Nationalism, Ethnic Conflict, and Rationality
Author(s): Ashutosh Varshney
Source: Perspectives on Politics, Vol. 1, No. 1 (Mar., 2003), pp. 85-99
Published by: American Political Science Association
Stable URL: http://www.jstor.org/stable/3687814
Accessed: 12/05/2010 15:21
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=apsa.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
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 [email protected].
American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to
Perspectives on Politics.
http://www.jstor.org
Articles I Nationalism,EthnicConflict,and Rationality
Nationalism,Ethnic
Conflict,
and
Rationality
By Ashutosh Varshney
in theworldreadyto dieassuicidebombers?
Doesa rational
calculuslie beneaththe
Whydo we haveso manyethnicpartisans
it
be
discovered
if
we
our
of
more
ThisartiCan
and
nationalist
creatively?
rationality
understanding
only
apply
pride passions?
of resistance.
It arguesthata focuson dignity,self-respect,
and
cleseeksto answerthesequestions
byfocusingon thenationalism
is
a
better
for
ethnic
and
nationalist
behavnotion
of
a
than
rather
self-interest,
understanding
prism
straightforward
recognition,
is not entirelyabsentasa motivation
in ethnicconflict.In theprocessof developing
thisargument,
a
ior,althoughself-interest
andrefinedfurther.
distinction
oncemadebyMaxWeber-betweeninstrumental
andvaluerationality-isrecovered
rationality
existthancenotaphs
andtombsof
emblemsof the moderncultureof nationalism
No morearresting
UnknownSoldiers.... Theyareeitherdeliberately
emptyor no oneknowswholiesinsidethem....
becomesevenclearerif onetriesto imagine,say,a Tomb
of suchmonuments
Theculturalsignificance
Isa senseof absurdity
avoidable?
Thereason
of theUnknownMarxistor a cenotaphforfallenLiberals.
withdeathandimmortality.
If thenationalis muchconcerned
norLiberalism
is thatneitherMarxism
thissuggestsa strongaffinitywithreligious
is so concerned,
ist imagining
imaginings...
1983
-Benedict Anderson,
Communities,
Imagined
re identities rational?Are identity-basedconflicts?If one
goes by the history of ideas, these questionshave mostly
beenansweredin the negative.Letme givetwo illustrations.
In a celebratedessay,CzechnovelistMilanKunderaarguesthat
scienceand novelswereborn together.Indeed,novelsweremade
necessaryby science,for "themore [man]advancedin [scientific
and rational]knowledge,the less clearlycould he see ... his own
self... and he plungedfurtherinto what ... Heideggercalled...
'the forgettingof being."''Followingthis argumentto its logical
end, Kunderaclaimsthat the questionof identity-Who am I?belongs to the realmof being,not to the domain of rationality.
The latterdealswith the question,How do I get whatI want?The
two questionsmay sometimesbe related-what I may be able to
get may beginto definehow I see myselfand my goals-but they
arenot the same.
Kundera'sargumentis aboutidentityper se, not aboutnational or ethnic identity. Is it that national identities are held for
rationalreasons?The customaryansweris even moreemphatically negative.IsaiahBerlin,arguablythe foremosthistorianof ideas
A
AshutoshVarshney
is an associate
professor
ofpoliticalscienceat the
He receivedmany
UniversityofMichigan([email protected]).
comments
on
earlier
He
this
helpful
draftsof
paper. wouldespeciallike
to
the
criticisms
and
ly
of Gabriel
acknowledge
suggestions
Benedict
Robert
Morris
Almond,
Anderson,
Bates,
Fiorina,Ira
Vibha
Katznelson,PratapMehta,
James
Pingle,RonaldRogowski,
Scott,KennethShepsle,
JackSnyder,SidneyVerba,the lateMyron
Weiner,ElisabethWood,CrawfordYoung,and threeanonymous
reviewers
of thisjournal.Needlessto add, not all of themagreed,so
thestandarddisclaimers
apply.
of the past century,informsus that the notions of nationalidentity and nationalismwerebornwith the Counter-Enlightenment.
As the FrenchEnlightenmentcelebratedthe triumphof rationality, emphasizingits universalityand objectivity,a reactiontook
place, especially in German-speaking Europe. The
Enlightenment intellectualsargued that "methods similar to
those of NewtonianPhysics... could be appliedwith equalsuccess to the fields of ethics, politics and human relationshipsin
general."2Johann Gottfried von Herder,who would later be
calleda leadingpatronof the Counter-Enlightenment,
disagreed:
"Germansmust be Germansand not third-rateFrenchmen;life
lies in remainingsteepedin one'sown language,tradition,local
feeling;uniformityis death."3The reactionwas both against"the
deathly embrace of impersonal, scientific thought" and the
Frenchculturalhegemonyof Europe.4In a similarvein, Johann
Gottlieb Fichte deliveredhis Addresses
to the GermanNation. It
was 1808-1809; the Germans,spreadamong Prussia,Bavaria,
Bohemia,and Silesia,wereyet to be born as a nation underone
political roof. Herder and Fichte, Berlin argues, were the
ideationalfathersof nationalism.5Indeed,nationalismas an idea
aroseas a romanticrevoltagainstthe universalizingand rationalizing thrustof the Enlightenment.
Can the child of the Counter-Enlightenment
be analyzed
in an Enlightenmentframework?
Has rationality,aftermore than
two centuriesof progresssince the Enlightenment,developed
capacities to deal with human passions, emotions, and
valuesin an illuminatingmanner?Is it thatbeneaththe nationalist
prideand passionslies a rationalcalculus,whichcan be discovered
if only we applyour understanding
of rationalitymorecreatively?
Recoveringa dualityfirstproposedby MaxWeber,I suggestthat
ethnicor nationalconflictis bestconceptualizedas a combination
www.apsanet.org 85
Articles I Nationalism,EthnicConflict,and Rationality
of "valuerationality"
and "instrumental
Both of these
rationality."
rationalitiesare expressionsof goal-directedbehavior,but their
conceptions of costs widely diverge. Instrumentalrationality
entailsa strictcost-benefitcalculuswith respectto goals,necessitatingthe abandonmentor adjustmentof goalsif the costsof realbehavioris producedby a
izing them aretoo high. Value-rational
conscious"ethical,aesthetic,religiousor other"belief, "indepenBehavior,when drivenby such
dentlyof its prospectsof success."6
can
embrace
values,
consciously
great personalsacrifices.Some
or
of
life
are
considered
so valuablethat theywould
spheres goals
not normallybe up for sale or compromise,howevercostly the
pursuitof theirrealizationmightbe. The meansto achievingthese
objectivesmight change,but the objectivesthemselveswould not.
The term value-rationaldoes not, of course, mean that the
values expressed by such behavior are necessarilylaudable.
Indeed, the values in question may range from pure pride or
prejudice(vis-a-vissome groupsor belief systems)to goals such
as dignity, self-respect,and commitment to a group or a set of
ideals. Likewise, value-rationalacts can range from long-run
sacrificesfor distant goals to violent expressionsof prejudiceor
status.
serious possibility of high costs that usually accompanythe
nationalismof resistancemake such an alignment extremely
difficult.
As scholarlywork proceeds further, the concept of value
rationalitywill need greaterunpacking.I takethe firststepshere
by concentratingon only one kind of nationalism:the nationalism of resistance.I am certainthatdignityand self-respectcannot
be the microfoundationsof all forms of ethnic or nationalist
behavior.Pendinglaterwork,for example,it is reasonableto suppose that the nationalismof exclusionis drivensubstantiallyby
hatredand/or deep-rootedcondescension:Afrikanernationalism
in SouthAfrica,the anti-Semitismof Hitler,and Hindu nationalism in India would be some examples.In what follows, the
claim of dignity and self-respectappliesonly to the nationalism
of resistance,not to the nationalismof exclusion.8
Terms and Distinctions
Let me startwith definitionsof the principalterms used here:
ethnicity,nation,and rationality.Not having the same meaning
for everyone,these termsneed clarification.
Ethnicityis used in two differentways.In the narrower,popularly understood sense, ethnic groups are racial or linguistic
groups.Thereis, however,a broadermeaningas well. As Donald
Horowitzsuggests,9all conflictsbasedon ascriptive(birth-based)
groupidentities,realor imagined-race, language,religion,tribe,
or caste-can be calledethnic.In thislargerusage,ethnicconflicts
can rangefrom (1) the Protestant-Catholic
conflict in Northern
Irelandand the Hindu-Muslimconflictin Indiato (2) the blackwhite conflict in the United States and South Africa and the
Malay-Chineseconflictin Malaysia,(3) the Quebecoisproblemin
Canadaand the Tamil-Sinhala
conflictin SriLanka,and (4) ShiaSunnitroublesin Pakistan.In the narrowerview,the firstof these
examplesarereligious,the secondracial,the thirdlinguistic,and
the fourthsectarian.The termethnichas customarilybeen usedin
the pastfor the secondandthirdtypesof conflicts,not for the first
and fourth.
Proponentsof the broaderusagedo not find the narrowerdistinctionsanalyticallyhelpful.They arguethatthe formtheseconflicts take-religious, racial,linguistic,tribal-does not change
their intensityor relativeintractability.
The broadermeaningof
ethnicis now increasinglyprevalentin the social sciences;I will
use the term in this way.
Also, for the purposesof this paper,the terms ethnicityand
nationcan be used interchangeably.
If the discussionwere about
some
ethnic
conflicts
remain
bounded
within the existing
why
state boundarieswhile others gravitatetowardindependence,a
distinction between the two terms would be essential.Ethnic
groups,as we know,can live withouta stateof theirown, making
do with some cultural rights (e.g., use of mother tongue in
schools)or affirmativeaction;but a nation meansbringingethnicity and statehoodtogether.'0This distinction,however,is not
necessaryfor our purposeshere,becausethe discussionis abouta
whole class of conflicts,which are framedin terms of national
Most of the time and in most places,ethnic or nationalmobilization cannot begin without value-rationalmicrofoundations.
Forit to be instrumentallyusedby leaders,ethnicitymust existas
a valued good for some. However,ethnic mobilizationcannot
proceedon value-rationalgroundsalone. Strategiesarenecessary;
coalitions must be formed;the responseof the adversary-the
state, the opposed ethnic group, the in-groupdissenters-must
be anticipated.And manywould join such mobilization,when it
has acquiredsome momentumand chanceof success,for entirely selfish reasons.The originsof ethnic mobilizationare thus
value-rational,and its evolutionmay contain a lot of strategic
behavior.
