Late Abrahamic reunion? Religious fundamentalism negatively predicts dual Abrahamic group

12 Pages·2014·0.5 MB·English

European Journal of Social Psychology,Eur. J.Soc. Psychol. 44, 337–348(2014)
Publishedonline16March2014inWileyOnlineLibrary(wileyonlinelibrary.com)DOI:10.1002/ejsp.2014
Research article
Late Abrahamic reunion? Religious fundamentalism negatively predicts dual
Abrahamic group categorization among Muslims and Christians
JONAS R. KUNST1*, LOTTE THOMSEN1,2,3 AND DAVID L. SAM4
1DepartmentofPsychology,UniversityofOslo,Norway;2DepartmentofPsychology,UniversityofCopenhagen,
Denmark;3DepartmentofPsychology,HarvardUniversity,USA;4DepartmentofPsychosocialScience,
UniversityofBergen,Norway
Abstract
Althoughextensiveresearchhasdocumentedtheeffectivenessofcommonordualin-groupsonimprovingintergrouprelations,
littleisknownabouthowindividual-differencevariablesaffectpeople’swillingnesstomakesuchre-categorizationsinthefirst
place.Here,wedemonstratethatindividualdifferencesinreligiousfundamentalismpredictwillingnesstocategorizeintermsof
thecommonAbrahamicreligiousoriginsofChristianityandIslamamongChristiansandMuslims.Study1(n=243Christians,
291Muslims)usesmultigroupstructuralequationmodelingandStudy2(n=80Christians)anexperimentalmanipulationtoshow
thatreligiousfundamentalismcauseslowerdualAbrahamiccategorization,which,inturn,predictsmorepositiveattitudestoward
therespectiveout-group,mediatingthenegativeeffectsofreligiousfundamentalismonreligiousintergroupbias.Whilemakingthe
generalcasethatindividualdifferencesmayplayimportantrolesfordualcategorizations,theseresultsalsohighlightthespecific
positive potential of dual ecumenical categorizations for improving interreligious relations. Research and societal implications
arediscussed.Copyright©2014JohnWiley&Sons,Ltd.
Fundamentalism as it is called is not confined to the Muslim andChristians(Fekete,2004;Zelizer&Allan,2002),focusing
world. It is something that we have seen in different parts of on “insurmountable cultural differences” (Kunst, Tajamal,
the world. Let us hope that a dialogue between the followers Sam, & Ulleberg, 2012, p. 519) and, arguably, on Muslims as
ofthethreegreatmonotheisticreligionscouldhelpinputting “Islamicfundamentalists.” Yet,fromatheologicalperspective,
anend to this. Islam and Christianity share several communalities, and scrip-
King Hussein I, 1999 tures and central religious figures emphasize this common
origin. Because both Islam and Christianity trace their origins
Throughout history, humans have justified the derogation, to the common progenitor Abraham/Ibrahim, Muslims as well
discrimination orevenpersecutionofmembersoffaithsother as Christians believe in an Abrahamic religion, and a number
thantheirown,oftenonthebasisofminusculetheologicaldif- of other religious figures, such as Noah/Nuh, Adam, Moses/
ferences (Gort & Vroom, 2002). And also in contemporary Musa,Jona/YunusandJesus/Issa,arecentralinbothreligions.
WesternEurope,theclimatetowardreligiousminoritygroups Indeed,thereisacertaintheologicalawarenessandacknowledg-
tends to be tense. About 14million Muslims presently live in ment of a common ecumenical Abrahamic group in both
WesternEuropeasaresultoflabormigrationintheaftermath religions (Boase, 2005). For instance, the Surat Al-Baqarah of
of World War 2 (Maréchal, 2002), and countries that were theQur’anstates:
relatively homogeneous in terms of religious beliefs have
Surelythosewhobelieve,andthosewhoareJews,andthe
becomemultireligiousoverthelastfewdecades.Thisprofound
Christians,andtheSabians,whoeverbelievesinAllahand
changehasnotbeenembracedbyeveryone.Inparticular,since
the last day and does good, they shall have their reward
the terror attacks of 9/11, religion has resurged as a salient
fromtheirLord,andthereisnofearforthem,norshallthey
social marker in the discourse on intercultural relations in the
grieve (Qur’an,2:62; also see Qur’an,5:69).
West (Poynting & Mason, 2007; Strabac & Listhaug, 2008),
andreportshavedocumentedariseinnegativeattitudestoward
In a similar vein, the deceased pope Paul VIstated:
Muslims and Islam (see, e.g., EUMC, 2005; EUMC, 2006;
Kunst,Sam,&Ulleberg,2013). The church regards with esteem also the Moslems. They
The public discourse in many western European countries has adore the one God, living and subsisting in Himself (…)
beeninordinatelypreoccupiedwithwhatdistinguishesMuslims and take pains to submit wholeheartedly to even His
*Correspondenceto:JonasR.Kunst,DepartmentofPsychology,UniversityofOslo,Postboks1094Blindern,0317Oslo,Norway.
E-mail:[email protected]
Copyright © 2014 JohnWiley& Sons, Ltd. Received23 December 2012, Accepted 16February2014
338 Jonas R. Kunst et al.
inscrutable decrees, just as Abraham, with whom the faith (Altemeyer & Hunsberger, 1992, 2004). Although a few
of Islam takes pleasure in linking itself, submitted to God studieshaveshownthatsocialdominanceorientationmoderates
(Pope Paul VI, 1965). the relation between common in-group categorizations and out-
groupbias(Esses,Dovidio,Jackson,&Armstrong,2001;Esses,
Unfortunately, little attention has been paid to these striking
Wagner, Wolf, Preiser, & Wilbur, 2006; also see Thomsen,
theological communalities between Christianity and Islam,
Green,&Sidanius,2008),itremainsunclearifandhowindivid-
although they may positively improve intergroup relations
ual-differencevariablescauseindividualstodifferentiallyendorse
between Christian and Muslim citizens—one of the most
dualgroupcategorizationsinthefirstplace.Here,wearguethat
pressing social issues in contemporary Europe: Because be-
religiousfundamentalismshouldnegativelyaffectbelievers’will-
longing to a group promotes intergroup bias (Tajfel, 1970;
ingnesstocategorizeintermsofdualreligiousgroups.
