof practice, the moment at which men and women enter with
greater or lesser degrees of consciousness into the making of their world.
For the neorealistintellectual,men and women, statesmenand entrepreneurs,
appear as mere supports for the social process that produces their will and
the logics by which they serve it. In particular,people are reduced to some
idealized homo oeconomicus, able only to carry out, but never to reflect
critically on, the limited rational logic that the system demands of them.
They are reduced in the last analysis to mere objects who must participate
in reproducingthe whole or, as the enlightenedintellectual knows, fall by
the wayside of history. True, neorealists would never admit that theory is
without "practicalrelevance."But for them, relevancefinds its measureonly
in terms of the technicaladequacyof the theorists'advice to agentsof power,
and technical adequacy consists solely in the enhancementof the efficiency
of means underobjective structuralconstraints.Nowhere in neorealistcate-
gories do we find room for the idea that men and women who are the
72. Hans J. Morgenthau,"The Escapefrom Power,"in Morgenthau'sDilemmas of Politics
(Chicago:Universityof ChicagoPress, 1958), pp. 244-45.
73. Aldous Huxley, BraveNew World(London:Vanguard,1952), p. 14. The forewordwas
authoredin 1946.
74. Thompson, Povertyof Theory,pp. 83-84.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 259

objectsof theorycan themselves theorizeabout theirlives; arein fact engaged

in a continuingstruggleto shape and redefinetheir understandingsof them-
selves, their circumstances,their agencies of collective action, and the very
categoriesof social existence;do indeed orient their practicesin light of their
understandings;and, thanks to all of this, do give form and motion to the
open-endedprocessesby which the materialconditionsof their practicesare
made, reproduced,and transformed.Neorealist structuralismcannot allow
this to be sb. For to do so would mean that neorealisttheory would itself
be a mere part of history,and not the intellectualmasterof historyit aspires
to be.
Third, for all its emphasis on "power politics," neorealism has no com-
prehensionof, and in fact denies, the social basis and social limits of power.
For the neorealist,as we have seen, power must be ultimatelyreducibleto
a matterof capabilities,or means, underthe controlof the unreflectiveactor
whose status as an actor is given from the start. No other position on power
could possiblybe compatiblewith neorealism'satomisticand utilitariancon-
ceptions of internationalorder. Yet such a position strictlyrules out a com-
petence model of social action. Accordingto a competencemodel, the power
of an actor, and even its status as an agent competent to act, is not in any
sense attributableto the inherent qualities or possessions of a given entity.
Rather, the power and status of an actor depends on and is limited by the
conditionsof its recognitionwithin a community as a whole. To have power,
an agentmust firstsecureits recognitionas an agentcapableof havingpower,
and, to do that, it must first demonstrate its competence in terms of the
collective and coreflectivestructures(that is, the practicalcognitiveschemes
and history of experience) by which the community confers meaning and
organizes collective expectations. It is always by way of performancein
reference to such collectively "known" (but not necessarily intellectually
accessible)generativeschemesthat actorsgainrecognitionand areempowered.
Thus, accordingto a competence model, buildingpower always has a com-
munity-reflectiveperformativeaspect.Thus, too, the powerof an actoralways
has its limits. Although an actor can play creatively off of given practical
schemes, and although an actor can sometimes offer up virtuoso improv-
isations that elicit novel orchestratedresponses to new circumstances,the
actorcan never exceed the limits of recognition.75 The authorof the "Melian
Dialogues" understood this dialectic of power and recognition.Neorealists
have forgottenwhat Thucydidesknew, in favor of a notion of powerwedded
to the IndustrialRevolution's faith in humankind'slimitless expansion of
control over nature.
Fourth, despite its spiritedposturingon behalf of politicalautonomy and
in opposition to the alleged economism of other traditions, neorealist his-

75. See especially Bourdieu, Outline, chap. 4, "Structure, Habitus, Power: Basis for a Theory
of Symbolic Power."

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
260 InternationalOrganization

toricism denies politics. More correctly,neorealismreducespolitics to those

aspects which lend themselves to interpretationexclusively within a frame-
workof economic action understructuralconstraints.In so doing,neorealism
both immunizes that economic frameworkfrom criticism as to its implicit
politicalcontent and stripspolitics of any practicalbasis for the autonomous
reflectionon and resistanceto strictlyeconomicdemands.It therebyimplicitly
allies with those segments of society that benefit from the hegemony of
economic logic in concertwith the state. Politics in neorealismbecomes pure
technique: the efficient achievement of whatever goals are set before the
political actor. Political strategy is deprived of its artful and performative
aspect, becoming instead the mere calculation of instruments of control.
Absent from neorealistcategoriesis any hint of politics as a creative,critical
enterprise, an enterpriseby which men and women might reflect on their
goals and strive to shape freely their collective will.
Takentogether,reflectionson these "fourp's" suggestthat neorealiststruc-
turalismrepresentsanythingbut the profoundbroadeningand deepeningof
internationalpoliticaldiscourseit is often claimed to be. Far from expanding
discourse,this so-called structuralismencloses it by equatingstructurewith
external relations among powerful entities as they would have themselves
be known. Far from penetratingthe surface of appearances,this so-called
structuralism'sfixed categoriesfreeze the given order, reducingthe history
and future of social evolution to an expressionof those interestswhich can
be mediated by the vectoringof power among competing states-as-actors.76
Far from presentinga structuralismthat envisions political learning on a
transnationalscale, neorealismpresentsa structurein whichpoliticallearning
is reduced to the consequence of instrumentalcoaction among dumb, un-
reflective, technical-rationalunities that are barragedand buffetedby tech-
nological and economic changes they are powerless to control.
Again, though, none of this is to say that neorealist "structuralism"is
without its attractions.For one thing, and most generally,there is something
remarkablycongenialabout a structuralismthat pretendsto a commanding,
objective portraitof the whole while at the same time leaving undisturbed,
even confirming,our commonsense views of the world and ourselves. As
compared to Wallerstein'sconception of the modem world system, for in-
stance, neorealist structuralismis far more reassuringas to the objective
necessityof the state-as-unit-of-analysisconvention among students of pol-
itics.77It thus relieves this particularniche in the academicdivision of labor
76. Some good examplesof the agenda-limitingeffectof neorealiststructuralismare pointed
up by CraigMurphyin his discussionsof StephenKrasner's"Transforming International
What the Third World Wants and Why," InternationalStudies Quarterly25 (March 1981),
pp. 19-48; and RobertW. Tucker'sThe Inequalityof Nations(New York:BasicBooks, 1977).
See Murphy,"Whatthe ThirdWorldWants:An Interpretation of the Developmentand Meaning
of the New InternationalEconomicOrderIdeology,"InternationalStudiesQuarterly27 (March
77. As Wallerstein,Hopkins,and others frequentlyurge,the modernworld system presents
itself as only one unit of analysis,an N of 1.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 261

of responsibilityfor reflectionon its own historicity.Its pose of Weberian

detachment can be preserved.
For another thing, this strange structuralismfinds much of its appeal in
the fact that it complements and reinforcesthe other three commitments of
the neorealistorrery.As alreadynoted, neorealism'satomisticunderstanding
of structuregives priorityto-and then reconfirms-the commitment to the
state-as-actor.One might also note that neorealismemploys the only form
of structuralismthat could possibly be consistent with its utilitarianand
positivist conceptions of internationalsociety. Anchored as they are in the
idealof rationalindividualactionundermeaningless,quasinaturalconstraints,
these conceptions would be radicallychallengedby modes of structuralism
that question the dualism of subject and object and thus highlightthe deep
intersubjectiveconstitutionof objective internationalstructures.Neorealism
is able to avoid this radical challenge. It is able to do so by restrictingits
conception of structureto the physicalist form of a clockwork, the philo-
sophical mechanism so dear to the heart of the Industrial Revolution's

3. The ghost of the old revolution

The "secret world," John le Carrewrites, "is of itself attractive.Simply by
turning on its axis, it can draw the weakly anchored to its center."78The
same, we can now note, might be said of the neorealist orrery of errors.
Havingseen its severalelements whiz by-statism, utilitarianism,positivism,
structuralism,and statism yet again-we sense that there is a strangeunity
of contrarietieshere. We sense that the whole machine exerts a centripetal
force that is difficultto defy.
To be sure, when we slow and examine the elements we find that errors
and absurditiesabound. We find, for example, that the utilitarianinterpre-
tation of internationalorderpresupposesa conceptionof the state-as-actor-
a conception that a utilitarianwould want to disown. We find, too, that
neorealiststatism runs contraryto any genuinelystructuralistunderstanding
of the internationalsystem. We find that neorealismappeals to a Weberian
interpretationof positivist method, a method that parades as the end of
ideology even as it subordinatesall criticism to a scientificallyindefensible
commitment to technical rationality'sobjectivity and neutrality. And we
find that despite neorealism's pretensionsto the status of a political struc-
turalism, neorealisttheory is as economistic as they come.
Yet the neorealistorreryis meant never to be held at rest. It presentsitself
only in motion. And thanks to this, its countlesserrorsbecome not damning
indictments but counterweightsto other errors,balancingand perpetuating