To illustratethis argumentin ampledetailand for tractability,
I shallrestrictmy analyticalfocus to only one kind of nationalist
or ethnicbehavior.A usefuldistinctionis often madebetweenthe
nationalismof exclusionand the nationalismof resistance.7
The
idea, of course, is quite old. The nationalismof anticolonial
movementswas nevercomparableto the nationalismof Hitler.
In the nationalismof exclusion,a dominant group within a
society-domestic or foreign-seeks to imposeits own valueson
the variousother groupswithin that societyor seeksto exclude,
sometimes violently, other ethnic groups from the portals of
power.Typically,this takesthe form of enforcinglanguage,religion, or culturevia controlof the state,or excludinggroupsfrom
poweron the basisof ethnic characteristics
only. In the nationalism of resistance,a dominatedgroup opposessuch a move and
seeksto preserveits culturalidentityand resistthe hegemonyand
powerof the dominantgroup.
I will arguethat dignity and self-respectform the microfoundations of the latter kind of nationalismor ethnic behavior.
Drivenby such values,resistingnationalistsarewilling to endure
very high costs-and for long periodsof time. The cost-benefit
calculusin such behaviordoes not workin a way that can be eas- identity or ethnicity.
What about our third key term, rationality?
ily alignedwith a standardaccount of instrumentalrationality.
In its standard
Indeed,long time frames,a radicaluncertaintyof results,and the economicusage,the termrefersto instrumentalrationality,and it
86
March2003 1 Vol. 1/No. 1
has two meanings.First,it meansconsistencyof choice:if I preferA over B and B overC, then I must preferA overC. The second meaningis identicalwith self-interest.Action is rationalif it
is aimed at realizingself-interest.If costs of an action outweigh
benefits,self-interestwill not be served;hence a cost-benefitcalculus accompaniesanalysisbasedon self-interest.
In philosophicaldiscussions, rationalityrefers to "reasoned
assessmentas the basis of action."'1Such an assessmentcan be
basedon self-interestbut also on largervalues.Self can be broadly defined in terms of group goals, national identity, religious
values, aesthetic considerations,and so on. This larger view
would also include what Weber called "valuerationality."In
Economyand Society,Webercategorizedsocial action into four
types: instrumental-rational, value-rational, norm-oriented
(basedon conventionsand traditions,without criticaldeliberation), and affectiveor impulsive(the expressionof anger,envy,
love, et cetera).
The alternativesto instrumentallyrationalbehaviorare thus
not simply emotional or irrationalbehavior.12Of the four
Weberiancategoriesof human action, the first two are goaldirected, only one of which is instrumental-rational,whose
uniquefeatureis a strictcost-benefitcalculuswith respectto goals
and means.Suchcalculusmayleadnot only to a changeof means
for the realizationof goals,but also to an alterationof goalsif the
costsof attainingthem areprohibitive.Value-rationality
is distina
continual
of
even
if
the
costs of realguishedby
pursuit goals,
izing them arehigh; it showsa high degreeof commitment.
Which of these categoriesof behavioris representedby the
term rationalchoiceoften usedin economicsand politicalscience?
Almost without exception, it is instrumentalrationalitywith
which rational-choicetheoristsidentify.They eitherdo not speak
of goals,concentratinginsteadon the means;or they assumethat
self-interestis the goal of human action. I will, therefore,use
these two terms-instrumental rationalityand rationalchoiceinterchangeablyin this paper.But I will not equate rationality
with rationalchoice.
These distinctionshavesome importantimplicationsfor a discussionof rationality.In a standardrational-choiceaccount,there
is considerableresistanceto the idea that differentmotivations
can underliebehaviorin differentspheresof life: that it may be
perfectlyrationalfor human beingsto be instrumentallyrational
while buying a car,but value-rationalwhile respondingto questions of nationalliberation,school choice for children,affirmative action, or multiculturalismin universities.13Moreover,
rationalchoice also remainshighly skepticalof the notion that
individual action can be rooted in group interests, not selfinterest. Value-rationalbehaviorwould not find identification
with groupinterestsirrational.
What else can we say about value-rationality?
Accordingto
Weber,as alreadynoted,value-rationalbehavioris pursued"indeThat notion, in my view,
pendentlyof its prospectsof success."14
is best seen as an ideal type, or a pure case of value-rationality.
Any reasonablenotion of value-rationalbehavior cannot be
insensitiveto costs. A more realisticreformulationof Weber's
notion is required.In orderto providethat, let me use the simple
economicconceptof elasticity.
From developmentmicroeconomics,we know that demand
for food is relatively,not absolutely,insensitiveto price-people
must eat, however expensive food might become-whereas
demandforTV sets and carsis remarkably
sensitiveto price,sugof
low
demand
for the formerand
gestingthereby priceelasticity
for
the
latter.
We
can
high elasticity
similarlyarguethat valuerationalbehavioris relativelyinelasticwith respectto costs. A
fullyinelasticbehavioras in the Weberianidealtype-with valuerationalbehavioron the horizontalaxisand cost/priceon the vertical-would be representedby a flat line, but low-elasticity
behaviorwould slope downward,like demandcurves,although
the slope would not be as steep, as in the case of highly elastic
goods such as cars. In this economic analogy,value-rational
behavioris more like the demand for food, and instrumentalrationalbehaviorlike the demandfor carsandTV sets.15
Thereis no doubt that an instrumentallyrational-or rational
choice-understandingof humanbehaviorhas maderemarkable
progressover the years, extending into newer directions and
fields. Behaviorcoveredby such reasoningand models ranges
from economic decision making of consumers and firms to
nuclearpolitics,legislativeand bureaucraticbehavior,and political mobilizationand ethics. Indeed, the list of topics to which
rational-choicemodelshavebeen appliedcontinuesto grow.16
In principle,one cannot objectto pushinga mode of analysis
to fields where it was not applied before. Indeed, severalnew
insights in the world of knowledgeare generatedpreciselythis
way.Much has been learnedon politicalmobilizationby exploring the ideathat the self-interestof individualsand the interestof
the group to which they belong are two differentthings: class
conflictmay thereforebe morelatentthan overt.17The prisoner's
dilemmagamehas taughtus betterthan manyothermodelsthat
rationallybehavingindividualsmay generatea macro outcome
that is suboptimalfor all. Similarly,how self-seekingpoliticaland
bureaucraticbehavior,as opposedto the selfishbehaviorof economic agentsin competitivemarkets,can lead to a wastefuluse
of society'seconomicresourcesand hampereconomicgrowthis a
problemwhere rationalchoice has been especiallyuseful as an
explanatorytool.18
The issue thereforeis not whether rational-choicetheories
explainhuman behaviorat all. More germaneis the questionof
whetherrational-choicetheoriesare especiallyrelevantto a specific classof problemsand a particularrealmof humanbehavior,
and if so, in what ways that realmmight be differentfrom others.19In this realm-specificspirit,20I ask whetherand how far
rational-choicetheoriescan accountfor ethnicbehaviorand conflict, dominatedas they often are by mass politics, not by the
institutionalizedformsof bureaucratic
or legislativepolitics.21
The Big Gap: Where Do Ethnic
Preferences Come From?
Beforeethnicconflictcan be explained,a rational-choice
analystis
confrontedwith a twofold task:providingmicrofoundationsof
ethnicbehaviorandexplainingethnicmobilization.To beginwith,
one has to accountfor why individualshave, or develop,ethnic
preferences.Can such preferencesbe explainedinstrumentallyi.e., as a means to a self-interested end (political power,
www.apsanet.org 87
Articles I Nationalism,EthnicConflict,and Rationality
economicbenefit,survival)?
And since it would be instrumentally
rational,given self-interest,for individualsto freeride,explaining
ethnic mobilizationrequiresspecifyingconditionsunderwhich it
would not make sense for individualsto free ride and, in fact, it
would be rationalto join an ethnicmovementor mobilization.
The standardrational-choiceaccounts assume that ethnicity
can be seen instrumentally.
They focus primarilyon how leaders
strategicallymanipulateethnicity for the sake of power.22This
argumenthas an intuitiveappealbecausethe behaviorof many,if
not all, politicalleaderscan be cited in support.
If presentedin this form, the instrumental-rational
argument
aboutethnicityrunsinto a seriousdifficulty.The elite mayindeed
gain powerby mobilizingethnic identitywithout believingin it
themselves,and could thereforebehaveinstrumentally.But if the
masseswere only instrumentalabout ethnic identity,why would
ethnicitybe the basisfor mobilizationat all?Why do the leaders
decideto mobilizeethnic passionsin the firstplace?Why do they
think that ethnicity,not the economic interestsof the people, is
the routeto power?And if economic interestscoincidewith ethnicity,why choose ethnicityas opposedto economicinterestsfor
mobilization?
In principle, a rational-choiceresolution of these problems
exists. Ethnicity can serve as a focal point, facilitatingconvergence of individualexpectations,and hence can be useful as a
mobilization strategy.The idea of focal points comes from
Thomas Schelling'sseminaltreatmentof the coordinationproblem in bargaining.In the famousSchellingexample:
Ethnic mobilizationfor politicalaction is not the same as ethnic
coordinationfor economic and social activities.By providinga
socialoccasion,festivalsmay indeedbringpeopletogethereven if
not everyone appreciatesthe ritual meaning of celebrationor
mourning;and by formingmutuallyconvergingtrust,geographically spreadethnic kinsmen are also known to have supplied
credit in long-distancetrade without a prior explicit contract
betweentradingpartners.
The analogyof a focal point, however,cannot be extendedto
group action when the costs of participationfor the massesare
very high. By its very nature,ethnic mobilizationin politics is
group action not only in favor of one's group but also against
some othergroup.More rights and power for my group often
mean a diminution in the abilityof some other group(s)to dictate terms, or a sharing of power and status between groups
where no such sharing earlierexisted; in the extremecases, it
may even entail the other group'sdisplacementfrom power or
status.Ethnicityin intragroupsocialor economic transactionsis
thus very different from ethnicity in intergrouppolitical conflicts. The former illustratesthe value of ethnicity as a focal
point; the latterpresentsproblemsof a differentorder.When an
individualprovidescreditto ethnic brethrenwithout an explicit
contract,incarceration,violence,injury,or deathis not likely the
cost he has to keep in mind.25But dependingon how the adversarialgroupor the statereacts,such costs arenot unlikelyin ethnic or nationalconflicts.
