Sheer Numbers: Critical Representation Thresholds and Women's Political Representation

Author(s): Karen Beckwith and Kimberly Cowell-Meyers

Source: Perspectives on Politics, Vol. 5, No. 3 (Sep., 2007), pp. 553-565
Published by: American Political Science Association
Stable URL: http://www.jstor.org/stable/20446503 .
Accessed: 25/06/2014 00:26

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

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

American Political Science Association is collaborating with JSTOR to digitize, preserve and extend access to
Perspectives on Politics.

http://www.jstor.org

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articles

Sheer Numbers: Critical Representation

Thresholds and Women's Political
Representation
Karen Beckwithand KimberlyCowell-Meyers

Studiesofwomen in legislatures indicatethatachievinga "critical

mass" ofwomenmay have theeffect of changingthelegislative
prioritiesofwomen, increasingthenumberof legislativeinitiatives dealingwithwomen and thepassagerateof such initiatives,and
alteringthe legislative ofmen. In theabsenceof a critical
priorities mass, "token"womenmay be so constrainedby their minority
statusas tobe unable torespondproactivelyto theirenvironment. Popularwisdom suggeststhata critical
massmay be necessaryfor
women tomake a difference aswomen ina legislature.
mass isbothproblematicand under-theorized
Yet, critical inpoliticalscienceresearch.
The critical
mass thresholdisdebated,the
mechanismof effectisunspecified,possiblenegativeconsequencesareoverlooked,and thepotentialforsmallnumbersof elected
women toeffect politicalchangeon behalfofwomen isneglected.Beyond sheernumbers, what are theconditionsthatgovernthe
abilityofwomen legislatorstomake a difference? We argue thattwomajor contextualfactors beyond thesheernumbersare likely
togoverntheextenttowhich femalelegislatorsserveto represent women. Relyingon thesecondaryliterature, thisarticlemaps
parliamentary and civilsocietycontextsto sheernumbersofwomen to locateconditionsinwhich femalelegislators aremost likely
tohavepolicysuccesses.

T o what extentdoeswomen'ssubstantive representa mass"ofwomenmay have theeffect of changingtheleg

tiondepend upon the sheernumbersof elected islative prioritiesofwomen, increasing thenumberof leg
women?Activists and scholars oftenasserta positive islativeinitiatives
dealingwithwomen and thepassage
relationship betweenthenumbersofwomen electedto rateof such initiatives, and alteringthelegislativepriori
officeandpublicpolicyoutcomesthatarewoman-friendly tiesofmen.2 Sheernumbersof electedwomen (descrip
and even feminist. The idea of a "critical mass" is the tiverepresentation) isexpectedtofacilitatepolicy-making
popularand compelling notionthatincreasing thenum inwomen'sinterests (substantiverepresentation);increas
bersofwomen inpolitics will starta chainreaction, lead ingthenumbers of electedwomen isexpectedtoenhance
ing to a new dynamicfavorabletowomen.1Studiesof thelikelihood ofwoman-friendly publicpolicy.
women in legislatures indicatethatachievinga "critical Yet,critical
mass as a theoryisproblematic and under
theorized inpoliticalscienceresearch.3 First,no threshold
numberhas been establishedthatmarks theboundaries
KarenBeckwithis theFloraStone MatherProfessor of betweennumbersofwomen too smalltohavean impact
PoliticalScienceat CaseWestern Research University and on and numberslargeenoughto securepolicy
legislation
Editor,withLisa Baldez,of Politics & Gender initiatives
representing women's interests. In the litera
(karen.
beckwith@case.edu).Her published workincludes ture,the threshold has been variouslyidentified as 15,
Women'sMovementsFacingtheReconfigured State 20, 25, or 30 percent.4Second, theconditionsunder
(Cambridge 2003, withLeeAnnBanaszakandDieter whichsomelarge numberofelected womencouldbe trans
Rucht),PoliticalWomen andAmericanDemocracy latedintosubstantive representationofwomen havenot
withChristina
(forthcoming, Wolbrecht andLisa Baldez), been theorizedor elucidated.Where representation
and articlesongenderandpoliticsin theEuropeanJour increases fromsmallto large minorities,as in theUnited
nal of PoliticalResearch,Politics
& Society,and Signs, Kingdom in 1997, specific mechanismsforproducing
amongothers. Kimberly Cowell-Meyers isAssistantProfes substantive,women-friendly, changehavenot
legislative
sorin the Department ofGovernment atAmericanUniver beenidentified.5Researchemploying critical
mass asa con
sity(kowell@american.edu). She isauthorof Religion cepthasnot clarified theprocessbywhich sheernumbers
and Politics:The PartyFaithfulin IrelandandGermany ofwomenmightwork to advancewomen's substantive
(Greenwood, 2002) and articlespublishedinWomen & representation. It isnot clearwhethersheernumbersof
Politics,NationalismandEthnicPolitics,and Irish women shouldhave a proportional impact,a curvilinear
PoliticalStudiesamongothers. Shehasworkedin the impact,or an absolutenumbersimpacton policy-making
BritishParliament and theUnitedStatesInstitute ofPeace. around women'sinterests.6 What isclearisthatcriticalmass

DOI:10.1017/Si53759270707154X 2007 1Vol.5/No.

September 3 553

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articles I Sheer Numbers

research cannotidentify a critical

numberafter whichevery Based on our theoretical assessment and empiricalcri
thingchanges.It isunlikelythattheNew ZealandHouse tique,we map theconditions underwhich femalelegisla
ofRepresentatives, forexample,inwhichwomen consti torsaremost and leastlikelytohavepolicysuccesses. We
tute29.2 percent of themembers, would experience amajor consider the range of potentialinteractionsbetween sheer
increaseinwomen'ssubstantive representation ifwomen numbersofwomen,on one hand,and thelegislative and
constituted an additional0.8 percent of themembership.7 politicalcontext, on theother,to identify theconditions
Similarly,no suchcriticalthreshold hasbeen identified in underwhich electedwomen mightmaximizewoman
theNordic countries, wherewomen'sparliamentary repre friendly policysuccesses. We constructa rangeof condi
sentation surpassedthe20 percentthreshold in the1970s tionsbetweenthetwoextremes: sheernumbers ofwomen
and the30 percentthreshold in the1980s.8 as thesole factorpredicting women'spolicy representa
Increasingthenumbersof electedwomen in a legisla tion,and legislative and civilsocietycontextsidentified in
ture may evenproducenegativeconsequences. The possi theliterature tohavebeenconducivefor women'ssubstan
bilitythatan increase inwomen'sdescriptive representation tiverepresentation. We situateresearch hypotheses regard
mightproducenegativeoutcomeshas not been consid ing theseconditionsand contextualfactorsforassessing
ered in the literature. Sheernumbersofwomenmight the impactof each, discussingthe interactive potential
conceivably generatea backlashfrom male gate-keepers, among sevenkeyvariablesthatwe hypothesizeshould
impedecross-party work amongfemalelegisla providethegreatestanalyticalleverageforassessingthe
legislative
tors,or serveto advanceindividual women uninterested impactof sheernumbersofwomen on women'spolicy
in (orhostileto)publicpoliciesconcerning women.Fur outcomes.In doing so,we map possibilitiesforfurther
thermore, an increaseinwomen's sheernumberscould developmentof theoryconcerning women'sdescriptive
increasethepartisan,ethnic,and racialheterogeneity of and substantive representation.
electedwomen in a legislature, with potentialconcomi
tantincreases inpartydiscipline, dividing womenbyparty,
potentiallyprovoking tensions involving thenecessary con Defining Women-Friendly
structions of cross-race, cross-ethnicity alliancesamong Public Policy
women.Electing morewomen couldmean electingfewer The most problematic componentof theorizing therela
womenmotivatedby theabsenceofwomen inpoliticsto tionship between women'sdescriptive and substantive
rep
"act for"otherwomen or itmay make no difference at resentation concernsestablishing what constitutes,in
all.9In short,theprocessesbywhich increasing thenum comparative women'sissues.11
politicalresearch, The idea
bersofwomen inparliaments will resultinwomenwilling ofwomen'ssubstantive representationhingeson thenotion
toadvance woman-friendly legislationhaveyettobe elab of somekindof sharedexperienceamongwomen that
orated,assumingelectedwomen are situatedin contexts fosters a senseof commonsocialor politicalinterests.12
that would permitthemtoundertake sucha project,even Becausewomen'sexperiences aresociallyconstructedand
ifwilling. because theyvarywidelybased on thespecificprocesses
mass"ofwomen toa nationallegis
Electinga "critical of construction, women's issuesdifferfromcontextto
laturemay be a necessary but insufficient causeofwomen's context,across and within statesand across time.13
substantive representation. It is also possible thatsheer Women's issuesin theU.S. in thenineteenthcentury
numbersofwomen arenot a necessary conditionforleg included,forexample,temperance, suffrage,
child-labor,
women'spolicy issuesand thatsmallnumbersof
islating public libraries, etiquette,and children's discipline.At
women who arewell-situatedmay be able to deliver the same time in Britain,women's issuesfocusedon
women's substantiverepresentation.10 What numbers, women s rightsas autonomous citizens inmarriage,
frameworks, conditions,and contextsgoverntheability marriedwomen'sproperty ownership, women's rightsin
ofwomen legislators tomake a difference? divorce,and child custody.In SoutheastAsia,women's
This articleexaminesthecomparative politicalresearch issuesinclude micro enterprise aimedatwomen; inparts
on women'sdescriptive and substantive representation. ofAfricaand theMiddle East, femalegenital mutilation,
First,we focuson the issueof sheernumberswith a veiling,citizenship rights,
and early marriageare thesub
reviewand critiqueof the literature thatsuggeststhat stanceof politicaland socialmovementson behalfof
numbersaredispositive. We reviewthetheoretical foun women. In contemporary India,women's issuesinclude
dationsof criticalmass inwomen'spublicpolicy-making dowryrequirements, protectionagainstdomesticabuse,
and considertheways inwhich sheernumbersof female and employment discrimination, amongothers.14Because
mightbe theorizedin connectiontowomen's women'ssubstantive
legislators representationdiffersacrosstimeand
substantive We thenturntoa critical
representation. assess space, what itmeans tobe successful on behalfofwomen's
ment of the empiricalliterature on the conditionsof interests isgovernedby dynamicsof cultureand history.
women electedtonationallegislatures and themajor fac It isalsoproblematicthattheeffects ofgender may not
torsthatfacilitatepolicy-making onwomen's issues. be distinctfromthoseof raceand class. In otherwords,

