You are on page 1of 24

Collective Identity and Social Movements

Author(s): Francesca Polletta and James M. Jasper
Source: Annual Review of Sociology, Vol. 27 (2001), pp. 283-305
Published by: Annual Reviews
Stable URL: http://www.jstor.org/stable/2678623
Accessed: 04/07/2010 23:34

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=annrevs.

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 support@jstor.org.

Annual Reviews is collaborating with JSTOR to digitize, preserve and extend access to Annual Review of
Sociology.

http://www.jstor.org
Annu.Rev.Sociol. 2001. 27:283-305
? 2001 byAnnualReviews.All rightsreserved
Copyright

COLLECTIVE IDENTITY AND SOCIAL MOVEMENTS

FrancescaPollettaland JamesM. Jasper2
'Department ofSociology,ColumbiaUniversity,
510 Fayerweather,
NewYork,
NY 10027, e-mail:Fap8@columbia.edu
2346 West15thStreet,
NewYork, NY 100]1; e-mail:jmJasper@juno.com

Key Words politics,
protest,
culture,
groupboundaries
* Abstract Sociologists tocollective
haveturned to fillgapsinresource
identity
andpolitical
mobilization processaccountsoftheemergence, andimpacts
trajectories,
of socialmovements. Collective
identityhas beentreatedas an alternative
to struc-
turally
giveninterests
inaccountingfortheclaimsonbehalf ofwhichpeoplemobilize,
an alternative
to selective
incentivesin understandingwhypeopleparticipate, an al-
toinstrumental
ternative inexplaining
rationality whattactical
choicesactivists
make,
andan alternative
to institutional
reforms in assessingmovements' impacts.Collec-
tiveidentity
hasbeentreatedbothtoobroadly andtoonarrowly, sometimesappliedto
toomanydynamics, atothertimesmadeintoa residualcategory withinstructuralist,
andrationalist
state-centered, accounts.

INTRODUCTION

Recenttrendsbothinsideand outsidetheacademyhaverevivedinterest in col-
lectiveidentity.In the1980s,battlesovercollegecurricula, multiculturalism, and
affirmativeactionwerebothjustified andattackedas "identity politics."Among
gayandlesbianandfeminist groupsandmovements forethnicandracialjustice,
effortstocontest cultural
representations andtocelebratealternative
identitiesvied
withmoretraditional suchas litigation
strategies andlobbying. Collectiveidentity
was hardto miss,and itsprominence in contemporary movements encouraged
sociologiststo assessitsrolein all movements, newandold.
Sociologistsofsocialmovements havealso beenattractedtocollectiveidentity
as a responseto gaps in dominant resourcemobilization and politicalprocess
models.Those modelssoughtto counterearliercollectivebehaviorist viewsof
protestersas irrationalindividuals propelledintoprotestby crowdcontagionor
systemstrain.Mobilizationandprocesstheorists focusedrather on thestructural
shiftsthatgavecollectiveactorstheresourcesto actcollectively on longstanding
grievances.Buttheiremphasisonthehowofmobilization overthewhyofit,their
focuson thestateas target ofaction,andtheirdependence on rationalistic
images
ofindividual actionleftimportant issuesunexamined.

0360-0572/01/0811-0283$14.00 283
284 POLLETTA E JASPER

In responsetotheselimitations, scholarsturned tocollectiveidentitytoanswer
fourkindsof questions.One was whycollectiveactorscome intobeingwhen
theydo. Resourcemobilization andpoliticalprocesstheorists cuttheirteethon
theAmerican civilrightsmovement, wherethefactthatinsurgents hadgrievances
was notparticularly mysterious; thechallengewas ratherto explainhow they
securedtheresourcesto do something collectively aboutthosegrievances. But
in othermovements, theveryfactthata groupformed aroundan issuedemanded
explanation. Forexample,whyhasabortion provoked suchintensemobilization in
thiscountry andnotinEurope?Focusingonidentity seemeda waytoexplainhow
interestsemergedratherthantakingthemas given.By examining theformation
ofcollectiveidentities, scholarswouldshedlighton themacrohistorical context
within whichmovements emerge.
A secondchallenging questionhadtodo withpeople'smotivations toact.Even
withan acknowledged interestinan issue,ofcourse,peopleoftenopttofreeride.
Butthosewhodo participate usuallydo so in theabsenceof selectiveincentives
orcoercion,Olson's (1965) solutionsto thefree-rider dilemma.Collectiveiden-
tityseemedto capturebetter thepleasuresandobligations thatactuallypersuade
people to mobilize.Identity was appealing,then,as an alternative to material
incentives.
A thirdquestionneglectedbymainstream modelshad to do withmovements'
strategicchoices.If peoplechooseto participate becausedoingso accordswith
whotheyare,theformsof protesttheychooseare also influenced bycollective
identities.Modelsofstrategic choicethathadmovement leadersselectingamong
strategies,tactics,andorganizational formsbyinstrumentally assessingenviron-
mentalopportunities and constraints missedthefactthatstrategic optionsmay
also be intrinsicallyappealing.Theyreflect whatwe believe,whatwe are com-
fortable with,whatwe like,who we are. Collectiveidentity thusrespondedto
theinadequaciesofinstrumental as an explanation
rationality forstrategicchoice.
Finally,collectiveidentity has beena wayto getat theculturaleffects of social
movements. Dominantmodelsofcollectiveactionhavebeenbetterat measuring
movement outcomessuchas policyreform or expandedpoliticalrepresentation
thanat gaugingimpactsoutsidetheformalpoliticalsphere.Butmovements also
transform culturalrepresentations, socialnorms-howgroupssee themselves and
are seen by others.Changesin collectiveidentity capturedmovement impacts
beyondinstitutional reform.
Thequestionsthatprompted socialmovement scholarstotheorize aboutcollec-
tiveidentity areimportant, andtheyhavegenerated strong claimsabouttheroleof
collectiveidentity in movements' emergence, trajectories,andoutcomes.Indeed
we arguethatcollectiveidentity has beenforcedtodo toomuchanalytically. The
termhas beenused to describemanydifferent dimensions and dynamicsof so-
cial protest:thesocial categoriespredominating amongactivists(say "women"
or "animalrightsactivists"),publicrepresentations of social categories(what
Johnston et al 1994 referto as "publicidentities"), activists'shareddefinition
of theirsituation,theexpressivecharacter of all action,theaffective bondsthat
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 285

motivate participation, theexperience ofsolidarity within movements, andothers.
As a resultofthisdefinitional catholicity, keyquestionshavebeenobscured.To
whatextentarecollectiveidentities constructed in andthrough protest ratherthan
preceding it?Is theidentity a groupprojectspubliclythesameone thatitsmem-
bersexperience? Arecollectiveidentities imposedongroupsorinvented bythem?
Do individuals choosecollectiveidentities to maximizetheirself-interest or do
interests flowfromidentities? How is collectiveidentity differentfromideology?
Frominterest? Fromsolidarity?
To avoidoverextension oftheconcept,we havedefined collectiveidentityas an
individual's cognitive, moral,andemotional connection witha broader community,
category, practice,or institution.It is a perception of a sharedstatusor relation,
whichmaybe imaginedratherthanexperienced directly, and it is distinctfrom
personalidentities, although itmayformpartof a personalidentity. A collective
identity mayhavebeenfirst constructed byoutsiders (forexample,as in thecase
of"Hispanics"in thiscountry), whomaystillenforceit,butitdependson some
acceptanceby thoseto whomit is applied.Collectiveidentities are expressed
in cultural materials-names, narratives, symbols, verbalstyles,rituals, clothing,
and so on-but notall culturalmaterials expresscollectiveidentities. Collective
identity doesnotimplytherationalcalculusforevaluating choicesthat"interest"
does.Andunlikeideology,collectiveidentity carrieswithitpositivefeelingsfor
othermembers ofthegroup.
Thereis another problemcommoninrecentanalyses.In relyingon collective
identity to fillthegaps in structuralist, rational-actor, and state-centered models,
thatis, to explaintheprocessesthosemodelsmiss,scholarshavesometimes ne-
glectedtherolecollectiveidentity playsintheprocessesthosemodelsforeground.
Theyhaveturned identityintoa kindofresidualcategory, describing whathappens
outsidestructures, outsidethestate,outsiderationalaction.We arenotarguing for
simplyreversing thecausalarrow, so thatcollectiveidentities as cultural construc-
tionsdetermine interests,relations, and structures [a claimmadeby somepost-
structuralist analyses,forexample,byLaclau & Mouffe(1985) andJoyce(1994)].
The bestrecentresearch,we believe,avoidsa prioriassumptions aboutcausal
mechanisms and allowsfora numberof different relationships betweencultural
anddiscursive practices ontheonehand,andlegal,political,economic,andsocial
structures ontheother. Theanalytical challengeis toidentify thecircumstances in
whichdifferent relationsbetweeninterest andidentity, strategy and identity,and
politicsandidentity circumstances
operate, thatincludecultural processesas well
as structural ones.
In thefollowing, we examinetheroleofidentity in fourphasesofprotest: the
creationof collectiveclaims,recruitment intomovements, strategic and tactical
decisionmaking,and movement outcomes.Because thescholarship thatbears
on collectiveidentity and social movements is huge and spansnumerousdis-
ciplines(sociology,politicalscience,psychology, law, anthropology, women's
studies,queertheory, andothers),we concentrate on sociologicaltreatments ex-
ceptwhereworkin otherdisciplines promisesto fillgapsin sociologicalmodels.
286 POLLETTA E JASPER

We omitdiscussionsof class andnationalidentities
(fortreatments of these,see
Calhoun1993,Cerulo1997,Hanagan1994,Krinsky 1999);as wellas discussions
oftherelationship
betweenpersonalandcollectiveidentities (see Goffman 1959,
Holstein& Miller1990,McCall & Simmons1978,Jenkins 1996; forreviewsof
identity
andsocialmovements generally,
see Johnstonetal 1994,Huntetal 1994,
Krinsky1999,Snow & McAdam2000, Snow 2001). We tryto addressa range
ofmovement dynamicsin whichidentity mayoperaterather thancoveringevery
importantworkin thefield.