Considerthe famous 1930 Salt March in India. The British
rulersmonopolizedthe manufactureand retailingof salt. Seizing
Whena manloseshis wifein a department
storewithoutanyprior a
symbolthat even the illiteratemassescould relateto, Mahatma
on whereto meetif theygetseparated,
thechancesare
understanding
Gandhi arguedthat the British insulted Indiansby not letting
goodthattheywillfindeachother.It is likelythateachwillthinkof them
freelymakeand sell somethingas basicas salt in their own
someobviousplaceto meet,so obviousthateachwillbesurethatthe
countryand by levyinga salt tax. He went on to lead a nonviootheris surethatit is obviousto bothof them.23
lent mobilizationagainstsalt laws and was later arrested.Civil
Schelling goes on to propose that without having an intrinsic disobediencecontinuedeven afterhis arrest.An Americanjourvalue for the couple, the lost-and-foundsection of the depart- nalistgavethe followingeyewitnessaccountof the earlyphaseof
ment store could be one such place. It will, however,not be a the movement:
focal point if there are too many lost-and-foundsectionsin the
The saltdepositsweresurrounded
by ditchesfilledwithwaterand
store.A focalpoint is distinguishedby its prominenceor uniquefourhundrednative... policein khakishortsandbrown
by
guarded
ness:it has the instrumentalpower of facilitatingthe formation
of mutually consistent expectations.Schelling then generalizes
the principle:
st
*
.
revoltmayreflectsimilarprinciples:
whenleaderscan
Spontaneous
easilybe destroyed,
peoplemayrequiresomesignalfortheircoordination,[whichis] ... so potentin its suggestionsfor actionthat
everyonecan be surethateveryoneelse readsthe samesignalwith
enoughconfidenceto act on it, thus providingone anotherwith
immunitythatgoeswithactionin largenumbers.24
o
O
(o
Ethnicity,in other words,can be viewed as one such focal point
for mobilization;it is not valued for its own sake. Its mobilizational potentialmay be deployedby leadersto extractgoods and
services from the modern sector, or to establish their own
power.
The idea of a focal point is not sufficient to explain ethnic
mobilization,for it does not distinguishbetweendifferentkinds
of collective action and what their respectivecosts might be.
88
March2003 I Vol. 1/No. 1
iL
':i.
A
-C
i
t::
__0
iil??:~~~~~~ii?~~~EBe"~~~~~ E
Gandhi on the Salt March, 1930
I
O~fl
*
_
turbans.Half a dozen Britishofficialscommandedthem. The police
carried... five-footclubs tippedwith steel....
In completesilence, the Gandhi men drew up and halteda hundred yardsfrom the stockade.A picked column advancedfrom the
crowd, waded the ditches, and approached the barbed-wire
stockade.... Police officials ordered the marchersto disperse....
The column silently ignored the warning and slowly walked
forward....
Suddenly, at a word of command, scores of... police rushed
upon the advancingmarchersand rainedblows on their headswith
theirsteel-shodclubs (lathis).Not one of the marcherseven raisedan
armto fend off the blows.They went down like tenpins.Fromwhere
I stood I heard the sickeningwhacks of the clubs on unprotected
skulls.
... In two or three minutes the ground was quilted with bodies.... Althougheveryone knewthatwithin a few minuteshe would
be beatendown, perhapskilled,I could detectno signsof waveringor
fear.... The marcherssimply walked forwarduntil struck down.
Therewere no outcries,only groansafterthey fell.
... I went backto the temporaryhospitalto examinethe wounded.
.. I counted320 injured,many still insensiblewith fracturedskulls,
otherswrithingin agony.26
Other examples of this kind of resolve can also be cited.
Consider the civil-rights movement of the United States in the
1960s. "In the Black community... going to jail was a badge of
dishonor."27And what kind of jails are we talking about?
Freedomriders,by all accounts,hada miserabletime in the jails.They
were crowded into small, filthy cells, forced to sleep on concrete
floors, fed unpalatablefood, preventedfrom maintainingpersonal
hygiene, intimidated,harassed,and sometimesbeatenby unfriendly
guards.28
As if these were small discomforts for black civil-rights activists,
we also have accounts of marches at night, even though "[u]nder
cover of darkness, a violent response by the police or by local vigilantes was almost assured. When civil-rights activists conducted
a night march in Marion, state troopers attacked and beat them
after the street lamps were intentionally blacked out."29
Finally, similar behavior can be noted in South Africa'shistory.
A violent repression or a harsh jail sentence was a near certainty,
once Nelson Mandela and his colleagues decided frontally to
challenge the apartheid regime on behalf of Africans. In the end,
Mandela himself and many of his colleagues were jailed in
Robben Island. The harsh and grim prison conditions did not
crush their spirit. The experience only clarified that-given the
objective of racial equality-the resolve to fight the dominant
group, the Afrikaners, would have to weather such suffering.
Mandela wrote:
Robben Islandwas without question the harshest,most iron-fisted
outpost in the South African penal system.... The warderswere
white and overwhelminglyAfrikaans-speaking,
and they demandeda
master-servantrelationship.They ordered us to call them "baas,"
which we refused.The racialdivide on RobbenIslandwas absolute:
therewereno blackwardersand no white prisoners.... J]ourneying
to RobbenIslandwaslike going to anothercountry.Its isolationmade
it not simplyanotherprison,but a worldof its own, farremovedfrom
the one we had come from. The high spirits with which we left
Pretoriahad been snuffedout by its sternatmosphere;we werefaceto
facewith the realizationthatour life would be unredeemablygrim.In
Pretoria,we felt connectedto our supportersand our families;on the
island,we felt cut off and indeedwe were.We had the consolationof
beingwith each other,but that was the only consolation.My dismay
was quickly replacedby a sense that a new and differentfight had
begun.30
After 27 years on Robben Island, Mandela did walk to triumph
and freedom; but in 1962, when he was jailed, there was a good
chance he would end up dying there. It was a life sentence after
all, and he knew it beforehand. The same was true of his many
fellow prisoners, if not to the same degree.
These examples illustrate a simple point, widely understood by
activists in such struggles. Ex ante possibility of violence or coercion almost always accompanies ethnic or national resistance.
Mobilization for ethnic or national protest cannot thus be equated
with solving problems of economic or social coordination through
the ethnic bond. It is a special kind of collective action, for the
costs of resistanceor mobilization are often known to be high.
Although exact estimates are hard to produce, it is generally
agreed that in this century, many more people have died for a
nation or an ethnic group-presumed or actual-than for joining a supranational economic collectivity, such as the European
Economic Community, the Association of Southeast Asian
Nations, or the North American Free Trade Agreement.31
Moreover, fighting for higher prices, subsidies, and wages, and
for more jobs, does not necessarily generate as much passion and
violence as does ethnic or nationalist mobilization. The masses
have often been much more willing to come out on the street for
ethnic issues than for economic ones.32 If they did not value
ethnicity, why would they respond so passionately to ethnic
appeals?
For something to be manipulated by a leader when death,
injury, or incarceration is a clear possibility, it must be valued as
a good by a critical mass of people, if not by all. A purely instrumental conception of ethnicity cannot explain why leaders mobilize ethnic or national identities at all. The point is analogous to
Jon Elster's famous objection to an instrumental conception of
norms: "Some argue that... norms... are tools of manipulation,
used to dress up self-interest in a more acceptable garb. But this
cannot be true.... If some people successfully exploit norms for
self-interested purposes, it can only be because others are willing
to let norms take precedence over self-interest."33
Epistemological Comforts of Home?
In the first available rational-choice work on ethnic conflict,
Alvin Rabushka and Kenneth Shepsle explicitly recognized that
microfoundations of ethnic behavior were hard to provide in a
strictly rational-choice framework. They argued:
[A] bothersomequestionremains.... Why... areconflictsin [plural]societiesnot organizedalong economiclines?Our answeris that
politiciansexertcontroloverthe definitionof alternatives,often relying on ethnic appeals.But why this particularchoice?
... If... the ethnic issuewerea facadefoistedupon an electorate
not receptiveto those issuessimplyto suit the motivesof strategically
www.apsanet.org 89
Articles I Nationalism,EthnicConflict,and Rationality
more intrinsicthan instrumentalwith respectto ethnicity (her
conscienceis her problem);and the latteris easyin smallgroups
but monumentallydifficultin largegroups,even when an institutionalizedregulationof individualbehavioris devised.If the
group action concernsmy caste or tribe in a village or even a
1. a formal explanation of the formation, development and town, I may rationallycoordinate:everyoneknowsme and I can
be monitored.But if the groupaction is aboutan imaginedethenduranceof valuesand preferences,and
2. a positivetheoryof politicalentrepreneurship.
nic or nationalcommunity-involving manyvillages,towns,and
states-I
can escapedetectionif I cheat.Lackingthe intimacyof
... With these two developments,then we could more persuasively
small
groups,how does one monitoran ethnicgroupor a nationaccountfor the preeminenceof ethnicityin the pluralsociety.34
ality?The Hardinproposalthus cannot be size independent.A
Three decades have passed since Rabushka and Shepsle wrote
nation is not an intimatelyknowable,face-to-facecommunity.It
is a large,imaginedcommunity.
their book. Do we now have an "explanation of the formation,
Second, why should ethnic or nationalmobilizationbe condevelopment and endurance of values" in a rational-choice
framework?
ceptualizedas a coordinationgame,whereasotherkindsof mobiadvantagedpoliticians, then one might expect successfulpolitical
recourseto be takenby the "losers."
