Muslim support for secular democracy - Harvard University support for... · muslim support for...

MUSLIMSUPPORTFORSECULARDEMOCRACYNORRIS5/3/20119:41AM

1

Muslimsupportforseculardemocracy

PippaNorris

McGuireLecturerinComparativePolitics VisitingProfessorofGovernmentandIRKennedySchoolofGovernment TheUniversityofSydneyHarvardUniversity DepartmentofGovernmentCambridge,MA02138 NSW,[email protected] [email protected] www.arts.sydney.edu.au

Abstract:Stirringsofpopularunrest inTunisia,Egypt,Libya,Bahrain,YemenandSyriaduring theArabspringhavereneweddebateabouttheregimesmostlikelytoreplacetraditionalautocraciesin the region. This development raisesmany questions, in particular, (i) what types of regimescurrentlygovernMuslimplurality societies,and (ii)what typesof regimesdoes thepublic living inthesesocietiesdesire? Inparticular,doMuslimpublicswantseculardemocracies,characterizedbyreligiouspluralism,separationofreligiousandgovernmentauthorities,toleranceofdiversereligiousbeliefsandpractices,moderateparties,andconstitutionswherenosingletypeoffaithisprivileged,following the Turkish model? Or do they prefer regimes where constitutions reflect Islamicprinciples, spiritual authorities predominate in politics, conservative Islamic parties prevail, andreligiouscourtsenforceSharialaw,reflectingtheIranianmodel?Or,alternatively,domanyMuslimpublicsfallsomewherebetweenthesetwopoles?

Toexplorethese issues,PartIoutlinestheconceptualframework.PartIIexaminesthedistributionof the typeof regimescurrentlygoverningMuslimplurality societies.Part IIIdescribes the surveymeasuresused to gauge popular support for each typeof regime. This study examines attitudesderived from theWorld Values Survey (19952007) conducted in 83 countries, including twentydiverseMuslimplurality societies.Part IVpresents the resultsof themultilevel analysis ofpublicregime preferences. The conclusion in Part V summarizes the key findings and considers theirimplications.

Keywords:Religion,secularization,regimetypologies,publicopinion,democracyanddemocratization

Paper for Spirited Voices from the Muslim World: Islam, Democracy and Gender Rights theUniversityofSydney,2930thApril2011.


2

TheArabspring,markedbycontagiouspopularunrestinTunisia,Libya,Egypt,Bahrain,YemenandSyria, has renewed speculation about the type of regime most likely to replace traditionalautocracies in theMiddleEastandNorthAfrica.Theseevents raisemanyquestions, inparticular,what typesof regimescurrentlygovernMuslimpredominantsocieties?Andwhattypesof regimesdoes the public in these societies desire? In particular, is there widespread support for seculardemocracies, characterized by religious pluralism where no one type of faith is constitutionallyprivileged,diversetypesofreligiousexpressionandpracticesarewidelytolerated,and Islamism isonefaithamongmany?HistoricallyTurkeyhaslongprovidedamodelforthistypeofstate,followingtheAtatrkreformsestablishingamodern,democratic,secularstate,wheretheconstitutionneitherrecognizes nor promotes any official religion, religious parties are banned, and faithbaseddiscrimination is prohibited. 1Or doMuslim publics prefer regimeswhere constitutions privilegeIslam as the official faith, states are governed by clergy, conservative parties promote strictinterpretationsof IslamandreligiouscourtsenforceSharia law,reflectingthemodelofthe IslamicRepublicof Iran?2Or,alternatively,domanyMuslimsocieties fallsomewherebetween these twopoles?

DebateaboutthecompatibilityofIslamwithdemocracyhascontinuedovermanydecades.SomeconcurwithHuntingtonspredictionofirreconcilabletensions.3Thisclaimhasbeenregardedasplausiblegiven thepersistenceofautocraticstates in theMiddleEastandNorthAfricadespiteprevious waves of democratization occurring elsewhere in the world.4 Many others challengeHuntingtonsview,however,believingthatthereisnoinevitableconflictbetweenIslamicteachingsand the principles of liberal democracy, religious pluralism, and social tolerance. 5 Scholars havequestionedtheexistenceofasingleIslamicculturestretchingallthewayfromJakartatoLagos,letaloneoneholdingvaluesdeeplyincompatiblewithseculardemocracy.6

Onereasonwhydebateremainsunresolvedisthatmanyscholarshavereliedprimarilyuponqualitative sources as indirect evidence used to interpret religious cultures, including personalinterviews, observations and direct experience, and traditional textual exegesis of the literature,religiousscriptures,andhistoricaldocuments.7 Yetrecentyearshaveseensubstantialadvances inthe empirical evidence available to analyze mass public opinion directly, as surveys based onrepresentative samples of the general population have expanded substantially in geographiccoverageandscope.8Thishasfacilitatedagrowing literaturecomparingtheempiricalevidenceforpublicopinion inmany (althoughnot all)predominatelyMuslim societies in theMiddle East andNorthAfrica,SouthEastAsiaandCentralAsia.ThisbodyofattitudinalresearchhasdrawnupontheresourcesoftheGallupWorldPoll,9aswellastheWorldValuesSurvey,10thePewGlobalAttitudesproject,11 Zogby International's 2002 Arab Values Survey,12 and the Afrobarometer and Arabbarometersurveys.13TheexpandingbodyofliteratureutilizingsurveydatahasexploredtheimpactofMuslimidentitiesupondiversetypesofattitudes,includingthestrengthofreligiosity,supportfordemocracy, feelings of nationalism, attitudes towards foreign policy, and orientations towardsgenderequality.ThisgrowingresearchseekingtounderstandMuslimpublicopinionrepresentsan0~ ^H =#~J =JJJ : = = = = = J / / / / re undertheorized.Moreoverthecomplexconceptofseculardemocracyneedstobeunpackedmoreclearlyasadistinctregimetype,alongwithitsalternatives.

Accordingly, building upon the previous research literature, Part I summarizes theconceptual framework anddiscusses thenotionof seculardemocracy as a regime type. Part IIapplies this framework toclassifycontemporary regimesaround theworldand tounderstand thedistributionofregimesinMuslimpluralitysocieties.Toexplorewhatregimesareactuallydesiredbycitizens inthesesocieties,Part IIIdescribesthesurveymeasuresmonitoringpublicsupportfortheprinciplesofseculardemocracy.Data isderivedfromthepooledWorldValuesSurvey(19952007)conducted in83countries.Part IVpresentstheresultsofthemultilevelanalysisofpublicopinion.TheconclusioninPartVsummarizesthekeyresultsandconsiderstheirimplications.


3

I:Theconceptualframework

As a preliminary step it is important to clarify the conceptual framework and what isunderstoodbythenotionofseculardemocracy.Figure1illustratestheregimetypologyusedinthisstudywhichemphasizestheinteractionoftwodimensions:liberaldemocracyandreligiousfreedom.

[Figure1abouthere]

Liberaldemocracyandautocracy

Onthehorizontaldimensionillustratedinthisfigure,statescanbeclassifiedonacontinuousscale,andasbinarycategories,accordingtohowfarconstitutionsmeettheprinciplesandpracticesofliberaldemocracy.Thecomplexnotionofdemocracyisopentomultipleconceptualizationsandtheemploymentofmanyalternativemodifiers,includingparticipatory,liberal,social,directanddeliberativeforms.14Democracy isunderstoodhere,mostsimply,tomeanthecapacityofpeopletoinfluenceregimeauthoritieswithintheirnationstate.IntheSchumpeteriantradition,minimalistnotions of representative democracy focus upon the provision of competitive elections as theprimarymechanisms guaranteeing the accountability of officials to citizens. Electoral democracyrequires leaders tobe recruitedon thebasisof freeand fairmultipartycontestsat theballotboxwhichmeetinternationalstandardsofintegrity.Neverthelessthisminimalistdefinitionfailstotakeaccountofthemanyotherinstitutionsrequiredtoensurethatdemocraticelectionsworkeffectively,includingtheprovisionoffreedomofexpressionandrespectforcivil liberties,checksandbalancesamongthecoreregime institutions,an independent judiciaryandaneffective legislature.Withoutthesesafeguards,andmanyothers,electoralmanipulationandfraudcanflourish.15

Thereforethisstudyadoptsabroaderunderstandingofliberaldemocracywhichreflectsthelong tradition of classical political thought from Hobbes and Locke to Bentham and J.S. Mill.Following Robert Dahls conception, liberal democracy rests upon the triple principles ofcontestation,participation,andhuman rights.16 Inpractice,Dahlsuggestedthat liberaldemocraticregimes (or polyarchies) canbe identifiedby thepresenceofcertainkeypolitical institutions:1)elected officials; 2) free and fair elections; 3) inclusive suffrage; 4) the right to run for office; 5)freedom of expression; 6) alternative information; and 7) associational autonomy.17 Liberaldemocraciesusecompetitivemultipartyelectionsto filloffices forthenational legislatureandthechief executive. Yet competitive elections alone, even if conducted according to internationalstandards of electoral integrity, are insufficient to establish liberal democratic states withoutfreedomofexpression,theavailabilityofalternativesourcesofinformation(freedomofthemedia),andassociationalautonomy (freedom toorganizeparties, interestgroupsand socialmovements).Liberaldemocracy includes freedomofworship and toleranceof religiouspluralism asbasic civilliberties,althoughtheseareonlyoneaspectofamuchbroaderphenomenon.