MOVEMENTEMERGENCE:IDENTITY AND INTEREST

Whydo movements emergewhentheydo? In the1970s,resourcemobilization
andpoliticalprocesstheorists stoppedaskingwhypeoplefeltfrustrated enough
toengageincollectiveprotest ratherthanorganizethrough conventional political
channels,andinsteadaskedwhenandhowtheysecuredtheresourcesto combat
theirexclusionfromthosechannels.Collectiveactors'"interests" wereimpliedby
theveryformulation; theylayin gainingaccesstothestablestructure ofpolitical
bargaining (Gamson1975,Tilly1978,McAdam1982).Collectiveinterests were
takentobe longstanding: Themodelpresumed an already-existingcollectiveactor
abletorecognizetheopeningofpoliticalopportunities andtomobilizeindigenous
resources forpoliticalpurposes.
Thatpresumption was challengedbyscholarsofthe"newsocialmovements,"
theprotestsaroundpeace, nuclearenergy, local autonomy, homosexuality, and
feminism thatseemedtobe displacing class-basedpoliticalmobilization inWest-
ernEuropein the1970s and 1980s (Touraine1981, 1985,Melucci 1985, 1989,
Offe1985,Castells1997,Laclau & Mouffe1985,Cohen1985,as wellas Larafta
et al's 1994 overview).New socialmovement theoristsarguedthatparticipation
insuchmovements couldnotbe predicted byclasslocation.Norwereparticipants
seekingto gainpoliticaland economicconcessionsfrominstitutional actors,to
further their"interests" in conventional terms.Rathertheysoughtrecognition for
newidentities andlifestyles.
Newsocialmovement theoristssawa profoundly changedsocialformation be-
hindthesenovelformsof collectiveaction,variouslydubbed"postindustrial,"
"programmed" (Touraine1981), "information" (Melucci 1996), or "network"
(Castells1997)society. InMelucci's(1996) account, "modernization" hasrequired
thatpeoplebecapableofprocessing theinformational resourcesonwhichsocieties
nowdepend,butexpandedindividual autonomy has beenaccompanied bystrong
pressurestowardnormative conformity. Social controlhas come to operatesi-
multaneously through self-regulation and through theincreasing penetration of
standards ofinstrumental rationalityintopeople'sbiologicalandemotionallives.
As a result,protestors havebeenless likelyto seek a redistribution of political
powerthanto seekto changedominant normative and culturalcodes bygaining
recognition fornewidentities (see also Pizzorno1978).
New social movement theoriesprovedbetterat raisingquestionsaboutthe
sourcesof movement identities thanat answering them.Theirexplanations for
COLLECTIVEIDENTITY,SOCIALMOVEMENTS 287

how shiftsin materialproduction haveaffected social movements werenoten-
tirelyclearand sometimesriskedtautology, withnew social movements taken
as bothevidenceandconsequenceofa newsocialformation (see Touraine1981
and Cohen's1985critique).Empirically, moreover, mostnewsocialmovements
havecombined politicalgoalswithmoreculturally oriented efforts. However, new
socialmovement theorists' arguments wereprovocative, and theydid encourage
sociologiststo highlight processesin older,class-basedso-
identity-construction
cialmovements-if onlytorebuttheclaimednovelty oftheirsuccessors(Calhoun
1995,Plotke1990,Buechler1990). Sociologistssympathetic topoliticalprocess
approachesalso beganto use collectiveidentity to explain"how structural in-
equalitygetstranslated intosubjective discontent" (Taylor& Whittier 1992:104;
see also Morris1992,Mueller1992).
Otherscholarshave agreedwithnew social movement theorists' claim that
effortsto define,celebrate,enact,and deconstruct identity are moreimportant
in recentmovements thantheyhave beenin thepast,buthave soughtdifferent
explanations forthatfact.Jasper(1997),forexample,pointstolegalinclusionas a
keydistinction.Unlikethecivilrightsandearlylabormovements, whichpursued
fullinclusionas citizens,post-citizenshipmovements arepeopledby thosewho
alreadyenjoymostor all ofthenormalrightsofcitizens,including theabilityto
mobilizelegallyandto putpressureon politicaldecisionmakers.Participants in
thesemovements do notusuallyhavean identity imposedon thembythepolitical
and legal systems;accordingly, theyhave morefreedomto engagein creative
reformulations ofwhotheyare.
New social movement theorists'
determination to historicize a contemporary
repertoireofprotesthasalsoencouraged toaccountinmacrohistorical
efforts terms
of
fortheconstruction contentious identities.
Some authors have looked to large-
scaleprocessessuchas industrialization, urbanization, andstateconsolidation, as
well as to theascendanceofnewcognitive to
paradigms, explain how particular
identitiesbecomethebasisforexclusionand/or discrimination butalso formobi-
lization.Forexample,D'Emilio (1983) tracestheemergence of a "homosexual"
to the
identity processes of urbanization
and industrialization that madepossiblean
autonomous personallife. Same-sex sexhas always existed and,indeed,has often
beenseverely punished,D'Emilio pointsout,butitwas only thebeginning
at ofthis
century thatitbecame not a
just deviant,immoral, illegal act but a deviant identity.
A homosexualwas a personwhosenature-acts,feelings, personaltraits,even
bodytype-was sharply distinguishable from"normal"heterosexuals. Thatshift,
aidedby a newpsychiatric modelof homosexuality, madepossiblebothheight-
enedrepression (one couldnowbe firedorprosecuted as a homosexualwhether
ornotonehadengagedin sex),andthecreationofa homosexualcollectiveactor.
In a comparabledynamic,thelegal institutionalization of racialprivilegein
theUnitedStatesandSouthAfricagenerated severeandpervasiveinequality and
eventually providedthebasis fordemandsby blacksforlegalequality.By con-
trast,theabsenceof legalizedracialcategories, agencies,and statistics in Brazil
impededblackmobilization (Marx1998).Tilly(1998) attributes theriseofidentity
politicsin nineteenthcentury Britain-theeclipseoflocal identities likespinner,
288 POLLETTA E JASPER

neighbor, ortenant ofa particular landlordbybroaderidentities suchas "citizen"
and"worker"-totheincreasedsalienceofthenationalstateinpeople'slivesand
thenewpatterns of claimsmaking thatresulted.Ratherthanappealingto a pow-
erfulpatronorunleashing theirragedirectly on theobjectoftheirdissatisfaction,
claimsmakers made
increasingly publicdemonstrations oftheirnumbers andcom-
mitment tobidforparticipation in a nationalpolity."Theydeclared'We existand
havea rightto exist.We have strength, coherence,and determination. National
politicsmusttakeus intoaccount"'(1998:14).Together, theseanalyseschallenge
viewsofraceandhomosexuality as transhistorical "natural"identities.
To explainthecreationofmobilizing severalauthors
identities, haveturned to
network analysis.Theyarguethatsuchidentities comenotfromfixedcategories
likerace,class,gender, ornation, butfromcommonpositionsinnetworks, whether
networks ofpatronage (Gould 1998),urbanresidence(Gould 1995),or political
affiliation(Mische1996).Forexample,Gould(1998) arguesthattheleadersofthe
1794Whiskey RebellioninWestern Pennsylvania differed fromothermembers of
thepoliticaleliteinonlyonerespect:Ata timewhenpowerwas shifting tofederal
authorities andtheeasternestablishment, thesepowerbrokers soughttopreserve
theirrelationships withbothwesternclientsand easternpatrons.By leadingan
insurgency, evenonelikelytofail,theycommunicated toclientstheirwillingness
to championtheirinterests, evenwhilesuggesting to easternelitesthattheywere
tryingto staveoffevengreater civilwarbyassumingtheleadership ofa bandof
angryfarmers. "Politicalidentificationsarenotmerely constrained bynetworks of
socialties,inthesensethatnetwork positionhelpstodetermine whichofa variety
ofexogenously availableself-understandings anindividual mightembrace;inthis
instance, at least,therelevant identification was definedin termsofthenetwork"
(38).
Otherauthorshavefocusedinsteadon theinstitutional contexts withinwhich
new identities are forged.Conceptsof "submerged networks" (Melucci 1989,
Mueller1994),"halfwayhouses"(Morris1984),"freespaces" (Evans & Boyte
1986), "havens"(Hirsch1990a), "sequesteredsocial sites" (Scott 1990), and
"abeyancestructures" (Rupp& Taylor1987) describeinstitutions removedfrom
thephysicalandideologicalcontrol ofthoseinpower,forexampletheblackchurch
beforethecivilrightsmovement (Morris1984) andliterary circlesin communist
EasternEurope(Johnston 1998). Suchinstitutions supplythesolidaryincentives
thatencouragemovement participation,buttheyalso represent a "freespace" in
whichpeoplecan developcounterhegemonic ideas and oppositionalidentities.
Whydo such sitesfacilitate thedevelopment of oppositionalidentities? Some
authors suggestthatitis simplytheirdistancefromthephysicalcoercionandide-
ologicalcontrol ofthoseinpower(Hirsch1990a).Otherssuggestthatis rather the
beliefsystemsthatare institutionalized in suchsitesthatare important (Polletta
1999).
The latterdovetailswithperspectives thatgiveculturean independent rolein
constituting thecollective identitiesaroundwhichpeoplemobilize.Inotherwords,
whatemergesfrom"freespaces"maynotmatchup with"objective"categories of
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 289