Althoughotherissuesmayaffectpoliticsin pluralsocieties,we assert
the preeminenceof ethnicity.We arenot able to explainits genesis.A
satisfactory
explanationof this problemawaitstwo developments:
In the most ambitious, sophisticated, and erudite rationalchoice work on ethnic conflict so far, Russell Hardin takes up the
challenge.35He seeks to provide such microfoundations and also
use them to explain ethnic mobilization and conflict. His proposal is threefold. First, "self-interestcan often be matched with group
interest"instrumentally.Identification with the group may be beneficial for two reasons: because "those who identify strongly with
the group may gain access to positions under the control of the
group" and because "the group provides a relatively secure and
comfortable environment." The identity between individual and
group interests, he argues, can only be "contingent," not "inherent," but it is enough to touch off ethnic mobilization. Second,
explanation of ethnic mobilization can'tbe reduced to the problem
of collective action where it is rational to free ride, or to a prisoners' dilemma where it is rational to defect. In ethnic mobilization,
"[t]he central strategic problem is merely one of coordination." So
as long as others in the group are cooperating, it is rational for me
to cooperate-for if all cooperate, the likelihood of the group gaining power (or group objectives) goes up tremendously. "[P]ower
based in coordination is superadditive, it adds up to more than the
sum of individual contributions to it." Third, all one needs to keep
the coordination game going is a "charismaticleader," a "focus,"
and a mechanism through which information about others cooperating is provided. "Coordination power is ... a function of reinforcing expectations about the behavior of others."36
Hardin's proposal entails serious difficulties. First, even if I
believe in group goals, contingently or inherently, it is not clear
why it is rational for me to cooperate when others cooperate with
one another. For if they are cooperating, and if "coordination
power" is "superadditive,"then my group is very likely to come
to power anyway and it is rational for me to take a free rideunless, of course, someone is monitoring my actions and the
nonparticipants will be excluded from the rewards of the group's
victory. Alternatively, my conscience could act as a monitor, giving me a sense of guilt or shame for not participating in group
action even though I believe that the group's interests are my
interests. Without these monitoring mechanisms, the situation
does not have a unique optimum, but two optima: both free riding and participating could be rational. In a purely logical sense,
Hardin's proposal thus requires monitoring of individual actions:
internallyor by others.The formerentailsan individualwho is
90
March2003 | Vol. 1/No. 1
lization-such
as peasant37or working-class mobilization38-are
moretypicalcasesof collectiveaction,crippledby free-riderproblems?Must the group in question have some specific qualities
that create"coordination"
as the "centralstrategicproblem,"preendemic
free
empting
riding?Can we accountfor this difference
in a rational-choiceframework,or is some other theoryrequired
to establishthe difference?If the latter question is chosen to
explorewhy ethnic action is differentfrom other group actions,
then it is potentiallydamagingfor rational-choicetheories,for it
mayshow that some kindsof preferencesemergein a nonrational
framework.
Hardinhas one suchproposalaboutethnicity:thatit mayprovide "epistemologicalcomforts of home" or, put alternatively,
securityof environment.This solutiononly re-statesthe problem.
Why does "ethnicity"providea home?Why can'ta tradeunion
or a politicalparty?The Communistexperimentwas, inter alia,
premisedupon the belief that the partywould supplantthe false
consciousnessof ethnicity and nation. After decadesof trying,
that experimentfailed, and ethnicityhas re-emerged-frighteningly so in severalplaces.Once we believethat ethnicitycan provide a home betterthan othergroupscan, we also acceptthat in
a basicsense,the microfoundationsof ethnicityarepsychological,
not rational.39
Thus, whetheror not I think that my interestsand my group's
interestsare different,the fundamentalpuzzle for instrumental
rationalityremainsas follows:why should I, behavingin a purely instrumental-rational
way,participatein groupactionbeforeit
is reasonablyclearto me that the groupis likelyto win?Consider
the structureof the problemdiagrammatically
(see Figure1). At
time T1, when my groupis not in power,my personalwelfareis
at a low level (W1); I expectthat at time T2, when my groupis
in power,my welfarewill riseto W2. The problemsimplyis that
at time T1, I don'tknow ex antehow far awayT2 is, and I also
don'tknowhow big the costsin the meantimewill be. Depending
on what the adversariesdo, the sacrificerequiredcould be low
(lookinglike S1) or high (S2). It is not rationalfor me to join at
time T1; I should let others join and when the movement or
mobilizationis alreadysubstantialand very likely close to T2, it
will be rationalfor me to participate.40
To sum up, the microfoundationsof the originsof ethnic mobilizationare differentfrom those that obtain once mobilization
Figure 1
W2 -----.-.--------.--------I
! i
WELFARE
,
wi
i
;
---------------s------..
'.,
I
',
I;
;?..
i
1
For the scientificmethod can teachus nothing beyondhow factsare
relatedto, and conditionedby, each other.... Yet it is equallyclear
that knowledgeof what is does not open the door directlyto what
shouldbe.... Objectiveknowledgeprovidesus with powerfulinstruments for the achievementof certainends, but the ultimategoal itself
and the longingto reachit must come fromanothersource.... Here
we face, therefore,the limits of the purelyrationalconceptionof our
existence....
To make clearthese fundamentalends and valuations,and to set
them fastin the emotionallife of the individual,seemsto me precisely the most importantfunctionwhich religionhas to performin the
sociallife of man.... 42
I
Einstein'sreasoningmay also help us understandwhy some of
the
most distinguishedscientistsof the centuryhavebeen greatly
\X o,j/
religious.Seenthisway,rationalityand religionbelongto two different realmsof human experience-the formerhaving little to
do with the ends of life.43For those uninspiredby religionand
TI
T1
T2
T1ME
some of its excesses,culture-a set of institutionsand normative
practiceswe live by, some coming from ethnic or nationaltradihas alreadyacquireda considerablefollowing,succcess,and visi- tions-has been a sourceof suchvalues.Culturereplacesreligion
dersas a mobi- in the agnosticor unbelievinghomes.
bility.To explainwhy ethnicityis privilegedby leac
lization strategy,we must thereforelook elsewllere. To make
A rational-choicescholarmay retortthat culturedoes not exist
greatersense of the supplyside of the story,we peerhapsneed an on its own; it is a creationof individuals.What appearsas an
inheritancetodaywas createdby individualactsin the past,makanalysisof the demandside as well.
ing it possible for analyststo explain the existenceof culture
instrumentally.In a fundamentalsense, this view cannotbe corAlternative Microfoundations
rect. Culturemay indeed have been createdby individuals,but
A searchfor alternativesmust startwith answeringtwo questions:
. .
elf suffice as a each individualengagedin such creationwas also actingin relaWhy can't instrumentalrationalityin and of its<
a
to an inheritedset of practices.In orderfor an individualto
th role
roleos
basis for human motivation?(What, for example,Iis the
of tion
,
de culture
culturoe create,affirm,deny,or innovatea set of culturalpractices-and a
cultureor religionin humanlife?)And how andwlnydoes
good dealof that happensin everydaylife-there has to be a preor religionbecome a sourceof group conflict?Once we answer
tationalism
of existingset of normativepracticesin the frameworkof which the
these questions, the microfoundationsof the rnationalism
aton
creation,affirmation,denial,or innovationacquiresmeaning.A
,
resistancewill become clear.My purposeis to shc)w
that nonin'
'
'
w that
ntonin- sentenceor wordhasno meaninguntil a languageexists.Cultural
strumentalconsiderationsare highly importantiri the nationalchoicesarethus differentfrombuyinga caror a houseon the one
n
telfreand
ism of resistance,lacedas they arewith notionsof self-respect
hand and forming political strategiesto defeat adversariesfor
dignity,not with a narrowlydefinedself-interest.
political office on the other. Rational-choicetheories may be
more applicableto marginaldecisions, less to decisions about
Why culture or religion?
Either instrumentalrationality,as alreadystated, is a concept how people choose fundamentalvalues.44
about the means and not about the ends, or the self-interestis
Anotherclarificationis necessary.Placingemphasison a preassumedto be the end of human action. In any case, a serious existing or inherited culture to explain ethnic behavior is
sometimesseen as an endorsementof the "primordial"
view of
problemarisesrequiringa discussionof ends.
Self-interestcan certainlygive us our immediateor intermedi- ethnicity.Accordingto this view, ethnicityis an ascriptivegiven,
ate ends, but can it also provide the ultimate elnds or values? existing for centuries and thereforestrongerthan modern or
Indeed, if seen as a foundationof human life or as its ultimate rational forms of human motivation or institutionaldesigns.
goal, self-interestcan promote,to paraphraseHoblbes,loneliness, Man, arguesa leading exponent of the primordialview, is an
nastiness,brutishness.It is not clear that any reglulatoryframe- ethnic being, or a "national,not a rationalanimal."45
workdesignedby humaningenuitycan fullycheckthe manyacts
The sense in which my account of alternativemicrofoundaof nastinessif self-interestis turnedinto a supremevalue.Forthis tions relieson culturemust be distinguishedfrom the primorreason,as well as for intrinsicmoralor culturalreeasons,human dial view.Volition in the realmof cultureand identity is indeed
beings cannot live without notions of right and wrrong,without possible. Culture,ethnicity,and the nation can be-and arenotions that can guide them about how to relatet<
o family,com- often "constructed."Peasantswere turned into Frenchmenin
and
loved
ones.41
in 1789 more than 50 percentof Frenchmendid not
France;46
munity,
Religion-not rationalityor its most monumental expression, speak Frenchat all, and "only 12-13 percent spoke it correctscience-is traditionallyconsideredto providesuchvalues.Albert ly."47Over a period of roughly a century and a half, a British
Einsteinarguedforcefully:
identity was created out of the English, Scottish, and Welsh
\
S2 /
!.
www.apsanet.org 91
Articles | Nationalism,EthnicConflict,and Rationality
scalemay accepttheir inferiorstatusas given.A hierarchybased
on birthcan existwithout causinggroupconflict.
We need, therefore,to ask a historicalquestion: when did
human beings begin to questionthe idea of an ascriptivegroup
at the time of the Italian Unification.49 And as for identities at
a level lower than the nation, some smaller castes in India, hierarchy?In a work that has attractedwide attention, Charles
First,in premodrespondingto the imperativesof an evolvingpoliticaldemocra- Taylorhas made two compellingarguments.57
ern
in
one's
came
a
of
fusion
to
form
Who
am
castes
times,
I, and where am I
identity-as in,
cy,
process
together
larger
from?-was
established
cultural
and
of
or
fixed
one's
centuries
divisions
by
coming
place in the hierargiven
changing
patterns
while
others
went
a
of
All
chical
social
structure.