ThestandardindicatorscommonlyusedtomeasureandcomparelevelsofliberaldemocracyincludePolity IVs scaleofdemocracyautocracy (from18002007)and theFreedomHouseGastilindexofpoliticalrightsandcivil liberties(from1972date).18Theemphasisonawiderangeofcivilliberties, rights, and freedoms in the FreedomHouse indexmeans that thismost closely reflectsnotionsofliberaldemocracyandithasbeenwidelyemployedbymanycomparativescholars.19Theindexmonitors theexistenceofpolitical rights in termsofelectoralprocesses,politicalpluralism,and the functioning of government. Civil liberties and political rights are defined by expertassessmentsofeachcountry,suchas levelsoffreedomofspeechandassociation,ruleof law,andpersonal rights. Each item is allocated a score from0 to4 and each is given equalweightwhenaggregated.Therawscoresforeachcountryarethenconvertedintoasevenpointscaleofpoliticalrightsandasevenpointscale forcivil liberties,and inturn thesearecollapsed tocategorizeeachregimeworldwideaseither free, partly free,or not free.Asa resultof thisprocess,FreedomHouseestimatethat in2009outof193nations,almosthalfor89(46%)couldbeclassifiedasfreewhile 58 nations could be classified as partly free.20 The index has the advantage ofprovidingcomprehensive coverage of nationstates and independent territories worldwide, as well as


4

establishinga long timeseriesofobservationsconductedannuallysince1972. TheGastil index isstandardizedtoa100pointcontinuousscaleforeaseofinterpretationinthisstudyandtheindexisalsodichotomizedaroundthemean,toclassifyregimesaseitherliberaldemocraciesorautocracies.

Secularandreligiousstates

Theconceptof secularstates isequallycomplex,as theverticaldimensionofthe regimetypology, along with related but distinct concepts of religious states, theocracy and statereligion.21 Both religionand secularizationaremultidimensionalphenomenaand it is importanttorecognizeanimportantdistinctionconcerningtheroleofreligionatthreelevels:(i)atmacrolevelin the state, (ii) atmezzolevel as a societal institution, and (iii) atmicrolevel, as an individualpractice.22Hence societieswhere thepublicdisplays intensepersonal faithand spiritualpracticesmay still be governed by secular constitutions. As with democracy, it is somewhat easier toconceptualize and define secular states than to identify the precise antithesis. The Americanconstitutionrecognizesaclearseparationofchurchandstate,butinmanyothercountriestherearegradations in the exact division and role of church and state. For example, many Europeandemocracies display the vestiges of religion in constitutional practices, including recognizing anestablishedstatereligion,whilefollowingmulticulturalpolicieswhichtolerateandrespectalltypesoffaith.HenceBritain isoneofthemoststrikinglysecularsocieties intheworldtoday intermsofmassobservanceofreligiouspracticesandtheexpressionofreligiousvalues.23Insuccessivedecadessincethe1960s,theBritishgovernmenthasofficiallyendorsedpoliciesofmulticulturalism,valuingdiversity in religious identities and practices, for example by funding faithbased schools forCatholics,Protestants,Jews,Muslimsandotherreligions.Neverthelessduetothelegacyofhistoricaltraditions, the Church of England remains the established religion in England, the constitutionalmonarch is required touphold theProtestant faith,dailyactsofworshipare required inall stateschools,andtheHouseofLordsstillcontainmorethantwodozenLordsSpiritual.

Despitethesecomplexities,secularstatesareunderstoodanddefinedhereasthosewhichareneutralor impartial towardsalldifferent typesof religion, forexampleasevidencedbystatespracticing religious pluralism and multicultural policies, separation of spiritual and governmentauthorities, tolerance of diverse religious beliefs and practices, with constitutions, laws, andregulatoryframeworkswherenosingletypeoffaithisfavoredoverothers,orindeedovernofaith.In secular democracies, legitimacy and authority derives from the ballot box. Themost secularstates follow policies of multiculturalism, valuing diversity, mutual respect, coexistence, andtoleranceofalltypesofreligious identitiesandpractices.Multiculturalpoliciesdiffersharply fromthose seeking assimilation or integration. The rise of secular states can be regarded as closelyintertwined with processes of democratization, although in Western Europe the expansion ofreligiousfreedomandtoleranceprecededthedevelopmentoftheinstitutionsof liberaldemocracyandtheexpansionofthemassfranchisebymanycenturies.

Followingasimilarlogic,ingeneralitisproposedtoconceptualizeanddefinenonsecularorreligiousstatesasthosewherethestateisnotneutraltowardsalltypesofreligion,exemplifiedbymultiplepractices,suchaswhereonetypeoffaith isprivileged inthenationalconstitution,thestate regulates or restricts freedom of worship or religious expression for some types ofdenominations, cults or sects, a single official state religion is recognized and favored, spiritualauthorities are closely related to governing authorities, the state appoints or approves religiousleaders, thestatesubsidizessome religions,thestatemandatescompulsory religiouseducation inonefaithinpublicschools,courtsseektointerpretandenforcereligiousprecepts,and,inthemostextremecaseoftheocracy,thegovernmentclaimstoruleonbehalfofGodorahigherpower,notderivingitslegitimacyandauthorityfromtheballotbox.Theocracyisthereforetreatedasasubsetand only one extreme type of religious state, not necessarily representative ofmoremoderatevariants.Aswithotherattemptsatregimeclassification,thedistinctionbetweensecularandnonsecular states can be treated as a binary division or, more realistically and accurately, as acontinuousscalerangingfromthemostseculartothemostreligiousstate.


5

To distinguish and classify the practices associated with secular and nonsecular statesaroundtheworld,thisstudydrawsupontheReligiousFreedomIndexdevelopedearlierforSacredandSecular.24This is constructedbasedon information foreach country contained in theUnitedStates State Department report on International Religious Freedom, 2002, a comprehensivecomparisonofstateregulationandrestrictionsofallworldfaiths.25Thescalesoughttoreplicatethemethodologyandexpandupon thecountrycoverageofferedby the1992ChavesandCann indexused inpreviousstudies tomeasure state regulation.26 The IndexofReligiousFreedom inSacredandSecular focusesupontherelationshipofthestateandreligiousorganizations, including issuessuchaswhethertheconstitutionlimitsfreedomofreligion,whetherthegovernmentrestrictssomedenominations,cultsor sects,andwhether thestate recognizesand favorsanestablishedchurch(seeAppendixA). All theseare features conceptualizedasdistinguishing secularandnonsecularstates.TheindexwasclassifiedaccordingtothecriterialistedinSacredandSecular,witheachitemcoded0/1.Thetwentypointscalewas then reversedso thatahigherscoreonthe20pointscalerepresentsgreatersecularity.27Theclassificationandmeasurementallowsustooperationalizetheregimetypologypresentedschematically inFigure1,combiningthe indicesofdemocratizationandsecularism,tocomparenationstatesworldwide.

II:Thedistributionofregimesworldwide

Seculardemocracies

Theresultsofapplyingthe indicesacrossbothdimensions,as illustrated inFigure2,showthattodaymanycountriescanbeclassifiedasseculardemocracies.Regimeslocatedinthetoprightquadrantofthescattergrampresented inFigure2arecharacterizedbycivilandpolitical libertiesand a state which is neutral towards different types of faith, tolerating religious pluralism andmulticulturalism. This includesmany established liberal democracies in postindustrial societies,exemplifiedbyCanada, theNetherlandsandtheUnitedStates,aswellasmanynewer thirdwavedemocracies,suchasSouthAfrica,SouthKoreaandBrazil.

[Figure2abouthere]

Most importantly,manyMuslimpluralitystatesalso fall into thiscategory,exemplifiedbyMali,Albania,BurkinaFaso,andNiger,aswellasTurkey,Malaysia,andIndonesia.Mali,forexample,since its firstdemocraticallyelectedpresident tookpower in1992,hashadaciviliangovernmentandenjoyedrelativepoliticalstability.Over90%ofthepopulation isMuslim,mainlySunni,withascatteringofChristiansand indigenous religions.The secularconstitutionprovides for freedomofreligion and the government respects these provisions in practice, for example with no recentreports of persecution, discrimination, or imprisonment on the basis of religious convictions oraffiliation.Mali'sconstitutionprovidesforamultipartydemocracy,withrestrictionsagainstpartiesbasedonethnicorreligiouslines.Publicschoolsdonotofferreligiousinstructionalthoughthereareno restrictions on private religious schools for Muslims and Christians.28 In short, despitewidespread and deeprooted poverty, Mali has enjoyed considerable religious freedom andtoleranceaccompanyingtheexpansionofdemocracy inrecentdecades,aMuslimpluralitysocietyexemplifyingastableandpeacefulseculardemocraticstate.