Tajfel, Billig, Bundy, & Flament, 1971), it should be the case
Religiousfundamentalismcanbedefinedas“thebeliefthat
thatwhen“membersofdifferentgroupsareinducedtoconceive
thereisonesetof religiousteachings thatclearlycontains the
of themselves as a single group rather than two completely
fundamental, basic intrinsic, essential, inerrant truth about
separategroups,attitudestowardformerout-groupmemberswill
humanityanddeity”(Altemeyer&Hunsberger,1992,p.118).
becomemorepositivethroughprocessesinvolvingpro-ingroup
Capturingthis,AltemeyerandHunsbergerfoundtheirreligious
bias” (Gaertner, Dovidio, Anastasio, Bachman, & Rust, 1993,
fundamentalism scale to predict prejudice and right-wing
p.6).Aseriesofexperiments(e.g.,Gaertner&Dovidio,2000;
authoritarianism so strongly that they speculated that religious
Gaertner, Dovidio, & Bachman, 1996; Gaertner et al., 1993)
fundamentalism may simply constitute a specific expression of
and correlational research (see Gaertner & Dovidio, 2000,
right-wingauthoritarianisminthereligiousdomain.Yet,studies
2005) support this common in-group identity model (CIIM;
haveshownthat,althoughbothconstructsarehighlyrelated,they
Gaertner & Dovidio, 2000) to great extent. In some studies,
accountforuniquevarianceinprejudice(Laythe,Finkel,Bringle,
however, efforts to establish a new, common in-group identity
&Kirkpatrick,2002;Laythe,Finkel,&Kirkpatrick,2001).
have led tomoreout-group bias (Hornsey& Hogg,2000; also
Fundamentalistsdifferfromtheirpeerswhoscorelowonreli-
seeDovidio,Gaertner,&Saguy,2007).Possibly,thisisbecause
giousfundamentalisminmanyways.Theyhavelessdoubtsabout
highidentifierscanexperiencetheestablishmentofaninclusive
their own religious belief and are less tolerant toward religious
common group as a threat to their subgroup identity (Dovidio
disagreement(Wrench,Corrigan,McCroskey,&Punyanunt-Car-
etal.,2007).Inthepresentcase,suchidentitythreatwouldlikely
ter,2006)andaremoreconcernedaboutotherpeoplenotliving
resultfromaskingMuslimsorChristians,whoidentifystrongly
uptotheirreligiousstandardsthanpeerslowonreligiousfunda-
with their religious faiths, to no longer identify as Muslims or
mentalism(Hunsberger,Alisat,Pancer,&Pratt,1996).Theyalso
Christians, but instead as “Abrahamics.” Instead, it should be
tendtobemorenegativetowardwomen(see,e.g.,Blogowska&
more beneficial to establish dual categorizations that allow
Saroglou, 2011; Hunsberger, Owusu, & Duck, 1999), homosex-
individuals to maintain pronounced subgroup identities within
uals (see Whitley, 2009 for a meta-analysis) and ethnic (Hall,
an overarching common group (Dovidio et al., 2007)—as
Matz,&Wood,2010)andreligiousout-groups(Johnson,Rowatt,
“Muslim-Abrahamics”and“Christian-Abrahamics,”sotospeak.
& LaBouff,2012; Rowatt,Franklin, & Cotton, 2005).Although
Indeed, such dual categorizations generally appear to reduce
this religious fundamentalism–prejudice link is empirically well
out-group bias to similar,or even better,degrees than common
established, so far research exploring its underlying processes
groupcategorizations(Eller&Abrams,2004).
hasonlyfocusedonvariouscognitivestyles.However,because
Here, we investigate whether dual categorizations, including
religiousfundamentalismisalsoacentralthemeofthepublicdis-
boththeAbrahamicandChristian/Muslimidentities,reducebias
course on interreligious group relations in the West, core
toward the respective Abrahamic out-group and whether
intergroupprocessesshouldalsoplayimportantrolesforitsen-
religious fundamentalism affects their willingness to consider
dorsement.Inparticular,theverycognitivestylesthatcharacter-
these common religious origins in the first place. Specifically,
izereligiousfundamentalistsshouldalsomakethemlessinclined
we expect that religious fundamentalism will cause lower dual
toendorsedualreligiousgroups,which,inturn,willhavesubse-
Abrahamic categorization, which, in turn, should mediate the
quentnegativeimpactonintergrouprelations,mediatingthefun-
negativeeffectsofreligiousfundamentalismoninterreligiousbias.
damentalism–prejudicelink.
Forreligiousfundamentalists,whotendtotakeholyscriptures
literallyandasthefinitetruth,religionprovidesaclear-cutandplain
RELIGIOUS FUNDAMENTALISM AS NEGATIVE
view on life and thereby “a sense of coherence in an otherwise
PREDICTOR OFDUAL CATEGORIZATION AMONG
chaoticworld”(Hood,Hill,&Williamson,2005,pp.17–18).This
RELIGIOUS GROUPS
notionissupportedbythefactthatreligiousfundamentalistshave
beenfoundtohavealesscomplexunderstandingofreligiousissues
Although there may be strong theological ground for catego- thanindividualslowinreligiousfundamentalism(Pancer,Jackson,
rizing related religions into supra-ecumenical groups, at the Hunsberger, Pratt, & Lea, 1995). Consequently, Christian or
individual level, believers may strongly differ in the degree Muslimfundamentalistsmaypaylittleattentionto,ignoreor
to which they endorse such dual groups. Despite extensive evenrejectcomplexintertwinedtheologicalrelationsbetween
researchontheCIIM,ithasseldombeeninvestigatedinlight ChristianityandIslamandconsequentlyexpresslowendorse-
ofindividual-differencevariablesthatareconsistentpredictors mentofadualAbrahamicgroup.
ofprejudice,suchassocialdominanceorientation(Sidanius& Similarly, their close-minded world view (Hunsberger &
Pratto, 1999), right-wing authoritarianism (Altemeyer, 1988) Jackson, 2005) makes fundamentalists vehemently reject
or—central in the present study—religious fundamentalism alternative ways to see the world (Brandt & Reyna, 2010;
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
Late Abrahamic reunion 339
Wrenchetal.,2006).Infact,theout-groupderogationtypical 2. dual Abrahamic categorization is related to less out-group
of religious fundamentalists may function precisely to protect bias, mediating the effect of religious fundamentalism on
their world views (Brandt & Reyna, 2010). As Hood et al. out-groupbias(Figure1).
(2005)state,“forfundamentalists,thereisbutonetrueavenue
BothstudieswereconductedinGermany,becausereligion
forfindingmeaning”(p.30)andthatinvolvesexclusivelyfol-
in Germany constitutes a salient social marker and because
lowing their own specific religion, which is seen as the one
interreligious tensions between Muslims and Christians in
and only path to salvation. Their religion, in turn, provides
Germanycanbedescribedastense(Kunstetal.,2012;Kunst
answertoanxiety-elicitingquestionssuchasthoserelatedtoone’s
etal.,2013).MorethantwothirdsoftheGermanpopulationis
own mortality (Friedman & Rholes, 2008; Vail et al., 2010).
Christian, and in contrast to other West-European countries,
Acknowledging that one’s own religion is related to other
thereisnojuridicaldivisionbetweentheChurchandtheState.
religionsorevenpartlyisbasedonthesametheologicalground,
Hence, compared with European countries practicing the
however,canbeseenasrelativizingtheabsolutenessandinfalli-
principle of Laïcité, such as France, Christian religious
bilityofone’sownreligiousmeaningsystem.Hence,forfundamen-
educationistaughtinschools,andlarge-scalepoliticalparties
talistMuslimsorChristians,endorsingcommonoriginswithanother
are declared Christian.