It
fission.50
of
was
not
thereby,
negotiated.The riseof moderthrough process
these identities were constructed,but the point to note is that nity has led to an increasingdecay of traditionalsocial hierarthey were not constructedon a clean slate.The acts of creation, chies-ideationally and/or structurally.As a result,for the first
innovation, or denial drew their rationale,negativeor positive, time in history a new individualmotivation has arisen:a selffrom an existing set of values. Culture,in this sense, is embed- awarenessof dignity. One does not take one's "station"as
ded in our life; it preexistsas a frameworkof meaning, within inevitable.Second, the pursuitof dignity and self-respectis not
which human deliberationand rationalityoperate.It is not just monological, but dialogical.The "dialogue"takes place in a
a privately underprovidedpublic good, but an "irreducibly social context. Hermits may define dignity monologically,but
social good."51
the more generalpursuitsof dignity requirerecognitionfrom
society. This is especiallyso because society is not a random
collectionof individuals;rather,it comeswith a historicalinherWhy is culture or religion a source of conflict?
If culture and religion provide values, how can they lead to conitance of perceptionsand misperceptions.Our identity as modflict? A simple answer would be that there are many such cultures
ern human beings
identities.48"Wehave made Italy,"said Massimod'Azeglioin a
legendarystatement;"nowwe have to make Italians."Only 2.5
percentof the populationspoke Italianas an everydaylanguage
and many religions, and their central tendencies clash. However,
as far as the nationalism of resistance is concerned, the issue is not
cultural or religious diversity per se, but a relationship of dominance, subordination, and differential worth that often gets historically built into many group relations, if not all.
Structurally speaking, groups in a society can be ranked or
unranked.52The hierarchical nature of the former is manifestly
clear: slavery in the United States and black-white relations in
South Africa during apartheid are among the best known examples. Similarly, in the Hindu caste system, the "lower"castes constitute an overwhelming majority but the tiny "upper" castes
have enjoyed ritual superiority and most of the power until
recently.
However, sometimes even if groups are structurally
unranked-in that a legal or deeply embedded ritual hierarchy
does not mark their interrelationship-domination or subordination could be discursive.53Some groups may argue that they are
the "sons of the soil," hence deserving of greater political, economic, or cultural privileges.54In Malaysia, the Malays make this
claim vis-a-vis the Chinese and the Indians; in Sri Lanka, many
Sinhalese do so with respect to the Tamils; Hindu nationalists in
India would like the Hindus to have a higher status than the
Muslims; and the followers of Le Pen would give more privileges
to French Europeans than to the North African immigrants in
France. Those who came earlier to a land have often argued that
they are more entitled to political privileges or to a preeminent
place in the national culture than those who came later.
By itself, of course, a structural or discursive hierarchy does not
engender ethnic or group conflict. Indeed, many from the disadvantaged groups may opt for what M.N. Srinivas called
"Sanskritization"-i.e., the attempt on the part of the ritually
subordinate Hindu castes to follow the life-styles of the upper
castes.55 Elsewhere, Antonio Gramsci spoke of "hegemony" to
describe how the subaltern may share the world view of the rich
and the powerful.56
Those ascriptivelyplacedloweron the social
92
March2003 [ Vol. 1/No. 1
is partlyshapedby recognition
or its absence,oftenby the misrecognitionof others,andso a personor groupsof peoplecansufferreal
damage,realdistortion,if thepeopleor societyaroundthemmirror
backto thema confining,demeaning,
or contemptible
pictureof
themselves.
or misrecognition
caninflictharm,can
Nonrecognition
be a formof oppression,
someonein a false,distorted,
imprisoning
andreducedmodeof being.58
Thus, even if structuralgrouphierarchyis absent,a discursive
hierarchy,laced with "confining,demeaning,or contemptible"
picturesfor some groups,maywell exist.Crudeilliberalprejudice
or hatredis, of course,an obvioussourcefor suchviews. But the
problemis much more complex.It is worth recallingthat until
this century,even well-meaningliberalsbelievedin group-based
notions of civilityand barbarism.In one of the foundingtextsof
liberalism,John StuartMill argued:
to a Breton,or a Basque
Nobodycansupposethatit is notbeneficial
of the FrenchNavarre,
to be broughtinto the currentof ideasand
feelingsof a highlycivilizedandcultivated
people-to be a member
of theFrenchnationality
... thanto sulkon hisownrocks,thehalfsavagerelicof pasttimes,revolvingin his own littlementalorbit,
withoutparticipation
or interestin the generalmovementof the
world.The sameremarkappliesto the Welshmanor the Scottish
asmembers
of theBritishnation.59
Highlander,
In the modernworld, thus, two differentnotions of dignity
and worth haveoften been at odds: one stemmingfrom the culturallyinheritedconceptionsof groupsas better or worse, and
anotherarisingout of a declineof socialhierarchiesand the rise
of equality.The latter seeks to underminethe formerby challengingthe inheritedstructureor discourseof grouphierarchy.
The questionof microfoundations-whereethnic or national
preferencescome fromin the nationalismof resistance-can now
be morepreciselyphrased.What arethe implicationsof a historicallyand culturallystructurednotion of ascriptivehierarchyfor
the individual-groupinteractionin moderntimes?How does an
individualfeel grouprelations?
An individualmay end up defining a core of her identity in
termsof hergroupbecauseshe is definedas such by society,a definition over whose origins she has no control but one whose
reorderingwill not take place unlesseffortsare made to compel
societyto changeits recognition.The questionis not simplyone
of waiting for othersto launch the effort and takinga free ride.
The individualwould like to participatein the effortbecauseshe
can'tlive a "reducedmode of being":she would feel less of a
humanbeing, or not able to respectherself,if she did not participate. Her self-respect,her dignity,is involved.
An account of the microfoundationsof ethnic or national
resistancethus requiressensitivityto historicallyinheritedattitudes and power relationsamong many groups, if not all. By
startingwith individualsand not the culturalor historicalinheritances and power relationswithin which individualsmay be
embedded,a typicalrational-choiceaccountmissesmuch of what
motivates ethnic or nationalistbehavior.In the process, it is
unableto accountfor some of the most importantand persistent
phenomenanoted by studentsof ethnicity.Why, for example,do
the minorities typically feel the group identity much more
stronglythan do the dominantgroups?People,whetherfromthe
dominant or the subordinategroup, are mere individualsin a
purely instrumentalframework.When Isaiah Berlin said that
Jews tend to "havelonger memories,"that "theyare awareof a
longercontinuityas a communitythan any otherwhich has suris what they historicallylacked,60he was
vived,"and "geography"
a
statement
about
his community that was incompremaking
hensiblein purelyinstrumentalterms.Why keepmemories?Why
should geographymatter?Why not change identity,instead of
findinggeographyto matchhistory?Structuredpatternsof dominanceand subordinationand a historyof sufferinghavecustomarily shaped answersto these questions,not pure instrumental
rationality.
Value Rationality and Ethnic Mobilization
The explanationabove exploresonly the microfoundationsof
ethnic resistance.It does not account for ethnic mobilization.
How are the microfoundationsand ethnic mobilizationrelated?
Three mechanismscan be specified.
First,a criticalmassof individualshavinga stronggroupidentification is all that one needs to explain the origins of ethnic
mobilization;strong identificationof all with the group is not
necessary.Value-rationalmicrofoundationsthus overcome the
principaldifficultyfaced by a purely instrumentalexplanation,
which was unableto explainthe originsof ethnic mobilization.
Second,dependingon how the dominantgroupsand the state
respondto the criticalmass, mobilizationitself can be identityforming for those who did not initially participate in it.
Hegemonymay give way to an assertionof self-respect.In 1919,
when thousandsof Indians(in defianceof a prohibitionon political meetings)organizeda protestmeetingin Amritsar,India,and
a Britishgeneralordereda massacreto implementthe law,a turning point was reached in India's national movement.61 The mas-
sacre changed Gandhi, convincing him that India'sself-respect
was not possibleuntil the Britishleft; it changedNehru from a
man who was "moreBritishthan the British"to one "homespun"
and capableof makingthe transitionfrom a life of privilegeand
luxury to one of personal sacrificefor the sake of a nation.
of digniIndeed,so manyIndiansexperiencedthe self-awareness
ty that afterthe Amritsarmassacreit becamepossibleto launcha
nationwidecivil-disobediencemovement.
Similarly,the Americancivil-rightsmovement in the 1960s
formed the assertiveidentity of a large number of African
Americans:"Whilethe studentsin their neat suits and demure
dressessat-in,marched,demonstrated,
sangandprayed,the police,
the sheriff'sdeputiesand the Klanrespondedto nonviolencewith
violence,meetingthe dovesof peacewith the policedogsof war."62
Elsewhere,barelya few yearsafterthe formationof Pakistan,the
EastPakistanisrealizedthat theirlinguisticidentitywas at stakein
a nation they joined for religiousreasons.They were told that
Urdu,the languageof MuslimmigrantsfromIndia,would be the
languageof the new nation, even though EastPakistanis,constitutinga majorityof the country,spokeBengali.A culturalcleavage
within the new nationwas thus born,givingroomto politicsand
mobilizationbasedon a linguisticidentity.As this politicsunfolded, the identityof the silentbystanderswas alsoformed.
Third, as is implicitabove,a conflictcannot takeplaceunless
we also factorin the behaviorof the dominantgroups.The dominantgroupstypicallyhavethreeoptions:defendpreexistingprivileges,with no adjustmentsmade;incorporatethe elite of the disadvantagedgroupsin the power structure;or renegotiateprivilege, acceptingsome notion of fairness.To defend preexisting
privilegesis a case of prejudice;to incorporatethe elite, one of
selectivecooptation;to renegotiateprivilege,one of fairness.In
no case,includingthe last, is conflictruledout.
A defenseof privilegeor prejudiceclearlyspellstrouble,once
the ideologicalhegemony of group hierarchyis broken and a
middleclasscapableof organizingthe groupdevelopsamongthe
previouslydisadvantaged.
Examplesarelegion.Dependingon the
nature of the political system, such conflict may be relatively
peacefulor violent. If the politicalsystemallowsthe freedomto
organize,ethnic mobilizationmay dominatedemocraticpolitics
but conflict may also be politicallyresolvedand violence overcome. However,if the politicalsystemis repressive,ethnic conflict may remainhidden or may not emerge in a routine way
(eruptingviolently,for instance,when the stateis weak).