Religiousautocracies

Figure2alsoshowsmanyreligiousautocracies inMuslimpluralitysocieties, located inthebottomleftquadrant,asexemplifiedbySaudiArabiaand Iranasthemostextremecases. Tunisia,Turkmenistan,Egypt,Algeria,BahrainandSudanaremoremoderateversionsofthistypeofstate.For example, at least prior to the Jasmine revolution, the Tunisian constitution stipulated thecountry'sdeterminationtoadheretotheteachingsofIslam,that Islam istheofficialstatereligion,and that the president is required to be Muslim.29 The population is 99 percent Muslim andoverwhelmingly Sunni.Groups that constitute less than1percentof thepopulation include Shi'a0~^H=#~J=JJJ


6

another religion, and the government does not require registration of conversion; however,government officials are reported to occasionally harass and discriminate against converts fromIslamtoanotherreligion.Thegovernmentcontrolsandsubsidizesmosquesandpaysthesalariesofimams (clerics). The president appointed the Grand Mufti of the Republic, who is the officialexpounderofIslamiclaw.The1988LawonMosquesstipulatesthatonlypersonnelappointedbythegovernmentmayleadactivitiesinmosques.BasedonIslamiclaw,thegovernmentforbadedomesticmarriagesbetweenMuslimwomenandnonMuslimmenand courtshaveappliedan Islamic lawbased interpretationofcivil law insomefamilycases.Atthesametime, inanattempttosuppressIslamicextremism,thegovernmentalsorestrictsthewearingof"sectariandress,"includingthehijab(Islamicheadscarf),andofficialsdiscouragemenwithtraditionalIslamicdressandbeards.Untiltherecentuprising,Tunisiahasheldelectionsatregularintervalsbuttherulingparty,theConstitutionalDemocraticRally (RassemblementConstitutionelDmocratique),hasexertedpredominantcontrol,sweepingup8590%ofthevote inthe2009PresidentialandAssemblyelections, facingweakanddividedoppositionparties.PresidentZinealAbidinBenAliruledformorethantwodecadesuntilhewas toppled and fled abroad in the January 2011 uprising, leaving Tunisia to face an uncertainfuture.Theoverall classification shows thatMuslimplurality societiesare ruledbymany typesofregimesbut it can alsobeobserved thatmostof the states ruledby religious autocracies are inMuslimpredominantsocieties.

Secularautocracies

Moreoverbasedontheseindicesmanystatescanalsobeclassifiedassecularautocracies,inthe toplefthand quadrant,where there are serious restriction onmany political rights and civilliberties,butwherethestateconstitutionsarealsoneutralorindifferenttowardsdifferenttypesoffaith,ratherthanfavoringone.Thiscategory includesdiversesocieties,bothMuslimpluralitysuchasChad,LibyaandEritrea,aswellasnonMuslimplurality,suchasAngola,TogoandZimbabwe.InChad,forexample,morethanhalfofthepopulationisMuslim,approximatelyonethirdisChristian,andtheremainderfollowsindigenousreligiousbeliefsorhasnoreligion.Thereligiouspopulationisalso divided by region. The Chad constitution is secular and it provides for freedom of religion;however,thegovernmentmonitorstheactivitiesoffundamentalistIslamicgroups.Thegovernmentis legallyobligated to treatall religiousgroupsordenominationsequally, forexample there isnoreligious instruction inpublicschools. Thegovernment is ledbyPresident IdrissDeby,whoseizedpower inamilitarycoupover twodecadesago. In theApril2011presidentialelections, themainoppositionpartiesboycottedthecontest,astheirdemandsforelectoralreformwerenotmet.Thelegislativecontestswerealsoflawedbyirregularitiesandlackofintegrity.Inshort,PresidentDebysrule in Chad exemplifies a country where the leader continues to lack legitimacy derived fromdemocraticmultipartyelections, rankedbyFreedomHouse in2010asoneof themost repressiveregimes in theworld,yetwitha state largely respecting religiouspluralism, reflecting thedividedsociety.

By contrast,with the single exception of theMaldives, there are few cases of religiousdemocracies identified in the worldwide comparison. Table 1 and Figure 3 summarize thedistribution and classification of regimes worldwide and also inMuslimplurality societies. Thusamong Islamic societies, two thirdsare religiousautocracies,onequarterare secularautocracies,andonly8%areseculardemocracies.Amongothertypesofsocietyworldwide,however,onlyonetenth are religious autocracies, one fifth are secular autocracies, while over half are seculardemocracies.Thusregimes inIslamicsocietiesareindeeddistinctivecomparedwiththerestofthe (0 ~ ^H =#~J =JJJ : = = = = = J / / / / , as well asrespectingthoseofnofaith. [Table1andFigure3abouthereabouthere]


7

III:Measuringregimepreferences

Given thisdistribution,whatexplainspublicpreferences towardsone typeof regimeoveranother?Isthetypeofregimecongruentwithpublicpreferencesineachcountry?And,inparticular,inMuslimpluralitysocieties,doespublicopinionfavorseculardemocracy,characterizedbyreligioustolerance,respectfordiversity,andadivisionofchurchandstate?ManyfactorswilldeterminethefinaltypeofconstitutionalsettlementandtheregimewhichdevelopsinthecountriesintheMENAregionwhicharecurrentlyexperiencingchallengestoautocracy.Beyondpopularpreferencesinthemasselectorate,thenewsettlementswillbeshapedbythetypeofpoliticalelite,thepowerandroleofthearmedforces,andexternalpressuresfromthe internationalcommunity,amongmanyotherfactors.Butpopularpreferencesinthemassculturearealsolikelytoplayaroleinanyconstitutionalsettlement,especiallywherethesearesubjecttopopularreferendum. It istherefore importanttodetermine whether the general public in Muslimplurality nations favors the Turkish model ofsecular democracy, the Iranian model of religious autocracy, or another type of constitutionalsettlementlocatedbetweenthesepoles.

A review of the previous empirical research literature analyzing Muslim public opinionsuggestsseveralpropositionsaboutpreferencesforseculardemocracy.RecentsurveyfindingsfromMuslimplurality countries provide evidence that Islamic religiosity does not lead to negativeattitudestowarddemocracy;insteadpreviousstudiesbasedonthe4thwaveoftheWVS/EVSfoundno significant differences between the publics living inWestern nations and inMuslimpluralitysocieties in approval of how democracy works in practice, support for democratic ideals, andapprovalof strongpolitical leadership. 30MarkTessler compared fourArab states (Egypt, Jordan,Morocco,andAlgeria),andreportedthatsupportfordemocracyasan idealformofgovernment isalmostuniversalinthesesocieties.31NeverthelessMuslimpublicsarealsofoundtodisplaygreatersupport fora strong societalandpolitical roleby religiousauthorities thanWesternpublics. 32 Ifclericalleadersarebelievedtobelegitimategoverningauthorities,thisposesamajorchallengeforprocesses of democratization, since clergy hold their position through appointment, charisma orinheritance, not through popular electoral mechanisms. Thus previous research suggests thatMuslim public opinion often displays tensions between the expressed desire for constitutionalarrangements respecting the principles and values of liberal democracy, which should include(0~^H=#~J =JJJ : = = = = = J / / / / a procedural understanding of democracywasmost common,many living inNorth African andMiddleEasternsocietieswithlittlehistoricalexperienceofdemocracy,suchasIraq,Jordan,IranandEthiopia,sawnocontradictionbetweenreligiousauthoritiesinterpretingthelawandtheprinciplesandcharacteristicsofdemocracy.33Toexaminepopularpreferences,theempiricalevidenceused inthisstudydrawsupontheWorldValuesSurveys/EuropeanValuesSurveys (WVS/EVS),pooledacross threewavesconductedfrom19952007.Thisprojecthasconductedrepresentativenationalsurveysofthebasicvaluesandbeliefs of the publics inmore than 90 independent countries, containing over 88 of theworldspopulationandcoveringallsixinhabitedcontinents.Thestudyutilizesthethirdwave,carriedoutin55nations in 19951996, the fourthwave, conducted in 59 countries in 19992001 and the fifthwavecovering55countriesandconducted in20052007.34Researchwhich focuses indepthupononlya fewMuslimpluralitycountrycasestudies isunabletodeterminewhetherattitudesaretheproductofthespecifictypeofpredominantfaithinanysocietyorelsethecharacteristicsofthetypeof cultural region, historical legacies, levels of human development, social structures, orconstitutionalcontexts.Bycontrast,aresearchdesignwhichexaminesattitudesworldwideamongdiversefaithsandtypesofsocieties,includingtwentyMuslimpluralitynationsfromdifferentglobalregions,providesabroaderandmorereliablebasisforcomparison.