Abrahamic group that interprets the same religio-historical events,
In Study 1, we test our mediation prediction using a
forinstance,theroleofJesus,inadifferentmannermayshakethe
multigroup structural equation model with Christians and
very fundament of their meaning system that provided ultimate
Muslims. Here, we alsocontrol for religious identity as it has
certainty.Consequently,ChristianandMuslimreligiousfundamen-
beenfoundtobehighlycorrelatedwithreligiousfundamental-
talists should reject dual categorizations insofar as it constitutes a
isminearlierresearch(Shaffer&Hastings,2007)andbecause
threattotheirfundamentalist,rigidandpreclusiveconceptions.
it may also predict individuals’ willingness to endorse com-
Lastly, and importantly, religious fundamentalists identify
monordualgroups(Gaertner&Dovidio,2000).Hence,con-
stronglywiththeir religious group(Shaffer & Hastings,2007) and
trolling for religious identity allows us to assess the unique
so may arguably perceive categorizing themselves into a dual
predictive role of religious fundamentalism. In Study 2, we
religious group as an identity threat (Dovidio et al., 2007). For
useanexperimentalsaliencymanipulationtoconfirmthatreli-
fundamentalists, belonging to their specific religion and religious
gious fundamentalism causes dual Abrahamic categorization
groupconstitutesacentralpartoftheirself-concept,whichmayhave
and, once again, using a structural equation model, test the
beenemphasizedfromthebeginningoftheirlives(Altemeyer,2003).
mediation model proposed—this timewith causal data.
Blurringthesegroupboundariesbyacknowledgingdualorcommon
groupmembershiptogetherwitharelatedreligionmayagainthreaten
individuals’ self-concepts. Thus, to maintain their clear group
STUDY 1
distinctionsbetween“them”and“us,”ChristianandMuslimfunda-
mentalistsshouldlikelyrejectanytypeofdualAbrahamicgroup.
This first study tests our proposed model with data obtained
from a sample of Christians and Muslims. In terms of group
Overview of theStudies
bias, we distinguish between affective and cognitive aspects
of out-group bias, as has been suggested in earlier studies
In the two studies reported here, we empirically test whether
(Stangor, Sullivan, & Ford, 1991; Tropp & Pettigrew, 2005),
1. religiousfundamentalismnegativelypredictsdualAbrahamic to gain a more comprehensive insight into the broadband of
groupcategorizationand the effects of dual Abrahamic categorization. Given that
Negative Out-Group
Emotions
+
- Negative Out-Group
Stereotypes
Religious
- -
Fundamentalism
-
Dual Abrahamic -
+ -
Categorization
-
Religious
+ -
Identity
Positive Out-Group
+ Stereotypes
+
Positive Out-Group
Emotions
Figure1. Hypothesizedrelations
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
340 Jonas R. Kunst et al.
subgroup identification in earlier research has negatively Unless stated otherwise, responses were rated on 6-point
predicted common or dual group categorization (Gaertner & Likert-type scales, with endpoints 1 (totally disagree) and 6
Dovidio, 2000), we control for the latter variable in the (totally agree). All scales showed satisfactory structural
model. Including both religious identity and religious equivalence(Table1),whichisaprerequisiteforcross-cultural
fundamentalismthathavebeenshowntobehighlycorrelated research. Valid group comparisons can only be conducted
in earlier research (Shaffer & Hastings, 2007), but seldom when equivalence in underlying factor structures of the
are considered together, also allows us to assess the measures has been established (Matsumoto & van de Vijver,
unique effect of the individual-difference variable religious 2012). Furthermore, reliability was satisfactory for all scales
fundamentalism. despitethenegativeout-groupstereotypesscale,whichshowed
acceptablereliabilityintheMuslimsample(Table1).
Methods
Religious Identity
Participants
An adjusted version of the identity subscale of the Collective
In all, 243 professed Christians and 291 professed Muslims Self-Esteem Scale (Luhtanen & Crocker, 1992) was used to
participated in the study. Most participants were young adults, measure theparticipants’ religious identity(Christians:α=.89;
M =23.62,SD =6.42,andnoagedifferencesbetweenthe Muslims:α=.85).Themeasurecanbeseenasassessingidenti-
age age
groupswereobserved,t(532)=1.47,p=.143.Thesampleswere fication with one’s religious group from a social identity per-
alsocomparableintermsofgender(Christians:46.5%women; spective,ratherthanmeasuringintrinsicorextrinsicreligiosity.
Muslims: 51.2% women) and educational status (Christians: Participants had to indicate their agreement with four items,
30.3%lowersecondaryschool,46.1%uppersecondaryschool, namely“Overall,myreligionhasverylittletodowithhowIfeel
23.9% university degree; Muslims: 38.8% lower secondary about myself” (reversed), “The religion I belong to is an
school, 38.8% upper secondary school, 22.3% university de- important reflection of who I am,” “The religion I belong to is
gree).AllMuslimparticipantshadaTurkishethnicbackground. unimportanttomysenseofwhatkindofapersonIam”(reversed)
Of the Christian participants, 68.3% described their ethnic and“Ingeneral,belongingtomyreligionisanimportantpartof
background as German, followed by 23.6% who indicated a myself-image.”
Europeanstateastheirculturalheritagecountry.AmongChris-
tians, 37.4% identified themselves as Catholics and 36.3% as Religious Fundamentalism
Protestants,whereasthe Muslimsampleprimarilyconsistedof
SunniMuslims(83.8%). We adopted the revised Religious Fundamentalism Scale
developed by Altemeyer and Hunsberger (2004) to measure
Procedure and Measures the degree to which the participants held fundamental beliefs
about their religion. To increase the applicability of the scale
Participants were recruited through social online networks, amongbothChristiansandMuslims,thefollowingadjustments
relevant webpages and snowball sampling in November weremade:First,wedeletedtwoitemsthatdealtwiththeexis-
2012.Allrespondentswereinformedthatthestudydealtwith tenceof“Satan,”becausetheconceptwasregardedastheologi-
interreligious issues accompanied by a picture showing a cally too variant. Second, expressions such as “God,” “God’s
circleofthesymbolsrepresentingthebiggestworldreligions. religion”and“sacredscripture”werereplacedby“God/Allah,”
Moreover, they were informed about the confidentiality and “Christianity/Islam” and“Bible/Qur’an,” respectively. Accord-
theirrighttowithdrawfromparticipation.Asafinancialincen- ingly,participantshadtoindicatetheiragreementwith10items,
tive,participantscouldparticipateinthedrawingofa€50gift suchas“Toleadthebest,mostmeaningfullife,onemustbelong
voucher. At the beginning of the study, participants were to the one, fundamentally true religion.” Exploratory factor
asked whether they were religious or not, and non-religious analyses conducted separately for both samples supported a
individuals were automatically excluded from participation. one-factorsolution.Toachieveaninvariantfactorstructurethat
Table1. Psychometricpropertiesandstructuralequivalenceforthestudyvariables
α ConstrainedCFA
Scale Items Christians Muslims CFI RMSEA χ2 p df
1.Fundamentalism 8 .897 .883 .993 .026 56.67 .065 42
2.Religiousidentity 4 .894 .846 1.000 .000 3.85 .697 6
3.DualAbrahamiccategorization 4 .875 .863 1.000 .009 6.27 .394 6
4.Negativeout-groupemotions 4 .926 .906 .981 .055 88.12 .000 34
5.Positiveout-groupemotions 4 .878 .847
6.Negativeout-groupstereotypes 5 .824 .642 .964 .039 134.67 .000 75
7.Positiveout-groupstereotypes 5 .872 .796
Note:Christiansn=243;Muslimsn=291.ConstrainedCFA:Measurementweightsandstructuralcovarianceswereconstrained.Two-factorstructureswere
obtainedforboththeemotionandstereotypemeasures.
