Religion and Religious Tolerance 1 Running Head: RELIGION AND RELIGIOUS TOLERANCE

When How You Believe Matters More than What You Believe: Religiosity and Religious Intolerance among Christians and Buddhists Ian G. Hansen and Ara Norenzayan University of British Columbia

Keywords: Buddhists, Christians, culture, tolerance, religion

Religion and Religious Tolerance 2 Abstract In three samples containing Buddhist and Christian participants from several nations, participants of both religions manifested a generally similar relationship between religiosity and intolerance. In all three samples, devotional religiosity (e.g. prayer frequency, faith devotion), was strongly correlated with coalitional religiosity (e.g. organized religious attendance, fundamentalism). Yet devotional religiosity independently predicted religious tolerance while coalitional religiosity independently predicted intolerance. Buddhists and Christians showed no differences in religious intolerance when devotional and coalitional religiosity and other demographic variables were accounted for. The results suggest that differences in mode (What do you mean here by “mode?”) and intensity (by “intensity”, are you referring to fundamentalism?) of religiosity are more important than differences in religious affiliation for predicting religious intolerance.

Religion and Religious Tolerance 3 When How You Believe Matters More than What You Believe: Religiosity and Intolerance among Christians and Buddhists "Faith is powerful enough to immunize people against all appeals to pity, to forgiveness, to decent human feelings.… Religious faith deserves a chapter to itself in the annals of war technology, on an even footing with the longbow, the warhorse, the tank, and the hydrogen bomb." --Richard Dawkins, The Selfish Gene It is a commonly-noted irony that religions, which are supposed to make people more prosocial and moral, are major players in violent world conflicts that are anything but pro-social or moral. The rising share of worldwide violence attributed to extremists of one religion or another appears symptomatic of a worldwide resurgence in religionrelated conflict and religiously-motivated intolerance and violence (Atran, 2002, 2003). (the above sentence seems somewhat confusing and redundant to me) This rise has brought renewed public and academic interest to the role of religion and culture in motivating occupation (?), insurgency, intolerance and violence. As a kind of prejudice, religious intolerance has long been thought of by social psychologists as being conceptually distinct from racial prejudice, yet many of the same devastating social consequences of racial prejudice attend religious intolerance: e.g. persecution, violence, and war (Appleby, 2000; Dennette, 2006; Dawkins, 2006; Juergensmeyer, 2003: Kimball, 2002; Nelson-Pallmeyer, 2003). Popular explanations for the apparent relationship between religion, intolerance, and violence are varied. Enthusiasts of one religion often note how rival religions fail to achieve peace, brotherhood and the moral edification of its members. These failures are typically blamed on the specific beliefs and practices of the other faiths. On the other hand, those without religious beliefs tend to blame religion in general for much of the violence and war in the world, as the

Religion and Religious Tolerance 4 opening quote from Dawkins illustrates. The ecumenically-minded often argue that all faiths have benign and toxic variations; the benign variations of all faiths embrace tolerance and peace, while the toxic versions foster prejudice and war. (new paragraph) The present paper argues for a nuanced version of the ecumenical position, though we ultimately conclude that tolerance-predicting religiosity and intolerance-predicting religiosity are difficult to distinguish from one another and tend to co-occur empirically. This paper also works from the lessons of past attempts to empirically investigate popular diagnoses regarding religion’s relationship to prejudice. (this sentence is unclear) Unfortunately, these investigations are generally considered inconclusive (Altemeyer, 1988; Kirkpatrick & Hood, 1990). Adding to the problem is the fact that much of the previous work on religion and prejudice has been done on racial prejudice and other forms of bigotry, and not on religious intolerance per se. Investigating religious intolerance, as a distinct form of prejudice, is the main focus of this research. In addition to the above, the present paper examines whether certain psychoreligious variables are related in similar ways in two very different faith groups (Buddhists and Christians), and whether these psycho-religious variables predict religious intolerance better than religious affiliation itself. The moral importance of understanding the relationship between religion and intolerance is relatively clear, though it is disputable whether tolerance is a greater virtue than other virtues, or whether the virtue of tolerance is always morally cost-free (e.g. being tolerant towards those who commit gross injustice, oppression or mass violence). Religious tolerance is not only a moral and theological matter, but also a matter of scientific concern: to understand the

Religion and Religious Tolerance 5 origins of many psychological processes, it is important to investigate how religion—a universal feature of human heritage—functions psychologically to unite and divide people, including divisions that are characterized by intolerance and violence. Buddhism and Christianity are two religions which are often considered to be either very similar or different depending on how and by whom they are compared. Buddhism and Christianity are both religions of assent rather than descent, meaning that whether one is a Buddhist or Christian depends more on one’s claimed adherence to Buddhist or Christian teachings and practice rather than whether or not one’s parents were Christian or Buddhist (Cohen & Rozin, 2001; Morris, 1996). Buddhist and Christian moral teachings often overlap, and their social and existential insights are potentially amenable to each other in tone if not in content (Hanh, 1995). The explicit religious teachings of Christians and Buddhists, however, are often widely divergent. There is fundamental disagreement among most influential Buddhist and Christian religious authorities on matters that appear to have tremendous religious importance (Smith, 1991/1958, Tweed, 2001), e.g. belief in a creator God, valuation of an eternal afterlife, and being “exclusivist” in one’s beliefs. Less belief in God, less enthusiasm for an eternal afterlife, and less willingness to be religiously exclusive might be considered, from a Christian perspective at least, to be an indicator of less religiosity among Buddhists. In fact, given the surface divergence of explicit beliefs like these, to class both Buddhism and Christianity as “religions” might be a serious error. Religion itself is often criticized as an ethnocentric Christian classification into which most so called “religions” (other than Christianity) do not fit very well. Derrida (1998) asserted that “the history of the word ‘religion’ should in