8

Mostimportantlyforourpurposes,thesurveyincludessystematicdataonpublicopinionintwentydiverse Islamic states containingMuslimpluralitypopulations, listed in Table 2, providingsomeof thebroadest global comparison available from anyexisting representative social survey.

SocietiesintheWorldValuesSurveyareclassifiedbytheirpredominantreligion,basedonestimatesof the religious population contained in the CIAWorld Factbook. Islamic nations are defined asthosewhere theMuslimpopulation is the largestplurality (althoughsocietiesmay,andoftendo,also contain substantialminoritiesofother faiths). It is important to compare awide variety ofsocieties toexamine thevarietyofattitudesandvalues foundamongdiverseMuslimnationsandIslamictraditionsaroundtheglobe.35TheWorldValuesSurvey includesArabstates,bothmajoritySunni(suchasJordan,Algeria,Morocco,andEgypt)andmajorityShia(suchasIranandIraq),aswellascountriesinAsia(Azerbaijan,Kyrgyzstan,Pakistan,Bangladesh,Malaysia),CentralEurope(BosniaHerzegovina, Albania), and in SubSaharan Africa (Mali, Nigeria, Burkina Faso). The survey alsocoversstatesthathaveadoptedIslamasthefoundationofpoliticalinstitutions(suchasSaudiArabiaand Pakistan), societieswhere Islam is the official or established state religion (including Egypt,Bangladesh, andMalaysia), and secular stateswhere the constitution is neutral towards religion~^H=#~J=JJJ:=====J////uding the top three in size, Indonesia,PakistanandBangladesh.The WVS/EVS surveys also covers societies with all levels of economic and humandevelopment, includingmanyaffluentWesterncountries,suchas theU.S., JapanandSwitzerland,withpercapitaannualincomesover$40,000;togetherwithmiddlelevelMuslimcountriesincludingMalaysia,TurkeyandSaudiArabia,aswellaspoorerMuslimsocieties,suchasBangladesh,MaliandBurkinaFaso,withpercapitaannualincomesof$500orless.Intermsofregimes,SaudiArabiaisanabsolutemonarchy,whichFreedomHouse (2009) classifiesasoneof theworldsmost restrictivestates in respect forcivil libertiesandpolitical rights,but theWorldValuessurveyalsomonitoredpublic opinion in the democratic states ofMali and Indonesia, as well as the secular state ofTurkey.36

Popularsupportfortheprinciplesofdemocracyorautocracy

Surveyshave sought to tappublic attitudes towardsdemocraticprinciples and autocraticformsofgovernance inseveralways.37Supportfordemocracy iscomplexandmultidimensional.Inparticular,ifovertapprovalofdemocraticgovernanceistapped,withoutconsideringalternativesortradeoff values, then the relative preferences fordifferent formsof rule cannotbedetermined.HencesupportfordemocracymaybewidelyendorsedbyJordanians,theEgyptiansorMoroccans,butitremainsunclearhowimportantthisistorespondentscomparedwith,forexample,thedesireto maintain social stability and order, the value of respecting traditional authorities, or thesubstantialrisksofregimechange. Inthesameway, ifpeopleareaskedbypollsterswhethertheyvaluehealthcareandalsowhethertheywantlowertaxes,thenitislikelythatbothstatementswillbewidelyendorsed.Iftradeoffquestionsareusedinstead,sothatpeopleareaskedwhethertheyprefermorepublicspendingonhealthcareormoretaxcuts,thenthispresentsrespondentswithmorerealisticandcomplexchoiceswheretheyareforcedtoprioritizeoptions.Amoreeffectivewaytoexplorewhetherdemocraticattitudesarerobust,andtomeasuremorenuancedchoices,istousetradeoff items where citizens are asked to express their preference for different types ofdemocraticandautocraticregimes.Forcomparability,toseewhethertheresultsremainrobustandconsistentwith other indicators, theWorldValues Surveymonitored preferences for democraticgovernance,militaryrule,rulebybureaucraticelites,andalsostrongman leadershipuncheckedbyparliamentandelections(seeAppendixA). Thestandardizeddemocraticautocraticvaluesscale isconstructedbyrecodingtheseitemstoreflecttheendorsementofdemocraticruleandtherejectionof autocratic forms of governance, and then combining these responses and standardizing theresultingscaleto100points(seeAppendixA).Itshouldbenotedthatresponsesarerelativetotheglobalaverage; forexample themajorityof thepublic inall societiesunder comparison supports


9

democraticvalues,butendorsement isfarmorewidespread incountriessuchasDenmark,AustriaandFrancethaninRussia,IndonesiaandthePhilippines.

Supportforsecularstates

The WVS/EVS survey contains multiple items monitoring religious values, beliefs andpractices, includingthe importanceofreligion,attendanceatreligiousservices,andthestrengthofreligious identities.Tomonitorpreferencesforsecularstates,asproxymeasuresthefollowingtwoitems in the survey were analyzed: How strongly do you agree or disagree with each of thefollowing statements? (i) Religious leaders should not influence how people vote in elections; (ii)Religious leadersshouldnot influencegovernmentdecisions.Agreementwitheachofthese itemson the 5point Likert type scalewere understood to reflect preferences for aminimal role for0~^H=#~J=JJJ:=====J////f church and state. The items werestrongly intercorrelated, forminga suitable scale.Agreementwithboth itemswere summedandcodedonacomposite10pointscale,whereahigherscorerepresentsgreatersupportforsecularity.Thisisfarfromaperfectwaytomonitorthecomplexconceptofreligiousorsecularstates,asmanyother indicatorswould be useful, such asmeasures of religious tolerance, support for religiousfreedoms, and approval ofmulticultural policies, but these two items are themost appropriatemeasuresavailableinthesurveyasthebasisforthepreliminaryanalysis.

Thedependentvariablesderivedfromtheseindicatorsarethereforethreefold:(i)approvalofdemocraticprinciplesandrejectionofautocracy;(ii)supportforsecularity(thedivisionofspiritualand government authorities); and (iii) the interaction of these scales, understood as reflectingendorsement of secular democracy. The analytical models explaining these attitudes involvemeasurementat twodistinct levels.A representativesampleof individual respondents (level1) isnestedwithinnationallevelcontexts(level2).TheWVS/EVSwasconductedamongarepresentativerandomsampleoftheadultpopulationwithineachcountry.TheuseofHierarchicalLinearModels,in particularmultilevel regression analysis, is themost appropriate technique for comparing theimpact of societallevel and individuallevel factors simultaneously. Themodels in this study userestricted maximum likelihood techniques (REML) to estimate direct and crosslevel effects forhierarchical data. Individual respondents are thus grouped into countries. Each country has adifferent set of parameters for the random factors, allowing intercepts and slopes to vary bynation.38Inhierarchical linearmodels,asiscustomary,allindependentvariableswerecentered,bysubtractingthegrandmean(whichbecomeszero).Thestandardizedindependentvariablesallhaveastandarddeviationof1.0.Thisprocessalsohelpstoguardagainstproblemsofcollinearity intheindependent variables in the OLS models. The independent variables were treated as fixedcomponents, reflecting the weighted average for the slope across all groups, while nation wastreatedasarandomcomponent,capturingthecountryvariability intheslope.Thestrengthofthebeta coefficients (slopes) can be interpreted intuitively as howmuch change in the dependentvariable is generated by a onepercent change in each independent variable. The multilevelregressionmodels used in this study usually generate small differences in the size of the slopecoefficient(b)comparedwiththeresultsofOLSmodels,buttheaveragestandarderrorsforlevel2variables tend tobe slightly larger.Theprocess is thusmore rigorousandconservative thanOLS,avoiding Type I errors (false positives, concluding that a statisticallysignificant difference existswhen,intruth,thereisnostatisticaldifference).IntheREMLmodel,bycontrast,SchwarzsBayesianCriterion(BIC)isused,wherethemodelwiththelowervalueisthebestfitting.

Level 1 in the coremodels includes individuallevelMuslim religious identities, and thestrength of religiosity, along with several other standard controls, described in the TechnicalAppendixA,includingmalegender(0/1),householdincomeusinga10pointscale,age(inyears),aneducation scale, marital status, and labor force participation. Level 2 includes nationallevelvariables, includingtheclassificationof IslamicornonIslamicsocieties,basedonthepluralityfaith


10

ineachcountryspopulation.Inaddition,thestudyalsocontrolsfortheregionallocation,classifiedasintheMiddleEastorelsewhereintheworld,totestwhetherglobalareasdiffer.