CFI,comparativefitindex;RMSEA,rootmeansquareerrorofapproximation.
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
Late Abrahamic reunion 341
isnecessaryincross-culturalresearch,wedeletedtwoitems(i.e., Stereotypes Toward the Out-Group
“When you get right down to it, there are basically only two
kinds of people in the world: The Righteous, who will be AnindexdevelopedbyStephanandStephan(1996)wasused
rewarded by God; and the rest, who will not” and “There is a to assess the participants’ stereotypes toward the respective
particularsetofreligiousteachingsinthisworldthataresotrue, out-group. In the first step, participants had to indicate how
you can’t go any ‘deeper’ because they are the basic, bedrock many percent of the out-group they perceived as possessing
message that God has given humanity”) that showed different 12 different types of traits. Next, participants were asked to
loadingsacrossthesamples.Theresultingscalehadaninvariant evaluatethedegreetowhichtheyfounditfavorableforaper-
one-factor structure and good reliability across the samples son to possess the different traits on a 9-point scale ranging
(Christians:α=.90;Muslims:α=.88;Table1). from1(verybad)to9(verygood).Finally,toconstructindex
itemsthattakeintoaccountboththeperceivedfavorabilityand
theprevalenceofthetraitamongtheout-group,wemultiplied
Dual Abrahamic Categorization
each percentage item with the respective valence item.
After deleting two of these index items (i.e., clannish and
Building on the dual identity conceptionalization of Dovidio
proud),factor analysis supported atwo-factorsolution, which
etal.(2007),sevenitems,ofwhichtwowerereversedworded,
showed acceptable structural equivalence across the samples
were developed specifically for this study. The measure
(Table 1). The first factor represented negative stereotypes
assessed the degree to which the participants considered
(i.e., ignorant, aggressive, undisciplined, unintelligent and
Muslims and Christians as belonging to a common group of
dishonest; Christians: α=.83; Muslims: α=.64), whereas the
Abrahamic religions while also constituting two distinct
secondfactorrepresentedpositivestereotypes(i.e.,respectful,
religious groups. Accordingly, the items acknowledged the
hardworking, friendly, reliable and clean; Christians: α=.87;
uniqueness and autonomy of each religious group while also
Muslims: α=.80).
emphasizing the common Abrahamic in-group. Exploratory
factor analysis yielded a one-factor solution in both samples.
Results
Yet, to achieve structural equivalence across the samples, the
two contrait items (i.e., “Islam and Christianity constitute
TestforDifferencesintheMeansoftheMainStudyVariables
totallydifferentreligions”and“MuslimsandChristiansbelong
to two groups, which couldn’t be more different”) and one
Muslim participants (M=4.35, SD=1.32) on average scored
protrait item (i.e., “Although Christians and Muslims belong
significantly higher on religious fundamentalism than did
to different religions, they are united in their belief in God”)
Christian participants (M=3.71, SD=1.38; see the note of
hadtobedeleted.Consequently,thefinalscalecomprisedfour
Table 2 for all effect sizes of intersample comparisons).
items: “Because Abraham/Ibrahim is the progenitor of both
Muslim participants also expressed a higher dual Abrahamic
IslamandChristianity,onecansaythatMuslimsandChristians
categorization(M=4.47,SD=1.28)thantheirChristiancounter-
belongtothesame‘family’ofreligions,”“Christiansaswellas
parts(M=3.91,SD=1.44),supportingearlierresearchshowing
MuslimsbelieveinanAbrahamicreligion,”“EventhoughIslam
that,inparticular,minoritymembersendorsedualidentitycate-
andChristianityaredifferentreligions,bothbelongtothesame
gorizations(e.g.,Ryan,Hunt,Weible,Peterson,&Casas,2007).
group of religions” and “Christianity and Islam have common
Christians showed more negative and less positive out-group
roots”(Christians:α=.88;Muslims:α=.86).
emotions (negative out-group emotions: M=3.11, SD=2.45;
positiveout-groupemotions:M=4.52,SD=2.05)thanMuslim
Emotions Toward the Out-Group participants (negative out-group emotions: M=2.44, SD=1.97;
positiveout-groupemotions:M=5.58,SD=1.94).
A scale developed by Stephan, Ybarra, and Bachman Among Muslims, differences in religious fundamentalism
(1999) measured the participants’ emotions toward the and dual Abrahamic categorization were observed in terms
respective out-group (i.e., Muslims or Christians). On a of education. Muslim participants with a university degree
9-point scale ranging from 1 (not at all) to 9 (extremely), (M=3.91,SD=1.50)scoredloweronreligiousfundamentalism
participants were asked to indicate the degree to which they than their peers who held a lower secondary school degree
felt 12 different types of emotions toward members of the (M=4.60, SD=1.14; F(2, 290)=5.92, p<.01, est η2=.04).
religious out-group. Moreover, Muslim participants with upper secondary school
Exploratory factor analysis supported a two-factor (M=4.70, SD=1.15) or university education (M=4.76, SD=
structure across the samples, where the first factor repre- 1.26) expressed a higher dual Abrahamic categorization than
sented negative and the second factor positive out-group theirpeerswithlowersecondaryeducation(M=4.07,SD=1.32;
emotions. After deleting four items (i.e., admiration, accep- F(2, 290)=9.52, p<.001, est η2=.06). In contrast, no
tance,superiorityandrejection) thathadsubstantial cross-load- differencesbetweendualAbrahamiccategorizationandreligious
ings (>.3), the factor structure was relatively equivalent across fundamentalismwereobservedforeducationamongChristians.
the samples (Table 1). Consequently, two sum scores were
created: one comprising four negative out-group emotions Structural Equation Model
(i.e., hostility, dislike, disdain and hatred; Christians: α=.93;
Muslims: α=.91) and one comprising four positive out-group Intermsofourgeneralanalyticalstrategy,wetestedthehypo-
emotions (i.e., affection, approval, sympathy and warmth; theticalmodel(Figure1)usingmultigroupstructuralequation
Christians:α=.88;Muslims:α=.85). modeling (SEM). Because chi-square difference tests are
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
342 Jonas R. Kunst et al.
susceptible to sample size, model fit was additionally
assessed using root mean square error of approximation ****
****
(RMSEA),comparativefitindex(CFI)andstandardizedroot 7. 694*4*1*4*
004665
mean square residual (sRMR). In terms of mediation, we (cid:1)...(cid:1)..(cid:1).
applied the mediation requirements proposed by Hayes
(2009), which have become widely accepted and more
common than the conservative requirements initially *** *
*** *
proposed by Baron and Kenny (1986; see Rucker, Preacher, 6. 23*5*5*8* 5*
01243 3
Tormala, & Petty, 2011 for a discussion). Here, we used (cid:1).(cid:1).(cid:1)..(cid:1). (cid:1).
bootstrappingwithasampleof5000andconfidenceinterval
of .95 to test for mediation. 01.