structurallyderivedinterests. What"worker" or "citizen"or"African American"
means,and whatbehavioralexpectations it entails,are partlya function of the
vocabularies, andimagesavailable.Somers& Gibson(1994: 67) arguethat
stories,
people'sexperiences as workers, forinstance, "wereinextricably interconnected
withthelargermatrix ofrelations thatshapedtheirlives-theirregionallocation,
thepracticalworkings ofthelegalsystem, family patterns-aswellas theparticular
stories(ofhonor, ofethnicity,ofgender, oflocalcommunity, ofgreed,etc.)usedto
accountfortheeventshappening tothem."(See alsoSomers1994,Steinberg 1996.)
Thecategories thatemergefromsuchprocessesmakefordiversemovements. For
example,labels suchas "worker"or the"workingclass" have promotedmore
pugnaciously anti-bourgeois labormovements thanhavelabelslike"citizen"or
"thepeople,"whichencouragecross-classalliances(Sewell 1980).
Ratherthanreadingoffinterests andidentities fromallegedlytransparent struc-
turalpositions,recentworkshave(a) revealedthehistorical construction ofwhat
seem"natural"identities suchas "workingclass," "black,"and "homosexual;"
(b) advancedmoresophisticated modelsofhowthesocial,economic,andpoliti-
inwhichpeopleparticipate
cal relations generate mobilizing identities;
(c) recog-
nizedtheindependent roleofculturein shapingthecollectiveidentities on behalf
ofwhichpeoplemakeclaims;and(d) identified thepoliticalconditions in which
identityclaimsarelikelyto be prominent in movements.
Once movements haveemerged, completewithorganizations, organizers, and
recruitment campaigns, strategicefforts
tocraftmobilizing identitiesbecomeim-
portant. Even identitiesthatare familiar,longstanding, andenforced bylaw and
customfrequently needtobe re-imagined bymovement Attheveryleast,
activists.
theymustbe integrated witha movement identity, i.e. a collectiveidentity based
on sharedmembership in a movement. We nowturnto theseefforts.

RECRUITMENTAND COMMITMENT:
IDENTITY AND INCENTIVE

Whywillpeoplejoin collectiveefforts whentheydon'tknowwhether theirpres-
ence willdo anygood andtheydo knowthattheycan ridefreeon theefforts of
others?Sociologistshavedevoteda greatdeal ofattention to thisquestion,posed
firstby Olson (1965). He arguedthatsharedinterests are simplynotenoughto
motivate individual
effortintheabsenceofselective rewards thatgo onlytopartic-
ipants.Butas Fireman& Gamson(1979) andothershavepointedout,individuals
sharepriorbondswithothersthatmakesolidaristic behaviora reasonableexpec-
tation."A personwhoselifeis intertwined withthegroup[through friendship,
kinship,organizational membership, informal supportnetworks, or sharedrela-
tionswithoutsiders]... has a bigstakeinthegroup'sfate.Whencollectiveaction
is urgent,thepersonis likelyto contributehis or hershareeveniftheimpactof
thatshareis notnoticeable"(22).
Activistsarenottheisolated,atomisticindividuals sociologistsoncetookthem
for.In manymovements rangingfromtheFrenchcommune(Gould 1995) and
290 POLLETTA E JASPER

theRussianrevolution (Bonnell1983)toNichirenShoshuBuddhism(Snowet al
1980) and the 1964 MississippiSummerProject(McAdam & Paulsen 1993),
recruitment has takenplace primarily through preexisting solidarities
(see also
Marwell& Oliver1993,Oberschall1973,Tilly1978). In thesecases,priorties
motivated participationthrough normsof obligationandreciprocity. "Collective
identity,"forsome authors,is shorthand fortheaffective connections one has
to membersof a groupthatoblige one to protestalong withor on behalfof
them.
However,this"loyalty"formulation raisesa numberof questions.First,what
is thecontent ofthoseties?Does collectiveidentity consistmainlyofmoralobli-
gation,altruism, and personalloyalty, or can it be self-interestedconcernwith
theopinionof others?Insteadof affective obligations, Chong(1991:50) argues
thatself-interested
"reputational concerns"motivate participation.
Since"people
expectconsistency fromus, we tendto obligeby forging and livingup to our
reputations.Andas Socratesadvised,theeasiestwayto maintain a reputation is
to becomethepersonyouwantothersto thinkyouare."Participation is a ratio-
nal bid to gainthebenefits thataccrueto thosewho sharea collectiveidentity.
Friedman & McAdam(1992) similarly connectcollectiveidentity toself-oriented
rationalaction.Highlyregarded roleswithincommunities maycometo be linked
withactivismin a waythatmakesparticipation a requirement ofthatrole.In the
earlycivilrightsmovement, activismwaslinkedwith-normatively required of-
churchgoers; in 1960,studentbecamelinkedto activist, becamea "prizedsocial
identity"whichsuppliedtheselectiveincentives toparticipate.Butarguments like
these,designedto showthatculturalmeaningsand emotionsare notlogically
incompatible withrational-actor
models,yieldconvoluted causalpictures:We try
to becomean altruistic personbecauseit is in ourinterest to seemone,yetit is
hardtoseemonewithout actuallybeingone.Whynotsimplyadmittheemotional
satisfactionsofcollectiveidentity(Jasper1997:23-29)?
Teske(1997:121)mediatesbetweentheloyaltyandself-interest models,argu-
ingthatwe errin seeingself-interested and moralactionas opposed.Activism
formanypeopleis a wayto construct a desirableself.Theydecideto participate
"neither primarilyon a quasi-quantitativecalculating of costsand benefits,as in
the rationalchoice approach to politics, nor on altruisticimpulses ... . Rather,
identityconstructionpointsto thequalitative concernsand thedesiresactivists
havethatcertainqualitiesbe instantiated in theiractionsand lives"(see also C.
Taylor1989). Lichterman (1996) makesa similarargument buthistoricizesit.
Lackingtheconnection to unifying ideologicaltraditionsliketherevolutionary
leftorreligiousradicalism,post-i1960s activiststurned insteadto a "personalized
politics"in whichtheindividualselfis thearbiterof moralchoices.This isn't
narcissism,Lichterman insists:a self-orientedpoliticscannurtureratherthancurb
civic engagement. An activistcollectiveidentity, thesemodelsimply,can be a
satisfyingaspectofpersonalidentity.
A secondquestionfortheloyaltymodelis whether collectiveidentities
nec-
essarilyprecedemobilization. Somemovements seemtoattract even
participants
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 291

in theabsenceofprioridentities andnetworks (Jasper& Poulsen1995). "Moral
shocks"produced, forexample,by a photograph of a tortured animalor thedis-
asterat ThreeMile Islandcan mobilizepeoplewho do notknoweach otheror
theorganizers (Jasper1997). Participants maysharedemographic or economic
traits-theytendto be middleclass, say,or are mostlymen-but thesedo not
add up toa perception ofthepreexisting "groupness" ofcollectiveidentity.Their
politicalactivityitselfprovidesthatkindofsolidarity: We arestudent radicals,we
arepeoplewhocareabouttheenvironment, we arecaring,criticalcitizens.These
"movement identities"maycometoservemuchthesamefunction as a preexisting
collectiveidentity [Jasper1997; Klandermans's (1997:95) conceptof "commit-
ment,"Hirsch's(1990b) of "solidarity;" andBuechler's(1990) socialmovement
"community" seemfunctional equivalents ofmovement identity].
Minkoff (1997)
arguesthatmanycollectiveactorssuchas women,theelderly, gaysandlesbians,
andthedisabledhaveinitially lackedtheinstitutional infrastructuresthathaveput
members of othergroupsintoregularday-to-day contact.In theabsenceof such
infrastructures,movement organizations havegenerated thecollectiveidentities
thatthencreatednetwork ties.
Since mobilization does notalwaysrequirepreexisting collectiveidentities,
activists'effortsto strategically "frame"identities arecriticalinrecruiting partic-
ipants."Frames"are theinterpretive packagesthatactivistsdevelopto mobilize
potentialadherents and constituents (Snow et al 1986,Gamson1988,Snow &
Benford1988,Benford1993,Tarrow1998). Whensuccessful,framesmakea
compelling case forthe"injustice"ofthecondition andthelikelyeffectiveness of
collective"agency"in changingthatcondition. Theyalso makeclearthe"identi-
ties"ofthecontenders, distinguishing "us" from"them"anddepicting antagonists
as humandecisionmakersratherthanimpersonalforcessuch as urbanization
(Gamson1988,1992,also Hunt& Benford1994,Huntet al 1994,Klandermans
1997).Organizers oftentrytobuilda movement identity on another,independent
collectiveidentity [whichmaycomefromprioractivism as wellas fromracialand
otherascribedidentities (Jasper1997:ch.8)]. ACT UP,forinstance, soughttocon-
vincelesbiansandgaymenthatprotest aroundAIDS was an essentialexpression
oftheirgayidentity.
Whileorganizers use considerable creativityin inventing newidentities or at-
tachingnewbehavioral requirements to old ones,suchprocessesmayalso occur
independently of organizers'strategic Polletta(1998a,b) foundthatthe
efforts.
storiestoldbystudent protesters in the1960lunchcountersit-inshelpedtoforge
an action-mobilizing collectiveidentity. Students'accountsturned unfamiliar and
potentially disturbing eventsintofamiliarepics of overcoming, withfrightened
students becomingtriumphant heroes(see also Hirsch1990band Fantasia1988
on howcollectiveidentities aredevelopedin andthrough protest).
Beyondrecruitment, identity workis crucialto sustaining solidarityandcom-
mitment. Taylor& Whittier (1992) showhowboundary-setting ritualsandinsti-
tutionsthatseparatechallengersfromthosein powercan strengthen internal
solidarity;theycall thissolidarity "collectiveidentity." But thereare liabilities
292 POLLETTA E JASPER