Religion and Religious Tolerance 6 principle forbid every non-Christian from using the name religion (36)”. Others have countered that as an ethnocentric classification, “religion” actually works remarkably well cross-culturally (Guthrie, 2001), perhaps because at least a few social and psychological dynamics among Christians might often reflect universal human patterns of thought and behavior likely to be found in any cohesive belief or practice-based group that makes frequent principled reference to the esoteric or supernatural. Whether empirical investigations will eventually remove all reasonable doubt on this matter or not, we acknowledge at the outset that when we say “religion” and “religiosity” we may mean little more than “culturally Christian understanding of religiosity.” To the extent what we call religiosity is only a culturally Christian understanding of religiosity, however, then it is likely not only that (a) Buddhists will be lower in our measures of religiosity than Christians, but more importantly that (b) the dynamics of what we call “religiosity” within Christianity will be radically different from those dynamics within Buddhism. Further, the aspects of religiosity predict religious intolerance by Christians may be radically different from the aspects that predict religious intolerance by Buddhists. The potential for radical cultural contingency in religious constructs has been recently demonstrated by Cohen & Hill (2006), who found that Allport’s classic constructs of intrinsic and extrinsic religiosity (Allport & Ross, 1967) were inversely correlated among Protestants, uncorrelated among Catholics, and positively correlated among Jews, with significant differences between the correlations. If the relationship between intolerance and the measures of religiosity used in this paper is as culturally contingent as that between intrinsic and extrinsic religiosity, then a radical and systematic difference should be found between Buddhists and Christians in the

Religion and Religious Tolerance 7 interrelationship of religiosity and intolerance. However, to the extent that the dynamics of our measures of religiosity and intolerance differ little between Christians and Buddhists, this would be strong evidence that our religiosity and intolerance constructs may travel well even outside Christian samples. (Ian – these last few paragraphs seem somewhat distracting and disruptive of your paper’s flow. Perhaps a better alternative (and more “saving” in terms of paper length) would be to simply refer in a sentence or two to some previous research or statements by others who have noted the issues between Christianity and Buddhism which you refer to here). Establishing the potential cross-religious utility of certain religiosity constructs is essential to addressing the much speculated-on relationship between religiosity and intolerance and violence. Whether taking a position like Dawkins’ and condemning all instantiations of religiosity as grounding violent intolerance (?) or taking an ecumenical position that distinguishes between some forms of religiosity as potentially intolerant and others as potentially tolerant, one assumes that the relevance of religiosity to intolerance does not vary much by religion. However, religious intolerance, religious violence and certain other undesirable religious attitudes and behaviors are frequently discussed as unique properties of certain religious groups (see, e.g. Manji, 2003, Nietzsche, 1895/1968). If inclinations to violence and intolerance are indeed religion specific, i.e. cultivated by some religions but not by others, then the relationship between religiosity and intolerance should differ categorically from religion to religion: for some religions, religiosity should be associated positively with religious intolerance, while for others it should be orthogonal or even negatively related. If there are any two religions in which we could expect to find such a difference, they would be Christianity and Buddhism.

Religion and Religious Tolerance 8 Many would argue that Christianity has historically used faith and religious sentiment to cultivate violence and intolerance in order to establish its relatively strong cultural influence in the world, while Buddhist influence has spread much more by the appeal of its teachings and the fecundity of its adherents (?) than by, say, violent crusades in the name of Buddha, or the killing and persecution of those who adopt non-Buddhist religious beliefs and practices. It is also possible however, that the way that religiosity relates to intolerance may have more to do with the kind of religiosity examined than the religious affiliation whose teachings and practices give doctrinal shape to one’s religiosity; that is to say that when predicting intolerance, how one believes may matter more than what one believes. (this seems to be your thesis statement and so should be quite strategically placed to punctuate, as much as possible, the “plunge” into the specifics of your study) Coalitional and devotional religiosity In the present paper, we rely on a theoretical distinction between two major constructs that we expect to see as distinguishable in both Christians and Buddhists: (1) coalitional religiosity or the defensive inclination to identify strongly with a certain group, ideology or traditional culture and (2) devotional religiosity or the passionate inclination to understand and engage life with an orientation to a numinous or “divine” entity, e.g. God or Buddha. Coalitional processes are natural to the human social animal and also play a significant role in the formation of religious groups (Atran & Norenzayan, 2004; Navarette & Fessler, 2005). For securing loyalty, widespread conformity to norms and effective coordination within a coalition, there are obvious psychological benefits to