0
Specifically, the analytic procedure involved the following ** ** <.
** ** p
steps: First, a constrained version of the hypothetical model 5. 130*4* 3*6* **
0146 25 *
was estimated to give initial information about the adequacy (cid:1)...(cid:1). (cid:1).. 1,
0
of the model across both study groups. To test for mediation .
<
in the next step, direct paths from religious fundamentalism p
*
*
and religious identity to the bias measures were added. This * *** 5,
* *** 0
allowed an estimation of the extent of the mediation (i.e., 4. 56*2* 8*5*3* <.
whether the relations were fully or partly mediated by dual (cid:1).0(cid:1).1(cid:1).3 (cid:1).4.3(cid:1).3 p*
Abrahamic categorization) when bootstrapping the indirect de.
si
effects while also controlling for religious identity. Last, to d
n
identify a final model that optimally fittedthe data,constraints ha
isinnydstithceeemsahttyoicpaaoldtlhdyefptoiacrtahclshmit-hsoaqdtuecalorewuldedrifeifmerrepedrnoucvceeesdathnpedamtuhsoibdnyegl.mpaothd,ifitceasttiiongn 3. (cid:1).46***(cid:1).23*** (cid:1).32***.32***(cid:1).25***.31*** slims,left-
u
Resultssupportedthefitoftheinitialmodel1withstructural M
e;
5lwRa3Mre4gig)eSh=EstsA5am8a=.np6d.l90e,c5spo0(v.<TaTar.in0haa0nek1ca,ec,swh1ich-9osicq8nhu7sta)ir.rsaeiAndnosetdteeu,sxnstpRuewsMcuataeRsldi=,snib.ge0ons8ttih3ifim,ciCaanntFittoI,en=rχsm.2w9s(72iot50hf,, 2. .67*** (cid:1).05.07(cid:1).09(cid:1).14*(cid:1).09 right-handsid
correlations(Table 2)andcoefficientsinthe model(Figure 2), ns,
a
religious fundamentalism was significantly and negatively sti
related to dual Abrahamic categorization, which supported the ******** ** Chri
first hypothesis. Dual Abrahamic categorization, in turn, was 1. 652014170618 er:
relatedtolessnegativeandmorepositiveout-groupbias on all .(cid:1)..(cid:1).(cid:1).(cid:1). ord
g
out-groupmeasures,alsogivingclear support for the second n
hmyepnotsthfeosrism(Fedigiautrieon2)w.Gerievefnultfihlelseedr(eHlaatyioenss,,2t0h0e9b)a.sHicernecqeu,iwree- ms 1.32)1.41)1.28)1.97)1.94)59.11)165.77) efollowi
sreeltigoiuotustofutnedstamtheentafulilslmmheaddiatsiiognnaifilccahnatinin,dtiersetcintgeffwehctesthoenr Musli 5(8(7(4(8(0(9( dinth
the bias measures that were mediated by dual Abrahamic 4.34.14.42.45.54.61.4 aye
cpmithdaaeeetteahnmgsstiuootbryrdeiezesatalwn,wtdieχoee2nrt(nhe.1e7rHae,doel5duirg3eet-i4,dog)wura=osneu4dfpe1usd.nm2tiidr4meea,acmapsttuee=rrndee.tl0saaa0ltiwim1som,enorssdeRaebcMnleodtiRnwntth=rewoee.lhn0loei6crdu5eht,l-figCodgrriiFor.oeIuuIc=npst studyvariables MSD() stians (1.38)(1.42)(1.44)(2.45)(2.05)(76.29)7(180.72)44 fincoefcientsdispl
f1u.n0d0a,mReMnStaEliAsm=.o0n60t,hneoosuigt-ngirfiocuapntmdieraescutreefsfewctesroeforeblsiegrivoeuds emain Chri 3.714.293.913.114.5278.8755.83 orrelatio
tt(S1celIafamuhhT.oqnpt2nnuhieepaode5tieolavnatc3reiimastnimnralfin<ieregndneoannctrgmtdlictiopyaraveeaa,ellusl.eiilblscnlDpym<mowteit,etdeh,aes.deswmp7ttihhfairbhs7tfeetete,ieiie7cohlassitchinnyta)rsge,tic,mpvmdsiilonuwioeatod.habtuldmhuielnecessrtelhionget,crw,awaaetnpwhlfsahegruaouieeseentsndrhsredneibedcifiwdnstoaasturhtmoteaethidhldtcleueseiAhwmgantsoblrwtioeturraoaedaetrAp-nhurleegeiblpgassrrcmtodimahlnueehifcplaohguaerfimitlmndeygstseihddehchoineagauttolsoimicytnutwdpyoraieeeeefipcernsfnfhr.dfitcoeiistctedarayimitvingechleenittncassdltsosiytmrttfim,rehualtecpihclgaoaaefitliuosntnnroriuoeattddefs-.fl Table2.Meansandcorrelationsforth a1.Fundamentalism2.Religiousidentityb3.DualAbrahamiccategorizationc4.Negativeemotionsd5.Positiveemotions6.Negativestereotypes7.Positivestereotypes4 Notenn:Christians=243;Muslims=291.C2(cid:1)<ηatp(532)=5.47,.001,est=.05;2(cid:1)<ηbtp(487.70)=4.76,.001,est=.04;2<ηctp(461.67)=3.41,.01,est=.02;2(cid:1)<ηdtp(532)=6.11,.001,est=.07.<<<ppp*.05;**.01;***.001.