554 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
 the intersectionalityof genderwith otheridentities may confront issues
withwhichwomenhavetraditionally been
limittheapplicability of a women's issuesparadigm.'5 concernedin theirroleas caregivers, suchas education,or
Much of thecriticism of first-world feminism by third theprotection of children."25 As Swerspointsout,how
worldactivists, forexample,focuses on theperceived com ever,thesedefinitions ofwomen's interest policiesare so
monalityof genderedexperiencesthat is expectedto broad thattheycould be read to includeall policiesor
transcend historiesof colonialism, and oppression, aswell none at all.26
as different culturesand economicexperiences."6 In addi There is also theproblemof distinguishing subjective
tion,women's experiences in thesame at
society the same from objective interestsand thisiswhymuch research
pointinhistory may be verydifferent, dependingon other on women'spolicy issuesconflates women'spolicywith
socialforces, andwomenmay not see themselves as sim feminist policy.27For example,Hernes uses the term
ilaror theirinterests as shared.Theymay evenorganize "woman-friendly" to referto genderequalityand social
againsteachother.'7 policies produced by state feminism.28 Gardiner and
In light
of thesechallenges, ofwomen'sissues Leijenaaralso definewoman-friendly
definitions as policydesigned
tendtowardthegeneral. Carrolldefines women'sissuesas to "improvethestatusofwomen, theadoptionof and
thosethat"disproportionately become theresponsibility implementation ofwhich reflecttheexistenceof a cul
ofwomen as a result of thesexualdivisionof labor"and as tureof equality."29Borchorst considerswoman-friendliness
issues"where policyconsequences arelikelytohaveamore todescribean atmosphereinwhich "patriarchal patterns
immediate anddirectimpacton significantly largernum of power" are debated.30Lovenduskidoes not distin
bersofwomen thanofmen"."8Lovenduski defineswomen's guishbetweenfeminism andwomen'smovementactiv
issuesas "thosethat mainlyaffect women, either for bio ism yetmeans only feminist activismwhen shediscusses
logicalreasons(suchas breastcancerscreening or repro women'sinterests inpublicpolicy.31 Gelb and Palleyalso
ductiverights) or forsocialreasons(sexequalityor child conflatefeminism andwomen'sactivism. They consider
carepolicy)"'.'9In theEuropeancontext, Hoskynsstudies thedifferent successratesof feminist publicpolicyinitia
women'spolicyas "theareaofpoliticalactionand activity tives,callingthemroleequityand rolechangeissues:
whichparticularly concerns(or targets) women orgroups
Role equity issuesare those thatextend rightsnow enjoyedby
ofwomen,orwhere issuesare forced onto theagendaby
othergroups (men,minorities) towomen. Role change issues
women".20
appear toproduce change in thedependentfemaleroleofwife,
In researchon legislatures,thedefinition ofwomen's mother, and homemaker,holding out thepotentialof greater
issuelegislationalso tendstobe broad.Thomas,forexam sexualfreedomand independencein a varietyof contexts.32
ple, uses a definitionof "policiesfavoringwomen, chil
dren and families" aswomen's issues.21 Brattondefines Mazur, althoughexplicitin acknowledging her focuson
women's interestlegislationas legislationthatwould feministpublic policy, limitsher study towomen's policy
"decreasediscrimination or countertheeffects ofdiscrim issuesthat involve at leastthreeof fivecriteria:
ination,orwould improvethesocio-economic statusof improvement ofwomen's rights,statusor situationto be in line
women".22Bystydzienski refersto carepoliciesand gen withmen's ... ; reductionor eliminationof gender-basedhier
der equalitypolicies in her studyofwomen'spolicy in archies. .. ; a focuson both thepublic and privatespheresor an
Norway, writingthatcarepolicies approach thatavoids distinctionsbetween thepublic and the
private;a focuson bothmen andwomen; and ideas thatcan be
readilyassociatedwith a recognizedfeminist group,movement
aim to improvethequalityof lifeofwomen and childrenvia
or individualactor in a nationalcontext."33
legislationfocusedin such areasas child and healthcare,educa
tion,and parentalleave.While genderequalitypolicies are con
cernedwith the redistribution
Not allwomen'sorganizations
of statusand power,carepolicies
orwomen'sactivismare
areconcernedwith theredistribution feministin thesenseof challenging
of goods, servicesand pub women
patriarchy;
lic funds.23 often organize around nonfeministand even anti
feminist women'sorganizations
goals.34Instead, can be
Heath,Schwindt-Bayer, andTaylor-Robinson considerthe distinguished
fromothertypes of organizations,
not by a
way thatLatinAmerica legislatures dealwithwomen's litmustestof feminism but by "theprimacyofwomen's
issuesin termsof committeesthat"deal explicitlywith women'sissues,andwomen'slead
genderedexperiences,
women and one of theprimarylocationsof gender ershipand decisionmaking."35 Much has beenwritten
inequality-thefamilyand home (e.g.,equalityin the on thenegativeconnotations of the"feminist"labeland
work place,protectionagainstviolencein thehome)."24 thisliteraturegoes somedistancetoexplainwhy individ
Swers,in her analysisofwomen in theU.S. Congress, uals andmovements may not identifywith the labelor
defineswomen'slegislativeissuesas those"thatarepartic with the framework of the broader agenda.36Some
ularlysalienttowomenbecausetheyseektoachieveequal women'smovements may deliberately(and strategically)
ityforwomen, theyaddress women'sspecialneeds,such avoid framingthemselves as feminist;37
in othercon
aswomen'shealthconcernsor child-careissues;or they texts, rights-based
approachestopressuringthestate
would