78. Le Carre, The Little Drummer Girl (New York: Knopf, 1983).

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
262 InternationalOrganization

the motion of the whole. The limits of positivism obscure the errors of
statism in a state-as-actorconception of internationalorder, which reduces
systemic analysisto a physicaliststructuralism,which in turn propelsus into
the utilitarianworldof technicalreasonand necessity,whichbringsus around
to positivism once again. Around and around it spins, eroding and then
consuming the ground upon which opposition would stand. Around and
around it spins, until we lose sight of the fact that it is only motion. Like le
Carres secret world, this neorealistorreryhas no center at all.
A much earlier study of the emergence of statist tendencies offers some
clues as to how such a centerless swirl could become so powerful. In The
EighteenthBrumaireof Louis Bonaparte,KarlMarx set out to discoverhow
in Franceof 1848 through1852 it becamepossiblefor "agrotesquemediocrity
to play a hero's part."His conclusionwas that Bonaparte'swieldingof power
was occasioned by a crisis-bornbourgeoisreaction,but that it could not be
wholly accounted for by materialcircumstances.Nor could it be attributed
to the intrinsic qualities of Louis Bonaparte.In large measure, Bonaparte
attainedpower because he was able to securerecognitionamong the French,
and he was able to secure recognitionbecause, amidst crisis, he helped to
make "the ghost of the old revolution walk about again."
... The French, so long as they were engaged in revolution, could not
get rid of the memory of Napoleon, as the election of December 10,
1848, proved. From the perils of revolution their longingswent back to
the flesh-pots of Egypt, and December 2, 1851, was the answer. They
have not only a caricatureof the old Napoleon, they have the old Na-
poleon himself, caricaturedas he would inevitably appear in the middle
of the nineteenth century.79
One may venture a similarinterpretationto account for the extraordinary
power of the neorealistorrery.As in the ascent of Bonaparte,the emergence
of neorealism is no doubt partly to be explained as a reaction to crisis. In
particular,a reasonablycomplete interpretationof neorealismin its context
would want to considerthe currentfiscaland legitimationcrisis surrounding
the Americanstate in its role as system-managerand guardianof the capitalist
accumulationprocess. Such an interpretationmight comprehendneorealism
as a crisis-promptedredeployment,from domestic to internationalpolitics,
of the "economistic"ideologicallegitimationshitherto evidenced primarily
and increasinglywith respect to the state's domestic performance.This ac-
count would grasp neorealism as a contributionto "statist economism," a
historical successor to internationallaissez-faireeconomism.80
Another part of the explanation of neorealism's success looks beyond
79. Karl Marx, "The Eighteenth Brumaire of Louis Bonaparte," in Robert C. Tucker, ed.,
The Marx-Engels Reader (New York: Norton, 1972), pp. 438-39.
80. I develop this interpretation at greater length in Richard K. Ashley, "Three Modes of
Economism," International Studies Quarterly 27 (December 1983).

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 263

materialcircumstancesto collective memoriesof the past. Despite its errors,

its idealism,and its emptiness, neorealismsucceedsin its illusionof greatness
because it at once cues and caricaturesthe ghosts of revolutions past, most
of all the behavioral revolution in the Cold War study of internationalre-
lations. Awakened is a remembranceof a naturalisticmodel of science, the
very model in whose name the behavioralrevolutionariesmarched.Awak-
ened, too, is a one-time revolutionary faith that the light of science will
illuminate the conditions of state action, thereby reducing the chances of
miscalculation, overcommitment, and their sometimes disastrous conse-
quences. Summoned forth once again is the sense of urgencyof the darkest
days of the Cold War. Objectivity,neutrality,detachment, "state-as-actor,"
a "technology of peace"-these were among the slogans. Recollectingthis
heroic revolutionaryproject,neorealismennobles its followers.Never mind
that the faith in naturalisticscience and the harmonizingforce of reason
implicit in these memories do violence to the very core of classicalrealism's
internationalistthought,includingits long-standingresistanceto behavioralist
methods. Never mind that the darklypessimistic side of Morgenthau,Carr,
Wight, and Herz does not square with the behavioralists'optimism. With
Gilpin, one can remember classical realism not as an embodiment of a
continuing dialectical struggle between absolutist darkness and bourgeois
Enlightenmentbut only as a product of the latter:
Embedded in most social sciences and in the study of international
relations is the belief that through science and reason the human race
can gain control over its destiny. Throughthe advancement of knowl-
edge, humanity can learn to master the blind forces and constructa sci-
ence of peace. Throughan understandingof the sources of our actions
and the consequences of our acts, human rationalityshould be able to
guide statesmen through the crisis of a decayingworld order to a reno-
vated and stable world order. The fundamentalproblem faced, this ar-
gument continues, is not uncontrollablepassions but ignorance.
Political realism is, of course, the embodiment of this faith in reason
and science. An offspringof modern science and the Enlightenment,re-
alism holds that throughcalculationsof power and national interest
statesmen can create order out of anarchyand thereby moderate the in-
evitable conflicts of autonomous, self-centered,and competitive
states.... [T]his faith that a 'science of internationalrelations'will ulti-
mately save mankind still lies at the heart of its studies.8'
Likethe Frenchgazingupon Louis Bonaparte,neorealism'sfollowersglimpse
in this caricatureof past revolutions an image that reflects well on them,
that calls to mind the best images of themselves.
Yet like the French, the followers quickly experiencea sickeningjerk as
the whole project is yanked too soon into closure. The neorealistcaricature
81. Gilpin, Warand Change,p. 226.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
264 InternationalOrganization

of sciencehas not deployedits revolutionaryimagesto "firethe imagination,"

or to "glorifythe new struggles,"but to seal discourse within a continuous
"parodyof the old." Its discourse is now frozen in acquiescenceto the Cold
War conditions of the revolution it recalled:competition among states mu-
tually preparingfor war. As Gilpin writes, "The advance of technologymay
open up opportunities for mutual benefit, but it also increases the power
available for political struggle.The advance of human reason and under-
standingwill not end this power struggle,but it does make possible a more
enlightenedunderstandingand pursuitof nationalself-interest."82 How painful
this is: the revolutionaryscience of peace has become a technology of the
state! Was it always so? By the time this awful realizationcomes, it is too
late for such reflections.The neorealistorreryhas spun its followersinto its
arc. A "grotesquemediocrity"reigns.
Such reflections suggest that a serious problem awaits the critic of neo-
realism.Despite its seriousflawsand totalitariannature,neorealistorthodoxy
will not be dislodged from its lofty status by the force of logical criticism
alone. For neorealism'spoweris largelydue not to its truthor the consistency
of its logic but to its capacity to elicit the collective recognitionof women
and men, scientists mostly, who must somehow organizetheir expectations
and coordinatetheirpracticesin lightof commonly rememberedexperiences.
True, as I have noted, neorealist theory is like all intellectualisttheory in
that it contains no terms that would allow it to reflect critically on the
practicalsocial basis and limits of its own power. Yet the fact remains, "the
ghosts of revolutions past" are neorealism's main allies. At least, so they
appear.Everywherethe collective remembranceof the study of international
politics is on neorealism's side.

4. Recovering a silenced realism

And then we spy an apparitionin the shadows, lodged deep in the recesses
of our rememberedexperience. When last we met these ghosts of classical
realism,we were introducedvia neorealists'terse interrogationsof theirheri-
tage. Neorealists asked the classical ghosts a few pointed questions: Is the
state the most important actor, yes or no? Is it not true that your central
concept is "national interest defined as power," yes or no? They elicited
testimony establishingneorealism'sstatus as classicalrealism'srightfulheir.
Then, as the ghostswerehurriedout of the probatecourtroom,the neorealist
interrogatorsexplainedto us why the interrogationhad to be so brief.Classical
realists had a few sound ideas worth remembering,it seems. But they "are
consideredto be traditionalists- scholarsturnedtowardhistoryandconcerned

82. Ibid.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 265

more with policy than with theory and scientific methods."83Sadly, their
work knew nothing of economic relationsand fell short of modern scientific
standards. They would often fall prey to the "analytic fallacy," engage in
"circularreasoning,"get their explanations "inside-out,"or even "drift to
the 'subsystem dominant pole.' "84 Hard as it is to believe, many classical
realists did not even know, until helped along by neorealists,that the logic
of their "frameworkis identicalto that used in neo-classicaleconomic theory
where it is assumed that firms act to maximize profits."85Worse, many had
been heardto echo Hans Morgenthau'sinsistenceon realism's"emancipation
from other standardsof thought," including economics.86Why, some had
even "insist[ed]that theorists'categoriesbe consonant with actors' motives
and perceptions."87Under the circumstances, surely, the classical realist
legacy is honored most and embarrassedleast by retainingits scientifically
redeemable statements and turninga deaf ear to its fatuities. Let the tired
old ghosts rest in peace.
As we edge closer to the dark corner, however, the haze lifts, and the
visages of classical realism appear, far more clear-eyed than we had been
led to believe. As their words become audible,we note that there is a definite
coherence here, a coherence born of a sustained, disciplined, and richly
developed perspectiveon internationalpolitics. Some of the wordsdo indeed
resonate with things that we have heard neorealists say. Others, however,
do not. Eavesdroppingon this conversationof honoredghosts we learnmany
things that the neorealist keepers of the flame have somehow neglected to
bring to our attention. As we listen, it begins to become clear why.

a. The practical structureof classical realism

The firstthingwe learnis in some respectsquite obvious. Classicalrealism,
in its method, is not at all the intellectualistor technocratictradition that
neorealiststructuralismwould aspireto be. On the contrary,classicalrealism
is very much animated by a practical interest in knowledge.88It is, if you
will, the ethnomethodology of the modern tradition of statesmanship.As
such, its approachis largelyhermeneutical.Its reality is the reality familiar
to those who are competent parties to the tradition of statesmanship. Its
remembrancesof thingspast arethe officialremembrancesof this community,

83. Waltz, Theory of International Politics, p. 63. In the quoted words, Waltz refers specifically
to Morgenthau and Kissinger.
84. The quoted words are all Waltz's. He includes nonrealists as well as classical realists
among his targets.
85. Krasner, Defending the National Interest, p. 37.
86. Morgenthau, Politics among Nations, pp. 12-14.
87. Waltz, Theory of International Politics, p. 62. Waltz refers specifically to "Aron and other
traditionalists" in this connection.
88. See Richard K. Ashley, "Political Realism and Human Interests," International Studies
Quarterly 25 (June 1981).