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
Late Abrahamic reunion 343
.08
Negative Out-Group
Emotions
Religious
Fundamentalism .4
0***-.5
.05 9***
Negative Out-Group
Stereotypes
*** 6
.6 DCuaatle Agobrriazhaatimonic -.49*** -.59***
.11
Positive Out-Group -.31
Stereotypes ***
Religious
.14
Identity .5
2
***
Positive Out-Group
Emotions
.13
Figure 2. Estimated hypothesized multigroup structural equation model with constrained structural weights and covariances. *p<.05;
***p<.001
indirectlypredictedlowerlevelsofpositiveout-groupstereo- PreliminaryDiscussion
types (Christians: β=(cid:1).15, p<.001; Muslims: β=(cid:1).15, p
.001) and emotions (Christians: β=(cid:1).15, p<.001; Muslims: The findings of Study 1 indicate that dual religious groups
β=(cid:1).15,p<.001),andhigherlevelsofnegativeout-groupste- have a substantial potential to reduce prejudice between
reotypes (Christians: β=.10, p<.001; Muslims: β=.13, p theologically related religious groups. However, irrespective
.001) and emotions (Christians: β=.12, p<.001; Muslims: of religious faith, religious fundamentalists, who are known
β=.14, p<.001). Hence, as all indirect effects but none of for holding dogmatic worldviews (Hunsberger & Jackson,
thedirecteffectsofreligiousfundamentalismreachedsignif- 2005;Wrenchetal.,2006),seemedlesswillingtoendorsesuch
icance,theresultsindicatedfullmediation. superordinateecumenicalgroups.Specifically,religiousfunda-
Although our constrained hypothesized model demon- mentalismpredictedlowerlevelsofdualAbrahamiccategori-
strated close data fit, stepwise releasing of the constraints zation,whichledtolessnegativeandmorepositiveout-groupbias.
for each path showed that the hypothetical model could However, we presented religious identity before religious
be improved by removing the constraint on the path fundamentalism and the out-group measures, assuming a role
between religious fundamentalism and dual Abrahamic for the religious identity construct early in the causal chain.
categorization, Δχ2=12.03, Δdf=1, p<.001. Moreover, Consequently, religious identity was made salient to partici-
modification indices showed that the fit of the model could pantsbeforeansweringallotherquestions, posingtheempiri-
be improved by adding a constrained path between cal question whether our results would hold without such
religious identity and negative stereotypes, Δχ2=12.03, salience of religious identity. On the one hand, this presenta-
Δdf=1, p<.001, and unconstrained paths between the tion order should satisfy the categorization saliency criteria
religious identity measure and the remaining bias measures, of Turner (1999) and arguably have led to more negative
Δχ2=20.47, Δdf=6, p=.002. No other modifications of out-groupbias,whichisinlinewithearlierresearch(Johnson,
the hypothetical model were made. The fitted model showed Rowatt,&LaBouff,2010),butincontrasttoouractualresults.
a very good fit to the data, χ2(17, 534)=13.20, p=.723, On the other hand, religious constructs have also been shown
sRMR=.027, CFI=1.00, with the lower boundary of the to prime pro-sociality (Pichon, Boccato, & Saroglou, 2007).
RMSEA even indicating an exact fit for the model, RMSEA Such an effect might, for instance, explain the unexpected
.001, 90% CI [.000, .047]. finding that religious identity predicted more positive out-
In the fitted model, relations generally remained the same, groupattitudesinourunconstrainedmodel.Inthenextstudy,
with the difference that religious identity now was related we address these issues in the sample in which the strongest
to less bias in both groups but particularly among Christians. relationshipbetweenreligiousfundamentalismanddualcatego-
Moreover, after removing the constraint of the relation rization was observed, namely Christians. We also use an
betweenreligiousfundamentalismanddualAbrahamiccatego- experimentalmanipulationofreligiousfundamentalismtofurther
rization, religious fundamentalism was now most strongly confirmthatitdoesinfactcausedualAbrahamiccategorizations,
andnegativelyrelatedtodualAbrahamiccategorizationamong as suggested by the superior fit of our proposed mediation
Christians(Figure3). model compared with a model that specifies that religious
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
344 Jonas R. Kunst et al.
.12/.12
Religious
Fundamentalism Negative Out-Group
Emotions
.3
7
.07/.11 /.31******/-.2-.57***
6
Dual Abrahamic -.25***/-.28*** Negative Out-Group ***
Categorization Stereotypes
*** ***.66/.66 .21/.06 /-.31-.47****** -.28***-.56/-.42*** ***
Positive Out-Group /-.1
Stereotypes 8
**
.16/.14 .5
0
***
/.5
1
***
Religious .21***/-.07 Positive Out-Group
Identity Emotions
.17/.14
Figure 3. Fitted estimated multigroup structural equation model. Coefficients displayed in the following order: Christian sample/Muslim
sample.**p<.01;***p<.001
fundamentalism instead mediates the effect of dual Abrahamic impossible, in this study, we simply varied the salience of
categorizationonout-groupbias. religious fundamentalism beliefs, using an order-manipulation
similar to earlier studies (e.g., Jost & Kay, 2005; Levin, 1996;
see also Pratto, Sidanius, & Levin, 2006; Shih, Pittinsky, &
STUDY 2 Ambady, 1999). Such a manipulation activates and increases
the accessibility of psychological constructs and,given a causal
relationship, may have an effect on dependent variables
While several studies have established the negative causal link
(Schwartz, Bless, Wänke, & Winkielman, 2003). In the present
between dual or common group categorizations and prejudice
study,participantswererandomlyassignedtoaprotraitcondition
(e.g.,Gaertner&Dovidio,2000;Gaertneretal.,1993;Gaertner
oracontrolgroup.Intheprotraitcondition,respondentsindicated
etal.,1996),thisstudyexperimentallytestedthenovelassump-
theiragreementwiththeprotraitreligiousfundamentalismitems
tion that individual difference variables—in this case, religious (e.g.,“Toleadthebest,mostmeaningfullife,onemustbelongto
fundamentalism—causallypredictdualgroupcategorization.In the one, fundamentally true religion.”) on a 6-point Likert scale,
other words, we set out to provide causal evidence for the first
forwhichhigheragreementrepresenthigherreligiousfundamen-
linkofourhypothesizedmediationalmodel(Study1).Toavoid
talism. We expected this condition to increase the salience of
confounding effects caused by religious identity that may have
religiousfundamentalism.Inthecontrolgroup,noreligiousfunda-
been possible in Study 1 and, furthermore, to test whether our
mentalismitemswerepresentedinthebeginningofthestudy,but
resultsholdintheabsenceofreligiousidentitysaliency,wedid
allwerepresentedattheend.Next,weaskedbothgroupstoindi-
notpresentreligiousidentitybeforetheothermeasuresthistime.
cate their agreement with the dual Abrahamic categorization
measure from Study 1 (α=.89), the out-group measures from
Participants Study1(i.e.,negativeout-groupemotions:α=.92;negativeout-
group stereotypes: α=.76; positive out-group emotions: α=.88;
In total, 80 Christians participated in the study. The majority positiveout-groupstereotypes:α=.78)andamanipulationcheck
ofparticipantswereyoungadults(Mage=23.63,SDage=5.16), (i.e.,thecontraitreligiousfundamentalismitems;α=.80).