IV:Publicregimepreferences

Wecanstartbycomparingthedescriptiveresultsatmacrolevelswithoutanycontrols,toseewhere thegeneralpublic fallson these scaleswithineach society,and thenexamine the fullmultilevel models seeking to predict support for secular democracy. Figure 4 illustrates publicattitudes in each society towards the proxy indicator of secularity (on the vertical axis) and theendorsementofdemocraticprinciplesand rejectionofautocracy,depictedon thehorizontalaxis.Table3comparestheproportionofthepublic ineachsocietysupportingeachtypeofregime.Thecomparison suggests that thepublic living inmany longestablished seculardemocracies stronglyapprove of the principles underpinning this type of regime. Hence citizens in France, Belgium,Denmarkand Japanare located in the top righthandquadrant, favoringdemocraticvaluesandaseparationofchurchand state.Somenewerdemocracies, suchasSlovakia,HungaryandEstonia,alsofall intothiscategory. Inthesecongruentcases,theregimepreferencesforthegeneralpublicare inaccordancewiththeconstitutionalprinciples ineachstate.AmongMuslimpluralitysocietiesundercomparison,thepublicsinTanzaniaandMoroccoalsoendorsetheseprinciples.Neverthelessit is also apparent that among themany postcommunist societies located in the top lefthandquadrant, such asRussia,Belarus andRomania,onbalance themajorityof thepublic still favorssecular democracy but they express less strong and overwhelming endorsement of democraticprinciples.Mali,IndonesiaandBurkinoFasoalsofallintothiscategory.

[Figure4andTable3abouthere]

Inthebottomlefthandcorner,severalcountriesexpresslessstrongsupportfordemocracyandfortheprinciplesofsecularity,notably inAlgeriaandMalaysia,aswellasinMexico,BrazilandVenezuela. These are all societies where religious values remain deeplyrooted, hence lessdistinction isdrawnbetweentheroleofspiritualandgovernmentauthorities.Lastly inthebottomrighthand corner,we find severaldeeplyreligious societies, including theUnited States,Uganda,Zambia,andGhana,wherethepublicstronglyendorsesdemocraticvaluesandprincipleswhilealsobelievingthatspiritualleadersshouldplayaninfluentialroleinpubliclife.InthecaseoftheUnited(0~^H=#~J=JJJ:=====J////liefsaboutthe importantroleofreligiousleadersinpublicaffairs.

[Table4abouthere]

Toanalyzethereasonsforthesupportoftheseprinciples,aseriesofmultilevelmodelsarerequired.AsshowninTable4,thesemodelsincorporateawiderangeofsocietalandindividuallevelcontrolswhich the research literature suggestsmightbeexpected to influence attitudes towardsdemocracyand secularity. Inparticular,manyprevious studiesassume thatMuslim identitiesandreligious values are given,when analystsproceed toexaminehow far attitudes arepredictedbyMuslimmajoritysocietiesatmacrolevel,orMuslim identitiesatmicro level.Yet identities ineverytypeoffaithmaybenominalandsuperficial,suchasspecifyingformalreligiousaffiliationsonofficialdocuments,ortheymayreflectdeeplyfeltmeaningsandorientations.Therefore it is importanttotap into religiousvaluesaswell,orthe importanceofreligion,which isexpectedtohaveastrongnegative impact on preferences for secular states. Themodels therefore enter three predictorvariablesMuslimreligiousidentities,livinginanIslamic(Muslimplurality)society,andthestrengthof religiosity. Themodels control at individuallevel for the standard sociodemographic factorswhich influence social andpolitical attitudes,namely age, gender, income,education, laborforceparticipationandmaritalstatus.Themodelsalsocontrolatnationallevel for theMiddleEastandNorth Africa, to see whether there is an additional effect arising from the regional culture


11

differentiatingMuslimpluralitysocieties,andeconomicdevelopment,measuredbyloggedGDPpercapitainpurchasingpowerparity.

The results of the three models are fairly consistent in the coefficients which provesignificantpredictors.Thusamong the individuallevel controlvariables, the factorwhich ismoststronglyandsignificantlyrelatedtosupportforseculardemocracyiseducation;thosewhoaremoreeducatedprovemostlikelytoendorsebothdemocraticvaluesandtheprinciplesofsecularity.Ageeffectsaresimilar;oldergenerationsaremoresupportiveofseculardemocracyanditscomponents.The other sociodemographic controls are less consistently significant in their performance. Atnationallevel,economicdevelopmentisalsoverystronglyrelatedtosupportfordemocraticvaluesandthussupportforseculardemocracy.This isnotsurprising;elsewherewehavedemonstratedastrong linksamongmany indicesofhuman securityanddevelopment,patternsof religiosity,andsupport for democracy. After controlling for these factors,we do not observe a separate effectarisingfromtheMiddleEast.

Lastly,intermsofthekeypredictorvariables,aftercontrollingforallthesefactors,Muslimsdisplay significantly less approval than nonMuslims for democratic values and rejection ofautocraticrule,secularism,andthusendorsementofthecompoundindicatorofseculardemocracy.Similarpatternscanbeobservedforthestrengthofreligiosity,withthemorefaithfullessapprovingof democracy and secularism, aswell as secular democracy. By contrast livingwithin an Islamicsociety proves negative but not statistically significant. It appears that regime preferences areshapedbothbythetypeofreligiousidentityaswellasbythestrengthofreligiosityandtheseeffectsoperateprimarilyatmicrolevel.Furtheranalysis isrequired,however, totestwhether theresultsremainrobustwhenusingalternativeattitudinalindicesofsupportfordemocracyandsecularism,aswell as whether the strength of these relationships vary among diverse Muslimpredominantsocieties, such as Turkey, Iran and Indonesia,with different types of contemporary regimes andpoliticalhistories.

V:Conclusionsandimplications

TheArabspringhasbeenmarkedbytherevolutionarydownfallandremovalofautocraticleadersinTunisia(PresidentZineElAbidineBenAli),Egypt(PresidentHosniMubarrak),andYemen(President Ali Abdullah Saleh). Major protests, strikes, and demonstrations have occurredthroughouttheregion,notablyinAlgeria,Armenia,Bahrain,Djibouti,Iraq,Jordan,Syria,andOman,whiletheoutbreakofcivilwarinLibyahasdestabilizedtheregimeofColonelGaddafi.TheseeventshaveencouragedmanycommentatorstohopethattheMiddleEastandNorthAfricaisexperiencingthegradualtransitionfromautocracytoseculardemocracies,wheremoderateIslamicparties,newconstitutions, and elected governments gradually expand human rights, including religiousfreedoms,pluralism,and toleranceofdiverse faiths.Neverthelesscontemporaryevents remain incontinuous flux, the final outcome of these developments are far from settled, and othercommentators fear that the instability caused by the popular uprisings will allow openings forconservativeandmoreextremistreligiousforcestocometopower,enforcingstricterinterpretationofIslamiclaw.

Theconceptual frameworkused in thispaper suggests thatcontemporary regimescanbeclassified intofourcategories:seculardemocracies,secularautocracies,religiousdemocracies,andreligious autocracies. The study operationalized these notions then explored the distribution ofregimesbothworldwideandalsoamongMuslimpluralitysocieties.The resultsofthecomparisonindicatethatMuslimpluralitysocietiesaregovernedtodaybydiversetypesofregimes, illustratedby the contrasts among Mali (exemplifying a secular democracy), Chad (classified as a secularautocracy),andTunisia(asareligiousautocracy).ThuswecanconcludethatthereisnobasisforthepopularperceptionthatmostMuslimpluralitysocietiesareruledbyreligiousautocracies,althoughat the same time it isalsoobserved thatmost religiousautocraciesare found inMuslimpluralitysocieties. Furtherexploration isneededtounderstandthecontrasts intheactualtypeofregimes


12

rulingMuslimpluralistsocieties,examiningthefactorscommonlyassociatedwiththepersistenceofautocracyandtheprocessofdemocratization.Hencestudiesneedtoanalyzetheroleofeconomicdevelopmentand societalmodernization;historicalprocessesandcolonial legacies;neighborhoodeffects, regional hegemonic states and international pressures; the impact of natural resources,conflict, and ethnic fractionalization; and the structure of the state in terms of powersharingconstitutionalarrangements.39

Tounderstandwhat regimesthepublic living inMuslimpluralitysocietiesactuallydesires,weturnedtotheanalysisofpublicopiniondata.Attitudestowards idealregimeprinciplesshowedthat the publics in these societies displayed diverse preferences; hence some countries, such asTanzania andMorocco, favored secular democracy,while others such as Albania and Indonesiaproved slightly less supportive of democracy although equally secular, and still others, such asAlgeria, favored religious autocracy.Overall, therefore, although some assume commonality, thecontrastsamongMuslimplurality societies standout.Themultilevel surveyanalysis suggests thatMuslims, and those populationswho aremost religious, are generally less supportive of seculardemocraticvaluesandprinciplesthanthoseofotherfaithsandthelessreligious.

The study provides insights into the political diversity which is evident among Islamicsocietiesaroundtheglobe.Whatalsoemerges,moreover,isthatalthoughMuslimpluralitysocietiesareparticularlypronetowardsreligiousautocracy,thisisfarfromtheonlytypeofregimegoverningthesecountries.TheexamplesofseculardemocracyinMali,Turkey,IndonesiaandAlbaniaprovidealternativemodelsofdevelopmentforIslamicsocietieselsewhere,especiallyincountriesfacingnewconstitutionalchoicesfollowingsuccessfulregimetransitionsduringtheArabspring.