Selectivecooptation may work if the elites so incorporated
continue to hold sway over the massesand are not outbid by
alternativeleadersrefusingto be co-opted. It may defuseethnic
conflict or even resolveit throughwhat Arend Lijphartcalls a
consociationalsystem.63Outbidding,however,is not uncommon
in ethnicconflicts.Consociationalism
worksunderwell-specified
institutionalconditions.64
Most interestingly,however,conflictcan occureven when the
leadersof the dominantgrouprenegotiateprivilege.The problem
simply is that the question of what constitutesfairnesshas no
uniquely acceptableanswer.Why should the membersof this
generationpay for the inequitiesof the past, in which they did
not directly participate?How much should they pay, if they
must? For how long? Multiple answersexist; the outcomes are
www.apsanet.org 93
Articles I Nationalism,EthnicConflict,and Rationality
politically determined. On affirmative action, such struggles are
universal.
Three Kinds of Ethnic and Nationalistic
Behavior
Central to the alternative account I have presented above are
notions of hierarchy, dignity, and recognition. Goal-oriented
thinking exists in this alternative account, but it is defined with
respect to the values so specified, not independently of these values. This conception of strategic behavior is different from the
one in which ethnicity itself is seen as a means to an end. If we
combine the two notions of rationality discussed above, we get
three different kinds of ethnic and nationalistic behavior, which
we should distinguish from one another.
Thepure case of value rationality
Martyrdom-suicide bombing, in these times-is the pure form
of value-rational behavior. In such cases, no cost (including
death) is considered too high by an ethnic partisan. If aimed at
enhancing group prospects, to kill may be a form of instrumental behavior-and likewise, being killed may result from someone else behaving instrumentally. But to die is not instrumentally rational for an individual, for whatever its benefits to the
group, the martyr will not be there to see his dreams fulfilled.
Such martyrdom, however, can be instrumentally beneficial for
the group, for it can touch off strong emotions, raising the level
of group consciousness. Indeed, collective martyrdom or martyrdom of an important leader of the group can be a tipping point
in group consciousness and mobilization.65
It is possible to argue that religious martyrdom is, in fact,
individually rational, for the motivations of the martyr extend to
life after death. Most religions have a notion of afterlife. Thisworldly martyrdom can pave the way for other-worldly glory.
But ethnic or national martyrdom, as opposed to religious
martyrdom, has no such notion of afterlife. Its aim is collective
benefit, pure and simple. Sri Lanka'sTamil Tigers repeatedly produced suicide bombers to increase group cohesion and to target
"enemies." In a number of national, or freedom, movements in
the developing world, there were many examples of men seeking
martyrdom or taking the risk of death.
Given the significance of death in nationalism, martyrdom can
also be instrumentally used by some-not, of course, by those who
die. Ethnic partisansare known to have killed important figures of
their own communities-so as to put the blame of death on the
adversaryand engineer in-group cohesion. This use of martyrdom
is instrumental-rational and must be distinguished from the
behavior of those seeking martyrdom. The latter is value-rational.
Thepure case of instrumental rationality
From an individual perspective, the instrumental benefits of participating in nationalist mobilization are obvious only under two
strict conditions: when nationalists are already close to capturing
power and much can be gained (or losses cut) by joining the bandwagon, and when law and order have broken down, ethnic animosities have soured group relations, and even neighbors-of long
standing but belonging to a different ethnic group-can't be trust94
March 2003
i Vol. 1/No. 1
ed, creating a "securitydilemma" for individuals and making preemptive violence against neighbors of a different ethnic group an
exercise in personal security.66Most ethnic conflicts do not reach
this last Hobbesian state of nature. It was typical of the former
Yugoslavia in recent times; of massacres in Rwanda; and of the
border states of India, especially Punjab, during the country's partition in 1947.
Combiningvalue rationality and instrumentalrationality
This is the category where a lot of ethnic conflict belongs. The
concept of rationality here can mean two things: seeing ethnicity
as a means to a self-interested end, or else selecting appropriate
means to realize group goals or choosing between competing
group goals. Enough has already been said about the first; why
might the second be necessary?
The fact that my identity gets tied up with my group does not
mean that I accept as right everything that the group (i.e., its
leadership on behalf of the group) does. I may have a different
version of group objectives and may even try to convince my
group that my version is right. My identity may be tied up with
my group, but my views may not be. Such intra-ethnic clashes on
what is valuable and what means are appropriate to achieve those
goals allow for a great deal of volition, intragroup strategizing,
and struggle. Indeed, if I have leadership ambition, I may even try
to retrieve my group's history purposively to show that I am historically more authentic than are my adversaries in the group,
while both my adversariesand I seek group betterment. Selective
retrievalof tradition is a standard strategy in nationalist struggles.
Alternatively, people may try to change the form of protest.
Sometimes, this means moving from nonviolent to violent means;
at other times, it simply entails exploring alternative nonviolent
strategies, as seen in the Indian freedom movement, of which the
Salt March was a component, and in the American civil-rights
movement. In many nationalist conflicts, however, even when the
ends are noble, the means are not. Violence is often used as an
instrument for ethnic ends. Our moral objections to violence
notwithstanding, it is undeniable that from the perspective of ethnic and national partisans, violence can represent a combination
of value rationality and instrumental rationality.
When asked by psychologist Sudhir Kakar why they killed
members of the other community, the wrestlers involved in communal violence in the Indian city of Hyderabad argued that they
were defending the quam (nation). They stopped killing, they
said, when they had killed more than the wrestlers of the other
community had killed. Indeed, after giving them tests to check
lies, falsehood, and dissimulation, Kakarhad to conclude-much
to his emotional dismay but true to his professional craft-that in
psychological terms, the killerswere "warriors,"not "murderers."67
Much of the dynamics and intensity of ethnic conflict cannot
be explained unless we understand how decisions are made about
which sections of the population-women, children, and old
people or the able-bodied men-are the targets of violence;
whether festivals and celebrations are disrupted; whether sacralized monuments and places of worship are attacked; whether
automatic weapons are used by a few or small weapons by a lot,
although each method may kill as many people. We are in a world
requires a great deal of
unpacking.A searchfor dig-
Table 1
Leaders
Masses
OUTCOME
PureValueRational
Pure InstrumentalRational
Value- and InstrumentalRational
nity and recognition may well
define the motivation under-
yes
yes
yes
yes
yes
lying the nationalism of resistance, but other forms of
nationalistic behavior-such
originsyes;
sustenanceno
originsno;
sustenancepossible
originsyes;
sustenance
yes
sustenance
yes
as those witnessed in the
nationalism of exclusion or
whereconsiderableplanningoften goes into the timing,type,and
targetsof violence, for symbolicviolence is often centralto ethnic conflict.Much, if not all, of this strategicbehavioris basedon
the groupgoals that ethnic or nationalpartisanshave. It will be
hardto provethatnationalistsmakesuchdecisionson purelyselfinterestedgrounds,without linking their strategiesto the interests of the groupthey seek to represent.
Table1 summarizesthe argumentso far.The purecaseof value
rationalitymayaccountfor the originsof ethnicmobilizationbut
not for its sustenance;the pure case of instrumentalrationality
cannot explainwhy ethnic mobilizationcommences,althoughit
may begin to explainbehavioronce mobilizationhas reacheda
criticalpoint; and the combinationof value and instrumental
rationalitycan explainboth why ethnic mobilizationbeginsand
how it is sustained.
Conclusions: Pluralizing
Microfoundations
Three conclusions follow. First, rational-choicetheories are
unableto answersome of the fundamentalquestionsin the study
of ethnicityand nationalism.They almostwholly concentrateon
why leadersmanipulateethnicity or nationalfeelings, ignoring
questionswithout which we can'tunderstandmobilizationfor
ethnic or nationalresistance:Why do the massesjoin ethnic and
nationalmovementswhen the costs of participationare almost
certainto be high?And why do minoritiesso often feel the group
To answerthesequesidentitymoreintenselythan do majorities?
tions, one has to pluralizethe conceptof rationality.A distinction
between value rationalityand instrumentalrationality,as proposed by Weber, will be a good starting point. The former
concept is considerablyless sensitiveto the notion of costs of
behaviorthan the latter.Some goals-national liberation,racial
equality,ethnic self-respect-may be deemed so precious that
high costs, quite common in movementsof resistance,are not
sufficientto detera doggedpursuitof such objectives.The goals
areoften not up for negotiationand barter;the meansdeployed
to realizethem may well be.
However-and this is the secondconclusion-once ethnicpreferencesare in placeand mobilizationhas reacheda criticalmass,
methraisingprospectsof success,one can use the rational-choice
ods to understandwhy manypeoplejoin ethnicor nationalmovements.The approachworksbestwhenethnicityis assumedto exist,
not if one wereto analyzewhereethnicpreferences
come from.
the
Weberian
idea
of
value
while
of generFinally,
rationality,
ic significancein analyzingethnic or nationalistbehavior,also
when majority groups develop
a minority complex-may be
motivations
other
than dignity and self-respect.
undergirded by
A pluralism of microfoundations is quite likely to be found as we
move further along this path of inquiry.
References
Almond, Gabriel. 1991. Rational choice theory and the social
sciences. In The Economic Approach to Politics, ed. Kristen
R. Monroe. New York: HarperCollins.
Alt, James, and Kenneth Shepsle, eds. 1990. Perspectiveson
Political Economy.New York: Cambridge University Press.
Anderson, Benedict. 1983. Imagined Communities. London:
Verso.
Bates, Robert. 1974. Ethnic competition and modernization
in contemporary Africa. ComparativePolitical Studies 6:4,
457-77.
. 1981. Marketsand Statesin TropicalAfrica.Berkeley:
University of California Press.
Berlin, Isaiah. 1982. Againstthe Current:Essaysin the History
of Ideas. New York: Penguin.
Booth, William James, Patrick James, and Hudson Meadwell,
eds. 1993. Politicsand Rationality:RationalChoicein Application. New York: Cambridge University Press.
Brass,Paul. 1975. Language,Religionand Politicsin North
India. Cambridge: Cambridge University Press.
Breton, Albert, Gianluigi Galeotti, Pierre Salmon, and Ronald
Wintrobe, eds. 1995. Nationalism and Rationality. New
York: Cambridge University Press.