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
266 InternationalOrganization

the way historywould have been and perhapsshouldhave been had it offered
recurringendorsement of this modern tradition of internationalpolitical
practice.Its problematicsare those that partiesto this tradition-competent
statesmen-are preparedto recognize as problematic,not as economic in-
dividualsand not as history-lessstates-as-actors,but as statesmenwho expect
that their understandingscan secure public recognitionwithin the overall
community of statesmen and againstthe backgroundof collectivelyremem-
bered experience.
What takes longer to see is that such an approachis above all systemic,
althoughnot of a form that neorealistphysicalismcan comprehend.In fact,
after a good deal of time watching and listening to neorealistcommentary
with an eye to elements of continuity, we begin to see that classical realist
scholarship already has a definite "structuralist" aspect. It is not the struc-
turalismof clockworkmetaphysics,to be sure. It is not a structuralism,like
Waltz's, anchored in an atomistic conception of global society. Instead, as
in competence models of social action, classical realism is enduringlycom-
mitted to a simple generative scheme, a practical cognitive structurethat
orients its discourseand in referenceto which all politicalpracticeis under-
stood. Before I introduce this scheme, a few general notes on its qualities
and its status in classical realist argumentare in order.
This scheme, it must be said, is a practicalscheme. At once subjective
and objective, necessary and contingent, the scheme exists as a preliteral
relation,almost a posturalrelation,which can be graspedonly in the objective
coherence of the actions it generates, in its uniting of otherwise seemingly
disparatepractices.It is not producedas a scientificpostulatein some sense
external to practice. Rather, it is learned, much as Thomas Kuhn would
insist a scientific paradigmis learned, through the practicaltransferenceof
quasiposturalrelations.89It can be grasped only by reliving, via ritual and
practice,the conflicts, rites of passage,and crises that bringthe scheme itself
strategicallyinto play, sometimesartfullyand sometimesineptly.Accordingly,
the meaning of the scheme is in its practicalstate, and the scheme is mis-
understood at the very moment that it is objectifiedor "captured"within
some conceptual system, formal logic, or set of rules externalto practice.90
89. See Kuhn'sconcludingcontributionto Lakatosand Musgrave,Criticismand the Growth.
All social scientistswho work with graduatestudentsin their researchprogramsare "familiar"
with such schemes. The generative scheme of a researchprogramis what our "brightest"
students-the ones who arereallycompetent-seem to graspthrougha kindof "fuzzyabstraction"
from our own researchpractices.It is what allows them to do, with minimaldirection,the kind
of independentresearchwe instantly recognizeas exactly the kind of work we would have
wanted to do but, for some reason, never thought to do. The generativescheme is also that
which our "second-rate"graduatestudentsnever quite graspwhen, in tryingto learnfrom our
own practices,they embarrassus by mimickingour practicestoo exactly under inappropriate
circumstancesor by followingour instructionstoo much by rote. It is that which we spend
hours tryingpatientlyto explainto graduatestudentsbut which, we know, always loses its life
once it is translatedinto a rule.
90. See Bourdieu'schapter,"GenerativeSchemesandPracticalLogic:InventionwithinLimits,"
in Outline.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 267

For the classical realist, however, this generative scheme is no less real
because it resists capture within a frozen category. On the contrary, this
scheme is graspedas the self-replicating"geneticcode of politicallife." Em-
bodied in all aspectsof internationalpolitics,from the sovereignstate through
the internationalsystem, this scheme is the principle"which allows [us] to
distinguishthe field of politics from other social spheres,to orient [ourselves]
in the maze of empiricalphenomenawhichmakeup that field,and to establish
a measure of rational order within it."9' For the classical realist, it is the
graspingof this scheme and the conditions of its successfulapplicationthat
makes competent political practicepossible. It is the indispensableelement
of internationalpolitical savoirfaire.
In moderninternationalsociety,the communityboundedby the consensual
recognition of this scheme defines the modern tradition of statesmanship
for the classical realist. Within this tradition, statesmanshipis not, as ob-
jectivism would have it, the "execution of a rule," or acting in accordance
with some externalobjectivenecessities,or mechanicalobedienceto a timeless
model for which all processes are reversible and time and tempo are no
matter.92 Nor is it reducible, as in neorealism, to rational choice, under
constraints, on the part of an actor whose status as such is pregiven and
unquestioned.Rather, statesmanshiprefersto practice, playing off the gen-
erative scheme in ways rangingfrom the awkwardand uninventive to the
artfuland creative-and alwayswith an eye to the problematicreproduction
of the state itself. On the one hand, it allows for the possibility of slips,
mistakes, and clumsy moments on the part of the maladroit.On the other
hand, it involves virtuoso improvisationsreflectinga perfect command of
the "artof living"and playingon "allthe resourcesinherentin the ambiguities
and uncertaintiesof behavior and situation in order to produce the actions
appropriateto each case, to do that of which people will say, 'There was
nothing else to be done.' "'9
What, then, is this scheme? One possible way of answeringis to refer to
"power,""interestdefined as power,"or, better, "balanceof power." I shall
refer to it as a balance-of-powerscheme.94All such labels are troubling,
however. They conceptualizethat which functionsto dispensewith concepts.
They invite a kind of fetishism among too-literalinterpreters,an ahistorical
reductionof the scheme to the manifest conditions and relationsthe labels
immediatelyconnote. For the communicationof the classicalrealisttradition
this has been a difficultproblem.
91. HansJ. Morgenthau,"TheCommitmentof PoliticalScience,"in his Dilemmasof Politics
(Chicago:Universityof ChicagoPress, 1958), p. 39.
92. Bourdieu,Outline,p. 24.
93. Ibid., p. 8.
94. 1will not defendthis choice of labelsat this point;my reasoningwill soon becomeevident.
RecallingErnstHaas's important1953 paper,"The Balanceof Power:Prescription,Concept,
or Propaganda?"WorldPolitics 5, 4, it may be an interestingexerciseto explore whetherthe
balance-of-powerscheme discussedhere could (undervariouscircumstances)generateall eight
of the meaningsHaas abstractedfrom the relevantliteratures.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
268 InternationalOrganization

Still,an answer,howevertentative,is required.Gropingtowardan answer-

and knowingthat any writtenanswerwill fail as much as it succeedsbecause
it idealizes that which has its true meaning in its practical site-I would
venture to introducethe balance-of-powerscheme as a "dialecticalunity of
pluraltotalities."An extremelysimple posturalscheme, it poises in unceasing
dynamic tension two opposed attitudesor interpretiveorientations.Lacking
better alternatives, let me call these opposed aspects the "particularityof
the universal"and the "universalityof the particular."
The particularityof the universal.Accordingto this side of the dialec-
tical relation, all universalizingclaims-all claims as to some universal
truth and objective necessity of law, morality,concept, theory, or politi-
cal order-are inherentlyproblematic.They are problematicin that
they reflect and conceal particularpoints of view and particularinter-
ests that cannot be reconciled with other points of view occupyingthe
same totality. Put somewhat differently,all universalizingclaims mask
an implicit hierarchyof social control relationscentered on some par-
ticular set of interests and subordinatingother opposing and legitimate
The universalityof the particular.Accordingto this side, all claims or
actions of a particularisticsort-interests expressed, actions undertaken,
and commitments made on the basis of immediate, contingent,and
specific experience-are inherentlyproblematic.They are problematic
because they bear implicit universalizingprojects, implicit claims on
what the whole world might be. Claims as to the truth of action or
commitments based on a unique heritageor on unreflectedunderstand-
ings of individual interests, for example, imply limits on what other as-
pects of the social totality can do, have, and be. They implicate the
social whole.
Such a scheme, it should be clear,in no sense impliesa necessaryreduction
of politics to the interplayof instrumentallogics (as in utilitarianmodels).
Quite the contrary. Among parties who recognize the scheme, including
classical realists, the scheme at once generatesand unifies all practicesin a
system of structures that are evaluative as well as cognitive. Against all
pretensesand aspirationsas to the existence of the possibilityof a universal
unity-be it the unityof unrestrainedreason,the anticipationof orderfounded
on law or morality, the expectation of logical unity implicit in positivist
science, the championing of universal empire, or a Cobdenite faith in the
harmonizinginfluenceof the market-the scheme commends recognitionof
countertendenciestowardpluralismor fragmentation,which will be animated
in reaction to just these universalisticand unifying aspirations. Likewise,
against all claims as to the rightnessor truth of a particularexperience or
point of view, the scheme commends attention to the universe of opposing
perspectivesthat might oppose or even negatethe first. In short, the scheme
legitimates a pluralityof perspectives on all relations in life, and it unifies