Religion and Religious Tolerance 9 exalting one’s own belief system, religious group or cultural tradition as uniquely true and good. Such an exclusive exaltation is likely, however, to be associated with intolerance (potentially including violence) towards those who fall outside the boundaries of one’s imagined coalition. Personality styles (which ones?) tapped into by scales like religious fundamentalism (Altemeyer & Hunsberger, 1992) and authoritarianism (Altemeyer, 1981, 1988, 1996) may once have contributed to adaptive coalitional processes (Kirkpatrick, 1999), though at present the most salient criterion they predict is prejudice (Hunsberger, 1995). “Religion” as it is commonly understood, especially by its opponents, appears highly inclined to cultivate this epistemologically and ideologically rigid grounding of coalitional cohesion. (difficult to follow; can you say more simply?) To define “religion” only in coalitional terms, however, is to neglect the aspect of religion that is directed to the transcendent, the divine, and the numinous. It was this aspect of religion that was originally investigated by early psychologists of religion for its potential to promote tolerance rather than intolerance (Allport, 1950; Allport & Ross, 1967)1. Critics of religion may be pessimistic about this divine-centric orientation, but there is some evidence that even with something like anti-gay prejudice, which the prejudiced often attempt to justify on religious grounds, there is a potentially negative relationship between religious devotion and prejudice when coalitional inclinations like authoritarianism are controlled for (Laythe, Finkel & Kirkpatrick, 2001). (again, wordy and difficult to follow) Our devotional vs. coalitional theoretical framework is based on recurring distinctions drawn by theorists from various fields between a kind of personallyfelt/natural religiosity and a propositionally-held/cultural kind (see Hansen &

Religion and Religious Tolerance 10 Norenzayan, in press, for a discussion). The framework is consistent in some ways with Allport’s distinction between intrinsic or personally felt religiosity and extrinsic or socially-derived/instrumental religiosity. Yet, unlike Allport, we do not treat devotional and coalitional religiosity as two orthogonal or inverse (opposing?) modes of religiosity, but rather as strongly correlated constructs that have the potential to make very different predictions. That is, we do not hypothesize a world in which all religious individuals or groups can be neatly divided into the devotionally religious and the coalitionally religious. Rather, we expect devotional and coalitional religiosity to co-occur in various religious groups and individuals. In general, we expect that to be more religious is to be both more devotionally religious and more coalitionally religious; correlations between what we consider devotional variables and what we consider coalitional variables have been found to be robust and reliable (e.g. Duck & Hunsberger, 1999; Kahoe, 1975; Leak & Randall, 1995; Moghaddam & Vuksanovic, 1990; Watson, Sawyers & Morris, 2003). The relationship between each form of religiosity and religious intolerance may be a different matter, however. The potential for overcoming religious intolerance may be encouraged by devotional religiosity even as it is discouraged by the coalitional religiosity that devotional religiosity co-occurs with. Devotional religiosity may be related both to the transcendence of narrow individual self-interest that makes coalitional cohesion between individuals possible, as well as to the transcendence of narrow group self-interest that makes tolerance between groups possible (Hansen & Norenzayan, in press). Stated metaphorically, devotional binding can be linked to coalitional breaking, but binding and breaking are very different, even opposite, processes. (hard to follow and vague)

Religion and Religious Tolerance 11 In sum, (I made a larger deletion here) we expect with the ecumenically-minded that all religions carry the potential for both tolerance and intolerance, peace and war, in spite of what their distinctive religious teachings are (see also Kimball, 2002). Moreover, while it may matter what one’s religious doctrines are in other ways, we expect that how one believes matters more than what one’s specific religious affiliation is when predicting intolerance. Present Research In sum, while the literature on religion and prejudice has clarified the relationship between religion and prejudice, it has overlooked a key type of prejudice—religious intolerance—a widespread form of prejudice which has been implicated in many cases of mass violence. Moreover, there has been very little cross cultural research on the robustness and nature of this relationship. The present research addresses these two neglected questions. Our specific hypotheses are (1) there will be little systematic difference between Christians and Buddhists in how coalitional religiosity, devotional religiosity and intolerance relate to each other, and (2) coalitional religiosity and devotional religiosity will be positively correlated, but coalitional religiosity will independently predict religious intolerance while devotional religiosity will independently predict religious tolerance. We investigated three samples to test the above hypotheses. Since we analyzed all samples at once to address each hypothesis, we introduce them together. Sample 1 Sample 1 consisted of Christian and Buddhist responses to a 10-nation survey conducted in 2004 by ICM research associates on behalf of the British Broadcasting

Religion and Religious Tolerance 12 Corporation (BBC) (2005). The survey asked representative samples2 in the U.S., U.K, South Korea, Indonesia, Israel, Lebanon, Nigeria, Russia, Mexico and India a range of questions about their religious beliefs. All items were translated by the BBC into the languages spoken by surveyed participants in each country. We examined items assessing (a) prayer frequency—a devotion measure3, (b) frequency of religious attendance—a measure of behavioral cohesion, and (c) exclusivity—also coalitional(confusing) —in predicting (1) religious scapegoating: the inclination to blame religious others for the world’s problems, and (2) sacrificial scapegoating: the combination of scapegoating with a willingness to die for one’s God or beliefs. These measures and how they were scored are listed in Appendix A. Since sacrificial scapegoating is a subset of scapegoating, we examined both measures as two different manifestations of intolerance. In another study (Ginges, Hansen & Norenzayan, 2006), we found that these measures of scapegoating were positively related to national demographics suggestive of intolerance and conflict: military spending and number of refugees fleeing the country. Participants Although there were 10,068 survey participants total, we limited our analyses to the responses of Buddhists and Christians. We analyzed all 254 Buddhist participants in the survey; 113 male and 141 female; 5 from India, 7 from Indonesia, 1 from Russia, 225 from South Korea, 7 from the U.K. and 9 from the U.S.A.; 47 were between 18 and 24, 68 were between 25 and 34, 54 were between 35 and 44, 52 were between 45 and 54, and 33 were 55 or older. We also analyzed all Christian participants in the survey; 2031 male and 2679 female; 31 from India, 156 from Indonesia, 14 from Israel, 294 from