September 2007 1Vol. 5/No. 3 555

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articles I Sheer Numbers

be inappropriate in theabsenceof a rights-based under Sheer Numbers and Critical Mass

standingof thestate/society relationship.Indeed,some Researchon theinfluence of sheernumbersofparliamen
women'smovementsand organizationsare pointedly tarywomen rely
primarily onRosabeth Moss Kanter'sfoun
anti-feminist (e.g., in theU.S., Daughtersof theAmeri dationalstudyofwomen in amale-dominated corporate
can Revolution,theEagle Forum; in Chile, El Poder environment. Because of thestructure ofKanter'sargu
Femenino;inNicaragua, theAssociationofNicaraguan ment, political scientistshave discussed the impactof
Women "LuisaAmandaEspinosa").Furthermore, itwould sheernumbersin terms of a "criticalmass"ofwomen and
be a mistake to attemptto apply objectivecriteriaof in termsof threshold percentagesthatarepredictedto
feminist contenttodetermineifamovementis feminist shift (thatis,increase)women'spolicyrepresentation. Kanter
or not or to parse out thegoals of different women's arguedthat womenwould have littleinfluence as individ
movementsforthepurposeof discerning separatepolicy uals or as a group inmostlyuniformsituations and that
processesthatpertainto each. because theyare stillsubjecttoperformance pressures,
Whilewe recognize thesedynamics andacknowledge that roleentrapment, and boundaryheightening as tokensin
womensmovements andwomen'sissuesshouldnotbe con "skewedgroups,"they would not be able tomusterinflu
flatedwith feminism, we also acknowledgethatnowhere encewithin the largercorporationuntil theyreacheda
in thecomparative literatureonwomen'ssubstantive rep representational threshold of 35 or 40 percent.41 Kanter
resentation isa theoreticallinkestablished between women's hypothesized thatin "uniform" majority/minority situa
movementsthatarenon-oranti-feminist and therepresen tionsand "skewed" groupcontexts, wherewomen consti
tationof issuesthatinan "immediate and directway,are tutelessthan15 percentof theoverallgroup,theirself
aboutwomen exclusively."38 Furthermore, no scholarship perception and theattitudes of theirpeerswould interfere
on electedwomen focuses onwomen in rightwing to
parties with theirability interact effectively andwould limit
or theorizesaboutthepolicyproducedbyrightwing female theirinfluence. In "tilted"groupcontexts, wherewomen
Therefore,the relationship
legislators. we aim tomap constitutebetween15 and 35 percent, Kanterhypoth
betweenthedescriptive representationofwomenand their esized womenwould be seenaspotential alliesand thattheir
substantiverepresentation hasmost commonly(ifnot influence would therefore increase.In "balanced"groups,
exclusively) been constructed aroundadvancing women's wherethegroupsaredivided,at the most extreme, between
statusand equality.For thepurposesof thisproject,we 60 and40 percent, Kanterargued, womenwould havetheir
narrowthescopeofwomen's issuesand operationalize bestopportunity toaffect priorities and policies.42
women-friendly publicpolicyasa subset ofpolicyadvanced Researchon women in nationaland regionallegisla
bywomen'sorganizations thatboth addressesissuesthat tureshas establishedthatwherewomen constituteless
affectwomen exclusively and directlyand thatsimulta than15 percentof a legislative body,women's influence
neouslyadvancestheirstatusin society. By thiswe mean will be constrained at best.There appearsto be general
specificallypolicies liberalizing divorceand reproductive agreement thata criticalrangeof between15 and 30 per
equalizingthecivilrights
rights; ofmen andwomen interms centofwomen ina nationalor regional body,from parties
of education,employment, pay,training,property owner acrossthepoliticalspectrum,isnecessaryforwomen to
shipand inheritance, marriage, mobility, andpoliticalrep influence theagendaor styleof business withinthatbody
providing
resentation; family and medical leave,subsidizing or itspolicyoutputs.43 The of
impact sheernumbersof
childcare,addressing domesticabuse,sexualassault, vio women at thelower boundaryof thiscriticalrangecan be
lenceagainstwomen;andproviding forwomen's health care; evaluated: women constitute 15 percentormore of the
amongothers.39 lower or singlehouse members in 81 of the185 nations
In short,our focus ison the interactionsbetween sheer about which theInter-Parliamentary Union reported data.44
numbersofwomen and conducivepolicy-making con For the of
purposes assessing the critical
mass argument,
texts,based on theempiricalpolitical researchto date. we employtheminimal 15 percentthreshold, whichwe
Our purpose is to theorizetheseinteractions to identify termthecriticalrepresentation threshold. For empirical
combinations ofconditionsthat mightbe expectedtooffer research purposes,settingany representational threshold
thegreatestpossibilities forconverting women'sdescriptive higherthan15percentbecomesslightly problematic, given
representation intosubstantive representation,in thecon theproblem ofsmallnumbers ofcases.Although53 nations
textof democraticpoliticalsystems. Empiricalevidence meet a criticalrepresentation threshold of 20 or higher,
supports a generalclaimthatthenumberofwomenelected onlyninenationshavea female delegationthatconstitutes
topoliticalofficehas a positiveimpactinterms ofwomen more than35percent oftheirlower house.45Moving towards
friendlypublicpolicy.40 We identify threemajor setsof higher percentages byquintile,thenumbers ofnationswithin
factorsthatprovidethegreatest analyticaland explanatory eachquintiledecreases(see table1).Nonetheless,thereis
leverageconcerning women'ssubstantive policyrepresen currently enoughvariation amongnationsinterms of their
tation:sheernumbers ofwomen,conduciveparliamentary percentage parliamentary representation ofelected women
contexts, andconducive civilsocietycontexts. toassesstheimpactof sheernumbers.

556 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
 Parliamentary Context
Table 1
'Which partyis ingovernment andhow electorally secure
Nations by percentage of women in lower
thatpartyismay affect thelikelihoodthatparliament will
house of parliament (2005)
produce women-friendly public policy.
The comparative
Women inLower House of Number of politicalliteratures
onwomen'ssubstantive representation,
Parliament Nations socialwelfarepolicy and feministpublic policy concuron
Percent < 15 104 thecontributions of leftwingpartiestoadvancing women's
15 ? percent < 20 28 The presence
issues.46 ofa leftwingpartyinparliament may
20 ? percent < 25 25
women'sissues
notbe a sufficientconditionforadvancing
25 < percent < 30 8
30 ? percent < 35 11 but, ingeneral,leftwing partieshavea betterrecordthan
35 < percent < 40 7 rightwing and supportinglegislation
partiesof initiating
Percent - 40 2 liberalizingdivorce,extending abortionrights,criminaliz
Source: www.ipu.org/wmn-e/classif.
htm ingviolence against
women, expandingemployment oppor
tunities,providing women's healthcare innovations, and
advancingsocialwelfareissues.47 As a result,we expect
thatleftwing partiesoffersheernumbers ofwomena better
It is likelythatthehigherthepercentage ofwomen contextforpromoting women-friendly policies.48
themore conducivethe
electedto a nationallegislature,
conditionsfor women'ssubstantive Inaddi
representation. H3: Where thegoverning party(orcoalitionofparties)is
tion,becausemost empiricalresearch has founda strong women'ssubstantive
leftwing, inpolicyterms
representation
relationshipbetweenwomen's representation within governedby centeror
will be greaterthanin legislatures
leftwing parliamentary
groupsor caucuses,we expectthat center-right parties.

highnumbersof leftwing women electedtonationalleg

islatureswill similarly
effect
women-friendly publicpol Second, it is important toconsiderthepositionof the
icy.These claimsin theliteraturecan be hypothesizedas governing party:is thegoverning partya long-standing
follows: governing partywith aweak opposition,or istheopposi
tiona crediblethreattothegovernment andwell-positioned
H 1:Where thepercentage ofwomen electedtoa national forthenextelection?'Where a leftwingpartyis securein
is 15 percentor higher,
legislature women's substantive government, in electoralterms,femalelegislatorsshould
representation inpolicytermswill be greaterthanin leg have theirbestchancesforenacting women-friendly pub
with a smallerpercentage
islatures women.
of elected licpolicies.Electoralsecurity can be operationalizedin
severalways: thepercentage marginof thevotewon by
H2: 'Where thepercentageofwomen representing leftwing thegoverning party/coalitionin thepreviouselection,the
partiesin a nationallegislatureis 15 percentor higher, marginof seatssecuredby thegoverning party/coalition
women'ssubstantive representationinpolicytermswill be in thepreviouselection,andpublicopinionapprovalrat
greaterthanin legislatureswith a smallerpercentage of ingsof thegoverning We subsumeall these
party/coalition.
women electedby leftwing parties. operationalizationsin thefollowing coveringhypothesis:

These,we argue,are thetwomost powerful variablescon H4: The stronger theelectoralpositionof a leftwinggov
cerning anycritical and theimpact erningpartyor coalition,thehigher
massof femalelegislators women'ssubstantive
of sheernumbersofwomen on publicpolicyoutcomes. representation in policy termswill be thanwhere the
government's electoralpositionisprecarious.
Political Contexts
Criticalmass and sheernumbers,however, do not take Civil SocietyContext
intoaccount thecontextwithinwhich electedwomen Women and politicsresearch indicatestheimportance of
We arguethatpoliticalcon
enactpoliticalrepresentation. an activefeminist movementin civilsocietyarticulating
textis thenecessary conditionfortranslatingsheernum women's issuesand insisting upon legislative action.49
bersof women intowomen-friendly public policy.To Feminist movementsarepositedtohave twoimpactsthat
map therelationship betweendescriptiveand substantive activatetheprocessofwomen's substantive representa
we need to take intoconsideration
representation, not tion.First,a feminist
movement militatingaround women's
onlythenumbersand characteristics women
of theelected contextfavorable
issuescancreatea legislative to theissues
but also thepoliticalcontext withinwhich thewomen itadvocatesbypublicizing women'sissues,settinga pro
arepositioned.
legislators onwomen'ssub
The literature posed legislativeagendaaroundthoseissues,and serving
stantiverepresentation suggeststwomajor contexts: par noticeof thepresenceof a [potential] mass sup
electoral
liamentary and civilsociety. portingthoseissues.Second, a feminist movementcan

September 2007 IVol. 5/No. 3 557

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articles I Sheer Numbers

Table 2
Conditions facilitatingwomen's substantive representation
CASE 1
Sheer Numbers CASE 2 CASE 3 CASE 4 CASE 5
and Positive Positive Numbers and Numbers and Sheer
PoliticalContext PoliticalContext CivilSociety Parliament Numbers
SHEER NUMBERS
Women inparliament2 15% x x x x
women inparliament2 15 %
Leftwing x x x x
CONTEXTS
Parliamentary context
Leftparty ingovernment x x x
No strongopposition x x x
Civil Society context
Active feminist
movement x x x
Weak or no opposingmovement x x x
Support inpublicopinion x x x
WOMEN-FRIENDLY PUBLIC POLICY Very likely Likely Possible N/A Very unlikely
Case 1 Scandinavia, 1960-present
Case 2 US, UK, FRG, late 1960s-early 1980s
Case 3 Italy,late 1960s-early 1980s
Case 4 Soviet bloc nations; no cases forparliamentary
democracies
Case 5 No examples

bringdirectpressureupon (orprovideimmediate support tion,than

wherepublicopiniondoesnotsupport
women's
to) electedwomen,providingthem with a political(and policy issues.
oftenfinancial)
base tostrengthen
their
capacitytoadvance
thoseissuesinparliament.
Mapping the Numbers/Context
H5A: The more activea feminist
movementinsupport of Intersection
women'spolicyissues,thehigher
women'ssubstantive
rep Mapping thesesevenfactors providesa way to assessthe
resentation,
thanwhere a feminist
movementisweak or impactof critical
mass againstsheernumbers,in specific
nonexistent. politicalcontexts.These sevenfactors aremapped inter
actively,albeitnot exhaustively, in table2.Althougheach
There are,however,negativeimplicationsof feminist factorshould facilitate substantive policies forwomen,
movementactivism,specifically
intheactivation
ofoppos notallarenecessary forsubstantive representationtooccur.
ingmovements.50 The emergence of an activecounter Instead,severalcombinations of intersecting factorsare
movementshouldbe expectedto impedetheefforts of likelytoeffectwomen'ssubstantive representation.
elected
women toadvancewomen'sissuesand tocreatea Table 2maps fiveexemplary setsof intersectingfactors
legislative
environmentlessconduciveto enactingthose with thepotentialtopredict women'ssubstantive repre
issues. we hypothesize
Therefore, that sentation.The first columnof table2 indicatesthemaxi
mumfacilitating conditions for
women'ssubstantive policies
H5B: The more activea countermovement inopposition
(Case 1).All sevenfacilitating factors arepresent:many
towomen'spolicy issues,the lower
women'ssubstantive
women are in office,especiallyin leftparties;thereis a
than in a nationwhere a countermove
representation,
secure,leftwing government, with strongsupportfroman
ment isweak or nonexistent.
activewomen'smovementand frompublicopinionmore
Finally,supportforwomen's issuesinpublicopinion, broadly.In short,Case 1 relieson largenumbers ofelected
especiallyamong themass ofwomen, shouldprovidea women in themost auspiciouscontext,hypothetically
positivecontextfortheirsubstantive Fem
representation. resulting inhighlevels ofwomen'ssubstantive representa
inist
movements oftenmake claimsbasedonwomen'spub tion.
Althoughtheintersection of thesemeasuresofnum
licopiniontoencouragelegislators
andpoliticalpartiesto bersand contextis likelytobe rare,and data on someof
adopt theirpolicyagenda.5' the conditionsare not available,Scandinavianparlia
mentssincethe1960s comeclosest meet thesecriteria.By
H6: The more supportwomen'spolicyissueshave inpub theearly1970s,Denmark,Finland,Norwayand Sweden
lic opinion, thehigherwomen's substantive
representa had allcrossedtherepresentational thresholdof 15 percent

558 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
women in thenationalparliament.52 Therewere relatively ment and by shiftsinpublicopinion,were able toenact
highnumbersof leftwing women in parliament, strong or sustainlegislation concerningcivildivorce,abortion,
leftpartiesingovernment, andpopularsupportfor women's and employment opportunities forwomen.58
equality.In eachcase,women'ssubstantive representation Case 4 isa situation ofnumbers andparliament inwhich
was theresult; eachnationintroduced lawsregarding con thesheernumbers ofwomen arepresentand theleft party
traception and abortion, parentalleavepolicies,equalpay is in a strongposition ingovernment, with littleor no
forequalwork,and equal opportunities atwork,among supportincivilsociety. Althoughwe map thistheoretical
otherwomen-friendly policies,fromtheearly1960s to possibility, no empirical examplesareavailableamongpar
themid 1980s. liamentary democracies. The soleempirical exampleappears
Case 2 isone of fewwomen but excellentconditions. tobe thatof theformer Sovietbloc statesundercommu
In thiscase,ofpositive politicalcontext, a strongleftwing nism.In thesecases,women'sequal legislative representa
partygoverns with littleeffective oppositionbutwith the tionwas guaranteedas a principleof communism,in the
support(or insistence) of a feministmovementencourag context of a singlegoverning leftparty;thepercentages of
ingwomen-friendly policy-making, andofwomen-friendly women sittinginnationallegislatures in theSovietUnion
publicopinion. Women'ssubstantive representationislikely and in the Warsaw Pactnationswere consistently higher
to occurunder theseconditions,even the in absence of than in other nations. Strictly speaking, no civil society
sheernumbersofwomen. Examplesof Case 2 can be existed,independent of statecontrol,andhencethere was
foundin theU.S. Congress,theGermanBundestag,and no socialmovementsector, and no feminist or opposing
theBritishParliamentin the late 1960s to themid movements.In most of thesestates,in fact,women's
1970s.53In eachexample,althoughthere were fewelected descriptive representation declinedsharplyafterthecol
women, leftwing partiescontrolledthe legislature, and lapseofcommunism, inpartbecauseitwas not supported
feminist movements were activeand visible.As a result, by sustained women'sactivism.59 These examples,how
legislationconcerning women'sequalitysucceededduring ever,falloutsidetheestablished democracies whose poli
thisperiod.In theFederalRepublicofGermany, abortion cieswe seek to explainand addresslegislatures thathad
lawandmarriageand familylawwere reformed (1969 littletono policyimpact;in thisregard, thecase ismore a
1976). In theU.K., theLabour government introduced theoretical construct thanan empirically operative one.A
social reforms thatincludedthe legalization of divorce, relatively largepercentage ofwomen areunlikelyto serve
abortion,birthcontroland homosexuality, theestablish in a democratically electednationallegislature, leftwing
of of
ment thesystem comprehensive education, and health government notwithstanding, in theabsence of support
In
carereform.54theU.S., Congressapprovedfederal fund from civilsociety. Similarly,women-friendly publicpolicy
ingforfamily planning(1967,1970),proposedtheEqual isnot likelytooccur inparliamentary democraciesin the
Rights Amendment(1972),55andapproved nationalchild absenceof pressures fromcivilsociety.
carelegislation (1971).56Ineachof theseinstances, women's Case 5, thatof sheernumbers only,represents a theo
descriptive representation neverexceededsixpercent.57 reticalideal-type forwhich, again, thereisno empirical
Case 3 isdistinctive byvirtueof thepresence ofwomen example.In thiscase,strong descriptive representation of
in twoarenas:sheernumbersof electedwomenwithin women is counteredby unfavorable parliamentary and
parliament, andextra-parliamentary women infem
activist civilsocietycontexts. We includethiscase fortwo rea
movementincivilsociety.
inist In thiscaseofnumbers and sons.First,theabsenceof suchempiricalexamplescan
civilsociety,women areactiveand positioned withinand be attributed to thecausalconnection betweensheernum
outsidegovernment, in theabsenceof a leftwing govern bersand thecontextvariables.Favorablecivilsocietyand
ingparty.Strictly speaking, electedwomen arepresentin parliamentary contextsare themselvesimplicated with
relativelyhighnumbersonlywithin leftwing parties,for favorable numbersconditions.In theabsenceof a femi
tworeasons.First,no scholarlyliterature theorizes about nistmovement,supportive public opinion,or a strong
thestrong presenceof rightwing womenvis-a-vis women's left-party to adopt the agendaof women's descriptive
publicpolicyoutcomes.Second,leftwing excluded representation,
parties, women are not likelyto be elected in
fromgovernment, areunlikelytohold a largenumberof largenumbersto theassembly. This relationship between
seatsandhenceareunlikelytohaveelectedenoughmem numbersand politicalcontext variablesiswell-established
bers,letalone female members,toenablewomen'srepre in theliterature onwomen'sdescriptive representation.60
sentation within theentirelegislature to surpassthe 15 Second,Case 5 completesthemappingof casesacross
percentrepresentational threshold. This dualpositioning, thesevenfactors and, to thatextent,isan important theo
not
although ideal, may nonetheless be to
sufficient wrest retical construct. The theoretical contribution of its
somewomen-friendly publicpolicy from is
nationallegisla empiricalinvisibilitythat, at some level,descriptive rep
tures. For example,in Italyin thelate1960sand intothe resentation and substantive representation ofwomen are
early1980s,activist women in leftwing parliamentary par deeplyconnected. Itisempirically impossible,although theo
ties,supported by a growing andmilitantfeminist move conceivable,tOhavelargenumbers
retically ofwomen ina