13

Table1:Thedistributionofregimes

Secular

democracy

Religious

democracy

Secular

autocracy

Religious

autocracy

Total

(%) (%) (%) (%) N.

Muslimpluralitysocieties 8.0 4.0 24.0 64.0 50

Othersocieties 52.5 13.5 19.9 11.3 137

Worldwide 40.8 11.0 20.9 25.1 187

Note:Fortheconstructionofallmeasures,seeAppendixA


14

Table2:MuslimpluralitycountriescontainedintheWorldValuesSurveys

Country %Muslims Globalregion1 Iraq 98.9 MiddleEast2 Algeria 98.0 NorthAfrica3 Iran 97.7 MiddleEast4 SaudiArabia 97.0 MiddleEast5 Jordan 96.9 MiddleEast6 Egypt 94.3 NorthAfrica7 Mali 93.0 WestAfrica8 Indonesia 92.2 SouthEastAsia9 Azerbaijan 91.0 CentralEurope10 Bangladesh 88.7 SouthAsia11 Turkey 81.7 CentralEurope12 Kyrgyzstan 74.3 CentralAsia13 Pakistan 69.6 SouthAsia14 Albania 64.3 CentralEurope15 Morocco 63.6 NorthAfrica16 Malaysia 57.4 SouthEastAsia17 BurkinaFaso 53.3 SubSaharanAfrica18 Bosnia 48.9 CentralEurope19 Tanzania 40.1 EastAfrica20 Nigeria 28.1 WestAfrica

Source:PooledWorldValuesSurvey,19952007


15

Table3:Publicpreferencesforregimes

PublicpreferencesforregimesSecular

democracyReligiousdemocracy

Secularautocracy Religiousautocracy

Denmark 71.2 18.7 8.5 1.6Andorra 69.7 11.2 12.6 6.5Malta 66.5 17.9 12.8 2.8Austria 64.3 18.4 12.2 5.0Montenegro 62.0 11.9 18.5 7.6Iceland 61.1 29.4 5.6 3.9France 57.6 13.2 23.1 6.1NewZealand 57.4 28.1 10.1 4.4Greece 56.8 36.1 3.7 3.5WestGermany 56.7 28.0 8.4 6.9Croatia 56.5 15.3 21.4 6.8CzechRepublic 55.4 23.8 14.5 6.3Japan 55.2 20.1 16.4 8.3Belgium 54.5 17.3 19.2 9.0Albania 54.4 19.8 16.4 9.4Portugal 54.4 9.8 20.5 15.3Italy 54.1 27.7 11.5 6.6NorthernIreland 53.9 28.8 8.8 8.4Ireland 53.0 28.4 12.4 6.2Estonia 52.6 16.8 19.5 11.1Spain 52.5 22.8 11.8 12.8EastGermany 51.1 23.0 15.9 10.0Tanzania 50.9 37.6 6.9 4.6Canada 50.8 25.6 14.5 9.0Sweden 50.5 31.1 9.9 8.5Slovenia 48.9 20.2 18.4 12.5Luxembourg 48.8 16.8 25.7 8.6Bangladesh 47.8 23.0 19.7 9.5Australia 47.5 27.6 14.0 10.9Hungary 46.8 17.7 23.2 12.2GreatBritain 45.3 28.1 15.7 10.9Netherlands 44.3 34.7 9.1 11.9Cyprus 44.2 14.9 26.3 14.6Serbia 42.5 17.6 22.7 17.2Finland 41.5 28.5 15.5 14.5Uruguay 39.9 24.9 17.6 17.7Argentina 39.4 23.4 20.5 16.7Bosnia 39.4 15.9 30.1 14.6Ghana 38.5 38.8 14.1 8.7Lithuania 38.2 11.1 37.8 13.0PuertoRico 38.0 30.4 15.4 16.2Morocco 37.8 39.3 13.6 9.3Slovakia 36.3 18.1 30.2 15.4Latvia 35.4 16.0 34.1 14.5Poland 35.1 9.6 41.7 13.6Bulgaria 33.9 8.1 42.4 15.5Belarus 33.1 10.2 39.4 17.3Taiwan 32.6 6.3 49.4 11.7Ethiopia 32.4 26.7 26.7 14.2TrinidadandTobago 32.3 33.7 16.1 17.9Chile 32.0 21.6 24.5 21.9BurkinaFaso 31.6 18.3 33.4 16.7UnitedStates 31.2 35.3 15.2 18.3


16

PublicpreferencesforregimesSecular

democracyReligiousdemocracy

Secularautocracy Religiousautocracy

SouthKorea 29.0 26.3 22.3 22.4Uganda 28.3 34.8 25.1 11.9Russia 28.1 10.2 40.4 21.3SouthAfrica 27.8 32.3 19.1 20.8VietNam 26.5 16.5 35.4 21.6Ukraine 26.2 11.5 35.8 26.4Armenia 23.8 19.6 27.3 29.4Romania 23.2 6.9 50.3 19.6India 23.2 14.0 36.4 26.4Zimbabwe 23.0 34.9 18.1 24.0Macedonia 21.8 7.3 49.4 21.5Moldova 21.7 16.0 31.7 30.6Venezuela 21.6 26.7 23.3 28.4Mali 20.2 12.3 45.4 22.1Kyrgyzstan 20.1 10.8 39.3 29.8Zambia 19.7 38.9 16.8 24.6Philippines 18.5 6.4 47.6 27.4Indonesia 18.4 9.8 47.6 24.2Mexico 16.1 11.2 31.4 41.3Brazil 14.0 13.7 34.4 37.9Guatemala 13.6 13.5 36.1 36.8Algeria 9.9 39.9 13.0 37.3Malaysia 8.3 17.8 32.1 41.8ALL 38.6 21.3 23.8 16.4Note:Fortheconstructionofallmeasures,seeAppendixA

Source:PooledWorldValuesSurvey,19952007


17

Table4:Multilevelmodelpredictingsupportforseculardemocracy Model1

DemocraticvaluesModel2

SecularismModel3

SecularDemocracyPREDICTORVARIABLES Muslimreligiousidentity .739*** .063*** .108*** (.125) (.014) (15.3)Islamicsociety 1.81 .114 50.9 (1.10) (.089) (107)Strengthofreligiosity .617*** .264*** 262*** (.093) (.011) (.12.3)INDIVIDUALLEVELCONTROLSAge(years) .801*** .032*** 86.5*** (.079) (.009) (10.5)Gender(male=1) .006 .026*** 11.3 (.069) (.008) (0.04)Householdincome10ptscale .100 .001 17.8 (.817) (.010) (10.6)Education9ptscale 2.01*** .118*** 233*** (.081) (.010) (10.7)Laborforceparticipation .097 .001 4.97 (.074) (.009) (9.73)Maritalstatus .255*** .009 13.4 (.072) (.008) (9.48)NATIONALLEVELCONTROLSMiddleEast 1.85 .124 10.98 (1.27) (.109) (131)LoggedGDPpercapita 4.12*** .112 3.97*** (.814) (.070) (.86)Constant(intercept) 70.0 7.54 53.1SchwartzBIC 454,657 206,495 765,586N.respondents 54,424 50,334 42,884N.nations 45 39 37Note:Allindependentvariableswerestandardizedusingmeancentering(zscores).Modelspresenttheresultsof the REML multilevel regression models including the beta coefficient, (the standard error below inparenthesis),andthesignificance.The100pointscalesareconstructedfromtheitemsdescribedinAppendixA.P.***>.000,**=.01,*>.05Source:PooledWorldValuesSurvey19952007.


18

Figure1:Schematicmodelofregimetypology

LowHighLowH

igh

Secularautocracy Seculardemocracy

Religiousautocracy Religiousdemocracy


19

Figure2:Operationalizingtheregimetypology

Notes:LiberaldemocracyismeasuredbytheFreedomHouseGastilindexofpoliticalrightsandcivilliberties,2008,standardizedto100points.ReligiousfreedomismeasuredbytheReligiousFreedomIndexona100ptscaledevelopedforSacredandSecular(NorrisandInglehart2004).SeeAppendixAfordetails.ReligiousculturesareclassifiedbythepredominantreligiouspopulationineachsocietyderivedfromtheCIAWorldYearbook.Fordetails,seeAppendixA


20

Figure3:Thedistributionofregimetypes

NotesLiberaldemocracyismeasuredbytheFreedomHouseGastilindexofpoliticalrightsandcivilliberties,2008,standardizedto100points.Democraciesandautocraciesaredichotomizedaroundthemean.

ReligiousfreedomismeasuredbytheReligiousFreedomIndexona100ptscaledevelopedforSacredandSecular(NorrisandInglehart2004).Secularandreligiousregimesarecategorizedaroundthemean.SeeAppendixAfordetails.