Religion and Religious Tolerance 13 Lebanon, 935 from Mexico, 646 from Nigeria, 774 from Russia, 337 from South Korea, 711 from the United Kingdom, and 812 from the U.S.A; 883 were between the ages of 18 and 24, 1019 between 25 and 34, 917 between 35 and 44, 797 between 45 and 54, and 1091 over 55. Sample 2 Sample 2 was drawn from Chinese Christians and Chinese Buddhists in Malaysia whose devotional and coalitional religiosity were assessed along with (1) disapproval of personal “multireligiosity” (anti-pluralism) and (2) antipathy towards religious others. Both of these were considered measures of religious intolerance. (with what instrument did you measure these constructs?) The religious psychological (would it be better to call these variables “psychoreligious”?) variables considered in this sample were (1) a religiosity scale measuring devotion to one’s faith, adapted from the Intrinsic Religious Motivation Scale (Hoge, 1972) (“faith devotion”), (2) Altemeyer’s (1999) Right Wing Authoritarianism scale (“authoritarianism”), (3) Altemeyer’s (1996) DOG scale measuring dogmatism (“dogmatism”)4, (4) Altemeyer and Hunsberger’s (1992) religious fundamentalism scale (“fundamentalism”), and (5) a scale that we devised to measure religious exclusivity—the conviction that only one religion or understanding of the divine can be true and the others must be false (“exclusivity”). (Do you have any sort of psychometrics on your scale? How reliable and valid is it?) The items for faith devotion are listed in Appendix B. Originally conceived as a more internally valid measure of “intrinsic religiosity”, the items of the scale are more intuitively understood as devotion to the beliefs, practices and values of one’s faith, as well as to God or the Divine (see also the comments on the scale

Religion and Religious Tolerance 14 by Batson et al, 1993). All scales were assessed on the same metric, with 1 indicating “totally disagree” and 9 indicating “totally agree” and 5 “neutral.” Items of the religious exclusivity scale are listed in Appendix C. All scales had some items amended to accommodate the multi-religious Malaysian sample. Some items were also simplified to be easier to understand since not all Malaysian speak English as their first language (though English is the first language for many Malaysian Chinese, who were the vast majority of the sample, and the language of instruction at the universities surveyed was English. For copies of the complete scales, please contact the authors.) Resistance to “multireligiosity” (anti-pluralism) was measured by how much participants disliked and expected damnation for the fictional character Piscine Patel, the central figure in Yann Martel’s (2001) international bestseller, Life of Pi. We extracted and edited a passage from this novel where Piscine describes his sense of religious devotion to God and to three religions—Hinduism, Christianity and Islam—and then later finds himself at the center of a heated verbal conflict between the authorities of those three religions—a Hindu pandit, Christian priest, and Muslim Imam. We measured religious antipathy by how negatively participants evaluated the personal qualities and likely afterlives of the characters in the story who did not share the same religious beliefs. Participants 192 participants completed a questionnaire packet for 8 Malaysian Ringgit: 138 women, 49 men, and 5 unidentified; 131 Chinese, 34 Malay, 12 Tamil/Indians, and 15 mixed or unidentified; 81 Buddhists, 48 Christians, 41 Muslims, 11 Hindus, and 11 other, unidentified or indicating “no religion”. Only Buddhist and Christian responses were analyzed. Participants were recruited either by posters advertising the study put up in

Religion and Religious Tolerance 15 public meeting areas in a university, or by invitations extended to students taking a course or members involved with a religious group, often with the help of willing professors and leaders of religious organizations. Some were recruited over the internet (the latter—seven participants total—were compensated by postal order). Participants completed everything as a questionnaire packet on their own time, and returned the questionnaire for compensation at a scheduled time. Materials and Procedure To accommodate Buddhists and others whose spiritual and religious beliefs do not necessarily include an agentic Creator God, we took out references in all scales to “God” and instead put “The Divine”. Thus, for instance, in the faith devotion scale, “Nothing is as important as serving God the best I know” becomes “Nothing is as important as serving the Divine as best I know.” Instructions in the questionnaire read, “Some of the statements that follow are related to religion, and since many people use different words to describe the source of truth they believe in and follow, the statements below often refer to ‘the Divine’. If you do not like the phrase ‘the Divine’ you may think of it as ‘Allah’ or ‘Buddha’ or ‘Christ’ or ‘God’ or some other name you have stronger feelings about.” We also asked participants to write down “the name or names you prefer for the Divine”. The scales were presented in the following order: faith devotion (Hoge, 1972), authoritarianism (Altemeyer, 1999), Religious Exclusivity, Altemeyer and Hunsberger’s (1992) Religious Fundamentalism Scale, Altemeyer’s (1996) DOG scale, and Neuberg & Newsom’s (1993) Personal Need for Structure scale. (what about your scale you mentioned above? Is this the same as the “Life of Pi” stuff you mentioned above?)