Buchanan, James. 1995. Economic science and cultural diversity. Kyklos48:2, 193-200.
Chong, Dennis. 1991. CollectiveActionand the Civil Rights
Movement. Chicago: University of Chicago Press.
Coase, Ronald. 1978. Economics and contiguous disciplines.
Journalof LegalStudies7:2, 201-12.
Colley, Linda. 1992. Britons:Forginga Nation (1707-1837).
New Haven: Yale University Press.
Connor, Walker. 1994. Ethnonationalism. Princeton: Princeton
University Press.
Davidson, Donald. 1963. Actions, reasons, and causes. The
Journal of Philosophy60:23, 685-700.
Einstein, Albert. [1954] 1982. Ideas and Opinions. New York:
Crown Publishers.
Elster, Jon, 1982. Marxism, functionalism, and game theory:
The case for methodologicalindividualism.Theoryand
Society11:4, 453-82.
www.apsanet.org 95
Articles I Nationalism,EthnicConflict,and Rationality
.
Nuts and Boltsfor the SocialSciences.New
1989.
Marx,Anthony W. 1998. MakingRaceand Nation:A ComYork:CambridgeUniversityPress.
parisonof SouthAfrica,the UnitedStates,and Brazil.New
Walter.
Nationalism
in
York:CambridgeUniversityPress.
a
1997.
mode:
Feinberg,
comparative
A responseto CharlesTaylor.In The Moralityof National- Mill, John Stuart. 1990. ThreeEssays.New York:Oxford
ism, eds. Robert McKim and Jeff McMahan.New York:
UniversityPress.
Oxford UniversityPress.
Miller,Webb. 1994. I FoundNo Peace.Sections reproduced
in The GandhiReader,ed. Homer Jack. New York:Grove
Fiorina,Morris. 1995. Rationalchoice, empiricalcontribuPress.Originalpublication 1936, New York:Simon and
tions, and the scientificenterprise.CriticalReview9:1-2,
Schuster.
85-94.
Clifford.
Old
New
States.
New
1963.
York:
Geertz,
Societies,
Monroe, KristenRenwick,ed. 1991. The EconomicApproach
Free Press.
to Politics.New York:HarperCollins.
Ernest.
Nations
and
Nationalism.
Ithaca:
1983.
Gellner,
Munck, Gerardo.2001. Game theory and comparativepoliCornell UniversityPress.
tics. WorldPolitics53:2, 173-204.
Gramsci,Antonio. 1971. Selections
from the PrisonNotebooks. Nozick, Robert. 1983. The Natureof Rationality.Cambridge:
London: Lawrenceand Wishrat.
HarvardUniversityPress.
Green, Donald, and Ian Shapiro. 1994. Pathologiesof
Olson, Mancur. 1965. The Logicof CollectiveAction.
RationalChoiceTheories.New Haven:Yale University
Cambridge:HarvardUniversityPress.
Press.
EthnicViolence:Fear,
Peterson,Roger D. 2002. Understanding
Hatred,and Resentmentin Twentieth-Century
Gurr,Ted Robert. 1993. Minoritiesat Risk:A GlobalView of
Eastern
EthnopoliticalConflicts.Washington,D.C.: The U.S. InstiEurope.New York:CambridgeUniversityPress.
tute of Peace Press.
Popkin, Samuel L. 1979. The RationalPeasant:The Political
Hardin, Russell. 1995. Onefor All: The Logicof GroupConEconomyof RuralSocietyin Vietnam.Berkeley:University
of CaliforniaPress.
flict. Princeton:PrincetonUniversityPress.
Posen, Barry.1993. The securitydilemma and ethnic conHarsanyi,John. 1976. Essaysin Ethics,SocialBehaviorand
Boston:
D.
Reid.
flict. Survival35:1, 27-47.
ScientificExplanation.
Hechter,Michael. 2000. Nationalismand rationality.Studies
Przeworski,Adam. 1985. Capitalismand SocialDemocracy.
in Comparative
InternationalDevelopment35:1, 3-20.
New York:CambridgeUniversityPress.
Hirschman,Albert. 1985. Against parsimony.Economicsand
Putnam, Hilary. 1981. Reason,Truthand History.New York:
Philosophy1:1, 7-21.
CambridgeUniversityPress.
Hobsbawm,Eric. 1990. Nationsand Nationalismsince 1780:
Rabushka,Alvin, and KennethA. Shepsle. 1972. Politicsin
UniverPlural Societies:A Theoryof DemocraticStability.
Programme,
Myth, Reality.Cambridge:Cambridge
Press.
Columbus, Ohio: CharlesE. MerrillPublishing.
sity
Horowitz, Donald. 1984. Ethnic Groupsin Conflict.Berkeley: Raines, Howell, 1977. My Soul is Rested.New York:GP
Putnam'sSons.
Universityof CaliforniaPress.
. 1987. DemocraticSouthAfrica?Berkeley:University
Rudolph, Lloyd, and SusanneRudolph. 1967. Modernityof
of CaliforniaPress.
Tradition.Chicago:Universityof Chicago Press.
. 2001. The DeadlyEthnicRiot. Berkeley:University
Schelling,Thomas. 1963. The Strategyof Conflict.New York:
of CaliforniaPress.
Oxford UniversityPress.
Kakar,Sudhir. 1996. Colorsof Violence.Chicago:University
Sen, Amartya.1973. Behaviourand the concept of preferof Chicago Press.
ence. Economica40:159, 241-59.
. 1982. Rationalfools. Choice,Welfareand MeasureKedourie,Elie. 1993. Nationalism.Oxford: Blackwell.4th,
ment. Cambridge:MIT Press.
expandeded.
Kolakowski,Leszek. 1990. Modernityon EndlessTrial,trans.
. 1992. On Ethicsand Economics.New York:
Stefan Czerniawski.Chicago:Universityof Chicago Press.
Blackwell.
Kundera,Milan. 1986. TheArt of the Novel, trans. Linda
Simon, Herbert. 1986. Rationalityin psychologyand ecoAsher. New York:Grove Press.
nomics. In RationalChoice:The ContrastsbetweenPsycholLaitin, David. 1986. Hegemonyand Culture:Politicsand Reliogy and Economics,eds. Robin Hogarth and Melvin Reder.
gious Changeamongthe Yoruba.Chicago:Universityof
Chicago:Universityof Chicago Press.
Srinivas,MysoreNarasimhachar.1966. Social Changein
Chicago Press.
. 1998. Identityin Formation:The RussianSpeaking
ModernIndia. Berkeley:Universityof CaliforniaPress.
Tarrow,Sidney. 1998. Powerin Movement:SocialMovements
Populationsin the Near Abroad.Ithaca:Cornell University
Press.
and ContentiousPolitics.New York:CambridgeUniversity
Arend.
in
Democracies
Plural
1977.
Societies:
A
ComPress.
2nd ed.
Lijphart,
New
Haven:
Yale
Press.
Charles.
1993. Reconcilingthe Solitudes:Essayson
parativeExploration.
University
Taylor,
Mandela,Nelson. 1994. Long Walkto Freedom.Boston:
CanadianFederalismand Nationalism.Montreal:McGillLittle, Brown.
Queen'sUniversityPress.
96
March2003 i Vol. 1/No. 1
. 1995. Irreduciblysocial goods. Philosophical
Arguments.Cambridge:HarvardUniversityPress.
Taylor,Charles. 1994. Multiculturalism:
Examiningthe Politics
of Recognition.Princeton:PrincetonUniversityPress.
Taylor,Michael. 1993. Structure,culture and action in the
explanationof social change. In Politicsand Rationality,
eds. William James Booth, PatrickJames, and Hudson
Meadwell.New York:CambridgeUniversityPress.
Tversky,Amos, and Daniel Kahneman.1990a. Judgment
under uncertainty.In Rationalityin Action,ed. Paul
Moser.New York:CambridgeUniversityPress.
. 1990b. Rationalchoice and framingof decisions.
The Limitsof Rationality,eds. KarenCook and Margaret
Levi. Chicago:Universityof Chicago Press.
Varshney,Ashutosh. 1995. Democracy,Developmentand the
New York:CambridgeUniversityPress.
Countryside.
. 2002. Ethnic Conflictand Civic Life. New Haven:
Yale UniversityPress.
Weber,Eugene. 1976. Peasantsinto Frenchmen.Stanford:
StanfordUniversityPress.
Weber,Max. 1978. Economyand Society:An Outlineof InterpretiveSociology,eds. Claus Wittich and GuentherRoth.
Berkeley:Universityof CaliforniaPress.
Weiner,Myron. 1978. Sons of the Soil: Migrationand Ethnic
Conflictin India. Princeton:PrincetonUniversityPress.
Young, Crawford.1983. The temple of ethnicity. WorldPolitics 35:4, 652-62.
Notes
1 Kundera1986, 3-4.
2 Berlin 1982, 1.
3 Ibid, 13.
4 The exchangewas, of course, not entirelypolite. Herder
called Voltairea "senilechild." He also wrote a vehement
poem (cited in Kedourie 1993, 53):
AndYouGermanalone,returning
fromabroad,
Wouldstgreetyour motherin French?
O spewit out,beforeyourdoor
Spewout theuglyslimeof theSeine
SpeakGerman,O youGerman.
5 Kedourieagrees.See Kedourie 1993, especiallychapters3
and 4.
6 Weber 1978, volume 1, 24-5.
7 For a recent statementof this distinction, see Feinberg
1997, 69-73.
8 Of course,all nationalisms,includingthe nationalismof
resistance,are to some extent exclusionary.Without the
notion of "us"and "them,"nationalismdoes not work.
However,despite not being entirelydevoid of exclusion,
nationalismof resistancetends to define communityin a
more inclusiveway than does the nationalismof exclusion.
For futureanalysis,the most difficultcategory-and a category differentfrom the above two-is going to be "majorities feeling like a minority"(the Sinhalesein Sri Lankauntil the 1970s; the Malaysin Malaysiauntil the 1970s; the
Pribumiin Indonesiaeven today).The microfoundations
drivingsuch behaviorare complex,requiringpainstaking
investigation.It is throughcumulativesteps that we will be
able to developan alternativetheory of where ethnic preferencescome from. For some thoughtfulpsychological
probings,see Horowitz2001 and Peterson2002.