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 269

all political practice in opposition to all attempts to reduce political life to

the singularrational unity that any one perspectivewould impose upon it.
It is not too much to say that the balance-of-powerscheme, far from being
a logical relation deduced from a prior structureof states in anarchy,is the
constitutive principleof a pluralisticstates system. Practicesoriented with
respectto the scheme produceand transformthe moderninternationalorder.
The reasoning,briefly summarized,is as follows.
Working amidst ever-shifting factors and forces, including all of those
traditionallycalled "elements of national power," statesmen never literally
possess power and never truly hold the reins of control. Rather,competent
statesmen are engagedin an unceasingstruggle,at once artfuland strategic,
to be "empowered."They succeed to the extent that they can strikebalances
among all aspects of power-e.g., industrialcapacity, populationdemands,
militarycapability,nationalistlabor,internationalistbankers,and the consent
and recognitionof other statesmen-to establishan at least momentaryequi-
librium that, in turn, defines the state and its interests.It is towardjust this
balance of forces, a balance of power embedded in and producingthe state,
that statesmen strive. This, and nothing else, is the national interest, the
"national interest defined as power."95This equilibrium,and nothing else,
is what defines the state's boundariesand secures the effective distinction,
however momentary,between domestic and internationalpoliticallife. The
state, to borrow from Foucault, is an "effectof power."96
Defining and achieving this balance is the art of statesmen. Theirs is not
the task, as in neorealism,of securingthe ends of an unproblematical,given
state. Theirs, rather,is the art of orchestratingthe (re)productionof the state
in a transnationalcontext of other statesmensimilarlyengaged.In their art's
work, statesmen must of course pick their way through an overwhelming
maze of problems, roadblocks,and dead ends, all of which are susceptible
to countless interpretationsfrom multiple points of view. Amidst all these
complexities, competent statesmen know "instinctively"to interpretall re-
lations with referenceto a scheme firstinscribedon internationalinstitutions
as the earliest modern states established their precariousfootings in the
dialectic of particular aristocratic and universalizing bourgeois interests.
Through long practice,they have internalizeda predispositionto orient all
practices with reference to the axial principle, the generative scheme, of
balanceof power.For statesmen,this simple dialecticalscheme has a genuine
economy of logic (though it is not the logic of economy) that will make their
own practicescomprehensiblein the eyes of other competentstatesmen,and
thanksto whichthey can understandtheirpracticesas well. Moreimportantly,

95. Morgenthau, Politics among Nations, p. 6.

96. Foucault,Power/Knowledge,pp. 73-74. In the specificquotation,Foucaultrefersto the
individualas an "effectof power,"but his overall argumentis applicableto the empowering
or constitutionof all subjective agents, includingthe state. His "Two Lectures"addressthe
relationbetween the two.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
270 InternationalOrganization

as judiciously applied by artful participants,the scheme orients the com-

prehension of interests and the undertakingof practices that promise to
optimize power, in its fullest sense, on behalf of the state.

b. The political conceptionof political concepts

As we listen furtherto the conversationamong classicalrealistghosts, and
as we begin to get a feel for the balance-of-powerscheme at workthroughout
classicalrealistdiscourse,anotherinsightcomes. If classicalrealistssometimes
appear slippery in their use of concepts, it is not always because of a lack
of discipline, an inductivist bias, or a failure to think in systemic terms.
Neorealists mistakenly read classical realism in this way because they do
not understand the tradition they purport to carry forward. They do not
understandthat, for the classical realist, the fruits of intellectuallabors are
no more immune to the dialecticallogic of the balance-of-powerscheme-
at once subjectiveand objective-than are the institutionsand practicesthey
study. They do not see that, thanks to classicalrealists'commitment to the
balance-of-powerscheme, classicalrealisttheoryinstitutionalizesa persistent
struggleto overcome problemsnoted earlierwith respectto neorealisttheory.
In particular,classicalrealisttheorizinglong soughtto avoid the tendency,
so evident in neorealism, to let the "impulse to theorize" manifest itself in
the production of a lifeless, antihistoricalenclosure. Deeply committed to
the balance-of-powerscheme, classical realists such as Martin Wight are
consciousof "a kindof recalcitranceof internationalpoliticsto beingtheorized
about."97Hans Morgenthaucautions that there is "a rational element in
political action that makes politics susceptible to theoretical analysis, but
there is also a contingentelement in politics that obviates the possibilityof
theoreticalunderstanding."98 Both warn against any attempt to arrive at a
complete, naturalisticaccountingin terms of finishedexternalstructuresthat
finally contain the political world. As both understand,to try to impose a
single conceptualorder on internationalpolitical life, even if that attempt is
inspired by the dialecticalbalance-of-powerscheme, is to deny the truth of
the scheme that inspiresthe attempt.
From the vantage point of classical realism, as from the vantage points
of statesmen participatingin the traditionof the modern states system, all
key concepts-power, nationalinterest,the sovereignstate, the statessystem,
and so on-must be understoodin referenceto the balance-of-powerscheme.
Seen as finding expression throughout all levels and in all things of the
political universe, the balance-of-powerscheme dictates an understanding
of all such concepts as reflecting (perhaps momentary) equilibria among
97. Martin Wight, "Why Is There No InternationalTheory?"in HerbertButterfieldand
Wight,eds., DiplomaticInvestigations(London:Allen & Unwin, 1967).
98. Hans Morgenthau,"The Intellectualand PoliticalFunctionsof Theory,"in Morgenthau,
Truthand Power.Essays of a Decade, 1960-1970 (New York:Praeger,1970), p. 254.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 271

opposed tendencies and opposed interests:monarchand church,nationalist

and internationalist,local and global,regressiveand progressive,traditionalist
and rationalist,fragmentingand universalizing,aristocracyand bourgeoisie.
Not even the structuresof the modern state or the states system, not even
the practicalefficacyof the balance-of-powerscheme itself, can be taken as
given. They are essentially political concepts because they are, in the words
of a recentdebate,"essentiallycontested"or "essentiallydisputed"concepts.99
And they are disputed not so much because of a lack of agreement but
because, at a deeperlevel, there is an agreementto disagree.There is a more-
or-less consensual recognitionof the truth of a dialecticalscheme that both
disallowsany finalclosureon a singular,contradiction-free truthand generates
the expectation that, for reasons unspoken, there will forever be pressures
to subsume the whole within a false unity.
In contrast to neorealism's closure, then, the tradition of practicerepre-
sented by classical realism (the tradition institutionalizedin the modern
states system) is never really "complete." Rather, it appears to each new
generationthat inheritsit as a lively, difficult,ambiguous,and nevercompleted
struggle.This strugglerequiresthe creativeinterpretationof new circumstances
and sometimesvirtuosoperformanceof system renewal,in lightof commonly
recognizedorganizingschemes embedded in past experienceand embodied
in currentsocial structures.The compromisesamong contendingforcesmust
ever be won again.The balancesmust ever be restruck.The strategicalliances
with various factions having contesting nationalist and internationalist
claims-alliances without which statesmen might not secure the autonomy
requiredto permittheir competent participationaccordingto classicalrealist
expectations-must ever be drawn anew. Always and everywhere,balances
are in jeopardy;always and everywhere,strategicartistryis required.

c. Power, recognition,and balance of power

At last we are beginningto graspthis balance-of-powerscheme. Although
we are still no doubt quite clumsy in its application-overcoming this clum-
siness does take practice and a sense of history-it does at least enable us
to make sense of the continuingconversationamong classicalrealists.Take,
for instance, their discussions of power. We can now see that the classical
realist conception, when compared to the neorealist conception, is consid-
erably richer and much more sensitive to power's social basis and limits.
We can sense that this is because the classical realistconception is founded

99. See W. B. Gallie, "Essentially Contested Concepts," Proceedings of the Aristotelian Society
56 (1955-56); StuartHampshire,ThoughtandAction(New York:Viking,1959);StephenLukes,
Power.A Radical View(London:Macmillan, 1974); K. I. Macdonald,"Is 'Power'Essentially
Contested?"BritishJournalof PoliticalScience6 (November 1976),pp. 380-82; Lukes,"Reply
to K. I. Macdonald,"ibid. 7 (1977), pp. 418-19; John Gray, "On the Contestabilityof Social
and PoliticalConcepts,"ibid.(August1977);and Shapiro,Languageand PoliticalUnderstanding.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
272 InternationalOrganization

on an implicit competence model of political practice,a model anchoredin

the balance-of-powerscheme. The scheme dictates a commitment to the
necessaryambiguityofpoliticalreality,and this in turnestablishesthe dialectic
of recognitionas a necessaryand irreducibleaspectof all socialpowerrelations.
This becomes clearerwhen one takes note of the interpretiveunity at work
behind classical realists'dual usage of the term "power,"both as a capacity
to influence(to "have power")and as a designationof competent,recognized
participants(to "be a power"). To be recognizedas a power, that is, as a
sovereignstate, a state must satisfycertainminimalrequisites(e.g., effectively
patrolledterritory);and, in general,these requisitesreflecta state's capacity
to makegood on the claim that its statesmen'spartiallyuniqueinterpretations
of the whole need to be recognized and granted a separate voice in the
dialogueamong mutually recognizingstates.'00To "increasepower," states-
men must engage in practicesthat serve to move their own vantage points
on the whole, and within the eyes of other powers, toward the collectively
recognized"central"vantage point-the view from the center of a plurality
of political orbits. To become a "dominant power" is to become the one
power recognized as the state whose vantage point, and interests defined
therefrom,defines the objective limits of politicaland social possibility;that
is, the limits that no power can violate without imperilingits status as a
Whethera stateis a minoror a dominantpower,though,its powerconfronts
its own limit, as contained in the anchoring scheme of balance of power
recognized by all competent parties to the system. Even among dominant
powers (perhaps especially so) statesmen must make good on the claim,
implicit in the modern concept of sovereignty,'?'that their own vantage
100. However, beforeone leaps to the conclusionthat there is some fixed set of operations
by which one translatescertainobjective requisitesinto "attributesof statehood,"it must be
stressed that the requisitesof statehood themselves depend upon collective recognition,are
essentiallyreinterpretable,are subjectto dispute, and have historicallyevolved.
101. This is one of the points where Ruggieand I part ways-or, perhaps,it is on this point
that Ruggieneeds to make more explicit what remains unclear,at least to me. In his review
of Waltz, Ruggiedrawsinterestingparallelsbetweenprivatepropertyrights,as conceptualized
by Locke, and sovereignty,as conceptualizedby Vattel. I agree that Vattel exhibits strong
parallelsto Locke's atomistic and contractarianviews; as Quincy Wrightnotes of Vattel, he
"adheredto the atomistictheorywhichholdsthat internationallaw is merelya seriesof contracts
betweenwholly independentstates" (Wright,A Study of War[Chicago:Universityof Chicago
Press, 1964], p. 230). I think, too, that the parallelsare provocative:Ruggie'sanalysiscauses
me to wonder if there might be more than an analogyat work here, if perhapsthere is more
than a coincidencebetween the emergenceof new relationsof laborand property,as justified
by Locke, and the new mode of sovereignty,as justifiedby Vattel.
It occursto me, however,that Ruggiemay have been drawninto a scholasticargumentwhich
causeshim to exaggeratethe surfaceparallelsbetweenintellectualrationalizationsat the expense
of an understandingof a real differencebetween private propertyrights and sovereigntyas
active principlesof practice.I wouldlike to offerthe view thatthe modernconceptof sovereignty
designatesthe collectively recognizedcompetenceof entities subject to internationallaw and
superiorto municipallaw. It thus involves not only the possessionof self and the exclusionof
others but also the limitationof self in the respectof others, for its authoritypresupposesthe
recognitionof others who, per force of their recognition,agree to be so excluded. In effect,