male (63.0%) and indicated Germany (74.1%) or another
Europeanstate(18.4%)astheirheritagecountry.Intermsofreli- Results
gious traditions, 34.6% of the participants were Protestants,
30.9%Catholicsand34.6%followedanotherChristiantradition. As predicted, participants assigned to the protrait condition
expressedasignificantlylowerdualAbrahamiccategorization
Procedure and Measures (M=3.35, SD=1.49) than their counterparts in the control
group (M=4.06, SD=1.38; t(79)=2.24, p=.028, est
Participants were recruited using the same procedure as in η2=.06). Individuals in the experimental group also scored
Study 1 during February 2013. As manipulating a construct lower (M=2.78, SD=1.43) on the manipulation check (i.e.,
that is deeply entrenched into individuals’ self-concepts, the contrait religious fundamentalism items) than those in the
such as religious fundamentalism, may be difficult or control group (M=3.34, SD=1.37), indicating an
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
Late Abrahamic reunion 345
Negative Out-Group
Emotions
.5
Negative Out-Group 0*** -.32**
FunCRdoaenmlitgreoionl utvassl .is m -.24* AbrDauhaalmic Stereotypes -.34** -.45***
Condition Categorization PosSititveere Ootuytp-Gesro up -.61***
.4
9***
Positive Out-Group
Emotions
Figure4. Estimatedexperimentalmediationmodel.*p<.05;**p<.01;***p<.001
experimentalassimilationeffectintheprotraitcondition,albeit achieve a common understanding that will lead “to a new
a marginally significant one (t(79)=1.79, p=.077, est way of presenting ourtwo religions, not in opposition, as has
η2=.02).Hence,itispossiblethattheexperimentalmanipula- happened too often in the past, but in partnership for the
tionnotonlyheightenedthesaliencebutevendirectlymanip- goodofthehumanfamily.”Asourstudieshighlight,acknowl-
ulated participants’ degree of religious fundamentalism. No edging religious communalities and the common Abrahamic
significant direct effects were observed for any of the out- heritage indeed has the potential to shape attitudes between
group measures (negative out-group emotions: t(78)=(cid:1).29, Christians and Muslims in favorable ways.
p=.770; negative out-group stereotypes: t(78)=.18, p=.858; Toequaldegrees,however,ourstudiesindicatethatreligious
positive out-group emotions: t(78)=.04, p=.968; positive fundamentalism is an obstacle to achieve awareness of such a
out-group stereotypes: t(78)=.32,p=.752). dual group. Our studies showed that the individual-difference
Next,weestimatedthehypotheticalmodelofStudy1using variable religious fundamentalism crucially limits the positive
SEM, with the difference that the religious fundamentalism potential of dual Abrahamic categorization among Muslims
variable was replaced by an experimental dummy variable and Christians. It can nevertheless be argued that individual-
(1=control, 2=protrait condition). Again, the hypothesized differencevariablesandsocialidentityconstructsmayoverlap.
model obtained a good fit to the data, χ2=0.97, p=.967; One may not consider religious fundamentalism as a construct
sRMR=.027, CFI=1.00, RMSEA<.001. In this model separable from religious identity but, in fact, as a construct
(Figure 4), the protrait condition negatively predicted dual representing an extreme form of religious identification,
Abrahamic categorization, which, in turn, was negatively re- involving strong (i.e., fundamentalist) adherence to religious
lated to negative out-group emotions and negative out-group group norms (Hogg, Adelman, & Blagg, 2010; Ysseldyk,
stereotypesandpositivelyrelatedtopositiveout-groupstereo- Matheson, & Anisman, 2010). In line with this, some authors
types and positive out-group emotions. Bootstrapping the have argued that “intergroup tension (rather than individual
resulting indirect effects, the protrait condition had a signifi- psychology) fuels fundamentalist attitudes” (Ysseldyk et al.,
cant indirect effect on positive out-group emotions (β=(cid:1).09, 2010, p. 65). Nevertheless, we included religious identity in
p<.05), positive out-group stereotypes (β=(cid:1).08, p<.05) thefirststudyexactlytodisentangletheeffectsofsocialidentity
and negative out-group feelings (β=.06, p<.05). Moreover, and the individual-difference variable. Here, results indicated
ithadamarginallysignificanteffectonnegativeout-groupste- thatitwasreligiousfundamentalism,andnotidentification,that
reotypes(β=.05,p=.055).Insum,experimentallymanipulat- limiteddualcategorization.
ing religious fundamentalism caused congruent changes in Moregenerally,althoughtheCIIM’spotentialforprejudice
dual Abrahamic categorization, which again mediated the ef- reduction is well documented, there is a lack of studies
fects of religious fundamentalism on out-group bias because investigatingre-categorizationintermsofindividual-difference
it,as in Study 1,was related to less out-group bias. variables and the few studies that exist have solely treated the
latter variables as moderators (e.g., Esses et al., 2006; alsosee
Thomsen, Green, & Sidanius, 2008). As the studies presented
GENERAL DISCUSSION inthispaperhowevershow,individual-differencevariables—in
this case, religious fundamentalism—may also function as
predictors of dual group categorizations early in the causal
Thepresentresearchsupportstheproposalthatreligiousfunda- chain.Futurestudiesshouldfurtherinvestigatewhetherother
mentalismnegativelypredictsdualidentification,which,inturn, individual-differencevariablespredictre-categorizationsinto
leads to more positive and less negative out-group bias. SEM common or dual groups. Clearly, religious fundamentalism,
with both cross-sectional and experimental data supported this involvingthebeliefthatthereisonlyonetruepathforbelievers
proposal,showingthatdualAbrahamiccategorizationmediates to follow, constitutes an individual-difference variable that is
theeffectsofreligiousfundamentalismonout-groupbias. particularly relevant to the dual religious group in the present
In his speech at a mosque in Damascus on 6 May 2001, study. When dual groups, however, are defined in terms of
Pope John II highlighted the importance for Muslims and ethnicity or nationality as mostly has been done in previous
Christians to “explore theological questions together” to studies,variablessuchassocialdominanceorientation(Sidanius
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)
346 Jonas R. Kunst et al.
& Pratto, 1999) and right-wing authoritarianism (Altemeyer, thereforebeawaytodiminishthebreedinggroundofreligiousfun-
1988)mayplaymoreofarole,astheyareknownforpredicting damentalism among Christians, which eventually may lead to a
out-groupbiasinthiscontext(e.g.,Whitley,1999). heightened awareness of Muslims’ and Christians’ common
Abrahamicorigin.
Societal Implications
Despite striking similarities between religions such as Islam REFERENCES
and Christianity, “most people tend to fixate on religious dif-
ferences rather than similarities” (Plante, 2009, p. 75). The
Altemeyer,B.(1988).Enemiesoffreedom:Understandingright-wingauthor-
present study highlights the positive potential of being aware
itarianism.SanFrancisco,US-CA:Jossey-Bass.
of common theological origins for interreligious relations in Altemeyer, B. (2003). Why do religious fundamentalists tend to be
the West, and it also shows that this potential is critically prejudiced? International Journal for the Psychology of Religion, 13(1),
17–28.DOI:10.1207/S15327582IJPR1301_03
limited by religious fundamentalism.