to strongand exclusiveformulations of identity. Manygroupsare tornbetween
assertinga clearidentityanddeconstructing it,revealingittobe unstable, fluid,and
constructed (J.Gamson1995;see also Epstein1987,Seidman1993,Phelan1989,
Fuss 1989).Wheresomemembers maysee destabilizing a collectiveidentityas
animportant goalinandofitself, withramifications beyondthegroup,othersmay
understandably see itas a threatto groupunityor as confusing to thepublic(we
return tothisissuein ourdiscussionofidentity as a strategy ofsocialprotest).
Ifidentitiesplaya criticalroleinmobilizing andsustaining they
participation,
also helpexplainpeople'sexodusfroma movement. One of thechiefcauses of
movement declineis thatcollectiveidentity stopsliningupwiththemovement. We
stopbelieving thatthemovement "represents" us (thetermsuggestsan expressive
dimension as wellas a strategicone).In somecases,cross-cutting identities
come
to thefore,justas thewomen'smovement came to griefpartlyon theshoalsof
class andrace [Echols 1989; Robnett(1997) shows,however, thatcross-cutting
identitiesare notinherently contradictory]. In othercases, people beginto see
theiridentitiesas sufficientlyrepresented in conventional politicalornonpolitical
arenas.
In sum,anysocialmovement groupmustcontinually manageitscollective iden-
tities,and evenidentities predating movements are subjectedto reconstruction.
Organizers oftenconcentrate on recasting constituents' identities to includepar-
ticipationas oneoftheresponsibilities orbenefits ofgroupmembership. Identities
needtobe integrated withinjusticeandagencyframesso as toclearlydistinguish
"us"fromopponents andbystanders. Finally,sustaining participants' commitment
overtimerequiresritualized reassertionsofcollectiveidentity andefforts toman-
age, without suppressing, difference.But in addition to identity management as
an internally directedtactic,movement leadersuse identity in a number oftactics
oriented towardtheworldoutsidethemovement, as we nowexamine.

TACTICALCHOICE: IDENTITY AND STRATEGY
How do activistschoosefromamongthestrategies, tactics,
targets,organizational
and
forms, deliberative stylesavailableto them?Earlyresourcemobilization and
politicalprocessaccounts tended to relyon a classicallyrationalmodel of deci-
sionmakingto answerthatquestion:activistsadaptstrategies to environmental
constraintsandopportunities on thebasisofa cost-benefit calculus(Barkan1979,
Kitschelt1986,McAdamet al 1988). Criticshavepointedout,however, thatac-
tivistsalso choose optionsthatconformto "who we are,"as pacifists, say,or
women, or revolutionaries.
Making decisionson thebasis of collective
identity
has beentreated accordingly as an alternativeto relyingon instrumentalcriteria;
an expressiveratherthana strategic
it reflects logic. Thereare twootherways
of relatingstrategy however.Ratherthanviewingit as at odds with
to identity,
strategy,makingidentity claimscanbe seenas a protest Andrather
strategy. than
viewingan instrumental logic operating exclusiveof identity concerns,we can
see thatinstrumentalcalculationoftendependson thecollectiveidentities thatare
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 293

widelyassociatedwithparticular strategies,
tactics, organizational forms, andeven
deliberative logics.We takeup thesethreeapproachesin turn.
Collectiveidentities can supplycriteria formakingdecisionsthatcompetewith
instrumentally rationalones.Forexample,members oftheantinuclear Clamshell
Alliancesawthemselves bothas an "opponent" oftheatomic-industrial establish-
ment, dedicatedtostopping nuclearpower,andits"opposite," seekingtoeradicate
domination withintheirownoperation(Downey1986,Epstein1991).The latter
identity wasresponsible fortheClamshell'sdistinctive ofconsensusde-
strategies
cisionmakingandnonviolent civildisobedience, as wellas fortheorganization's
eventualdemiseas theincreasingly heterogeneous groupwas paralyzedby the
requirement ofconsensus.Butto suspendtheconsensusrequirement wouldhave
been,formany,to destroythegroup'sidentity. For the"Green"environmental
activiststhatLichterman (1996) studied,sustaining theorganizations thatmade
up the movement was not ofparamount concern. They would rather see an orga-
nizationcollapsethancompromise theiroverriding commitments to democratic
process.None of theseactivistsabjureconsiderations of instrumental efficacy;
theyseekrather tobalancethemwiththeprincipled commitments thatdefinewho
theyare. Strategicchoicesare notsimplyneutraldecisionsaboutwhatwill be
mosteffective, inthisview;theyarestatements aboutidentity (see also Kleinman
1996).
People developa "taste"forcertaintactics,partlyindependently of theiref-
ficacyin attaining formalexternalgoals (Jasper1997). Some mayenjoystaying
within theboundsoflegality, othersstepping outsidethem.Somemaypridethem-
selves on theirmoderatedemandsand tactics,otherson beingavant-garde or
radical.Theymaydevelopcollectiveidentities basedon thosetacticaltastes.Tac-
ticalandorganizational identitiesoftencoincide,as organizations embodyforms
ofaction."Organizational forms maybe a sourceofsharedidentity," saysClemens
(1997:50)."Theanswerto'whoarewe?' neednotbe a qualityornoun;'we arepeo-
plewhodo thesesortsofthings inthisparticular way'canbe equallycompelling."
Important to understanding tacticalchoicewithinmovements is theoperation of
numerous withvarying
identities, salience.Activists mayidentify primarily witha
movement organization,
affinitygroup,styleofprotest, ordegreeofmoderation or
radicalism. Jasper(1997),forinstance, distinguishes among"activist," "organiza-
tional,"and"tactical"identities. Thefirstinvolvesa history ofpoliticalactivity that
is usuallybroaderthana specificmovement. An organizational identity involves
loyaltyto a singleorganization anditsfellowmembers, evensomething as small
as an affinity group.Thosewithtacticalidentities maydefinethemselves as on the
cutting edge,or theymaybe proudofparticular stylesof actionsuchas nonvio-
lenceor civildisobedience.Such identities mayexistalongsidebothmovement
identities andpreexisting collectiveidentities,interweaving withthemincomplex
ways.An individual mightidentify herselfas a nonviolent feminist, ecologicalac-
andmember
tivist, oftheaffinity group"Matrix,"eachoftheselabelscarrying an
identification withsomebroadercollectivity. In a similarscheme,Gamson(1991)
distinguishes betweensolidary, movement, andorganizational identities.
294 POLLETTA * JASPER

Tacticaltastesmayoriginate incollective identitiesthatexistoutsideandpriorto
themovement (Ennis1987)orwithin it.Whittier (1995) describes "micro-cohorts"
inthewomen'smovement ofColumbus, Ohio:groupswhoentered radicalfeminist
organizations together everyyear or two. Their experiences before they entered the
movement and within it provided a collective identityand frame of reference for
theirunderstandings offeminism andpolitics,anda basisfortheirstrategic pref-
erences.OtheraccountsoftheClamshellAlliancehaveattributed itsdeclinenot
so muchtothe"egalitarian/instrumental" tensionthatDowneydescribesas to an
old guard/new guardconflict whereby newermembers oftheorganization bidfor
statuswithinthegroupbychallenging veterans'commitment to fullydemocratic
practices(Cohen1988). In thiscase, as in Ross's (1983) description ofbattlesin
Studentsfora DemocraticSocietyoverorganizational structure, tacticalprefer-
encesand thecollectiveidentities theyexpressedoriginated notin newcomers'
priorexperiencesbutin and through a sharedexperienceof marginality in the
organization.
In a secondapproach,sociologists haveshownhowactivists construct, decon-
struct,celebrate, and enactcollectiveidentities as strategies ofprotest. For East
Germanchallengers totheHoneckerregimein 1989,callingthemselves "thepeo-
ple" notonlyinspiredgreater participation thaniftheyhadusedsomeotherlabel
butprevented a regimethatalso associateditselfwith"thepeople"fromattacking
themas outsiders. Itmayalso havediscouraged policerepression (Pfaff1996).In
thiscase,insurgents' publicconstruction oftheiridentity limitedtheactionsthat
theiropponents couldtake.
Activists maydefinetheiridentities indifferent waysdepending onthestrategic
situation.If theyarerepresenting theirgroupto a publicaudience,theymaycast
themselves as moreunified andmorehomogeneous thantheywouldina setting of
fellowactivists. Pulido(1996) foundthatnonwhite environmental-justice activists
routinely invoked"peopleofcolor"as a primary identitywhentargeting thestate
or a polluterbutnarrower racialandethnicidentifications in theirinternal move-
mentdeliberations. Theiridentities, she concludes,were"situational" (see also
Lichterman 1999). Another factorin determining howa groupstrategically con-
structs itsidentity maybe thekindof oppositionit confronts. Bernstein (1997)
foundthatgay and lesbianactivistscampaigning forantidiscrimination statutes
deployedstrategies of "identityforcritique," in whichtheycastigated thehomo-
phobicpractices ofmainstream society, whentheyfacedorganized opposition and
whentheywereled byexclusiveorganizations uninterested in coalition-building.
By contrast, movements withstrongorganizational infrastructures or access to
politicaldecisionmakerstendedto seekpolicychangeandemphasizetheirsimi-
laritiesto themajority, usingless controversial strategiesof "identity foreduca-
tion,"in whichtheysuppressed rather thancelebrated theirdifferences fromthe
mainstream.
As Bernstein'sworkindicates,movement leadersmuststrategize not only
againstsingleopponents, butwithina "multiorganizational field"ofallied,com-
peting,andoppositional movement organizations, authorities, media,andfunders
(Huntet al 1994,Gamson1988,Klandermans1997). Whatis therelationship
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 295