Religion and Religious Tolerance 16 For each statement, participants were instructed to indicate how true they thought the statement was on a one to nine scale (from “totally untrue” to “totally true”). Following these scales participants completed the dependent measures assessing tolerance for religious others. Participants first read a heavily edited excerpt from Yann Martel’s (2001) Life of Pi, encompassing descriptions and events from Chapter 16 through Chapter 23, though condensed for this sample into three single-spaced pages, with some rewriting, primarily for clarity. A descriptive summary of the passage is included as Appendix D. For the full re-written passage, please contact the authors. After reading the passage, participants responded to a series of questions all answered on a 1 to 9 scale. The first asked how positively participants felt about each character (from 1, “extremely negative” to 9, “extremely positive”). For each question, participants were reminded that Piscine was “the main character of the story, who wishes to be close to God through Hindu, Christian and Muslim belief and practice”) that the Pandit “wishes Piscine to be Hindu only”, that the Priest “wishes Piscine to be Christian only”, and the Imam “wishes Piscine to be Muslim only” and that Piscine’s parents are not religious. Participants also indicated tolerance by stating how inclined they would be to forbid the following from teaching their beliefs to friends and family: People like Pi who practice three religions at once, Hindus like the Pandit, Christians like the Priest, Muslims like the Imam, and people without a religion like Piscine’s parents. Participants were then asked to “imagine that at the moment Piscine is asked to choose between the faiths he practices, the story is interrupted by a bomb that explodes, killing Piscine, his parents, the priest, the imam and the pandit.” Participants then indicated how likely it was that each character would go to heaven or to hell in this

Religion and Religious Tolerance 17 scenario (from 1, “definitely hell” to 9, “definitely heaven”). As a measure of potentially violent religious aggression, we asked as how sad/joyous the Divine would feel at the death of each character (from 1, “extremely sad”, to 9, “extremely joyous”).(interesting stuff…). Finally, participants completed a demographics measure. Dependent measures: anti-pluralism and religious antipathy We measured intolerance in two ways in this sample. A measure of hostility to multireligiosity (“anti-pluralism”) was based on three of the four dependent measures, since only three correlated significantly with each other. The three that correlated were (1) disliking Piscine, the multireligious main character, (2) belief that he would go to hell rather than heaven, and (3) desire to forbid rather than allow him to teach to one’s friends and family. Feelings at the imagined death of Piscine did not correlate with these items. When the three measures of anti-pluralism were included with measures of hostility towards Pi’s non-religious parents and the Hindu Pandit (religious and cultural outsider to Buddhists and Christians in Malaysia), the nine items together, which we call antipathy towards religious outsiders (“religious antipathy”). Since anti-pluralism was a subset of religious antipathy, we examined each dependent measure separately. Sample 3 Sample 3 had similar measures to sample 2 but was conducted in Canada, allowing us to examine the difference between Buddhists and Christians in tolerance within a North American context. Dependent measures of intolerance included not only anti-pluralism and antipathy towards religious outsiders (the forms of intolerance measured in Study 2), but also indifference to outsiders’ civil rights, aggressive feelings towards outsiders, and general sympathy with religious violence against “the wicked”.

Religion and Religious Tolerance 18 Participants 194 psychology undergraduates from the University of British Columbia completed the study for one academic credit. Participants were 128 women and 64 men, 109 of East Asian ethnicity (Chinese, Japanese, Korean, Vietnamese), 53 Caucasians, 32 other or unidentified; 73 indicating “no religion” or agnostic, 71 Christians, 24 Buddhists, 22 other or unidentified, and 4 atheists. We examined only the Christian and Buddhist responses. Materials The predictor variables for this sample included all the religiosity variables used in Sample 2: faith devotion, exclusivity, fundamentalism, dogmatism5 and authoritarianism. In addition to faith devotion we used another measure of devotional religiosity that is less well-tested but is comparable in tone to faith devotion: it also has the advantage of having reversed items that non-religious people will be more likely to endorse than the reversed items of Hoge’s (1972) scale, which are derived from Allport & Ross’s (1967) measure of “extrinsic” religiosity, though functionally the three items included in Hoge’s scale are more indicative of a lack of devotion than an alternative form of religiosity per se. This alternative devotion measure was adapted from items used by Fiorito & Ryan (1998) to assess the more spiritual aspect of religious belief. Items from this scale, which we call devotion to the divine (“divine devotion”), are included in Appendix E. We used the same adaptation from Life of Pi as used in Sample 2, though we modified our previous outcome measure of intolerance (called “religious antipathy” in this sample), included a validated measure of civil intolerance (Sullivan, Pierson &

Religion and Religious Tolerance 19 Marcus, 1982), and included measures more explicitly tapping aggressive antipathy towards the characters: feeling joy at the killing of the characters, and four stronglyworded items indicating support for “killing the wicked” as a measure of general religious or moral aggression. These measures are listed in Appendix F. Note that whereas in Sample 2, we asked for feelings at the death of the characters (a measure that did not correlate with any other intolerance measures), in this sample we asked instead for feelings at the killing of the characters. We did this to assure that the measure would only tap aggressive antipathy and not be too contaminated by the belief that the Divine will be joyous at the death of one’s co-religionists—i.e. joyous to welcome the righteous dead into heaven. Study 1a Study 1a examined whether there was a broadly similar pattern of relationships between the variables and constructs of devotional religiosity, coalitional religiosity, and intolerance for Christians and Buddhists, or whether there were any consistent differences in these patterns of inter-relationship. Factor Analyses We performed exploratory factor analyses as a preliminary investigation of whether devotional religiosity, coalitional religiosity, and intolerance could be conceived as distinct constructs related in a comparable way across studies and religions. We used principle axis factoring with direct oblimin rotation, producing a one factor solution for the variables in Sample1 and a two factor solution for variables in Samples 2 and 3 (with the exception of Christians in Sample 2). We have used brackets to indicate which items we theoretically grouped together as devotional religiosity, coalitional religiosity, and