September 2007 1Vol. 5/No. 3 559

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articles Sheer Numbers

legislature
wheretheotherconditionsfor women'ssubstan As suggestedin thediscussionof thecases,above,it is
tiverepresentationarehostile.At somelevel, womenactu unlikely that a subset of cases can be mapped to every
allyembodywomen's policyissues,because whenconditions combinationof sheernumbersand favorablecontexts.
arehostiletowomen'spublicpolicyissues, women arenot Nonetheless,we advance table2 as a site forplotting
electedtoparliaments.Case 5-of sheer numbers ofwomen more specific hypotheses and applyingcasedata.Among
ina politicallyhostilecontext-wouldbe theleastfavor thefiveideal-type caseswe discuss,only twoare likelyto
able,and leastlikely,caseforwomen-friendly publicpolicy. predict women'spolicysuccess:Cases 1 and 2, themaxi
Note thatone finalpossibility has been excludedfrom mumfacilitating conditions and positive politicalcontexts.
thistable:caseswherenoneof theconditions obtain.That An arrowdiagramof thesetwocases revealsthreepossi
is,nowomen sitinparliament, no leftpartiesgovern,there bilitiesof strongrelationships betweennumbersand
isno feminist
movement, andpublicopiniondoesnotsup contexts.
portwomen'sissues. We haveexcludedthissetof casesfor
which,whileperhapsobvious,deserve
severalreasons men ARROW DIAGRAM 1:DIRECT IMPACT OF CIVIL
tion.First,thereisno explicittheorizing in theliterature SOCIETY FACTORS ON WOMEN-FRIENDLY PUB
on women'spolicy issuesorwomen'selectiontoparlia LIC POLICY
mentconcerning outcomesregarding women'spublicpol
Civil SocietyContext *Women-Friendly PublicPolicy
icyunderthese unfavorable conditions.Itwouldbepossible
todo so, forexample,insofaras theseconditions might ARROW DIAGRAM 2 (Case 1):CIVIL SOCIETY FAC
map to fasciststates, which actuallylegislatein regardto TORS AS ANTECEDENT TO SHEER NUMBERS +
women'sissues butnot inthedirection we predictforfavor POLITICAL CONTEXT, WITH THAT COMBINA
ableconditions.61 Hypothesizing about theabsenceof the TION DIRECTLY EFFECTING WFPP
sevenfactorswe haveidentified intheliteratureandmapped
Civil SocietyContextm SheerNumbersofWomen
intable2would require something otherthana simplestate
mentofnullhypothesis; a different
setof theorizing,
which + PoliticalContext
no one hasyetundertaken, would be required.62 Second, Women-Friendly PublicPolicy
theabsenceof such theorizing would nonethelessrequire
thedevelopment of anothersetof hypotheses suggesting ARROW DIAGRAM 3 (Case2): CIVIL SOCIETY FAC
thattheabsenceof any facilitating conditions would not TORS AS ANTECEDENT TO POLITICAL CON
predictanabsenceofwomen's publicpolicy, butwould pre TEXT ALONE,WHICH THEN DIRECTLY EFFECTS
dictinstead awaveofanti-women/anti-feminist publicpol WFPP
icyaroundissues ofabortion,reproduction, marriage,rape, Civil SocietyContextm PoliticalContext
wife-beating, purdah,and othersimilarissues,a project Women-Friendly PublicPolicy
beyondthescopeof thispaper.
In these models,we understand civilsocietycontextas
antecedent both to sheernumbersand to politicalcon
Conclusions text; we alsomodel civilsocietyas directly affecting pub
We havedevelopeda theoretical model thatisparsimoni lic policy.Two major paths, therefore, predictwomen's
ous, testablein all democracies, consistent with existing publicpolicyoutcomes:one directlythroughcivilsoci
empiricalresearch, and supportedby theavailabletheo ety, where thepressurefrompublicopinionand feminist
rizing.There aremany othervariablessuggestedin the movements,in theabsenceofopposition, persuade(male)
literaturesuch as incumbency and newness,levelsand legislators to act forwomen (Case 2); and a second path
capacityof partydiscipline,ethnicand racialdiversity, indirectly fromcivil societythroughcombinationsof
access to leadership and legislativecommittee positions, agencyrepresented by favorable politicalconditionsof
attitudinal diversityamongthewomen and their willing leftwing governments with little oppositionor sheernum
ness to representotherwomen,politicaland electoralsta bers of women (Case 1).
bility,institutional newness,design,and culture,the As data aremapped to these diagrams and to the table,
presenceor absenceofwomen'spolicymachineryin the competingimpacts concerning sheernumbersand critical
formof an executive agencydedicatedtowomen'sissues, mass explanations should become apparent. A criticalmass
thatmight provideadditionalexplanationforwomen's impact,relying on a percentagethreshold, shouldemerge
public policyenactment.63 The seven we have selected as a sudden or dramatic increase inwomen's public policy;
are,however, most likelytoprovidethestrongest analyt as thepercentage of elected women increases fromone elec
ical leverage
on thequestionof critical mass and on the tionto thenext(inparliamentas awhole or simplyin a
relationshipbetweennumbersofwomen (descriptive rep major leftwing party),a stepchangeshouldnotbe evident
resentation)andwomen-friendly policyoutcomes(sub until thepercentage of elected womenmeetsor surpasses
stantiverepresentation). the critical
representational threshold.Incontrast, theimpact