ReligiousculturesareclassifiedbythepredominantreligiouspopulationineachsocietyderivedfromtheCIAWorldYearbook.


21

Figure4:Publicattitudestowardsidealregimeprinciples

Notes:SeeAppendixAfordetailsofeachscale.

Source:WorldValuesSurveys


22

AppendixA:ConceptsandMeasures

Variable Definitions,codingandsources

PerCapitaGDP Measuredinconstantinternational$inPurchasingPowerParity.Variousyears.Source:TheWorldBank.WorldDevelopmentIndicators.

Typeofreligion V184:Doyoubelongtoareligiousdenomination? [IFYES]Whichone?Coded:No,notamember;RomanCatholic;Protestant;Orthodox(Russian/Greek/etc.);Jewish;Muslim;Hindu;Buddhist;Other.Source:WorldValuesSurveys.

Typeofpredominantreligionineachsociety

Theclassificationofthemajorreligion(adheredtobythepluralitygroupsinthepopulation)inall193statesaroundtheworldisbasedontheCIA.TheWorldFactbook,2009.(Washington,DC:CentralIntelligenceAgency).Source:http://www.cia.gov/cia/publications/factbook

Levelandtypeofdemocracy

FreedomHouseGastilindexofPoliticalRightsandCivilLiberties,FreedomintheWorldwww.freedomhouse.org

Islamicsocieties BasedonthesocietieswithMuslimpluralitypopulations,basedontheabovesource,aslistedinTable1.ItshouldbenotedthatthroughoutthepaperthetermIslamicisusedtorefertoMuslimpluralitysocieties,anditisnotusedtodescribetheofficialreligionorpoliciesofthestate,ortherelationbetweenreligiousandpoliticalauthorities.MuslimreferstoindividualswhoidentifywiththeMuslimfaith.

Supportforsecularity(separationofchurchandstate)

Howstronglydoyouagreeordisagreewitheachofthefollowingstatements?(i)Religiousleadersshouldnotinfluencehowpeoplevoteinelections;(ii)Religiousleadersshouldnotinfluencegovernmentdecisions.(Agreecodedhighona10pointscale).Source:WorldValuesSurveys19992007.

Supportfordemocraticprinciples

I'mgoingtodescribevarioustypesofpoliticalsystemsandaskwhatyouthinkabouteachasawayofgoverningthiscountry.Foreachone,wouldyousayitisaverygood,fairlygood,fairlybadorverybadwayofgoverningthiscountry?(i)Havingademocraticpoliticalsystem.(ii)Havingastrongleaderwhodoesnothavetobotherwithparliamentandelections;(iii)Havingexperts,notgovernment,makedecisionsaccordingtowhattheythinkisbestforthecountry;(iv)Havingthearmyrule.Source:WorldValuesSurveys19952007

OccupationalClass

Codedfortherespondentsoccupation.Inwhichprofession/occupationdoyou,ordidyou,work?Thescaleiscodedinto4categories:Professional/manager(1);Othernonmanual(2);Skillednonmanual(3);UnskilledManualWorker(4).Source:WorldValuesSurveys

Paidworkstatus V220.Areyouemployednowornot?Codedfulltime,parttimeorselfemployed(1),other(0).Source:WorldValuesSurveys

Education V217.Whatisthehighesteducational levelthatyouhaveeverattained?Codedona9pointscalefromnoformaleducation(1)touniversitylevelwithdegree(9).Source:WorldValuesSurveys

Age Agecodedincontinuousyearsderivedfromdateofbirth.Source:WorldValuesSurveys.

Agegroups:Younger(1829),middle(3049),andolder(50+).

Religiosity V192HowimportantisGodinyourlife10pointscale

Household V253Onthiscardisascaleofincomesonwhich1indicatesthelowestincome


23

Income decileand10thehighestincomedecileinyourcountry.Wewouldliketoknowinwhatgroupyourhouseholdis.Please,specifytheappropriatenumber,countingallwages,salaries,pensionsandotherincomesthatcomein.(Codeonenumber).Source:WorldValuesSurveys

Educationscale V238. Whatisthehighesteducationallevelthatyouhaveattained?[NOTE:ifrespondentindicatestobeastudent,codehighestlevels/heexpectstocomplete]:(1)Noformaleducation;(2)Incompleteprimaryschool;(3)Completeprimaryschool;(4)Incompletesecondaryschool:technical/vocationaltype;(5)Completesecondaryschool:technical/vocationaltype;(6)Incompletesecondary:universitypreparatorytype;(7)Completesecondary:universitypreparatorytype;(8)Someuniversityleveleducation,''''''''''*GZZJ*'',>JLLJLLLMD''

Freedomofreligionscale

Constructedfromthefollowingitems:1. Theconstitutionlimitsfreedomofreligion.2. Theconstitutiondoesnotrecognizefreedomofreligion.(Orthelaw

doesnotrecognizefreedomofreligion,incountrieswithoutawrittenconstitution).

3. Asingleofficial(established)statechurchexists.4. Thestatefavorsonereligion.5. Religiousorganizationsmustregisterwiththestateorbedesignatedby

ittooperatelegally,orthegovernmentimposesrestrictionsonthoseorganizationsnotregisteredorrecognized.

6. Thestateissueslegalpermitsforreligiousbuildings.7. Thestateappointsorapproveschurchleaders,churchleadersappoint

orapprovegovernmentofficials,and/orchurchleadershavespecificpositionsinthegovernment.

8. Thestatepayschurchsalariesdirectly.9. Thestatesubsidiessome/allchurches.10. Thestateprovidestaxexemptionsforsome/allchurches.11. Thestatebansclergyfromallorsomespecifiedreligionsfromholding

publicoffice. 12. Thestateownssomechurchpropertyandbuildings. 13. Thestatemandatessomereligiouseducationinstateschools,even

thoughstudentscanbeexemptedfromthisrequirementwithaparentsrequest.

14. Therearereportsofforcedreligiousconversions.15. Thestaterestrictssomedenominations,cults,orsects.16. Thestaterestricts/banssomemissionariesenteringthecountryfor

proselytizingpurposes.17. Thestaterestricts/censorssomereligiousliteratureenteringthe

countryorbeingdistributed.18. Thestateimprisonsordetainssomereligiousgroupsorindividuals.19. Thestatefailstodeterseriousincidentsofethnoreligiousconflictand

violencedirectedagainstsomeminoritygroups.20. Thestateisdesignatedacountryofparticularconcernforfreedomof

religionbytheUnitedStatesstatedepartment.Source:NorrisandInglehartSacredandSecular

Note:FulldetailsoftheWorldValuesSurveycodebooksandquestionnairescanbefoundathttp://worldvaluessurvey.org/