Religion and Religious Tolerance 20 intolerance to assess to what degree factor analyses justify treating these as distinct constructs. Table 1 shows the pattern matrix for Buddhists and Christians on these constructs in Sample 1. Table 2 shows loadings for Buddhists in Sample 2. Table 3 shows the loadings for Christians in Sample 2. Table 4 shows loadings for Buddhists in Sample 3. Table 5 shows the loadings for Christians in Sample 3. Where there appears to be a pattern of loadings that indicates either identity or distinction between devotional religiosity, coalitional religiosity and intolerance, boxes are placed around groups of loadings to highlight this identity or distinction. There were inconsistencies in the factor analyses both across studies and between Christians and Buddhists. However, the differences between Christians and Buddhists did not appear to be systematic. In Sample 1, with the largest Ns in the three samples, the factor analytic patterns were broadly similar between Christians and Buddhists. For both Buddhists and Christians the three “religiosity” variables loaded substantially (between .4 and .8) on the one factor, with sacrificial scapegoating not loading substantially on the factor for either Buddhists or Christians (approximately .2 in both cases). If using simple scapegoating instead of sacrificial scapegoating, the overall pattern was similar for both Buddhists and Christians, though scapegoating loaded even more weakly on the summary factor (neither scapegoating nor sacrificial scapegoating could be represented as a separate factor using principal axis factoring since this extraction method precludes one item factors). In Samples 2 and 3, where two factor solutions emerged three times out of four, Buddhists showed a different factor analytic pattern from Christians in each study, but the nature of the difference was not the same across studies. Interestingly, in the two

Religion and Religious Tolerance 21 samples that were relatively large—81 Buddhists in Sample 2 (Table 2) and 71 Christians in Sample 3 (Table 5)—the patterns had broad similarities. For both of these larger samples the first factor appeared to be a “religiosity” factor encompassing devotional and coalitional religiosity equally, and the second factor appeared to be an intolerance factor on which devotional religiosity variables loaded (if at all) positively and on which coalitional religiosity variables loaded (if at all) negatively. Among the 23 Buddhists in Sample 3, the factor analytic pattern was different (with coalitional religiosity/intolerance as the first factor and devotional religiosity/rejection of violence as the second) but still the theme illustrated was largely similar: coalitional religiosity ran along the grain of religious intolerance and devotional religiosity potentially ran against that grain. The factor analysis with Buddhists should be interpreted with caution, however. There are too many items examined in the analysis than can be reliably measured with such a small subsample. With Buddhists in Sample 2, if examining antipathy towards Pi, Hindus and the non-religious together (instead of only the anti-pluralist antipathy towards Pi chosen for Table 2), the factors reversed themselves, with intolerance coming first and religiosity second. There was also a low loading for exclusivity on the religiosity factor (.15) but otherwise the outcome was comparable in how devotional religiosity, coalitional religiosity and intolerance were related. Examining religious antipathy for Christians in Study 2 did not manifest a substantially different result from that shown with antipluralism in Table 3. These factor analyses provide no evidence of systematic or coherent differences between Buddhists and Christians in the interrelationships of the constructs under investigation. However, these patterns also provide little evidence for treating devotional

Religion and Religious Tolerance 22 religiosity, coalitional religiosity and intolerance as three empirically distinct constructs. Christians in Sample 2 and both Buddhists and Christians in Sample 1 produced one factor rather than two factor solutions. Even where there were two factor solutions, there was always a factor on which all the variables of at least two constructs loaded heavily in the same direction (devotional and coalitional, coalitional and intolerance, or devotional, coalitional and intolerance together). Certainly there were no clean three factor solutions with all and only devotional variables on one factor, all and only coalitional variables on another and all and only intolerance variables on the other. The three constructs under investigation thus appeared quite closely empirically linked and not readily distinguishable. In spite of the empirical overlap of the constructs, the thematic patterns consistently observed in the two factor solutions (with coalitional and devotional religiosity variables showing potentially opposite relationships to intolerance) provide some warrant for considering the constructs separately, at least with regard to predicting intolerance. Regression slopes We compared regression slopes across the three studies, examining how the latent variables of devotional and coalitional religiosity respectively predicted the latent variable of intolerance, as measured in each of the studies. For Sample 1, the only construct measured by more than one item was coalitional religiosity, so this latent construct was simply represented by the mean of exclusivity and religious attendance items. For Samples 2 and 3, the first principle components (PCs) for each construct were obtained from principle components analyses including the summary scores of variables

Religion and Religious Tolerance 23 theoretically included as part of that construct, .e.g. for coalitional religiosity we took the first PC of the four scale indices of authoritarianism, exclusivity, fundamentalism and dogmatism. When only one scale was used for a construct (e.g. faith devotion in Sample 2), we simply used the mean of the items of that scale to represent the construct. Table 6 lists the unstandardized regression coefficients and standard errors for Buddhists and Christians in Sample 1, based on logistic regression and in samples 2 and 3, based on linear multiple regression. Table 7 gives a conceptual summary of the respective contributions of devotional and coalitional religiosity to predicting intolerance across religions, samples, and measures of intolerance. Where there was a significant difference found between the regression coefficients of Buddhists and Christians it is marked in the table. Only one significant difference was found out of 10 comparisons of coefficients, not sufficient to show systematic differences between Buddhists and Christians. Moreover, the difference found (Buddhist devotion less negatively predictive of intolerance than Christian devotion in Sample 3) is not replicated even directionally in the other samples. In Samples 1 and 2, the direction is actually the opposite in all but one case where the two regression coefficients are exactly the same. In sum, although there was variation in significance levels and nominal magnitude of effects between samples and between religions in any one sample, the general pattern of effects across studies and samples was similar to the factor analyses: coalitional religiosity was nominally or significantly positively related to all forms of intolerance, while devotional religiosity was nominally or significantly negatively related to all forms of intolerance (exception: predicting sacrificial scapegoating, B = .03). In some samples (Sample 1 and Sample 3) significant effects following this pattern were best detected