betweenthisorganizational contextand the identitiesthatoperatewithinit?
Carroll& Ratner(1996) arguethatcertainbroadidentities, forexamplethatof
a victimof materially grounded injustice,are able to linkdiverseorganizations.
Does itworktheotherway?Areorganizationally diversemovements better ableto
developencompassing identities?Ferree& Roth's(1998) studyofa failedstrike
byGermanday-careworkers showsthattheorganizational insularity ofpotential
coalitionpartners suchas unions,womenin thegoverning legislativecoalition,
andgrassroots feminists led themto see daycareworkers as "difficult anddiffer-
ent"(643) fromtheirusualconstituencies ratherthanas offering an opportunity
to developnewallies.The predominance of "exclusionary" identities discourag-
inga coalition,in turn, resultedfromthelackoforganizational linkagesto other
movements (see also Gordon& Jasper1996).The samekindofdynamiccan op-
eratewithin a movement group.Roth(1998) foundthattheexistenceofa feminist
caucuswithin anACT UP groupeffectively "compartmentalized" women'sissues
to thecaucusbecauseno one else woulddeal withthem.Tarrow(1998:ch.7) ar-
guesthatmovements attheendofprotest cycles(presumably, anymovement with
dwindling appeal)oftencompensate fortheirlackofmembership, allies,andbroad
appealbydefining theiridentitiesnarrowly andrejecting alliancesas "sellingout"
(see also Gitlin1995). Suchexclusiveness can helpto sustainthecommitment of
theremaining stalwarts.
How successfully groupsframetheiridentities forthepublicthusaffects their
abilityto recruit membersand supporters, gaina publichearing, makealliances
withothergroups,anddefuseopposition. The studieswe havecitedindicatethat
how a groupframesits identity (exclusiveor inclusive,involuntary or chosen,
challenging or conventional) dependson thesetting andtheaudienceto whichit
is speaking, thekindofopposition itconfronts,andtheorganizational linkagesit
has toothergroupsandmovements.
A thirdapproachto therelationsbetweenidentity and strategic choicebreaks
witha view of activiststrying to juggle strategicimperatives and identity con-
cernsbypointing tothewaysthatidentity informs eventhemostself-consciously
strategiccalculation. Collectiveidentities arealreadyembeddedin strategies, tac-
tics,claims,organizational forms, anddeliberative styles,andtheyinfluence how
suchoptionscan be used."Embeddedin" can meandifferent things, though.For
example,activists seekinglegalchangeon behalfofwomenandminorities often
struggle todecidewhether toplayup ordownthedifferences on whichtheirdis-
advantages rest.Discrimination casesbrought bywomenhavebeenlimitedbythe
implicitlymalestandard towhichtheymustanalogizetheirownsituation. "Differ-
ence,"whether itis thebiologicalcapacitytogetpregnant (Scott1988)ora dislike
forhigh-pressure salesjobs (Scott1988,Milkman1986),is seen as "deviance,"
andactivists mustdecidebetweentheequallyunacceptable alternatives oftrying
tobe "like"menortojustify"special"treatment withitsimplications ofinferior-
ity.Groups'strategic effortsarethusconstrained notonlybytheirownperceived
identitiesbutbythedefinitions contained innotveryobjectivebutlegallyenforced
definitionsofequality:male,white,able-bodied, heterosexual, andso on (Minow
1990,Crenshaw1990).
296 POLLETTA * JASPER

Clemens(1997) depictsa less formalprocess.Certainorganizational forms
havebeenwidelyseen as "appropriate forwomen"or "middle-class," in a way
thatinfluenceswhomaylegitimately usethem.Morebroadly, ourveryconceptions
of whatis instrumental,strategic,
efficacious, andpoliticalreston theidentities
withwhichtheyareassociated.Forexample,Bordt(1997) showshowcollectivist
stylesoforganizationcametobe seeninthe1970sas feminist ina waythatmade
theiradoptionby new feminist groupsa matterof commonsense.Earlier,the
sameformshad cometo be seenas whitein thesouthern civilrightsmovement
and,forthatreason,hadbecomeincreasingly unappealing toAfricanAmericans
(Polletta1997).Thislineofinquirymesheswithrecentneo-institutionalist theo-
rizingon organizations'
propensitytomimicorganizational formsthatarewidely
seenas cuttingedge(DiMaggio& Powell1991).Thequestionthatneoinstitution-
alistshavenotadequatelyanswered, to strategic
also relevant innovationin social
movements, is whethersuchimitation benefitstheorganization or
strategically
whether theinnovationis assumedtobe strategicintheabsenceofanycompelling
evidence.
In sum,recentidentity argumentsrejectthecommonplace oppositionbetween
identityas expressiveand strategyas instrumentalin orderto demonstrate that
deployidentities
activists andthatstrategic
strategically optionshavemeaningby
referencetothegroupswithwhichtheyareidentified.

MOVEMENTSUCCESS: IDENTITY AS OUTCOME
How successfularemovements? Andhowdo theyaffect individuals,groups,and
broaderstructures? In accounting formovement outcomes,theorists havetended
to treatidentity undertheheadingof culturalimpactsratherthaninstitutional
ones. Yet thereare manykindsof movement impacts-institutional and extra-
institutional-in whichidentity playsa role.In somecases,theimpactis intended,
in others,a byproduct ofotheraims.
First,
changing identities
is oftena primarymovement goal.Thismaybeclearest
in religiousor self-helpmovements, butmanymovements have it as one goal
alongside others.The of
development grouppride is a form of identity work.
Identitytalkwithinmovements maybe aimednotonlyat buildingsolidarity but
also atchanging selvesandrelationships inwaysthatextendbeyondthemovement
(Lichterman 1999,Breines1989,Epstein1991).
Second,participation usuallytransforms activists'subsequentbiographies,
marking theirpersonalidentitiesevenafter themovement ends,whether ornotthis
is an explicitgoal (McAdam1988,Fendrich1993,Rogers1993,Andrews1991,
Whalen& Flacks1989,Taylor& Raeburn1995,Whittier 1995).Thisis notonly
trueofpeoplewhoseactiveparticipation was of longduration or highintensity,
butalso ofmanycasualparticipants. Mansbridge (1995),forinstance, arguesthat
beinga feminist doesnotrequiremembership ina feminist organization, butonlya
senseofaccountability toan idealoffeminism. Itsbehavioral requirements differ
acrosssocialandhistorical contexts, butthecorecollectiveidentity continues to
shapean individual's senseofself.
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 297

Outsideofpublicinstitutions, identity workwithinsmallcirclesoflike-minded
people is criticalto sustaining "abeyancestructures" duringperiodsof limited
politicalopportunities (Taylor1989,Whittier 1995). Identitiesnurtured within
thesenetworks contribute to thespillovereffectfromone movement to another
(McAdam1994,Meyer& Whittier 1994,Tracy1996). Broadidentities suchas
radicalpacifistor anarchist can also be preservedin popularculturalmaterials
rather thanorganizations, thusbecomingavailableforsubsequent wavesofprotest
(Eyerman& Jamison1998).
Of specialinterest becauseitchallengesthetendency to separateidentity and
powerorientations (Rucht1992), demanding recognition fora new or changed
identity canbothsecureconcessions andpermanently changetheterrain ofpolitical
conflict. Putativelyblack,orwomen'sorGreeninterests nowhavetobe reckoned
withbypolicymakers (Mueller1987,Costain1988). The formation of women's
caucuses,centers, programs, andsupport groupswithin mainstream legal,medical,
economic,religious, andmilitary institutions hasbeenanenduring outcomeofthe
women'smovement (Katzenstein1998). Scientistsin the 1960s who struggled
to squaretheiridentities as activistsand as nonpartisan truthseekersfounded
publicscienceorganizations liketheUnionforConcerned ScientistsandtheCenter
forSciencein thePublicInterest thatcontinuetoday(Moore 1996). Collective
identities developedwithinmovements mayhavelastingimpacton institutional
politicalarenasandorganizational forms.
In another kindofimpact,a movement's associationin theeyesof thepublic
witha particular strategy, tactic,organizational form,or stylecan influence sub-
sequentuses ofit.Whenpro-life activistssing"We ShallOvercome,"or sit-inat
abortion clinics,theybenefit fromthepopularidentification ofthosetacticswith
thecivilrights movement (Eyerman & Jamison 1998).No progressive grouptoday
wouldappropriate thegoosesteportheswastika.A feminist groupthatadopteda
bureaucratic styleoforganization wouldbe interpreted as signalingitsdeparture
from1970sfeminism-perhaps initsideologicalcommitments as wellas itsorga-
nizationalform(Bordt1997).Symbolsandstrategies resonatewiththeidentities
ofpriorusers.
Finally,thecreation ofa strong movement identityusuallyleadstoa backlash,
as thoseportrayed as theenemymaybe angeredorfrightened intocounterorganiza-
tion.Sometimes thecountermobilization outstrips theoriginalprotest movement.
Forinstance,severalyearsofpublicity andvictoriesby theanimalrightsmove-
mentpushedthebiomedicalcommunity intoforging a new (and veryeffective)
publicidentity foritself,emphasizing aid to sick individuals,especiallychildren,
ratherthantheabstractions of scientific progress(Jasper& Poulsen1993). The
Americannuclearpowerindustry, too,beganfighting backonceitrealizeditwas
underattackbya nationalmovement (Jasper1990:ch7).
Ratherthanviewingcollectiveidentity exclusivelyas a kindofcultural move-
mentimpact,separatedfromthe domainof institutional impacts like legal
reform andpolicychange,theseanalysespointtothewaysinwhichnewlypromi-
nentor reformulated identities can transform theinstitutionalpoliticalplaying
field.
298 POLLETTAu JASPER