24

1MuratSomer.2007.ModerateIslamandsecularistoppositioninTurkey:implicationsfortheworld,Muslimsandseculardemocracy.ThirdWorldQuarterly28(7):12711289.2J.L.EspositoandR.K.Ramazani.Eds.2001.IranattheCrossroads.NewYork:Palgrave;N.Keddie.2003.ModernIran:Rootsandresultsofrevolution.NewHaven,CT:YaleUniversityPress;G.Kepel.2002.Jihad:ThetrailofpoliticalIslam.Cambridge:HarvardUniversityPress;K.Alamdari.2005.ThepowerstructureoftheIslamicRepublicofIran.ThirdWorldQuarterly26:12851301.3SamuelP.Huntington.1996.TheClashofCivilizationsandtheRemakingofWorldOrder.NewYork:SimonandSchuster.4M.I.Midlarsky.1998.DemocracyandIslam:Implicationsforcivilizationalconflictandthedemocraticprocess.InternationalStudiesQuarterly.42(3):485511.5EdwardSaid.2001.AClashofIgnorance.TheNation;B.M.Russett,J.R.ONealandM.Cox.2000.Clashofcivilizations,orrealismandliberalismdjvu?Someevidence.JournalofPeaceResearch.37(5):583608;TedGurr2000.PeoplesversusStates.WashingtonDC:USInstituteforPeacePress;S.R.Hofmann.2004.Islamanddemocracy:Microlevelindicationsofcompatibility.ComparativePoliticalStudies37:65276.6NiazFaiziKabuli.1994.DemocracyaccordingtoIslam.Pittsburgh,PA:DorrancePublications;JohnL.EspositoandJohnO.Voll.1996.DemocracyandIslam,NewYork:OxfordUniversityPress;AnthonyShadid.2001.Legacyoftheprophet:despots,democrats,andthenewpoliticsofIslam.Boulder,Co:WestviewPress.7See,forexample,BernardLewis.2002.Whatwentwrong?WesternimpactandMiddleEasternresponse.NewYork:OxfordUniversityPress.8MarkTesslerandA.Jamal.2006.PoliticalattituderesearchintheArabworld:Emergingopportunities.PSPoliticalScience&Politics39(3):433437.9JohnL.Esposito&DaliaMogahed.2007.WhospeaksforIslam?WhatabillionMuslimsreallythink.NewYork:GallupPress;JohnL.Esposito.2010.TheFutureofIslam.NewYork:OxfordUniversityPress;MoatazA.Fattah,2006.DemocraticvaluesintheMuslimworld.Boulder,Colo.:LynneRiennerPublishers.10PippaNorrisandRonaldInglehart.2011.SacredandSecular:PoliticsandReligionWorldwide.NewYork:CambridgeUniversityPress.2ndedition;M.StevenFish.2011.AreMuslimsDistinctive?NewYork:OxfordUniversityPress.11M.N.Shafiq.2010.DoeducationandincomeaffectsupportfordemocracyinMuslimcountries?EvidencefromthePewGlobalAttitudesProject.EconomicsofEducationReview29(3):461469.12P.A.FuriaandR.E.Lucas.2008.ArabMuslimattitudestowardtheWest:Cultural,social,andpoliticalexplanations.InternationalInteractions34(2):186207.13MarkTesslerandE.Gao.2005.GaugingArabsupportfordemocracy.JournalofDemocracy16(3):8397;MarkTessler.2002.DoIslamicorientationsinfluenceattitudestowarddemocracyintheArabworld?EvidencefromEgypt,Jordan,Morocco,andAlgeria.InternationalJournalofComparativeSociology43(35):229249;MarkTessler.2002.IslamanddemocracyintheMiddleEast:TheimpactofreligiousorientationsonattitudestowarddemocracyinfourArabcountries.ComparativePolitics34(3):337+.14SeeDavidHeld.2006.ModelsofDemocracy.3rdEdition.Stanford:StanfordUniversityPress.15R.MichaelAlvarez,ThadE.Hall,andSusanD.Hyde(Editors).2008.ElectionFraud:DetectingandDeterringElectoralManipulation.Washington,DC:BrookingsInstitutionPress.16RobertA.Dahl.1971.Polyarchy:ParticipationandOpposition.NewHaven,CT:YaleUniversityPress;RobertA.Dahl.1989.DemocracyanditsCritics.NewHaven:YaleUniversityPress.17RobertDahl.1989.DemocracyanditsCritics,NewHaven:YaleUniversityPress.p.22118GeraldoL.MunckandJayVerkuilen.2002.Conceptualizingandmeasuringdemocracy:Evaluatingalternativeindices.ComparativePoliticalStudies.35(1):534.19LarryDiamond.1996.DevelopingDemocracy:TowardConsolidation.Baltimore:JohnsHopkinsUniversityPress;RobertJ.Barro.1999.Determinantsofdemocracy.JournalofPoliticalEconomy107(6):158183;


25

RonaldInglehartandChristopherWelzel.2005.Modernization,CulturalChange,andDemocracy:TheHumanDevelopmentSequence.NewYork:CambridgeUniversityPress.20FreedomHouse.2010.FreedomintheWorld2010.Washington,DC:FreedomHouse.www.freedomhouse.org.21WilliamSafran.Ed.2003.Thesecularandthesacred:nation,religion,andpolitics.London:FrankCass.22SeeKarelDobbelaere.1981.Secularization:amultidimensionalconcept.CurrentSociology.29(2):121.SeealsoS.Hanson.1997.Thesecularizationthesis:Talkingatcrosspurposes.JournalofContemporaryReligion12:15979;WilliamH.Swatos,JrandKevinJ.Christiano.1999.Secularizationtheory:Thecourseofaconcept.SociologyofReligion.60(3):209228;L.Shiner.1966.Theconceptofsecularizationinempiricalresearch.JournalfortheScientificStudyofReligion.6:20720.23GraceDavie.1994.ReligioninBritainsince1945:BelievingwithoutBelonging.Oxford:Blackwell24PippaNorrisandRonaldInglehart.2004.SacredandSecular:PoliticsandReligionWorldwide.NewYork:CambridgeUniversityPress.25USStateDepartment.InternationalReligiousFreedom,2002.Washington,DC.http://www.state.gov/g/drl/rls/irf/ThereportisproducedduetotheInternationalReligiousFreedomActof1998(U.S.PublicLaw10592)andmonitoredbytheUSCommissiononInternationalReligiousFreedom.http://www.uscirf.gov.TheUSStatedepartmentreportlargelyreflectstheevaluationsofthestateofreligiousfreedomsreportedbyhumanrightsorganizationssuchasFreedomHouseandAmnestyInternationalandbycomparativestudies.See,forexample,KevinBoyleandJulietSheen.Eds.1997.Freedomofreligionandbelief:Aworldreport.NewYork:Routledge;PaulMarshall.Ed.2000.Religiousfreedomintheworld:Aglobalreportonfreedomandpersecution.Nashville,Tenn:BroadmanandHolman.26MarkChavesandDavidE.Cann.1992.Regulation,pluralismandreligiousmarketstructure.RationalityandSociety.4:272290.27BruceLedewitz.2007.Americanreligiousdemocracy:comingtotermswiththeendofsecularpolitics.Westport,Conn:PraegerPublishers.28USStateDepartmentInternationalReligiousFreedomReport2010.Mali.http://www.state.gov/g/drl/rls/irf/2010/148703.htm29USStateDepartmentInternationalReligiousFreedomReport2010.Tunisia.http://www.state.gov/g/drl/rls/irf/2010/148847.htm30PippaNorrisandRonaldInglehart.2011.SacredandSecular:PoliticsandReligionWorldwide.NewYork:CambridgeUniversityPress.2ndedition;M.StevenFish.2011.AreMuslimsDistinctive?NewYork:OxfordUniversityPress.31MarkTesslerandE.Gao.2005.GaugingArabsupportfordemocracy.JournalofDemocracy16(3):8397;MarkTessler.2002.DoIslamicorientationsinfluenceattitudestowarddemocracyintheArabworld?EvidencefromEgypt,Jordan,Morocco,andAlgeria.InternationalJournalofComparativeSociology43(35):229249;MarkTessler.2002.IslamanddemocracyintheMiddleEast:TheimpactofreligiousorientationsonattitudestowarddemocracyinfourArabcountries.ComparativePolitics34(3):337+.32PippaNorrisandRonaldInglehart.2011.SacredandSecular:PoliticsandReligionWorldwide.NewYork:CambridgeUniversityPress.2ndedition.33PippaNorris.2011.DemocraticDeficit.NewYork:CambridgeUniversityPress.Ch8.34FullmethodologicaldetailsabouttheWorldValuesSurveys,includingthequestionnaires,samplingprocedures,fieldworkprocedures,principleinvestigators,andorganizationcanbefoundat:http://worldvaluessurvey.org/35MansourMoaddel.2007.ValuesandperceptionsoftheIslamicandMiddleEasternpublics.NewYork:PalgraveMacmillan.36Theseregimesareclassifiedaccordingtothe2009FreedomHouseassessmentsofpoliticalrightsandcivillibertiesFreedomHouse.2009.FreedomintheWorld.www.freedomhouse.org.


26

37WilliamMishlerandRichardRose.1995.Trajectoriesoffearandhope:supportfordemocracyinPostCommunistEurope.ComparativePoliticalStudies28:55381;RichardRoseandWilliamMishler.1996.TestingtheChurchillhypothesis:popularsupportfordemocracyanditsalternatives.JournalofPublicPolicy16:2958;RichardRose,Richard,WilliamMishler,andChristianHaerpfer.1998.DemocracyandItsAlternativesinPostCommunistEurope:TestingtheChurchillHypothesis.Cambridge:PolityPress;RodericCamp.ed.2001.CitizenViewsofDemocracyinLatinAmericaPittsburgh:UniversityofPittsburghPress;MichaelBrattonandRobertMattes.2001.SupportfordemocracyinAfrica:IntrinsicorInstrumental.BritishJournalofPoliticalScience31(3);MichaelBratton,RobertMattes,andE.GyimahBoadi.2004.PublicOpinion,Democracy,andMarketCambridge:CambridgeUniversityPress;RussellJ.DaltonandDohChullShin,eds.,2006.Citizens,DemocracyandMarketsaroundthePacificRimOxford:OxfordUniversityPress;LarryDiamondandMarcF.Plattner.2008.HowPeopleViewDemocracy.Baltimore:TheJohnsHopkinsPress;YunhanChu,LarryDiamond,AndrewJ.Nathan,andDohChullShin.Eds.2008.HowEastAsiansViewDemocracy.NewYork:ColumbiaUniversityPress;PippaNorris.2010.DemocraticDeficit.Cambridge:CambridgeUniversityPress.38StephenW.RaudenbushandAnthonyS.Bryk.2002.HierarchicalLinearModels(2nded).ThousandOaks,CA:Sage;RobertBickel.2007.MultilevelAnalysisforAppliedResearch:ItsJustRegression!NewYork:theGuilfordPress.39PippaNorris.2009.DrivingDemocracy.NewYork:CambridgeUniversityPress.

Muslim support for secular democracy - Harvard University support for... · muslim support for...

Documents

Transcript of Muslim support for secular democracy - Harvard University support for... · muslim support for...