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 273

points on the whole are indeterminate,ambiguous,and allow for the plurality

of possible interpretationswithin the whole that they themselves interpret.
Playingoff the balance-of-powerscheme, the powerfulstatesmanmust make
good on the claim that his interpretationof the whole allows room for the
interpretivevantagepoints of even the lowliestof "minorpowers"recognized
by the system of states and from whom the dominant power expects rec-
ognition in return. For to the extent that a power statesman fails to make
good on this claim in the eyes of others-perhaps because he has become
captive of moralzeal, for instance,or becausehe has surrenderedto "national
interests" in the most "egoistic" sense of the term-his practiceswill take
on a special significance when read in reference to the balance-of-power
scheme. His practices will signify the state's incompetence as a partnerto
the dialogue of internationalpolitics. The statesman representingthe power
in question will be viewed as unwilling, or perhaps as structurallyunable,
to learn from diplomatic interaction. His diplomatic practiceswill be seen
as nothing more than instrumentsof "nationalistuniversalism,"the attempt
on the part of a single society to bend the whole world to the logic and
demands of its unique national experience. Such a state will actually lose
power, or what we may call "symbolic capital."'02
In fact, there is a strong sense in which a master-slavedialecticof power
obtains: one in which the aspirationto absolute power yields absolute im-
potence, and the demand for a totally gratifiedego negates being itself. At
the extreme, the high-capabilitystate whose leaders performincompetently
and allow themselves to be understoodas attemptingto make globally de-
terminate their own univocal interpretationsof order will be denied recog-
nition, and hence existence within the communityof states, altogether.From
the point of view of the tradition,such statesmenwill be seen not as competent
partnersin a communityof sovereignstates,and leastof all as worthyleaders,
but as threats and dangersto be opposed, taught to behave if possible, and
expelled or destroyed if necessary.

d. Rudiments of a theory of internationalpolitical practice

The conversationamong classicalrealistghosts continues. And as it does,
classical realism's status in internationalrelations discourse, and its impli-

sovereigntyis a practicalcategorywhose empiricalcontents are not fixed but evolve in a way

reflectingthe active practicalconsensus among coreflectivestatesmenwho are ever struggling
to negotiateinternaland externalpressuresand constraintsand who, if competent,orienttheir
practicesin respectof the balance-of-powerscheme. Thus, one cannot say flatlythat sovereign
states exhibit a "form of sociality characteristicof 'possessive individualists,'for whom the
social collectivity is merely a conventionalcontrivancecalculatedto maintainthe basic mode
of differentiationand to compensate for the defects of a system so organizedby facilitating
orderly exchange relations among the separateparts" (Ruggie,p. 278). One has to say that
practiceso normalizedmay be associatedwith a particularform of sovereigntyunderspecific
historicalcircumstancesyet to be explained.
102. See especiallythe last chapterof Bourdieu'sOutline.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
274 InternationalOrganization

cations for scientific theory, begin to be seen in a differentlight. It is clear

that classical realism, owing to its practical interests and its reliance on a
competence model, provides a richer political frameworkthan does neo-
realism: at once broader in the scope of critical questions it can entertain
(includingself-reflectivecriticism),superiorin its generativepotential, more
elegant and less demanding in its assumptive and conceptual bases, and
more sensitive to the deep politicallimits and dilemmas of competentinter-
national practice.At the same time, there is somethingtroublingabout this
conversation-especially if we measure it as a scientificdiscoursethat might
provide a viable alternativeto neorealisttheory. Thanks to its commitment
to "the priorityof practiceto theory," classical realism is vulnerableto the
charge,advanced by neorealists,of circularity.'03 It is ensnaredin the "her-
meneutic circularity"of the traditionof practiceit interprets.
Evolving its theory while peering "over the shoulder of the statesman
when he writes his dispatches,"'04classical realism can advance its theory
no fartherthan competent statesmen,in the courseof their practice,are able
to theorize about themselves and their circumstances.Classicalrealismthus
cannot pose questions that competent professionalsin the traditionwill not
recognize, including those questions that they have a structuredpolitical
interest in not recognizing. Where competent statesmen are prepared to
recognizeproblems, classicalrealism will give voice to problems. But where
competent statesmen have an interest in silence, classical realism will be
silent, too. Among these problems are those that would call into question
the traditionwithin whose context statesmendemonstratetheir competence,
securerecognition,and orchestratethe empoweringof states.For the classical
realist, as for the competent statesman, such questions are not literallyfor-
gotten. Rather, they inhabit the domain of "that which must not be said."
They are unspoken and unrecognizedby competent parties as a condition
of their competence.'05
Consideringthis tradition-boundcircularityof the classicalrealistdialogue,
we are forced to conclude that, as a theoreticalalternativeto the neorealist
orreryof errors, classical realism fails. It fails as a theory of world politics,
in the firstplace, becauseit is so deeply immersedin the traditionit interprets
that it lacks any independenttheoreticalstandardsfor the criticism of that
tradition'slimits. It fails, in the second place, because it honors the silences
of the traditionit interpretsand thus containsno categoriesthat would allow
one to specify, or even advance hypotheses about, the historicalconditions
103. Martin Wight, "The Balanceof Power and InternationalOrder,"in Alan James, ed.,
The Bases of InternationalOrder(London:OxfordUniversity Press, 1973).
104. Morgenthau,Politics among Nations, p. 5.
105. This interestin silence, it must be stressed,is not an instrumentalinterest,not a relation
that competentstatesmenconsciouslyconceptualizein logicalor means-endsterms.It is, instead,
one that statesmen do not necessarilyconceptualize,one best served when it is universally
internalizedwithout conceptualization.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 275

that make that traditionpossible or the deep and transnationalsocial power

structuresthat tradition-boundpractice at once presupposes, reproduces,
and mystifies. It fails, in the third place, because it is unable to contemplate
the historicallyspecificdevelopments that threatenthe unspokenconditions
upon which the dominance of that traditionhistoricallyrests. It is therefore
unable to grasp the deeper dimensions of crisis in the world polity. And it
fails, in the fourth place, because in refusingto comprehendits own limits
it must refuse as well to engage and learn from opposing theories and ar-
guments (includingtheories of imperialism,transnationalism,and interde-
pendence)that would call these conditionsinto question.Thus, while classical
realism is rich with insights into political practice, it fares no better than
neorealism as a scientific theory of internationalpolitics. Though it closes
on an understandingthat is far truer to the traditional practice of world
politics, it is no less closed. Though it affirmsthe hegemony of a balance-
of-power scheme ratherthan the hegemony of a logic of economy, it is no
less self-affirming.
To say that classical realism should be denied the status of a scientific
theory, though, is not to say that it should be banished from the theoretical
discourseof internationalrelations.On the contrary,its practicalsignificance
should be accommodated within any reasonablycomplete theory of inter-
national politics. True to its own practicalcommitments, it should be con-
ceptualized within theory as the "organicintellectuality"of a transnational
tradition of statesmanship. It should be conceptualizedas the ideological
apparatusof a global professionalcommunity, the communityof competent
statesmen, which administersthe recognizedpublic sphere of international
life, which is able to rememberand interpretits past, which can learn from
its experiencesand theorizein limited ways about itself and its performance,
and which can to some degree transformits practicesand its institutionsin
light of its remembrancesand its theories.'06So viewed, classical realism's
lapses, circularities,silences, and omissions cannot be regardedas pretexts
for "rejecting"and then ignoringclassicalrealistargumentor, as in neorealism,
for developinga "new,improved,scientificallytested"version.Suchreactions
forget that classical realism and its lapses and omissions at once mirrorand
ideologicallyreproducea traditionthat constitutes importantaspects of the
world we study. Instead, classical realism, its generative potential, and its
limits and distortions should be addressed, interpreted,explained, and de-
ployed as part of the explanation within a theory of modern international
political practice.
To think of classical realism in this way is to begin to look beyond the
simple competence model implicit in the conversation among ghosts. It is
106. The term "organic intellectual" is due to Antonio Gramsci in his path-breaking studies
of hegemony. See Quintin Hoare and Geoffrey Nowell Smith, eds., Selections from the Prison
Notebooks ofAntonio Gramsci (New York: International Publishers, 197 1), especially pp. 5-14.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
276 InternationalOrganization