Altemeyer,B.,& Hunsberger, B.(1992).Authoritarianism,religious funda-
Believers, who in addition to identifying with their mentalism,quest,andprejudice.InternationalJournalforthePsychology
religious group agreed with belonging to the overarching ofReligion,2(2),113-133.DOI:10.1207/s15327582ijpr0202_5
Altemeyer,B.,&Hunsberger,B.(2004).Arevisedreligiousfundamentalism
Abrahamic group, were substantially less prejudiced toward
scale:Theshortandsweetofit.InternationalJournalforthePsychologyof
theAbrahamicout-group.Theseencouragingresultsindicate Religion,14(1),47-54.DOI:10.1207/s15327582ijpr1401_4
thatthesupra-ecumenicalcategorizationwasnotperceivedas Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable
athreattothedistinctivenessofbelievers’religioussubgroup distinction in social psychological research: Conceptual, strategic, and
statistical considerations. Journal of Personality and Social Psychology,
identification, which might have led to worse out-group atti- 51(6),1173-1182.DOI:10.1037/0022-3514.51.6.1173
tudes (see, e.g., Bianchi, Mummendey, Steffens, & Yzerbyt, Bianchi, M., Mummendey, A., Steffens, M. C., & Yzerbyt, V. Y. (2010).
What do you mean by “European”? Evidence of spontaneous ingroup
2010; Waldzus, Mummendey, Wenzel, & Weber, 2003). In-
projection. Personality and Social Psychology Bulletin, 36(7), 960-974.
deed,morerecentevidenceshowsthatdualAbrahamiccatego- DOI:10.1177/0146167210367488
rizationaffectsnotonlyattitudinalbutalsobehavioralformsof Blogowska,J.,&Saroglou,V.(2011).Religiousfundamentalismandlimited
prosocialityasafunctionofthetarget.JournalfortheScientificStudyof
bias as it leads believers to altruistic and real monetary
Religion,50(1),44-60.DOI:10.1111/j.1468-5906.2010.01551.x
donationstomembersofAbrahamicout-groupsinneed(Kunst Boase,R.(2005).EcumenicalIslam:AMuslimresponsetoreligiouspluralism.
&Thomsen,2014). InR.Boase(Ed.),Islamandglobaldialogue:Religiouspluralismandthe
pursuit of peace (pp. 247-266). Farnham, United Kingdom: Ashgate
Although dual categorization seems to contribute to more
Publishing.
harmonious interreligious relations, religious fundamentalism Brandt, M. J., & Reyna, C. (2010). The role of prejudice and the need for
emerged as a key obstacle to such dual group endorsement. closure in religious fundamentalism. Personality and Social Psychology
Bulletin,36(5),715-725.DOI:10.1177/0146167210366306
Beinganindividual-differencevariablethatregardscorecharac-
Dovidio,J.F.,Gaertner,S.L.,&Saguy,T.(2007).Anotherviewof“we”:
teristicofpeople’sfaith,religiousfundamentalismcanbeseenas Majorityandminoritygroupperspectivesonacommoningroupidentity.
difficult to change. Yet, the marginally significant manipulation European Review of Social Psychology, 18(1), 296-330. DOI: 10.1080/
10463280701726132
check in Study 2 suggests that the construct has at least some
Eller,A.,&Abrams,D.(2004).Cometogether:Longitudinalcomparisonsof
degreeofflexibilitysothatinterventionsmaypossiblychangeit. Pettigrew’s reformulated intergroup contact model and the common
Here,our first study highlightedthe potential of education, as it ingroupidentitymodelinAnglo-FrenchandMexican-Americancontexts.
wasnegativelyrelatedtoreligiousfundamentalismandpositively European Journal of Social Psychology, 34(3), 229-256. DOI: 10.1002/
ejsp.194
relatedtodualAbrahamiccategorization,atleastamongMuslims.
Esses,V.M.,Dovidio,J.F.,Jackson,L.M.,&Armstrong,T.L.(2001).The
It is possible that Muslim participants with higher formal immigration dilemma: The role of perceived group competition, ethnic
education had obtained more knowledge about the complex prejudice,andnationalidentity.JournalofSocialIssues,57(3),389-412.
DOI:10.1111/0022-4537.00220
intertwined relations between their own and others’ religious
Esses, V. M., Wagner, U., Wolf, C., Preiser, M., & Wilbur, C. J. (2006).
traditions, which challenge the less complex views on religious Perceptions of national identity and attitudes toward immigrants and
issuesheldbyreligiousfundamentalists(Panceretal.,1995). immigrationinCanadaandGermany.InternationalJournalofIntercultural
Relations,30(6),653-669.DOI:10.1016/j.ijintrel.2006.07.002
However, education did not significantly affect religious EUMC.(2005).Majorities’attitudestowardsminorities–Keyfindingsfrom
fundamentalism among Christian participants. Nevertheless, it the Eurobarometer and the European Social Survey. Vienna, Austria:
wasinthisgroupthatreligiousfundamentalismemergedasthe EuropeanMonitoringCentreonRacismandXenophobia.
EUMC. (2006). Muslims in the European Union: Discrimination and
strongestpredictorofdualgroupendorsement.Ontheonehand,
Islamophobia. Vienna, Austria: European Monitoring Centre on Racism
thisfindingchallengesthebiasofthepublicdiscourseinmany andXenophobia.
countries,where“themediahavenotscrupledtotarallMuslims Fekete,L.(2004).Anti-MuslimracismandtheEuropeansecuritystate.Race
with the same fundamentalist brush” (Fekete, 2004, p. 23), &Class,46(1),3-29.DOI:10.1177/0306396804045512
Friedman,M.,&Rholes,W.S.(2008).Religiousfundamentalismandterror
whereasself-declaredChristianfundamentalists,suchasAnders management.InternationalJournalforthePsychologyofReligion,18(1),
Behring Breivik (the Norwegian mass murderer), seldom are 36-52.DOI:10.1080/10508610701719322
Gaertner, S. L., & Dovidio, J. F. (2000). Reducing intergroup bias: The
connectedwithfundamentalism(Rozbahani &Just,2012).On
Common Ingroup Identity Model. Philadelphia, US-PA: Psychology
the other hand, it begs the question of how Christian religious Press.
fundamentalistconceptionscanbealtered.Studieshaveshown Gaertner, S. L., & Dovidio, J. F. (2005). Understanding and addressing
thatthreatperceptionscontributetomajoritymembers’negative contemporary racism: From aversive racism to the common ingroup
identity model. Journal of Social Issues, 61(3), 615-639. DOI: 10.1111/
attitudestowardMuslimminorities(e.g.,González,Verkuyten, j.1540-4560.2005.00424.x
Weesie,&Poppe,2008;Schlueter&Scheepers,2010)andcan Gaertner,S.L.,Dovidio,J.F.,Anastasio,P.A.,Bachman,B.A.,&Rust,M.
C.(1993).Thecommoningroupidentitymodel:Recategorizationandthe
leadtohigherlevelsoffundamentalism(Salzman,2008;Ysseldyk
reduction of intergroup bias. European Review of Social Psychology,
et al., 2010). Reducing perceptions of symbolic threat might 4(1), 1-26. DOI: 10.1080/14792779343000004
Copyright©2014JohnWiley&Sons,Ltd. Eur.J.Soc.Psychol.44,337–348(2014)