Religion and Religious Tolerance 24 among Christians. In Sample 2 they were best detected among Buddhists. Samples most clearly showing the pattern also had sufficient sample sizes to minimize standard error, however, which suggests that the reliability of the pattern is N-dependent rather than religion-dependent. Internal reliability and intercorrelation We calculated the intercorrelations and internal reliabilities of scales used to measure the constructs of (a) devotional religiosity (b) coalitional religiosity and (c) religious intolerance, separately for Buddhists and Christians. The full table is available from the authors. Alphas for all multi-item scales ranged between .76 and .97 for both Christians and Buddhists, with absolute value of the differences between Buddhist and Christian alphas ranging between .02 and .11, suggesting that the scales held together similarly among Buddhists as among Christians. For devotional religiosity in Sample 3 (the only sample where two measures were used), faith devotion and divine devotion correlated .68 among Buddhists, and .81 among Christians. For coalitional religiosity, exclusivity and attendance (Sample 1) were correlated .29 among Buddhists and .30 among Christians; exclusivity, authoritarianism, dogmatism and fundamentalism (Samples 2 and 3) were correlated between .30 and .94 among Buddhists, and between .21 and .88 among Christians. For intolerance in Sample 3 (the only sample where several scales correlated to form the construct) correlations between religious antipathy, civil intolerance, aggressive antipathy and religious violence ranged between -.05 and .74 for Buddhists, and between .25 and .52 for Christians. -.05 was the only nominally negative correlation for Buddhists; all others were nominally or significantly positive. Of 20 possible comparisons between Buddhist and Christian correlations, only 2 showed a

Religion and Religious Tolerance 25 significant difference. For one of these, a greater Christian relationship between dogmatism and authoritarianism in Sample 2, the nominally reverse comparison was evident in Sample 3. The other, a greater relationship between religious antipathy and aggressive antipathy among Buddhists in Sample 3, is unlikely to be a systematic difference because some intolerance measures were more nominally correlated among Buddhists and some among Christians in that sample. Table 8 lists the intercorrelations and internal reliabilities of the summary constructs (devotional religiosity, coalitional religiosity and religious intolerance). Internal reliabilities were largely similar between Buddhists and Christians. There was no consistent difference between Christians and Buddhists in the way that devotional and coalitional religiosity correlated with religious intolerance. In Sample 2, both devotional and coalitional religiosity were significantly more related to intolerance for Christians than for Buddhists. In Sample 3, however, this pattern was reversed. There were no significant differences in the prediction of intolerance for Sample 1. While most differences found between Christian and Buddhist correlations were idiosyncratic to one sample, there was a notable and consistent difference in the pattern of correlations such that devotional and coalitional religiosity were always more strongly correlated for Christians than for Buddhists (significantly so in Sample 1, and nominally in Samples 2 and 3). This weaker relationship for Buddhists may be indicative of greater Buddhist success at disentangling the valuation of divinity from a sense of defensive bounded coalitional identification.

Nevertheless, the success is a modest reduction

rather than a miraculous reversal. For both Buddhists and Christians coalitional and devotional religiosity were strongly intercorrelated. Moreover, there did not appear to be

Religion and Religious Tolerance 26 consistent differences between Buddhists and Christians in the way coalitional and devotional religiosity respectively related to intolerance. The lower correlations among Buddhists may also reflect more restricted range among Buddhists: in all three samples, Buddhists were lower than Christians in the summary constructs of both devotional and coalitional religiosity, all ps < .01. In sum, there do appear to be some magnitude differences between Buddhists and Christians in the interrelation of the constructs, though only one difference found generalized across samples. There appears little evidence for claiming that Buddhists and Christians have radically different understandings of the constructs under investigation. There was a surprising consistency in the broad patterns of the measures and constructs across studies and across religions. Specifically, for both religions in all studies, coalitional and devotional religiosity were strongly positively related, but coalitional religiosity was always nominally more positively (or less negatively) related to every measure of intolerance than devotional religiosity. Moreover, in spite of the strong relationship between devotional and coalitional religiosity in all studies and both religions, coalitional religiosity consistently had a more positive independent relationship with intolerance, while devotional religiosity had a more negative independent relationship. The relative importance of religiosity (coalitional and devotional) vs. religious affiliation in predicting intolerance is assessed in Study 1b. Study 1b Study 1b, using the same three samples, examined the degree to which religious affiliation (Buddhist vs. Christian), individual variance in distinct conceptual kinds of religiosity (coalitional and devotional) and other demographic variables were important

Religion and Religious Tolerance 27 in predicting religious intolerance. We submitted all samples to multiple regression (logistic regression for Sample 1, linear regression for Samples 2 and 3). Tables 9 through 11 list the results of the regressions when analyzing all scales separately. The tables list both the standardized regression coefficients and standard errors, as well as the variance inflation factor (VIF), an indicator of multicollinearity for each variable. Higher VIFs indicate greater potential for multicollinearity to distort the apparent independent effect of the variable. In every regression (except for that predicting sacrificial scapegoating in Sample 1), at least one item of devotional religiosity and one of coalitional religiosity was an independent predictor of each sample’s measure of intolerance, while religious affiliation (Christian vs Buddhist) was never an independent predictor. Moreover, when performing the same analyses with coalitional and devotional summary constructs only while still controlling for all demographics, the results were remarkably consistent across samples. In Sample 1, coalitional religiosity positively predicted scapegoating, Wald = 4.36, OR = 1.25 [95%CI = 1.01 – 1.54], p = .04, and sacrificial scapegoating, Wald = 125.94, OR = 5.22 [95%CI = 3.91-6.97], p < .001, while devotional religiosity negatively predicted scapegoating, Wald = 23.39, OR = .66 [95%CI = .56 - .78], p < .001, and was orthogonal to sacrificial scapegoating, Wald < 1, ns. In Sample 2, coalitional religiosity positively predicted anti-pluralism, 7.39 p < .001, and religious antipathy,