to anticipatethe development of a dialectical competencemodel-a model

that would overcome classical realism's failingsand provide a viable alter-
native to neorealism'seconomic model of choice under constraints.Several
features of such a model merit notice.
First, such a model would be developed to account for the emergence,
reproduction,and possible transformationof a world-dominantpublic po-
litical apparatus:a tradition or regime anchored in the balance-of-power
scheme and constitutive of the modern states system. The regime should
not be construedto organizeand regulatebehaviorsamong states-as-actors.
It insteadproducessovereignstates who, as a condition of their sovereignty,
embody the regime. So deeply is this regime bound within the identitiesof
the participantstates that their observations of its rules and expectations
become acts not of conscious obedience to something external but of self-
realization,of survival as what they have become.'07We may refer to this
regimeas a balance-of-powerregime.We may understandit to be the tradition
of statecraftinterpretedby classical realism. Again, classical realistsare the
"6organicintellectuals"of this regime,the reigningintelligentsiaof the world-
wide public sphere of modern global life.
Second, such a model would situatethis balance-of-powerregimein terms
of the conditionsmakingit possible:the social,economic, and environmental
conditions upon which its practical efficacy depends. One such condition
can be inferredfromclassicalrealists'notorioussilenceon economicprocesses
and their power-politicalramifications.As Hedley Bullsays of MartinWight,
so can it be said of classical realists and regime-boundstatecraft:they are
".notmuch interestedin the economic dimension of the subject."''08 How is
it possible for the balance-of-powerregimeto maintainsuch a silence?Under
what historicalsocial, economic, and environmentalconditionsis it possible
for the balance-of-powerregime,as the publicpoliticalsphereof worldsociety,
to maintainsilence on matterseconomic while at the same time coordinating
and orienting practicesin ways reaffirmingthe regime itself? One possible
answer is that the regimepresupposescapitalistrelationsof productionand
exchange.It presupposesa deep consensus grantingcontrolover production
to a sphere of "private" decisions that are themselves immunized from
public responsibility-a practicalconsensus that thereby producesa sphere
of "economy" operatingaccordingto technical rationallogics of action. In
turn, such a consensus further presupposes capitalist labor and property
relations. This consensus, together with the worldwide power bloc whose
dominance it signifies and secures, might be called the modernglobal he-
gemony. The balance-of-powerregimeis its publicpoliticalface. The silences

107. See Richard K. Ashley, The Political Economy of War and Peace: The Sino-Soviet-
American Triangleand the ModernSecurityProblematique(London:FrancesPinter, 1980),
pp. 38 and 279-86.
108. Hedley Bull, "MartinWight and the Theory of InternationalRelations:The Second
MartinWightMemorialLecture,"BritishJournalof InternationalStudies2 (1976), pp. 101-16.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 277

of the regimeon matterseconomic at once reflectand reinforcethe dominant

power bloc's control over productionindependentof public responsibility.
Third, such a model would necessarilyaccount for the balance-of-power
regime's orientation and coordination of political practices such that, on
balance(andas an unintendedconsequence),they tend to directcommitments
of resourcesand the developmentof ideologicallegitimationsin ways securing
the possibilityconditionsof the regime.The model might show, for example,
how the competent statesman's interest in accumulatingsymbolic capital,
or symbolic power, by playing off the balance-of-powerscheme, effects a
"double standard"of politicalaction. That double standard,in turn, secures
the politicalpreconditionsof globaldominationon the partof a transnational
capitalistcoalition,the dominantpowerbloc of the modernglobalhegemony.
Fourth, such a model would explore the learningpotentialof the balance-
of-power regime. In particular,along the lines of Pierre Bourdieu'sargu-
ment, '09it might furtherdevelop its specificationsof the processof symbolic
capital accumulation. It might explore how symbolic capital, accumulated
throughthe ambiguousand "disinterested"performancesof competent heg-
emonic statesmen,providesa kind of "creativereserve,"a basis in authority,
for the exerciseof leadershipin the orchestrationof collective improvisations
in response to crisis.
Fifth, such a model would offer an account of crisis. It would specify the
tendenciesthreateningto undermineor transformthe conditionsupon which
the practicalefficacyof the balance-of-powerregime depends. It might spe-
cificallyconsider those tendencies that threatento eradicatethe statesman's
latitudefor ambiguous,intrinsicallyequivocalpoliticalperformanceshonoring
the balance-of-powerscheme and not immediatelyreducibleto expressions
of economic interests.''? Owing to this loss of latitude for ambiguous per-
formances, it might be shown, the competent statesman is deprived of the
ability to accumulate symbolic capital and, with it, a reserve capacity for
learningand changein responseto systemcrisis.Suchreasoningwould suggest
the possibility of a world crisis-not just one more cyclical economic crisis,
but an epochal crisis of world political authority,a crisis involving a degen-

109. Bourdieu,Outline,chap. 4.
110. A number of tendenciesare relevant in this connection. Most can be associatedwith
late capitalistdevelopment:"post-industrial"formsof state legitimationaccordingto whichthe
state legitimatesitself, not on traditionalgrounds,but increasinglyas an economic dysfunction
manager;the fiscal crises of modem states strugglingto justify themselves in these terms;the
internationalization of capitaland the emergenceof newly industrializedcountries,resultingin
a malalignmentof worldindustrialcapacitywith political-coercive meansand traditionalsymbols
of politicalpower;the globalizationof the world polity such that hegemonic"responsibility"is
ostensibly universal,with no "externalareas" remainingfor the externalizationof costs; the
contradictionsexposed through encountering"limits to growth";the emergenceof socialist
movements aimingto institutionalizethe public politicaldeterminationof productionand ex-
change but which are also under pressureto rationalizetheir politics;the Cold War, which
institutionalizesthe totalizationof politicalcompetition;and nuclearweapons, which institu-
tionalizethe possibilityof totalizedwarfare.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
278 InternationalOrganization

eration in the learningcapacity of the regime and, consequently, a loss of

politicalcontrol. Understood in the context of the modernglobalhegemony,
such a crisismight be expectedto be markedby the economizationof politics
and the resulting loss of political autonomy vis-a-vis economic and tech-
nological change. As if internationalpolitics were the last frontier of the
progressiveworld rationalizationtendency delineatedby Weber,hegemonic
practice might come under increasingpressuresto find its rationalenot by
playingequivocallyoff the balance-of-powerscheme, but by measuringevery
gesture in terms of the ultrarationalisticlogic of economy.
Sixth, such a model would not view the modern global hegemony in iso-
lation. Nor would it mistake it for the whole of world politics."' It would
insteadregardit as the dominantworldorderamonga multiplicityof mutually
interpenetratingand opposed world orders,some of which might escape the
logic of the modern global hegemony and assert alternativestructuringpos-
sibilities under circumstancesand by way of oppositionalstrategiesthat can
in principlebe specified. For example, the modern global hegemony might
be understoodto contest with-and, as a kind of "pluralisticinsecuritycom-
munity,"to contain-totalitarian communist, collectivistself-reliance,Euro-
communist, Muslim transnationalist,and corporatistauthoritarianworld
order alternatives."2Developing such a niodel would involve exploringthe
strategies by which oppositional movements representingthese and other
alternativesmight take advantageof the indeterminateand ambiguousqual-
ities of regime-boundstatecraft,while exploiting its traditionalsilences, to
transformits conditions of dominance, producethe conditions of their own
self-realization,and secure the widening recognitionof their own ordering
principlesas the active principlesof practice."I3
These anticipations of theory are, of course, rudimentaryat best. They
do, however, suggestsome possibilitiesfor the development of a model that
would preserve classical realism's rich insights into internationalpolitical
practicewhile at the same time exposingthe conditions,limits, and potential
for change of the tradition in which classical realism is immersed. Fully
developed, such a model would more than surpassneorealism.It would offer
an interpretationof neorealism, finding in it a historicallyspecific reaction
to crisis that refusesto comprehendthat crisisbecauseit cannot acknowledge
the richnessof the traditionthat is endangered.It would interpretneorealism,

11 1. Ashley, Political Economy of War and Peace, pp. 294-98; Hayward R. Alker Jr., "Can
the End of Power Politics Possibly Be Part of the Concepts with Which Its Story Is Told?" in
Alker's "Essential Contradictions, Hidden Unities" (in progress).
1 12. See Hayward R. Alker Jr., "Dialectical Foundations of Global Disparities," International
Studies Quarterly 25 (March 1981).
1 3. As might be inferred from this description, the capitalist power-balancing order addressed
in this dialectical competence model is not understood to exhaust the totality of international
political reality worthy of theoretical examination. On the contrary, while it is arguably the
dominant mode of order, it is but one point of entry into the theoretical analysis of an international
reality that consists of the dialectical interplay and interpenetration of multiple world orders.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 279

in other words, as an ideologicalmove towardthe economizationof politics.

And it would underscorethe possibly dangerousconsequencesshould this
move succeed.For from the point of view of such a model, the economization
of internationalpolitics can only mean the purgingof internationalpolitics
of those reflective capacities which, however limited, make global learning
and creativechangepossible.It can only mean the impoverishmentof political
imaginationand the reductionof internationalpolitics to a battlegroundfor
the self-blind strategicclash of technical reason against technical reason in
the service of unquestionedends.

e. The classical realist repudiationof neorealism

We do not need to await a dialecticalcompetence model's critical inter-
pretationof neorealism'sflawsand dangers,however.Such a critiquealready
exists. It is alreadypresent in the very literatureto which neorealismclaims
to owe its intellectual debts. In classical realism's revered texts, we find
recurringwarningsagainst the very practicesthat neorealism has made its
own. In the name of realism, neorealism commits specific errors against
which classical realists specificallywarned.
Listen, for instance, to the sixteen-year-oldwords of Hans Morgenthau,
whom Gilpin calls "the leadingmodem spokesmanfor politicalrealism."" I4
What characterizescontemporarytheories of internationalrelationsis
the attempt to use the tools of modern economic analysis in a modified
form in order to understandinternationalrelations.... In such a theo-
retical scheme, nations confront each other not as living historic entities
with all their complexities but as rationalabstractions,after the model
of 'economic man.' ...
Hear Morgenthau'swords of nineteen years ago:
... This theorizingis abstractin the extreme and totally unhistoric.It
endeavors to reduce internationalrelationsto a system of abstract
propositionswith a predictive function. Such a system transformsna-
tions into stereotyped'actors' engagingin equally stereotypedsymmet-
ric and asymmetric relations. What Professor[Martin]Wight has noted
of internationallaw applies with particularforce to these theories:the
contrast between their abstractrationalismand the actual configurations
of world politics. We are here in the presence of still another type of
progressivisttheory. Its aim is not the legalizationand organizationof
internationalrelations in the interest of internationalorder and peace
but the rational manipulationof internationalrelations... in the inter-
est of predictableand controlled results. The ideal toward which these
theories try to progressis ultimately internationalpeace and order to be
achieved through scientificprecision and predictabilityin understanding
and manipulatinginternationalaffairs.
114. Gilpin, Warand Change,p. 213.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
280 InternationalOrganization