560 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
ofsheernumbers canbe evincedintwo ways:withina nation 15 See Mohanty, Russo, and Torres 1991; Lovenduski
acrosstime,or acrossnations.First,as thepercentage of and Norris 1993;W?gnerud 2000; Youngs, Jones,
women ina singlenation'slegislatureincreases
acrosstime, and Pettman 1999.
thereshouldbe a concomitant increaseinwomen'spublic 16 See Mohanty, Russo, and Torres 1991.
policy.Second,amongnationswith a highpercentage of 17 See Baldez 2002; Luker 1984;Waylen 1996.
publicpolicyinwomen'sinter
women intheirlegislatures, 18 Carroll 1992, 28; 1994, 15.
estsshouldbehigh;nationswitha smallpercentageofelected 19 Lovenduski 2001, 745.
womenshouldhavelittlesuccessinlegislating onwomen's 20 Hoskyns 1996, 6.
policyissues.
Similarly,
women-friendly publicpolicyisnot 21 Thomas 1994, 7.
likelytooccurinparliamentary democracies intheabsence 22 Bratton 2002, 123.
ofpressuresfromcivilsociety,
particularlyintheabsenceof 23 Bystydzienski 1995, 72.
an organizedfeministmovement.64 24 Heath, Schwindt-Bayer, and Taylor-Robinson 2005,
When and underwhat conditionsdoeswomen'ssub 421.
stantiverepresentationoccur?Drawingfromthecompar 25 Swers 2002, 34.
onwomen-friendly
ativeliterature publicpolicy,we have 26 Ibid.
proposeda means to predictwomen's representation in 27 W?gnerud 2000; Jones 1990; Molyneux 1985.
policy terms.In doing so,we taketheclaimsof critical 28 Hernes 1987.
mass theory and contextualizetheminotherexplanatory 29 Gardiner and Leijenaar 1997, 61.
variables,showingthelimitsof thesheernumbersclaim 30 Borchorst 1994.
but demonstrating a way to approachthe relationship 31 Lovenduski 1986; especially 207-208.
between women'sdescriptive and substantive representa 32 Gelb and Palley1996,6.
tion.Our purpose is to advance the theoretical discus 33 Mazur 2002, 30-31.
sionof thelinkbetweenwomen'selectedrepresentation 34 See Baldez 2002; Beckwith 1996, 2000.
and women-friendly public policybeyond the idea of 35 Beckwith 2000, 437.
mass and open up avenuesforfurther
critical research 36 See, e.g., Conover 1988.
intohowwomen's interests are advanced in thepolicy 37 See Conover and Sapiro 1993; Guenther 2003.
process. 38 Reingold2000.
39 Note that there are also women's organizations that
are not
expressly feminist but that have mobilized in
Notes support ofwomen-friendly public policies concern
1 Dahlerup 1988. ing childcare and maternal benefits, among other
2 Flammang 1985; Saint-Germain 1989; Skard and policies. These include theNational Council of
Haavio-Mannila 1985. See interalia Flammang Women (Ireland; see Connolly 2002, 254, n6); the
1985; Thomas 1991; Thomas andWelch 1991; Committee ofMothers (Russia; seeWhite 2000);
Yoder 1991. and housewives' associations in theNordic countries
3 Sarah Childs and Mona Lee Krook (2005) examine (Dahlerup and Gulli 1985); see also theWomen's
the implications and limitations of criticalmass Alliance (Iceland; Sigurdjarnard?ttir 1998), the
theory as well as the popular history of the concept. Women's Cooperative Guild of England, theUnion
4 See inter alia Dahlerup 2005; Studlar and Matland f?minisme civique et sociale of France and the League
1996; Thomas 1991. ofWomen Voters in theU.S. (Black, 1989).
5 Childs 2001b;Cowley andChilds 2003, 2001; 40 See inter alia Bystydzienski 1992, Dahlerup 1988,
Dahlerup 1988; Lovenduski 2001. Darcy, Welch and Clark 1994, Haavio-Mannila
6 See Crowley 2004. 1985, Kathlene 1998, Mansbridge 1999, 2005;
7 Grey 2002. Swers 2002, Thomas andWilcox 1998.
8 Dahlerup 1988.We thankDrude Dahlerup for 41 Kanter 1977, 966, 972-984.
to our attention. 42 Although Kanter hypothesized solely about women
drawing this example
9 SeeCarroll2001; Childs 2001a; Childs andKrook in her work, she suggested that a similar logicwould
2005; Dahlerup 1988. apply to racialminorities in corporate settings
10 See Crowley 2004; Childs and Krook 2005; Dahl (1977,966).
erup 1988, 287; Grey 2006; Tremblay 2005. 43 See Studlar and McAllister 2002, 235-238; Bystydz
11 See Beckwith 2000; Mohanty, Russo, and Torres ienski 1992, 15.
1991; Molyneux 2003, 1985. 44 "Women inNational Parliaments," as ofOctober
12 Phillips1995, 1998. 30, 2005. www.ipu.org/wmn-e/classif.htm
13 Mohanty 2003; Weldon 2006. 45 Nations include Costa Rica, Spain Cuba, theNeth
14 Basu 1995. erlands, Denmark, Finland, Norway, Sweden, and

September 2007 1Vol. 5/No. 3 561

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
Articies I Sheer Numbers

Rwanda, only two ofwhich (Sweden and Rwanda) only twice: in 1976 (17.6 percent) and 1983 (20.3
meet a 40 percent representational threshold.Note
percent).
that eliminating the Scandinavian countries leaves 59 See Matland and Montgomery 2003; Nechemias
only 5 nations in the 35 percent group, www.ipu. 1994.
org/wmn-e/classif.htm 60 See Katzenstein and Mueller 1987; Kittilson 1999;
46 See, e.g., Banaszak, Beckwith, and Rucht 2003; Bystydzienski 1988; Gelb 1989; Studlar and Mat
Bashevkin 1985; Katzenstein and Mueller 1987; land 1996; Lovenduski and Norris 1993; Norris
Lovenduski and Norris 2002; Nelson and 1993.
Chowdhury1994;Threlfall1996. 61 See, for example, fascist ItalyunderMussolini and the
47 Beckwith 1987; Staggenborg 1991, but seeHtun development of rape provisions in theCodice Rocco.
2003; Jenson 1987;Weldon 2002. 62 See, however, Htun 2003.
48 Any categorization of parties as leftwing is time- and 63 SeeChilds 2001a; Beckwith2007; Jeydel
andTaylor
nation-specific; we do not attempt such classification 2003; Bratton 2002; Swers 2002; Cowell-Meyers
here. For an attempt to identify left-and right-wing 2003, 2001; Dobrowolsky 2003; Dodson 2006;
parties, see Gabel and Huber 2000. Considine and Deutchman 1996; Kathlene 1994;
49 Baldez 2002; Basu 1995; Banaszak, Beckwith, and Kenney 1996; Rosenthal 1998; Mazur 2002. Note
Rucht 2003; Howell and Mulligan 2006; Htun that our project focuses on explaining women's
2003; Katzenstein and Mueller 1987; Mueller 1988; substantive
representation,
not women's
descriptive
Threlfall1996;
Weldon 2002, 2006. representation or the election of numbers ofwomen,
50 See Beckwith 2002, 2003; Studlar and McAllister which would involve a list of antecedent variables
2002; Luker 1984; Mansbridge 1986; Baldez 2002; well-established in the literature. See, e.g., Carroll
Buechler 1990, Meyer and Staggenborg 1996; Tar 1994; Darcy, Welch, and Clark 1994; Matland
row 1998. 1998; Lovenduski and Norris 1993;Welch and
51 Mueller 1988, Mansbridge 1986. Studlar 1990, 1996.
52 See Skard and Haavio-Mannila 1985, table 4.3, 64 SeeWeldon 2002; Chappell 2002.
62-63. In Finland, women accounted for at least 17
percent of all parliamentary seats by 1966.
53 In theU.S., theDemocrats lost control of the execu References
tive in 1968 in the context of the ongoing Vietnam s
Baldez, Lisa. 2002. Why Women Protest:Women Move
War, which though divisive, did not divert public ments in Chile.
Cambridge: Cambridge University
attention from other social reform movements or Press.

martialize civil society. Banaszak, Lee Ann. 1996. WhyMovements Succeed or

54 Note that the British example is not a perfect in Fail: Opportunity, Culture, and theStrugglefor Woman
stance of this case as therewas
significant and grow Suffrage.Princeton: Princeton University Press.
Conservative to the Labour Banaszak, Lee Ann, Karen Beckwith, and Dieter Rucht,
ing opposition
governments led byHarold Wilson. eds. 2003. Womens Movements Facing theReconfigured
55 The ERA never received sufficient support from the State. Cambridge: Cambridge University Press.
states to achieve ratification. Bashevkin, Sylvia. 1985. Women and Politics inWestern
56 President Nixon vetoed the legislation,which Con Europe. London: Frank Cass.
was unable to override. -. 1996a. common Feminists,
gress Losing ground:
57 The U.S. Congress averaged less than 3 percent conservatives and public policy inCanada during the
women from 1965-1973; the British Parliament Mulroney years. Canadian Journal ofPolitical Science
averaged approximately 4 percent from 1964 to 29 (2): 211-43.
1974; and the FRG's Bundestag mean representation -. 1996b. Tough times in review:The British
was approximately 6 percent from 1969 to 1976. womens movement the Thatcher years. Com
during
58 Beckwith 1987; d?lia Porta 2003. Note, however, parative Political Studies 2% (1): 525-52.
that women's
parliamentary representation within Basu, Amrita, ed. 1995. The Challenge ofLocal
leftparties varied by party.The Italian Communist Feminisms. Boulder: Westview Press.
Party (PCI) consistently elected a higher percentage Beckwith, Karen. 1987. Response to feminism in the
ofwomen than did the Italian Socialist Partywhich, Italian Parliament: Divorce, abortion, and sexual
from 1963 through 1979, never managed to elect violence legislation. In The Womens Movements of
more than one woman (for a
"high" of 1.8 percent Western Europe and theUnited States: Consciousness,
of the PSI delegation). In addition, from 1968 to Political Opportunity, and Public Policy, ed. Mary
1992, women in the PCI delegation met or ex Fainsod Katzenstein and Carol McClurg Mueller.
ceeded the 15 percent representational threshold Philadelphia: Temple University Press.