CONCLUSION
Whatis collectiveidentity? How do collectiveidentities matter to social move-
ments?And whatdon'twe knowyet?We concludewithone morecutat these
questions.
Collectiveidentity describesimaginedas well as concretecommunities, in-
volvesanactofperception andconstruction as wellas thediscovery ofpreexisting
bonds,interests, andboundaries. It is fluidandrelational, emerging outof inter-
actionswitha number ofdifferent audiences(bystanders, allies,opponents, news
media,stateauthorities), rather thanfixed.Itchannelswordsandactions,enabling
someclaimsanddeedsbutdelegitimating others.Itprovidescategories bywhich
individuals divideup andmakesenseofthesocialworld.
Whatis notcollectiveidentity? Collectiveidentities arein constant interplay
withpersonalidentities, buttheyare neversimplytheaggregateof individuals'
identities.Ifcollectiveidentity describeswhatmakespeopleoccupying a category
similar,personalidentity is thebundleoftraits thatwebelievemakeus unique.Nor
is collectiveidentitycoextensive withculture;therearemanyculturalmeanings
thatdo notimplyimagesofboundedgroups.Collectiveidentity is notthesameas
commonideologicalcommitment. One canjoin a movement becauseone shares
itsgoals without identifying muchwithfellowmembers (one can even,in some
cases,despisethem).Likewise,peoplecandevelopcollectiveidentity onthebasis
of theirdistinctive know-how or skills,butsuchknow-how and skillscan have
influence evenintheabsenceofcollectiveidentities aroundthem.Thoseskilledin
explosivesmayfavorbombingas a protest tactic,butthisdoesnotnecessarily give
thema sharedcollectiveidentity. Movements contain,symbolize, andritualizeall
kindsofpeopleandattributes; onlysomeofthemarecollectiveactors.Collective
identitiesareone particular formofculture, although theymaybe builton other
forms.
How does collectiveidentity matter to socialmovements? Payingattention to
thecauses and consequencesof collectiveidentity can move us beyondsome
theoretical impasses.The proliferation of workon thetopicsuggeststhatmany
sociologistsrealizeas much.But too oftencollectiveidentity has beeninvoked
simplyto fillgapsleftby structuralist, state-centered,or rationalchoicemodels,
in theprocessreproducing theverydichotomies theconceptis supposedto chal-
lenge.Specifically, we shouldnotassumethatidentity is theoppositeof interest
(withidentity-oriented movements opposedto interest-based ones),thatit is the
oppositeofincentives (withself-regarding actioncontrasted to altruistic
action),
thatitis theoppositeof strategy (withexpressive criteria forchoosingstrategies
contrasted withinstrumental ones),orthatitis theoppositeofpolitics(withmove-
mentimpactson individual selvescontrasted withthoseon institutional politics).
Instead,theworkwe havehighlighted hereshowsthatstructural interestsareoften
recentin origin;thatwe mayengagein moralprotest to developthekindof self
we want;thatwhatis considereda good strategy is oftenbased on whatgroups
it is symbolically associatedwith;and thatmovements promotenew identities
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 299

as a wayto gainpoweras well as transform selves.The mostinteresting recent
workon identity has inquiredintothemacrostructural processesby whichnew
collectiveidentitiesdevelopandintothemicro-interactional processesbywhich
peoplecometosee themselves as obligedtoprotest. Ithasemphasized organizers'
capacitytoredefine old identities andcreatenewones,andthepressures on them
todo so.
Thatsaid,thereis stilla lotthatwe do notknowaboutcollectiveidentities. We
havelittleevidenceabouthowindividuals sortoutandcombinedifferent sources
of identity,or aboutthepsychological mechanisms behindcollectiveidentities.
People have a rangeof groups,roles,and positionsavailableto them,and we
knowlittleabouthowthey juggleandchooseamongthem;therelationship between
personalandcollectiveidentities is a stapleof socialpsychology thatstudents of
social movements haveyetto incorporate (Tajfel1981,Stryker 1980,Burke&
Reitzes1991).In addition, we knowlittleabouttheemotions thataccompany and
shapecollective identity.Collectiveidentity is notsimply thedrawing ofa cognitive
boundary; itsimultaneously involvesa positiveaffect towardothergroupmembers
(Jasper1998).
Takenas a whole,theliterature on collectiveidentity stillleavesfuzzythere-
lationsbetweenidentity and an individual'scalculusof self-interest. Is identity
or interestthebedrockof individualchoice?This questionunderpins severalof
thebroaderissueswe haveaddressed,and scholarshaveanswereditbothways.
For some,individuals chooseidentities thatwill maximizetheirpreferences. In
Gould's(1998) account,forexample,someWestern Pennsylvanian elitebrokers
choseto identify witha cause thatwouldgainthemallies whether theywon or
lost.ForChong(1991),actingsolidaristically is a waytoimprove one'sreputation
andthebenefits thatflowfromit.Forotherauthors, identities settheveryterms
of individual and strategic calculation.Pizzorno(1986) arguesthatthecategory
ofinterest is meaningless without thatofidentity, in otherwords,without recog-
nitionof theselfdoingtherationalchoosing."Circlesof recognition" notonly
validateactionson behalfof alreadyestablished interests,buthelpto constitute
new identities and theinterests thatflowfromthem(see also Emirbayer 1997,
Calhoun1991).In a sense,thedebatecanbe seenas a kindofsociologicalchicken
and egg questionakinto whether individualor societycomesfirst. However,an
alternativetackaskswhether interest or identity is moresalientin different con-
texts.Alongtheselines,Ringmar(1996) arguesthatactionsdrivenby identity
ratherthancalculations ofinterest areespeciallylikelywhenpolitical,economic,
orsocialchangehasdestabilized prioridentities. Duringsuchformative moments,
one acts-and "one" can be nationsas wellas persons-inordertoreassert who
one is.
buildingblocksthatareusedtoconstruct
We stillknowlittleaboutthecultural
Laws andpoliticalstatushavebeenstudiedas a source,but
collectiveidentities.
we shouldlearnmoreabouthowintellectuals andgroupleadersuse nostalgiaand
otherelements ofcollective
memory a pastfora group.Whatareother
toconstruct
toolsand rawmaterialsof identitywork?How important is place forexample?
300 POLLETTA * JASPER

Whataboutbodilydifferences andbodilyneeds?To whatextent aremetaphors and
imagescreatedoriginallythrough nationalism
centraltoothercollective
identities?
moreattention
Finally, tohistorical
andnon-Western movements wouldexpose
us to different
understandings of therelationshipbetweenselfand other,and to
differentdynamicsofcollectiveidentity
formationandcontestation.Notleast,they
shouldhelpus movebeyondsimplyasserting theconstructednessofidentitiesby
showingthevariety offormsthatidentities takeandtheverydifferent behaviors
theyrequire.Like theothergapswe havenoted,thisone shouldspurus to better
specifyourconceptsandquestions, andtobegintesting competing answers.

ACKNOWLEDGMENTS
ThankstoMaryBernstein,JohnKrinsky,
KellyMoore,VertaTaylor,andCharles
Tillyforcomments
on earlierdrafts.

VisittheAnnualReviewshomepageat www.AnnualReviews.org

LITERATURECITED
AndrewsM. 1991.Lifetimes of Commitment: UnitedStates.New Brunswick, NJ:Rutgers
Aging, Politics, Psychology.New York: Univ.Press
Cambridge Univ.Press BurkeP, ReitzesD. 1991. An identity theory
BarkanS. 1979.Strategic, tacticaland organi- approachto commitment. Soc. Psychol.Q.
zationaldilemmasin theprotestmovement 54:239-51
againstnuclearpower.Soc. Probl. 27:19- CalhounC. 1991. Indirectrelationships and
37 imaginedcommunities: large-scalesocialin-
BenfordRD. 1993. 'You could be the hun- tegrationandthetransformation ofeveryday
dredth monkey':collectiveactionframes and life.In Social Theoryfora ChangingSoci-
vocabulariesof motivewithinthe nuclear ety,ed. P Bourdieu,JColeman,pp. 95-121.
disarmament movement. Sociol.Q. 34:195- New York:RussellSage Found.
216 CalhounC. 1993. Nationalismand identity.
BernsteinM. 1997. Celebration and suppres- Annu.Rev.Sociol. 19:211-39
sion: the strategicuses of identity by the CalhounC. 1995. 'New social movements' of
lesbianand gay movement. Am.J. Sociol. theearlynineteenth century.In Repertoires
103:531-65 and CyclesofCollectiveAction,ed. M Trau-
BonnellV. 1983.RootsofRebellion:Workers' gott,pp. 173-215.Durham,NC: DukeUniv.
Politicsand Organization in St. Petersburg Press
and Moscow,1900-1914. Berkeley:Univ. CarrollWK, RatnerRS. 1996.Masterframing
Calif.Press andcross-movement networking in contem-
BordtRL. 1997.How alternative ideasbecome porary socialmovements. Sociol.Q. 37:601-
institutions:
thecase offeminist collectives. 25
NonprofitVol.Sect.Q. 26:132-55 CastellsM. 1997. ThePowerofIdentity. Ox-
BreinesW. 1989.Community andOrganization ford,UK: Blackwell
intheNewLeft1962-68: TheGreatRefusal. CeruloKA. 1997. Identityconstruction: new
New Brunswick, NJ:RutgersUniv.Press issues, new directions.Annu.Rev. Sociol.
BuechlerS. 1990. Women'sMovements in the 23:385-409
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 301