9 Horowitz 1984, 41-54.
10 Gellner 1983, 1.
11 Sen 1982, 105. To sample the varietyassociatedwith rationality in philosophicaldiscussions,also see Nozick
1983, especiallythe chapterentitled "InstrumentalRationality and Its Limits,"133-81; Putnam 1981, especially
the chapter"TwoConceptionsof Rationality";and
Davidson 1963.
12 Mention should also be made of the concept of rationality in psychology,sometimescalled "boundedrationality."
See Tverskyand Kahneman1990a; Tverskyand Kahneman 1990b; and Simon 1986.
13 On whether the same instrumentalrationalityis applicable to spheresbeyond commercialbehavior,see not only
Sen 1982 (cited above), but also Coase 1978 and
Buchanan1995.
14 Also see Almond 1991.
15 Some of the leading scholarsof rationalitywould not entirely approveof such an analogy.AmartyaSen, while
writing about behaviorbased on commitment as opposed
to self-interest,drawsa distinction between preferences
and metapreferences.
The formerconcept is basicallywhat
Paul Samuelsoncalled "revealedpreference,"representing
choices evident in marketbehavior,such as when we buy
cars or footwear;the latter speaksof the largerpsychological and culturalprocessesthat undergirdthe actuallyobserved marketchoices. See the discussionin Sen 1973, as
well as Sen 1982; also see Hirschman1985. Strictlyfor
the purposesof this paper,although not more generally,
this criticismcan basicallybe viewed as a dispute over appropriateanalogies.Whether or not value-rationalitycan
be seen as a deeper set of metapreferences
generatingobservedchoices in behavior,the basic claim that it is different from instrumentalrationalityis not underminedby
an argumentabout metapreferences.
16 For overviewsof rational-choicemodels of politics, see Alt
and Shepsle 1990; Green and Shapiro 1994;
Monroe 1991; and Booth et al. 1993.
17 Olson 1965.
18 Bates 1981.
19 Critiquesemergingfrom within the rational-choiceparadigm are very helpful in understandingthe limits of
rational-choicetheories.Among the most thoughtfulselfcritiquesare Jon Elster 1989 and MichaelTaylor 1993.
Elster'sargumentis that rational-choicetheory is inapplicable in the following situations:(1) when multiple optima exist, (2) when the choice set has incommensurable
options, (3) when no reliableprobabilityestimatescan be
made, subjectively or objectively, because of insufficient
evidence, and (4) when it is not even clear how much
www.apsanet.org 97
Articles | Nationalism,EthnicConflict,and Rationality
20
21
22
23
24
25
evidence should be collected before such judgmentscan
be made.
For some reflectionsof the domain specificityof
rational-choicearguments,see Munck 2001.
MorrisFiorina,a rational-choicescholarof Americanpolitics, acceptsthat elite and mass politics have very different implicationsfor a rational-choiceanalysis:"Rational
Choice Models are most useful where stakesare high and
numberslow. . . . Thus in works on mass behaviorI utilize minimalistnotions of rationality. . . whereasin
works on elites I assumea higher order of rationality."
Fiorina 1995, 88.
There are two kinds of works on instrumentalconceptions of ethnicity.The works that follow the rationalchoice method self-consciouslyinclude Rabushkaand
Shepsle 1972, Hardin 1995, and Hechter 2000. The idea
of the instrumentaluse of ethnicity,however,goes beyond
the rational-choiceliterature.It is implicit in much of the
literatureon ethnic conflict. See, e.g., Brass 1975 and
Bates 1974. Sometimes,Gellner 1983 is also seen as a
major instrumentaltext. Gellner'sbasic argumentis that
industrializationled to nationalismin history.The "low"
oral cultures,he argues,could not have producedthe
standardizationnecessaryto run an industrialeconomy;
only "high"cultureswith standardizedmodes of communication could have. I read Gellner more as a functionalist than as an instrumentalist.For a clear statementof
differencesbetween functionalismand rationalchoice, see
Elster 1982.
Schelling 1963, 54.
Ibid, 74.
Unless, of course, the Mafia is involved in the transac-
original.
35 As Hardin 1995 was published,anothercollectionof essays
addressingthis problemcame out. See Bretonet al. 1995.
The opening lines of this book are worth noting: "The literatureon nationalismis enormous.Economists,historians,
philosophers,politicalscientists,sociologistsand other
scholarsas well as lay observersand commentatorshave all
broughttheir particularskills and methods to bearon the
phenomenonwhich, it would be easy to argue,has dominated human affairsfor a good part of the nineteenthcenMarch2003
40
41
tion.
26 Miller 1994, 250-3. This is not to say that demonstrations do not often dissolve in the face of coercion.That,
however,is less surprisingthan the fact that so many ethnic movementspersistdespite coercion.
27 Raines 1977, 56.
28 Chong 1991, 85.
29 Ibid, 25-6.
30 Mandela 1994, 387.
31 Gurr 1993 makes a statisticalattempt.
32 For how economic and ethnic mobilizationscan dramatically vary,see Varshney1995 and 2002.
33 Elster 1989, 118.
34 Rabushkaand Shepsle 1972, 64-5. Emphasisin the
98
36
37
38
39
I Vol. 1/No. 1
42
43
tury and throughoutthe twentieth.The contributionof
what we may call the rationalchoice paradigmhas, however,not been large"(ix).
Hardin 1995, 5, 36-70.
Popkin 1979.
Przeworski1985.
Hardin'sapproachis abstractand philosophical.In a more
empiricalvein, there is also Laitin 1998 on the formation
of a new identity, "the Russian-Speaking
Populations,"in
four republicsof the formerSoviet Union: Kazakhstan,
Estonia, Latvia,and Ukraine.In theory,this work could
have answeredthe question posed above:how do people
develop, or maintain,ethnic or nationalpreferences,especially when the costs of expressingthose preferencesare
ex ante so high? Laitin'sempiricalapproach,however,allows him to focus only on the formationof new and
pragmaticidentitiesof (primarily)Russiansin areaswhere
conflict did not take place. There is little variationon the
dependentvariable.As such, the empiricalmaterialsof
Laitin are unable to answerquestionsraisedabove about
the nationalismof resistance.Had Laitin'sfocus included
the Chechens,we would have learnedmuch empirically
about the source of nationalisticpreferenceseven in the
face of high-cost conditions.
Indeed, even close to time T2, as arguedearlier,so long
as the benefits of group power are nonexcludable,I
should not join for I will get the benefits anyway.Thus,
time T2 also has a problemof indeterminacy,requiring
ethnic leadersto set up mechanismsto ensurethat benefits are distributedaccordingto participation.For the sake
of parsimony,however,let us assumethat instrumentalrationality at time T2 means participation.
This, of course, raisesthe question of whether "rational
ethics"can exist and whether it can be embeddedin society. See Sen 1992 and Harsanyi1976.
Einstein 1982, 41-2. Also see Kolakowski1990, especially
"Modernityon EndlessTrial"and "The Revengeof the
Sacredin SecularCulture."
The conflict, Einstein adds, begins when rationalityclaims
it can pronounceauthoritativelyupon the ends of human
life and religion claims that it can explain empiricalrelationships.
44 Laitin 1986, 148-9, makes a roughly similar argument:
"Rationalchoice [theory]... is a theory of marginaldecisions. It cannot tell us if ultimatelybutter is better than
guns; it can tell us that at a certain point the production
of a small number of guns will cost us a whole lot of
butter and at that point it is probablyirrationalto produce more guns. Within a political structure,individuals
constantlymake marginaldecisions. Neo-Benthamitetheories can give us a grasp on how individualpolitical actors are likely to make choices within that structure.
Microeconomictheory cannot, however,handle long-term
and non-marginaldecisions. When marketstructuresare
themselvesthreatened,and people must decide whether
to work within the new structureor hold on to the
old-without an opportunityfor a marginaldecisionmicroeconomictheory is not applicable."Also see Elster
1989, 40.
45 Connor 1994. The primordialview is often associated
with Geertz and Connor. See Geertz 1963. This view was
fashionablein the 1960s. In the 1970s, the "instrumental" view arose as a reactionto the primordialview. For a
review of the debate, see Young 1983.
46 Weber 1976.
47 Hobsbawm 1990, 61.
48 Colley 1992.
49 Hobsbawm 1990, 60.
50 Rudolph and Rudolph 1967.
51 Taylor 1995.
52 Horowitz 1984, 22-36.
53 The implicationhere, it should be clarified,is not that
ethnic groups are alwaysranked,either structurallyor discursively.Many unrankedethnic relationshipsin both
senses can, and do, exist. The Jews, Irish, and Italianstoday have an unrankedrelationshipwith the WASPsin the
United States;that was not true in the late nineteenth
century.The relationshipof the Parsisand Sikhs with the
majorityHindu community in twentieth-centuryIndia is
unranked,unlike India'scaste system, which continues to
be discursively,if not legally,ranked,although its ranking
is being vigorouslychallengedin currentpolitics. Another
interestingexampleof a rankedrelationshipturning unrankedcomes from South Africa.The English and
Afrikanerstoday are unrankedwith respectto each other,
although right until the early decadesof this century,the
relationshipwas ranked.For how this happened,see Marx
1998.
54 Weiner 1978.
55 Srinivas1966.
56 Gramsci1971.
57 See Taylor 1994. Tayloris not only a leading political
philosopherof our times, but also a political activistdealing with the politics of nationalismin Quebec. For his
Quebec-focusedwritings, see Taylor 1993.
58 Taylor 1994, 25.
59 Mill 1990, 385-6.
60 Berlin 1982, 252.
61 The massacrewas orderedin a walled park that had only
one opening to the road, servingboth as an exit and as
an entrance.The generalbroughthis forces in, closed the
exit-cum-entrance,and orderedhis troops to shoot unarmed men and women assembledfor a peacefulprotest
meeting. The crowd could not leave the park, even as the
bullets rained in.
62 Tarrow1998, 130.
63 Lijphart1977.
64 Horowitz 1987.
65 The significanceof death in nationalismis broughtout
forcefullyby Anderson 1983. The epigraphto this paper,
focusing on the idea of the tomb of unknown soldiers,
capturesone of the basic ideas.
66 Posen 1993.
67 Kakar1996.
www.apsanet.org 99