562 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
-. 1996. Lancashire women against pit closures:
Women's standing in a men's movement. Signs 21 (4):
1034-68.
-. 2000. Beyond compare?Women's movements in -.
comparative perspective. European Journal ofPolitical
Research 37 (4): 431-68.
-. 2007. Numbers and newness: The descriptive Cowley, Philip, and Sarah Childs. 2001. "Critical but
and substantive representation ofwomen. Canadian
Journal ofPolitical Science 40 (1): 27-49.
Black, Naomi. 1989. Social Feminism. Ithaca, NY: Cor
nell University Press.
Borchorst, Anette. 1994. The Scandinavian welfare
sates: Patriarchal, gender-neutral or woman-friendly?
International Journal ofContemporary Sociology 1 (8):
3-21.
Bratton, Kathleen. 2002. The effectof legislative diver
sityon agenda setting: Evidence from six state legisla
tures.American Politics Research 30 (2): 115-42.
Buechler, StevenM. 1990. Women'sMovements in the
United States.New Brunswick and London: Rutgers
University Press.
Bystydzienski, Jill. 1988.Women in politics inNorway.
Women & Politics 8: 73-95.
-, ed. 1992. Women TransformingPolitics: World
wide Strategiesfor Empowerment. Bloomington: Indi
ana University Press.
-. 1995. Women inElectoral Politics: Lessonsfrom
Norway. Westport, CT: Praeger.
Carroll, Susan. 1994. Women as Candidates inAmerican
Politics. Bloomington, IN: Indiana University Press.
-. 2001. The Impact ofWomen inPublic Office.
Bloomington, IN: Indiana University Press.
Chappell, Louise. 2002. Gendering Government: Feminist
Engagement with the State inAustralia and Canada.
Vancouver: University of British Columbia Press.
Childs, Sarah. 2001a. Attitudinally feminist'? The new
Labour women MPs and the substantive representa
tion ofwomen. Politics 21 (3): 178-85.
-. 2001b. In their own words: New Labour
and the substantive representation ofwomen. British
Journal ofPolitics and International Relations 3 (2):
173-90.
Childs, Sarah, and Mona Lee Krook. 2005. "The Sub
stantiveRepresentation ofWomen: Rethinking the
'CriticalMass' Debate." Presented at theAmerican
Political Science Association meetings, Washington,
DC, September 1-4.
Conover, Pamela. 1988. Feminists and the gender gap.
Journal ofPolitics 50 (4): 985-1010.
Conover, Pamela, and Virginia Sapiro. 1993. Gender,
feminist consciousness, and war. American Journal of
Political Science 37 (4): 1079-100.
Considine, Mark, and Iva Deutchman. 1996. Instituting
gender: State representation inAustralia and theU.S.
Women & Politics 16 (4): 1-19.
This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM
All use subject to JSTOR Terms and Conditions
Cowell-Meyers, Kimberly. 2001. Gender, power and
peace: A preliminary look atwomen in theNorthern
Ireland Assembly. Women & Politics 23 (3): 55-88.
2003. Women inNorthern Ireland: Gender and
the politics of peace-building in theNew Legislative
Assembly. Irish Political Studies 18 (1): 72-97.
not Rebellious? New Labour Women in theHouse of
Commons." Presented at theAmerican Political Sci
ence Association onWomen and
Special Session
Politics, San Francisco, August 29.
Cowley, Philip, and Sarah Childs. 2003. Too spineless to
rebel?New Labour's women MPs. British Journal of
Political Science 33(3): 345-65.
Crowley, Jocelyn. 2004. When tokensmatter. Legislative
Studies Quarterly 24 (1): 109-36.
Dahlerup, Drude. 1988. From a small to a largeminor
ity.Scandinavian Political Studies 11 (4): 275-98.
-. 2005. "The Theory of a 'CriticalMass' Revis
ited." Presented at theAnnual Meeting of theAmeri
can Political Science Association,
Washington, DC,
September 1.
Dahlerup, Drude, and Brita Gulli. 1985.Women's
organizations in theNordic countries: Lack of force
or counterforce? In
Unfinished Democracy: Women in
Nordic Politics, ed. Elina Haavio-Mannila. Oxford
and New York: Pergamon Press.
Darcy, R., SusanWelch, and JanetClark. 1994. Women,
Elections and Representation. 2d ed. Lincoln: Univer
sityofNebraska Press.
d?lia Porta, Donatella. 2003. The women's movement,
the Left, and the state:Continuities and changes in
the Italian case. InWomens Movements Facing the
Reconfigured State, ed. Lee Ann Banaszak, Karen
Beckwith, and Dieter Rucht. Cambridge: Cambridge
University Press.
Dobrowolsky, Alexandra. 2003. Shifting states:Women's
constitutional
organizing
across time and space. In
women Womens Movements Facing theReconfigured State, ed.
Lee Ann Banaszak, Karen Beckwith, and Dieter
Rucht. Cambridge: Cambridge University Press.
Dodson, Debra L. 2006. The Impact ofWomen in Con
gress.Oxford: Oxford University Press.
Flammang, Janet. 1985. Female officials in the feminist
capitol: The case of Santa Clara County. Western
Political Quarterly 38: 94-118.
Gabel, Matthew J., and JohnD. Huber. 2000. Putting
parties in their place: Inferringparty left-rightideo
logical positions from partymanifestos data. American
Journal ofPolitical Science 44 (1): 94-103.
Gardiner, Frances, and Monique Leijenaar. 1997. The
timid and the bold: An analysis of thewomen
friendly-state in Ireland and theNetherlands. In Sex
Equality Policy inWestern Europe, ed. F. Gardiner.
London: Routledge Press.
September 2007 1Vol. 5/No. 3 563
 Artilcles I Sheer Numbers

Gelb, Joyce. 1989. Feminism and Politics. Berkeley: and Western Europe. Philadelphia: Temple University
University of California Press. Press.

Gelb, Joyce, and Marian Palley. 1996. Women and Public
Policies: Reassessing Gender Politics. Charlottesville:
University Press ofVirginia.
Grey, Sandra. 2002. Does sizematter? Critical mass and
New Zealand's women MPs. Parliamentary Affairs 55:
19-29.
-. "Numbers and Beyond: The Relevance of
2006.
Critical Mass inGender Research." Presented at the
Annual Meeting of theAmerican Political Science
Association, Washington, DC, September 1-4.
Guenther, Katja. 2003. A case study of local feminist
mobilization in Eastern Germany, 1990-2000. Jour
nal ofWomen's History 15 (3): 143-49.
Haavio-Mannila, Elina. 1985. Unfinished Democracy:
Women inNordic Politics. Oxford and New York:
Pergamon
Press.
Heath, Roseanna Michelle, Leslie Schwindt-Bayer, and
Michelle Taylor-Robinson. 2005. Women on the
sidelines:Women's representation on committees in
Latin American legislatures.American Journal ofPolit
ical Science 49 (2): 420-36.
Hernes, Helga Maria. 1987. Welfare State and Woman
Power: Essays in State Feminism. London: Norwegian
University Press.
Hoskyns, Catherine. 1996. IntegratingGender: Women,
Law and Politics in theEuropean Union. London:
Verso.
Howell, Jude and Diane Mulligan. 2006. Gender and
Civil Society. London: Routledge.
Htun, Mala. 2003. Sex and theState: Abortion, Divorce,
and theFamily Under Latin American Dictatorships
and Democracies. Cambridge: Cambridge University
Press.
Kenney, Sally. 1996. New research on gendered political
institutions.Political Research Quarterly 49: 445-66.
Kittilson, Miki Caul. 1999.Women's representation in
Parliament: The role of political parties. Party Politics
5 (1): 79-98.
Lovenduski, Joni. 1986. Women and European Politics:
Contemporary Feminism and Public Policy. Sussex:
Wheatsheaf.
-. 2001. Women and politics: Minority representa
tion or criticalmass? Parliamentary Affairs 54:743-58.
Lovenduski, Joni, and Pippa Norris. 1993. Gender and
Party Politics. London and Thousand Oaks: Sage.
-. 2002. Westminster women: The politics of pres
ence. Political Studies 51 (1): 84-103.
Luker, Kristin. 1984. Abortion and thePolitics ofMother
hood. Berkeley, CA: University of California Press.
Mansbridge, Jane. 1986. Why We Lost theERA. Chicago:
University of Chicago Press.
-. 1999. Should blacks represent blacks and women
representwomen? A contingent "Yes." Journal of
Politics61 (3): 628-57.
-.
problems: Essentialist dangers and
2005. Quota
potential safeguards. Politics & Gender 1 (4): 622-38.
Matland, Richard E. 1998.Women's Representation in
National Legislatures: Developed and Developing
Countries. Legislative Studies Quarterly 23 (1):
109-125.
Matland, Richard E,. and Kathleen A. Montgomery,
eds. 2003. Womens Access toPolitical Power inPost
Communist Europe. Oxford: Oxford University Press.
Matland, Richard E., and Donley Studlar. 1996. The
women candidates in
contagion of single-member
district and electoral
proportional representation