Chong D. 1991. CollectiveActionand the Social Movements. Cambridge,UK: Cam-
CivilRightsMovement. Chicago,IL: Univ. bridgeUniv.Press
ChicagoPress FantasiaR.1988.Cultures ofSolidarity.
Berke-
ClemensE. 1997.ThePeople'sLobby.Chicago, ley:Univ.Calif.Press
IL: Univ.ChicagoPress Fendrich JM.1993.Ideal Citizens.TheLegacy
CohenE. 1988.Ideology,InterestGroupFor- oftheCivilRightsMovement. Albany:State
mation,and theNew Left.New York:Gar- Univ.New YorkPress
land FerreeMM, RothS. 1998.Gender,class,and
CohenJL. 1985.Strategy or identity:newthe- theinteraction betweensocialmovements: a
oreticalparadigmsand contemporary social strikeofWestBerlindaycareworkers. Gen-
movements. Soc. Res. 52:663-716 derSoc. 12:626-48
CostainA. 1988. Representing women:the FiremanB, GamsonWA. 1979. Utilitarian
transitionfromsocial movement to interest logic in theresourcemobilization perspec-
group.In Women, Power,and Policy,ed. E tive.In TheDynamicsofSocial Movements,
Boneparth, E Stoper,pp. 26-47. New York: ed. M Zald, J McCarthy, pp. 8-44. Cam-
Pergamon. 2nded. bridge,MA: Winthrop
CrenshawK. 1990. A Black feminist critique FriedmanD, McAdam D. 1992. Collective
ofantidiscriminationlawandpolitics.In The identityand activism:networks choicesand
PoliticsofLaw: A Progressive Critique,ed. thelifeofa socialmovement. See Morris&
D Kairys,pp. 195-218.NewYork:Pantheon. Mueller1992,pp. 156-73
2nded. FussD. 1989.Essentially Speaking:Feminism,
D'EmilioJ.1983.SexualPolitics,SexualCom- Nature,and Difference. New York:Rout-
munities. Chicago,IL: Univ.ChicagoPress ledge
DiMaggioPJ,PowellWW.1991.Introduction. Gamson J. 1995. Must identitymovements
In The New Institutionalism in Organiza- A queerdilemma.Soc. Probl.
self-destruct?
tionalAnalysis,ed. WW Powell,PJDiMag- 42:390-407
gio, pp. 1-38. Chicago,IL: Univ.Chicago Gamson WA. 1975. The Strategyof Social
Press Protest.Homewood,IL: Dorsey
DowneyG. 1986. Ideologyand theclamshell GamsonWA.1988.Politicaldiscourseandcol-
identity:organizationaldilemmasin the lectiveaction.Int.Soc. Move.Res. 1:219-44
anti-nuclearpowermovement. Soc. Probl. GamsonWA. 1991. Commitment and agency
33:357-71 in social movements. Sociol. Forum6:27-
EcholsA. 1989.DaringtoBe Bad. Minneapo- 50
lis: Univ.Minn.Press Gamson WA. 1992. TalkingPolitics. Cam-
Emirbayer M. 1997.A manifesto fora relational bridge,UK: Cambridge Univ.Press
sociology.Am.J.Sociol. 103:281-317 Gitlin T. 1995. The Twilightof Common
Ennis J. 1987. Fields of action:structure in Dreams.New York:HenryHolt
movements' tacticalrepertoires.Soc. Forum Goffman E. 1959. ThePresentation ofSelfin
2:520-33 Everyday Life.GardenCity,NY: Doubleday
EpsteinB. 1991.PoliticalProtestand Cultural GordonC, JasperJM. 1996. Overcoming
Revolution.Berkeley:Univ.Calif.Press the 'NIMBY' label: rhetorical and organi-
EpsteinS. 1987. Gay politics,ethnicidentity: zationallinksforlocal protestors. Res. Soc.
thelimitsofsocialconstructionism. Socialist Move.Confi.Change19:159-81
Rev.17:9-54 Gould RV. 1995. InsurgentIdentities:Class
Evans S, BoyteH. 1986. Free Spaces. The Community and Protestin Parisfrom1848
SourcesofDemocraticChangeinAmerica. totheCommune. Chicago,IL: Univ.Chicago
New York:Harper& Row Press
EyermanR, JamisonA. 1998. Music and Gould RV. 1998. Politicalnetworksand the
302 POLLETTA * JASPER

local/national boundaryin the Whiskey JoyceP. 1994. DemocraticSubjects. Cam-
Rebellion.See Hanaganetal 1998,pp.36-53 bridge,UK.: Cambridge Univ.Press
HanaganM. 1994. New perspectives on class KatzensteinM. 1998. Stepsisters:feminist
formation: culture,reproduction,andagency. movement activismin differentinstitutional
Soc. Sci. Hist.18:77-94 spheres.In TheSocialMovement Society,ed.
HanaganM, MochLP, teBrakeW,eds. 1998. D Meyer,S Tarrow, pp. 195-216.Lanham,
ChallengingAuthority. Minneapolis:Univ. MD: Rowman& Littlefield
Minn.Press KitscheltH. 1986. Politicalopportunity struc-
HirschEL. 1990a. UrbanRevolt:EthnicPoli- turesand politicalprotest. Br.J. Polit.Sci.
ticsin theNineteenth-Century ChicagoLa- 16:57-85
borMovement. Berkeley:Univ.Calif.Press Klandermans B. 1997. TheSocial Psychology
HirschEL. 1990b.Sacrifice forthecause:group ofProtest.Oxford, UK: Blackwell
processesrecruitment andcommitment in a KleinmanS. 1996.OpposingAmbitions: Gen-
studentsocial movement. Am. Sociol. Rev. derandIdentity inan Alternative Organiza-
55:243-54 tion.Chicago,IL: Univ.ChicagoPress
HolsteinJ,MillerG. 1990.Rethinking victim- Krinsky.1999.Working classification:
identity
ization.Symb.Interact.13:103-22 in social movements and class formation.
HuntSA, BenfordRD. 1994. Identity talkin AvailableonlinefromCIAO: ColumbiaInt.
thepeaceandjusticemovement. J.Contemp. Affairs Online(www.columbia.edu/cu/libra-
Ethnog.22:488-517 ries/indexes/ciao.html).
HuntSA, Benford RD, SnowDA. 1994.Iden- Laclau E, MouffeC. 1985.Hegemony and So-
tityfields:framingprocessesand the so- cialistStrategy.London:Verso
cialconstruction ofmovement identities.
See LarafiaE, Johnston H, GusfieldJR, eds.
Larafiaet al 1994,pp. 185-208 1994.New Social Movements: FromIdeol-
JasperJM. 1990. NuclearPolitics.Princeton, ogy to Identity.Philadelphia,PA: Temple
NJ:Princeton Univ.Press Univ.Press
JasperJM. 1997. The Artof Moral Protest. Lichterman P. 1996. The Searchfor Political
Chicago,IL: Univ.ChicagoPress Community: American Activists
Reinventing
JasperJM. 1998.The emotionsof protest: re- Commitment. New York:CambridgeUniv.
activeand affective emotionsin and around Press
social movements. Sociol. Forum13:397- Lichterman P. 1999.Talkingidentity inthepub-
424 lic sphere:broadvisionsandsmallspacesin
JasperJM, PoulsenJ. 1993. Fightingback: sexualidentity politics.TheorySoc. 28:101-
vulnerabilities,blunders,and countermobi- 41
lizationbythetargets in threeanimalrights Mansbridge J.1995.Whatis thefeminist move-
campaigns. Sociol.Forum8:639-57 ment?In FeministOrganizations, ed. MM
JasperJM, Poulsen JD. 1995. Recruiting Ferree,PY Martin, pp. 27-34. Philadelphia,
strangers and friends:moralshocksand so- PA: TempleUniv.Press
cial networksin animal rightsand anti- MarwellG, OliverP. 1993.TheCriticalMass
nuclearprotest. Soc. Probl.42:493-512 inCollective Action:A Micro-SocialTheory.
JenkinsR. 1996. Social Identity. New York: Cambridge, UK: Cambridge Univ.Press
Routledge Marx AW. 1998. MakingRace and Nation.
Johnston H. 1998. Subcultures and theemer- A Comparisonof the UnitedStates,South
genceoftheEstoniannationalist opposition Africa,and Brazil. New York:Cambridge
1945-1990.Sociol.Prospect.41:473-97 Univ.Press
Johnston H, LarafaE, Gusfield JR.1994.Iden- McAdamD. 1982. PoliticalProcess and the
tities,grievances,andthenew social move- Developmentof Black Insurgency, 1930-
ments.See Larafiaetal 1994,pp. 185-208 1970. Chicago,IL: Univ.ChicagoPress
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 303