A quarter-centuryago:
... [T]he contemporary political scene is characterized by the inter-
action between the political and economic spheres....
Yet what political science needs above all changes in the curricu-
lum-even though it needs them too-is the restorationof the intellec-
tual and moral commitment to the truth about matters political for its
own sake. That restorationbecomes the more urgentin the measure in
which the general social and particularacademic environment tends to
And thirty-six years ago:
The very appearanceof fascism not only in Germany and Italy but
in our very midst ought to have convinced us that the age of reason, of
progress,and of peace, as we understood it from the teachingsof the
eighteenth and nineteenth centuries, had become a reminiscenceof the
past. Fascism is not, as we preferto believe, a mere temporaryre-
trogressioninto irrationality,an atavistic revival of autocraticand bar-
baric rule. In its mastery of the technologicalattainmentsand
potentialitiesof the age, it is truly progressive-were not the propa-
ganda machine of Goebbels and the gas chambersof Himmler models
of technical rationality?-and in its denial of the ethics of Westerncivi-
lization it reaps the harvest of a philosophy which clings to the tenets
of Western civilization without understandingits foundations. In a
sense it is, like all real revolutions, but the receiver of the bankruptage
that preceded it."5
Other famous lines could be recalled, includingthe contrastatisttheme in-
troduced as part of Morgenthau'ssix-point manifesto of political realism:
"While the realist indeed believes that interest is the perennialstandardby
which political action must be judged and directed, the contemporarycon-
nection between interest and the nation state is a productof history, and is
thereforebound to disappearin the course of history."'16
Now it is abundantlyclearwhy neorealistinterrogatorshave been so abrupt
in theirquestioningof the classicaltradition.These extensive remarkssuggest
that classicalrealists,given a chance to speak, would be among neorealism's
sternestcritics. As Morgenthaustressedon many occasions, utilitarian,pos-
itivistic,and rationalistcommitments-commitments presentin neorealism-
tend to pitch social science on an unhistoricand apolitical attitude toward
politics, an attitude too often dangerouslyallied with the state. Such com-
mitments threatento producea form of pseudo-politicalunderstandingthat
falselyreducesthe inherentlydialecticalcharacterof politicsto the monothetic
orientation of economic reason, an orientation in which all perspectives,
115. Morgenthau, "Common Sense and Theories," in Truthand Power, p. 244; "The Intellectual
and Political Functions of Theory," p. 248; "The Commitments of Political Science," p. 48;
Scientific Man versus Power Politics (Chicago: University of Chicago Press, 1946), pp. 6-7.
1 16. Morgenthau, Politics among Nations, p. 10.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 281

even the measure of power and its changes, are thought to be ultimately
collapsible into a singular, internally consistent scale of universally inter-
convertible values. Such commitments permit no real sense of political di-
lemmas, no real appreciationof the autonomy of the political sphere. They
tend toward a one-sided rationalism,a view tragicallyflawed for its lack of
a sense of the tragic, a half-truththat must ultimately be transformedinto
the opposite of itself and must produce its own ruin in the vain search for
a universaldomain. These commitments are not just politicallynaive. They
are positively dangerous.
I do not mean to glorify classical realism. It is, as I have said, a tragic
tradition. It is a tradition whose silences, omissions, and failures of self-
critical nerve join it in secret complicity with an order of domination that
reproducesthe expectationof inequalityas a motivatingforce,and insecurity
as an integratingprinciple.As the "organicintellectuality"of the worldwide
public sphere of bourgeois society, classical realism honors the silences of
the traditionit interpretsand participatesin exemptingthe "privatesphere"
from public responsibility.It thus disallows global public responsibilityfor
"economic forces" that will periodically disrupt and fragment the global
public sphere. Herein, I think, are the seeds of realismas a traditionforever
immersed in the expectationof politicaltragedy,an expectationthat realists
can explain only euphemistically,in terms of the antinomies of fallen man.
My aim, rather,is to underlinethe fact that neorealism is not worthy of
its name. Neorealismscoffsat classicalrealism'swarningsand sense of limits,
misstates its interests, deadens its ironies, empties its concepts, caricatures
its rich insights, reduces practice to an endless serial performanceof con-
strained economic choices on the part of one-dimensional characters,and
casts the whole of it up beforea flat historicalbackdropdevoid of perspective,
contradiction,and life. Once again,the memory of TheEighteenthBrumaire
comes to mind: "Hegel remarkssomewhere that all great, world-historical
facts and personagesoccur, as it were, twice. He forgotto add: the firsttime
as tragedy,the second as farce.""7

5. A concludingself-critique

E. P. Thompson concluded ThePovertyof Theorywith an "obligatory"auto-

critique.I, too, sense the obligation.Having played the critic, I want to offer
a few self-criticalremarksin conclusion. Much that I have said needs to be
criticized-so much, in fact, that I fearthat the severalrespondentswill have
neither the time nor the space to give me all the swats I deserve. Let me
concentrateon five criticisms that I think are important.
First, my argument may not have been sufficientlyattentive to the sig-

117. Marx, "Eighteenth Brumaire," p. 436.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
282 InternationalOrganization

nificanceit will take on when readagainstthe backgroundof the greatbattles

of the past. In particular,much of my argumenthas crossed the venerable
battlegroundwhere once the entrenched soldiers of "tradition"confronted
"science" on the march, and I may have left room for the conclusion that
I have enlistedin the ranksof "tradition"as over againstneorealist"science."
In fact, such a conclusionwould be mistaken.For the burdenof my argument
in this respect is not to condemn science in favor of tradition. Nor is it to
suggest the dilution of scientific standards by broadeningliberal scientific
toleranceto embracetraditionas part of science. What is called for, instead,
is a methodologicallymore demanding science: a science that expands the
range of allowable criticism, and sharpens the standardsof theoreticalad-
equacy, by institutionalizingthe expectationof continuouscriticalreflection
on the historicalsignificanceand possibilityconditions of our own attempts
to arrive at objectivist conclusions. That is, a dialectical science is called
Second, I think there is some merit to the charge that I am engagingin
that sort of criticism which, if successful, leaves an aching void in the soul
of a discipline. In attackingneorealism,I have moved not only againstneo-
realist theory but also against predominantunderstandingsof method that
deflect criticismand obscureneorealism'smany theoreticalflaws.Yet I have
only brieflyoutlined some possibilitiesfor an alternativetheory, and I have
not even begun to sketch the implicationsof my argumentfor an alternative
method. I have failed to outline a dialectical methodology that recognizes
the inevitable opposition of theory and practiceand attempts to internalize
that opposition in its method. I have failed to stress adequatelythe need to
approach all aspects of internationalsystems not only as reflectionsof an
objective whole but also, and at the same time, as potentiallybearingcom-
peting wholes, competing ordering principles strugglingto find unfettered
expression. I have insufficientlystressed that practice must be grasped not
just as conformity to the norm but also as a continuingstrategicstruggleto
produceor escapealternativeforms of order.If my critiqueis to be successful,
these gaps will need to be filled."'
1 8. Bourdieudescribesthree modes of knowledge-phenomenological,objectivist,and di-
alectical. For a discussion of these three modes of knowledge in the study of international
politics, see Ashley, "Realist Dialectics."
119. One major projectunder way is the "Dialecticsof WorldOrder"work of HaywardR.
Alker Jr., Thomas Biersteker,Ijaz Gilani, and Takashi Inoguchi. See, for example, Alker,
"DialecticalFoundationsof Global Disparities";Alkerand Thomas Biersteker,"The Dialectics
of WorldOrder:Notes for a FutureArchaeologistof InternationalSavoirFaire" (Deliveredat
the September 1982 meeting of the American Political Science Association, Denver, Colo.);
and Alker,Biersteker,and Inoguchi,"FromImperialPowerBalancingto People'sWars:Searching
for Orderin the TwentiethCentury"(Presentedat the April 1983 meetingof the International
Studies Association, Mexico City). The World Order Models Project,under the directionof
Richard Falk and Saul Mendlovitz, can be said to exemplify a dialectical methodologyof
normativeclarificationby whichcompetingworldorderdevelopmentalconstructs,representing
various social and culturalpoints of view, are exposed, confronted,and elaborated.See for
example, Richard Falk, A Study of Future Worlds(New York: Macmillan, 1975). Papersby

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 283

Third, it may be reasonablyarguedthat I have unfairlysingled out neo-

realismto take the bruntof a critiquethat could easilybe expandedto include
other targets. Neorealism is in many ways just part of a trend. I think it is
evident, for example, that the "economization"of political theory is not a
phenomenon unique to internationalrelations theory in the United States.
The current"legitimationcrisis" in the U.S. polity may help to account for
the suddenreinvigorationof microeconomictheoriesof politics,game theory,
exchangetheory, rationalchoice theory, and public choice theory, but such
economic theories of domestic politics have always been near the center (if
not always the surface)of the postwar scientific study of politics. I think it
is evident, too, that even among internationalrelationstheories, neorealism
is not the only perspective meriting such an attack. In importantrespects,
much that I have said here could as easily be applied to Immanuel Wall-
erstein's world systems perspective. I think that Wallerstein'sneo-Marxist
structuralismexhibits many of the positivistic and physicalistic qualities
earlierascribedto neorealism.I think, too, that Wallerstein'sperspectiveis,
in some ways, just as guilty of statism. I would be inclined to account for
this replicationof flaws in terms of Wallerstein'ssharinga Weberianlineage
(in his case via Merton and Parsons)with neorealism.'20