Jeydel,Alana, and Andrew J.Taylor. 2003. Are women systems:Canada and Norway. Journal ofPolitics 58
legislators less effective?Evidence from theU.S. (3): 707-34.
House in the 103rd-105th Congress. Political Research Mazur, Amy. 2002. Theorizing Feminist Policy. Oxford:
Quarterly 56 (1): 19-27. Oxford University Press.
Jones,Kathleen B. 1990. Citizenship in a woman Meyer, David S., and Suzanne Staggenborg. 1996.
friendlypolity. Signs 15 (4): 781-813. Movements, countermovements, and the structure of

Kanter, Rosabeth Moss. 1977. Some effectsof propor political opportunity. American Journal ofSociology
tions on group life: Skewed sex ratios and responses 101: 1628-60.
to tokenwomen. American Journal ofSociology 82 (5): Mohanty, Chandra. 2003. Feminism without Borders:
965-90. Decolonizing Theory, Practicing Solidarity. Durham,
Kathlene, Lyn. 1994. Power and influence in state legis NC: Duke University Press.
lative policy-making: The interaction of gender and Mohanty, Chandra, A. Russo, and L. Torres, eds. 1991.
position in committee hearing debates. American Third World Women and thePolitics ofFeminism.
Political Science Review 88 (3): 560-76. Bloomington, IN: Indiana University Press.
-. 1998. In a differentvoice:Women 1985. Mobilization without eman
and the policy Molyneux, Maxine.
process. InWomen and Elected Office: Past, Present and cipation?
Women's interests, the state, and revolution

Future, ed. Sue Thomas and Clyde Wilcox. New inNicaragua. Feminist Studies 11 (2): 227-54.
York: Oxford University Press. -. 2003. Womens Movements in International Per

Katzenstein, Mary Fainsod, and Carol McClurg Mueller, spective:Latin America and Beyond. London: Institute
eds. 1987. The Womens Movements of theUnited States of Latin American Studies.

564 Perspectives on Politics

This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM

All use subject to JSTOR Terms and Conditions
 Mueller, Carol McClurg, ed. 1988. The Politics of the
Gender Gap: The Social Construction ofPolitical Influ
ence.Newbury Park, CA:
Sage Publications.
Nechemias, Carol. 1994. Democratization and women's
access to seats:The Soviet case, 1989
legislative
1991. Women & Politics 14 (3): 1-19.
Nelson, Barbara J., and Najma Chowdhury, eds. 1994.
Women and PoliticsWorldwide. New Haven: Yale
University Press.
Norris, Pippa. 1993. Conclusions: Comparing legisla
tive recruitment. In Gender and Party Politics,
ed. Joni Lovenduski and Pippa Norris. London: Sage.
Phillips, Anne. 1995. The Politics ofPresence. Oxford:
Oxford University Press.
-. 1998. Democracy and representation: or,Why
should itmatter who our representatives are? In Femi
nism and Politics, ed. A. Phillips. Oxford: Oxford
University Press.
Reingold, Beth. 2000. RepresentingWomen. Chapel
Hill and London: University ofNorth Carolina Press.
Rosenthal, Cindy Simon. 1998. When Women Lead.
New York: Oxford.
Saint-Germain, Michelle. 1989. Does their difference
make a difference?The impact ofwomen on public
policy in theArizona legislature. Social Science Quar
70 (4): 956-68.
terly
Sigurdjarnard?ttir, Sighrudur Helga. 1998. On their
own
premises: The political project of the Icelandic
Women's Alliance. In Is There aNordic Feminism? ed.
Drude von de Fehr, Bente Rosenbeck, and Anna G.
J?nasd?ttir. London: UCL Press.
Skard, Torild, and Elina Haavio-Mannila. 1985.Women
in Parliament. In Unfinished Democracy: Women in
Nordic Politics, ed. Elina Haavio-Mannila. Oxford
and New York: Pergamon Press.
Staggenborg, Suzanne. 1991. The Pro-Choice Movement:
Organization and Activism in theAbortion Conflict.
New York and Oxford: Oxford University Press.
Studlar, Donley T. and Richard E. Matland. 1996. The
contagion ofwomen candidates in and multi
single
member district systems:Canada and Norway. Jour
nal ofPolitics58 (3): 707-33.
Studlar, Donley T, and IanMcAllister. 2002. Does a
criticalmass exist?A comparative analysis ofwomen's
legislative representation since 1950. European Journal
ofPolitical Research 41 (2): 233-53.
Swers,Mich?le L. 2002. The DifferenceWomen Make:
The Policy Impact ofWomen in Congress. Chicago:
University of Chicago Press.
Tarrow, Sidney. 1998. Power inMovement. 2d ed. Cam
bridge: Cambridge University Press.
This content downloaded from 185.2.32.106 on Wed, 25 Jun 2014 00:26:04 AM
All use subject to JSTOR Terms and Conditions
Sue. 1991. The women on state
Thomas, impact of
legislative policies. Journal ofPolitics 53 (4): 958-76.
-. 1994. How Women Legislate. Oxford: Oxford
University Press.
Thomas, Sue, and SusanWelch. 1991. The impact of
on activities and
gender priorities of state legislators.
Western Political Quarterly 44: 445-56.
Thomas, Sue, and Clyde Wilcox. 1998. Women and
Elected Office: Past, Present and Future. New York:
Oxford.
Threlfall, Monica, ed. 1996.Mapping theWomens Move
ment. London: Verso.
Tremblay, Manon. 2005. "UnderstandingWomen's
Descriptive and Substantive Representation: Critical
Mass or Surrogate Representation?" Presented at the
Annual Meeting of theAmerican Political Science
Association, Washington, DC, September 1-4.
Tremblay, Manon, and Rejean Pelletier. 2000. More
feminists or more women? Descriptive and substan
tive representations ofwomen in the 1997 Canadian
Federal Elections. International Political Science Re
view 21 (4): 381-405.
W?gnerud, Lena. 2000. Testing the politics of presence:
Women's representation in the Swedish Riksdag.
Scandinavian Political Studies 23 (1): 67-91.
Waylen, Georgina. 1996. Gender in Third World Politics.
Boulder: Lynne Rienner.
Welch, Susan, and DonleyT. Studlar. 1990. Multi
member districts and the representation ofwomen:
Evidence from Britain and theUnited States. Journal
ofPolitics52 (2): 391-412.
-. 1996. The opportunity structure forwomen's
candidacies and electability in Britain and theUnited
States. Political Research Quarterly 49 (4): 861-74.
Weldon, S. Laurel. 2002. Protest,Policy and theProblem
ofViolence againstWomen. Pittsburgh: University of
Pittsburgh Press.
-. 2006. The structure of intersectionality:A com
parative politics of gender. Politics & Gender 2 (2):
235-248.
White, Anne. 2000. "New Mothers' Campaigning
Organizations inRussia." In Feminisms and Womens
Movements in ContemporaryEurope, ed. Anna Bull,
Hanna Diamond, and Rosalind Marsh. New York:
St. Martin's Press.
Yoder, JaniceD. 1991. Rethinking tokenism: Looking
beyond numbers. Gender & Society 5 (2): 178-92.
Youngs, G., K.B. Jones, and J.J.Pettman. 1999. New
spaces, new politics: International feminist directions.
International Feminist Journal ofPolitics \ (1): 1-13.
September 2007 1Vol. 5/No. 3 565