McAdam D. 1988. FreedomSummer.New ginsofwomen'sliberation. See Larafiaet al
York:OxfordUniv.Press 1994,pp. 234-63
McAdamD. 1994. Cultureand social move- MuellerCM. 1987. Collectiveconsciousness,
ments.See Larafiaet al 1994,pp. 36-57 identitytransformation, and the rise of
McAdamD, McCarthyJD, Zald MN. 1988. womeninpublicofficein theUnitedStates.
Social movements. In HandbookofSociol- In The Women'sMovements of the United
ogy,ed. N Smelser,pp. 695-737. Newbury Statesand Western Europe,ed. M Katzen-
Park,CA: Sage stein,C Mueller,pp. 89-108. Philadelphia,
McAdamD, PaulsenR. 1993. Specifying the PA: TempleUniv.Press
relationshipbetweensocialtiesandactivism. MuellerCM. 1992.Buildingsocialmovement
Am.J.Sociol.99:640-67 theory. See Morris& Mueller1992,pp.3-25
McCall G, SimmonsJ. 1978. Identitiesand OberschallA. 1973.Social Conflict and Social
Interactions.New York:FreePress Movements. EnglewoodCliffs, NJ:Prentice-
MelucciA. 1985. The symbolicchallengeof Hall
contemporary movements. Soc.Res.52:789- OffeC. 1985.Newsocialmovements: challeng-
816 ing theboundariesof institutional politics.
MelucciA. 1989.NomadsofthePresent.Lon- Soc. Res. 52:817-68
don:Hutchinson Radius OlsonM. 1965.TheLogicofCollective Action.
MelucciA. 1996.Challenging Codes: Collec- Cambridge, MA: HarvardUniv.Press
tiveActionin theInformation Age. Cam- PfaffS. 1996.Collectiveidentity andinformal
bridge,UK: Cambridge Univ.Press groupsin revolutionary mobilization:East
MeyerD, Whittier N. 1994.Social movement Germany in 1989.Soc. Forces75:91-118
spillover.Soc. Probl.41:277-98 PhelanS. 1989.Identity Politics.Philadelphia,
MilkmanR. 1986. Women'shistoryand the PA: TempleUniv.Press
Searscase. Feminist Stud.12:375-400 PizzornoA. 1978.Politicalexchangeand col-
MinkoffD. 1997.Producing socialcapital:na- lectiveidentity in industrialconflict.In The
tionalsocial movements and civil society. Resurgence ofClass Conflict inWestern Eu-
Am.Behav.Sci. 40:606-19 ropesince1968,ed. C Crouch,A Pizzorno,
Minow M. 1990. MakingAll theDifference. pp. 277-98. London:Macmillan
Ithaca,NY: CornellUniv.Press PizzornoA. 1986. Some otherkindsof other-
Mische A. 1996. Projectingdemocracy:the ness:a critiqueof 'rationalchoice'theories.
construction ofcitizenship acrossyouthnet- In Development Democracyand theArtof
worksinBrazil.In Citizenship, and
Identity, Trespassing:Essays in Honor ofAlbert0.
Social History, ed. C Tilly.Cambridge, UK: Hirschman, ed. A Foxley,MS McPherson,
Cambridge Univ.Press G O'Donnell,pp. 355-72. NotreDame,IN:
MooreK. 1996. Organizingintegrity: Ameri- Univ.NotreDame Press
can scienceandthecreationofpublicinter- PlotkeD. 1990. What'sso newaboutthenew
estorganizations, 1955-1975.Am.J.Sociol. socialmovements? SocialistRev.20:81-102
101:1592-627 PollettaF. 1997. Cultureand its discontents:
MorrisA. 1984.TheOriginsoftheCivilRights recenttheorizing on cultureandprotest. Soc.
Movement. New York:FreePress Inq. 67:431-50
MorrisA. 1992. Politicalconsciousnessand PollettaF. 1998a. 'It was like a fever':narra-
collectiveaction. See Morris& Mueller tiveandidentity insocialprotest.Soc. Probl.
1992,pp.351-73 45:137-59
MorrisA, MuellerCM, eds. 1992.Frontiers in PollettaF. 1998b.Contending stories:narrative
Social Movement Theory.New Haven,CT: in socialmovements. Qual. Soc. 21:419-46
Yale Univ.Press PollettaF. 1999. 'Freespaces' incollectiveac-
MuellerC. 1994.Conflict networks andtheori- tion.TheorySoc. 28:1-38
304 POLLETTA * JASPER

PulidoL. 1996. Developmentof the 'people resonance, andparticipant mobilization. Int.
of color' identity in theenvironmental jus- Soc. Move.Res. 1:197-217
ticemovement of theSouthwestern United SnowD, McAdamD. 2000.Identity workpro-
States.SocialistRev.26:145-80 cesses in thecontextof social movements.
Ringmar E. 1996.Identity,Interest,andAction: In Self,Identity,and Social Movements, ed.
A CulturalExplanationof Sweden'sInter- S Stryker,TJOwens,R White.Minneapolis:
vention intheThirty YearsWar.Cambridge, Univ.Minn.Press
UK: Cambridge Univ.Press SnowDA, Rochford EB Jr,Worden SK, Ben-
RobnettB. 1997. How Long? How Long? fordRD. 1986.Framealignment processes,
African-American Women intheStruggle for micromobilization, andmovement participa-
CivilRights.New York:OxfordUniv.Press tion.Am.Sociol.Rev.51:464-81
RogersKL. 1993.Righteous Lives:Narratives SnowDA, Zurcher LA, Ekland-Olson S. 1980.
oftheNew OrleansCivilRightsMovement. Socialnetworks andsocialmovements: a mi-
New York:New YorkUniv.Press crostructuralapproachtodifferential recruit-
RossR. 1983.Generational changeandprimary ment.Am.Sociol.Rev.45:787-801
groupsina socialmovement. InSocialMove- SomersM. 1994.The narrative constitution of
mentsoftheSixtiesandSeventies, ed.JFree- identity:a relationalandnetwork approach.
man,pp. 177-88.New York:Longman TheorySoc. 23:605-49
Roth B. 1998. Feministboundariesin the SomersM, GibsonGD. 1994.Reclaimingthe
feminist-friendly organization: thewomen's epistemological 'other':narrativeandtheso-
caucusofACT UP/LA.GenderSoc. 12:129- cial constitution
ofidentity.In Social Theoty
45 and thePoliticsofIdentity, ed. C Calhoun,
RuchtD. 1992.Thestrategies andactionreper- pp. 37-99. Cambridge, MA: Blackwell
toiresofnewmovements. In Challenging the Steinberg MW. 1996."Thelabourofthecoun-
PoliticalOrder,ed. R Dalton,M Kuechler, tryis thewealthofthecountry": class iden-
pp. 156-75.New York:OxfordUniv.Press tity,consciousness,andtheroleofdiscourse
RuppL, TaylorV. 1987.Survivalin theDol- in themakingoftheEnglishworking class.
drums:TheAmerican Women's RightsMove- Int.Lab. Work. -ClassHist.49:1-25
ment,1945 to the1960s.New York:Oxford StrykerS. 1980. SymbolicInteractionism.
Univ.Press MenloPark,CA: Benjamin-Cummings
ScottJW.1988.Deconstructing equalityversus TajfelH. 1981.HumanGroupsandSocial Cat-
difference.Fem.Stud.14:33-50 egories.New York:Cambridge Univ.Press
ScottJ. 1990.Dominationand theArtsofRe- TarrowS. 1998. Power in Movement:Social
sistance.New Haven:Yale Univ.Press Movements, CollectiveAction,and Politics.
SeidmanS. 1993. Identityand politicsin a New York:Cambridge Univ.Press.2nded.
'postmodern'gay culture:some historical TaylorC. 1989.SourcesoftheSelf:TheMak-
and conceptualnotes.In Fear of a Queer ingoftheModernIdentity. Cambridge, MA:
Planet:QueerPoliticsandSocialTheory, ed. HarvardUniv.Press
M Warner, pp. 105-42. Minneapolis:Univ. TaylorV. 1989. Social movement continuity.
Minn.Press Am.Sociol.Rev.54:761-75
Sewell W. 1980. Workand Revolutionin TaylorV, RaeburnNC. 1995.Identity politics
France. Cambridge,UK: CambridgeUniv. as high-risk activism:careerconsequences
Press forlesbian,gay,and bisexualsociologists.
Snow DA. 2001. Collectiveidentity. In Inter- Soc. Probl.42:252-73
nationalEncyclopediaoftheSocial andBe- TaylorV, Whittier NE. 1992.Collectiveiden-
havioralSciences,ed NJSmelser, PB Baltes. tityinsocialmovement communities: lesbian
London:ElsevierSci. feminist mobilization.See Morris& Mueller
SnowDA, Benford RD. 1988.Ideology,frame 1992,pp. 104-29
COLLECTIVE IDENTITY, SOCIAL MOVEMENTS 305

TeskeN. 1997.PoliticalActivists in America. of social movements.Soc. Res. 52:749-
TheIdentity ConstructionModelofPolitical 87
Participation.Cambridge,UK: Cambridge TracyJ.1996.DirectAction:RadicalPacifism
Univ.Press. fromtheUnionEightto theChicagoSeven.
TillyC. 1978.FromMobilization toRevolution. Chicago,IL: Univ.ChicagoPress
Reading,MA: Addison-Wesley WhalenJ, Flacks R. 1989. BeyondtheBar-
TillyC. 1998.Politicalidentities.
See Hanagan ricades.TheSixtiesGenerationGrowsUp.
etal 1998,pp. 3-16 Philadelphia,
PA: TempleUniv.Press
TouraineA. 1981.TheVoiceandtheEye.Cam- Whittier N. 1995. FeministGenerations:The
bridge,UK: Cambridge Univ.Press Persistenceof theRadical Women'sMove-
TouraineA. 1985.An introduction tothestudy ment.Philadelphia,
PA: TempleUniv.Press