Terence Hopkins indicate an important effort toward the development of dialectical perspectives
within neo-Marxist world systems analysis. See Hopkins, "World Systems Analysis: Method-
ological Issues," in Barbara Kaplan, ed., Social Change in the Capitalist World Economy (Beverly
Hills, Calif.: Sage, 1978); and Hopkins and Immanuel Wallerstein, "Cyclical Rhythms and
Secular Trends of the Capitalist World Economy," Review 2, 4 (1979). Johan Galtung's The
True Worlds as well as his most recent methods text, Methodology and Ideology, outline and
richly illustrate a dialectical approach.
120. The mention of Wallerstein reminds me to amend my earlier remarks on the commitment
to "actor models" implicit in Weberian solutions to the problem of subjectivity and meaning.
There is a partial exception to the claim that social scientists rooted in this tradition will generally
reject as "meaningless" social analyses which do not come to rest in an "actor model." That
exception is social analyses which come to rest in "the market"; I term it a "partial" exception
because market and exchange relations are generally taken to be individualist in origin within
bourgeois ideology, and hence all analyses in terms of market relations can themselves be
thought ultimately to come to rest in an actor model. Despite his radical intentions, Wallerstein's
analysis seems now stuck in this box. His model of the capitalist world system seems to amalgamate
a market-based model of production and exchange relations (one which refuses to close its eyes
to the reproductive hierarchy of the global division of labor) with an actor model of state practice,
a joining that has some sorry consequences. Wallerstein is left to oscillate between-without
ever transcending-the poles of market force and state purpose. Worse, when called upon to
account for creative moments in the system's evolution (moments that cannot be reduced to
market "dynamics" within the center-periphery hierarchy) he is left only two avenues: either
(a) that instrumentalist form of economism according to which the state conspires with (or is
totally enslaved to) a dominant power bloc or segments of capital who themselves are close to
omniscient, or (b) that idealist form of statism which credits the state with an all-seeing awareness
of its situation in history, and the will and ability to change the system while perpetuating its
essential structures. As this suggests, Wallersteinians offer us the choice between economistic
accounts and what turn out to be, on close inspection, neorealist accounts (which, we have
seen, are themselves economistic in an important sense). I do not think, by the way, that this
trap is escapable via a Parsonian move in the treatment of the states system, such as the one
promoted by John Meyer. It seems to me that escaping this trap will require reexamining the
position that locks Wallerstein into it, namely, the Weberian position on subjectivity and meaning

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
284 InternationalOrganization

Fourth,in this connection, it may be arguedalso that my neglectof Waller-

steinian and other radical perspectives has deprived my critique of a full
sense of the epistemo-politicalcontext of neorealism.This is a fair criticism.
A more complete critique would have stressed the opposition between
Wallersteinianheterodoxyand neorealistorthodoxyagainstthe background
of crisis.'2'On the one side, Wallerstein'sperspectiveis an instance of het-
erodoxy, a mode of theory that consciously sides with the dominated in a
social system. Occasionedby the currentcrisis, it strugglesto find words for
that which the dominantwould preferto excludefrom the realmof conscious
political discourse, namely, the political content of core-peripheryrelations.
In PierreBourdieu'swords, it strugglesto "expandthe domain of discourse"
and expose "the arbitrarinessof the taken for granted."'22On the other side,
neorealist theory is an instance of orthodoxy, implicitly on the side of the
dominant. It emerges under the condition of a crisis that calls into question
the self-evidence of the given order by severing the once near-perfectcor-
respondencebetween the objective order and the subjectiveprinciplesof its
organization.Under these conditions, and with the help of heterodoxy, the
arbitraryprinciplesof the prevailingorder can begin to appear as such. It
becomes necessaryto develop orthodoxies to "straightenopinion" by "nat-
uralizingthe given order," an order that, prior to crisis, was simply taken
for granted. Neorealist orthodoxy does just that. It develops "a system of
euphemisms, of acceptable ways of thinking and speakingthe natural and
social world, which rejects heretical remarks as blasphemies."'23It serves
primarily,as I noted earlier,to constrain the domain of discourse. I regret
that this line of argument could not have been more fully developed; had
we followed it further,we mighthave come to understandsome of the strange
asymmetriesin debates between neorealistsand neo-Marxistworld systems

in social reality. Having said that, let me distance myself from a fashion current among neorealists:
the ritual slaying of Immanuel Wallerstein (usually coupled with the celebration of the totemic
figure of Otto Hintze). I want to state plainly my own intellectual debt to Wallerstein's pioneering
work: like many American international relations theorists trained in the 1970s, I owe much
to Wallerstein, not just for his theory but for the exemplary boldness of his enterprise, and his
willingness (so threatening to neorealism) to punch holes in the convention-made walls of our
On Wallerstein's error of anchoring his analysis in market-based explanations, see Robert
Brenner, "The Origins of Capitalist Development: A Critique of Neo-Smithian Marxism," New
Left Review no. 104 (1977). See also John W. Meyer, "The World Polity and the Authority of
the Nation-State," in Albert Bergesen, ed., Studies of the Modern World-System (New York:
Academic Press, 1980); John Boli-Bennett, "The Ideology of Expanding State Authority in
National Constitutions, 1870-1970," in Meyer and Michael T. Hannan, eds., National Devel-
opmentand the WorldSystem:Educational,Economic,and PoliticalChange,1950-1970 (Chi-
cago: University of Chicago Press, 1979).
121. This treatment of the opposition between heterodoxy and orthodoxy in the context of
crisis is due to Bourdieu, Outline, pp. 159-71.
122. Ibid., p. 165.
123. Ibid., p. 169.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
The poverty of neorealism 285

Finally,I must confess to naggingdoubts about tone. Had I the neorealist's

gift for theorizing,I would certainlyaspire to write the kind of critiquethat
wouldcausemy oppositionto concedeerroron the spot-a deft, dispassionate,
and surgicallyprecisearticulationof a logicalflawthat would bringneorealists
to their knees with technical certitude. I am not so gifted, however, and so
my lesser approachhas been to circle the neorealistorreryof errors,try to
glimpse its self-affirmingand antihistoricalclosure, and then set the whole
of it before the eyes of the classical realist traditionupon which neorealism
claims to improve. Along the way, I have called names, I have poked fun,
I have stolen some of Morgenthau'sangrierwords to hurl at neorealists,and
I have triedto expose the implicitpoliticalcontent of this purportedlyneutral
enterprise.I have said, in effect, that this supposedlyscientificrealismis bad
science and worse realism. In the manner of Thompson, I have whacked
away at the bungs of structuralistbarrels lined up in the academy, and I
have done so in the hope that some "minds might get out." Such a tone is
hardly calculated to win friends, I know. Worse, it makes me appear the
aggressor,and leaves me open to a calculated strategyof "witheringnon-
chalance" in response.
In my defense,let me say that I am drivento these lengthsby a combination
of concernand hope. My concernis that, amidst the wrenchingof economic,
social, and epistemic crisis, social scientistswho study internationalrelations
will mistake neorealism's anticriticalclosure for a much needed pillar of
certainty,security,and, most of all, collective understanding.I am concerned
that the faculties that above all distinguish science from nonscience-the
reflective exercise of criticism-are thus being deadened at just the time
when their potential is most needed and most likely to burst forth. I am
concerned that, as a result, the scientific study of internationalpolitics in
the United Statesis gravitatingtowarda reactionarypole ratherthaninvolving
itself in the expansion of the field of political discourseand, with it, oppor-
tunities for the creative evolution of world society. And I am especially
concerned about graduate students and younger scholars who are told to
think criticallyand creativelybut whose freedomto think criticallyin public
depends to a very considerabledegree upon their linking their accomplish-
ments to collectivelyrecognizedfoundations.Insofaras neorealistlore comes
to occupy the collectively recognizedfoundationsof the discipline,the urging
of criticism-consciousnesscan only be a cruel hoax.
My hope is encapsulated in the words of Sartre:"Words wreak havoc
when they find a name for what had up to then been lived namelessly."'24
My hope and my hunch is that the present polemic amounts to little more
than a putting into words of what many have already "lived namelessly."
I suspect that I am not the first to wonder why neorealistargumentsalways

124. Jean-Paul Sartre, L idiot de la famille (Paris: Gallimard, 1971), 1: 783, and quoted in
Bourdieu,Outline,p. 170.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions
286 InternationalOrganization

come to us not as ideas that pry open beloved concepts and make room for
new scientificadventuresbut as case-hardenedconceptualdevicesthat enclose
the senses in an all-encompassingfinality.I imaginethat othershave recoiled
at the feeling, upon encounteringthese neorealistdepictions, that one is an
innocent fallen victim of a vast and diabolicalmachine, a perpetualmotion
machine that bends every attempt to escape it into a reaffirmationof itself.
And I would guess that others have been troubled by the eerie sense of
completeness about neorealisttheory, as if there is no more of consequence
to be said, save a defense of the edifice here, a demonstrationof its efficacy
there. If I am right, and I hope I am, then most internationalrelations
scholars have long sensed what I have tried to put into words.
Let us then play havoc with neorealistconcepts and claims. Let us neither
admire nor ignore the orreryof errors,but let us instead fracturethe orbs,
crackthem open, shake them, and see what possibilitiesthey have enclosed.
And then, when we are done, let us not cast away the residue.Let us instead
sweep it into a jar, shine up the glass, and place it high on the bookshelf
with other specimens of past mistakes.

This content downloaded from on Fri, 04 Sep 2015 08:01:25 UTC
All use subject to JSTOR Terms and Conditions