= .80 S = .09 t = 8.62 p < .001, while devotional

religiosity negatively predicted anti-pluralism, religious antipathy,

= .75 S = .10 t =

= -.19 S = .10 t = -1.87 p = .06, and

= -.15 S = .09 t = -1.66 p = .1. In Sample 3, coalitional religiosity

positively predicted the first PC of the four intolerance measures, p < .001, while devotional religiosity negatively predicted it,

= .75 S = .13 t = 6.23

= -.41 S = .12 t = -3.19 p

Religion and Religious Tolerance 28 = .002. Also, the multicollinearity evident in some of the religiosity items (especially those listed in Tables 10 and 11) was eliminated when using summary constructs (all VIFs < 3 when using summary constructs). It thus appears clear that some aspect of devotional or coalitional religiosity or both is always an active independent ingredient in religious tolerance/intolerance, while religious affiliation itself is a relatively unreliable criterion to use for predicting such intolerance. However, since coalitional and devotional religiosity are correlated but opposite predictors of intolerance, it is difficult to speak to the relationship between religion and intolerance without information about both predictors. Coalitional religiosity appears to be a predictor of stronger magnitude than devotional religiosity, since in our samples devotional religiosity was a less consistently detectable predictor. For scapegoating in Sample 1, however, devotional religiosity was the stronger predictor (negative). Attempts to distinguish tolerant religiosity from intolerant religiosity have generally relied on zero-order correlations between a religiosity measure and an intolerance measure. This zero-order correlation approach may have led to the failure to identify a measure of religiosity that is both consistently coherent (internally reliable) and consistently predictive of tolerance6. Since the aspect of religiosity that predicts intolerance and the aspect that predicts tolerance are so strongly positively correlated, zero-order correlations may hide the pro-tolerance predictive role of devotional religiosity or, less often, the anti-tolerance predictive role of coalitional religiosity. As in Laythe et al (2001) multiple regression is a more reliable method to bring the potential contrast between devotional and coalitional religiosity into relief. General discussion

Religion and Religious Tolerance 29 Analysis of the three samples points to there being little consistent difference between Buddhists and Christians in the way devotional and coalitional religiosity relate to intolerance, thus supporting out first prediction. Moreover, while devotional and coalitional religiosity are consistently positively related, the independent relationship of coalitional religiosity to intolerance is positive, while the independent relationship of devotional religiosity to intolerance is negative. This supports our second prediction. Among Buddhists and Christians in several nations, the constructs of devotional religiosity, coalitional religiosity and religious intolerance are potentially overlapping yet potentially distinct. There are no apparent differences between Buddhists and Christians in religious intolerance when devotional and coalitional religiosity are accounted for along with other demographics. It thus appears that, at least with regard to intolerance among Buddhists and Christians, how one believes matters more than what one believes. Limitations Our assertion that religiosity and intolerance are not systematically different in their inter-relationship for Buddhists and Christians is a weak assertion that differences found in one sample will likely be cancelled by differences in other samples, rather than a strong assertion that in all possible samples of Buddhists and Christians, no significant differences will ever appear. (hard to follow) There were significant differences found in Buddhist vs. Christian interrelationships in some samples, but the opposite differences could be found in other samples. For example, religiosity (both devotional and coalitional) seemed more linked to intolerance for Christians in Malaysia compared with Buddhists in Malaysia, but religiosity was less linked to intolerance for Christians in Canada than Buddhists in Canada.

Religion and Religious Tolerance 30 Our assertions about the relationship of religiosity variables and constructs (confusing – which constructs are you referring to? Perhaps what you say bellow clarifies it…) is an even weaker one: while there may be systematic religious differences in the correlation between devotional and coalitional religiosity and the variables that compose them, these differences will be differences of magnitude rather than difference in kind. Devotional and coalitional religiosity were more strongly positively correlated for Christians than for Buddhists in every sample, but in every sample, Buddhists were also significantly lower than Christians on the means of devotional and coalitional religiosity summary scores. In another study (Hansen & Norenzayan, 2006a), we performed correlations between exclusivity (coalitional) and prayer frequency (devotional) among Catholics, Protestants, Eastern Orthodox, Jews, Muslims, Atheists, Buddhists and Hindus and found that in every group the two items were positively correlated, though this was least obvious among Atheists, Buddhists, Hindus and Muslims. In each case of low correlation, though, there were potential restricted range issues: Atheists and Buddhists were uniformly low in exclusivity and prayer frequency, Hindus showed variation on prayer but were uniformly low in exclusivity, and Muslims were uniformly high in both. To the extent our measures of devotional and coalitional religiosity are measures of Christian or more generally monotheist religiosity, it is unsurprising that Buddhists are lower on both and thus show lower correlations due to restricted range. It is perhaps more surprising that the Buddhist relationship between devotional and coalitional religiosity looks as much like the Christian relationship as it does. With a culturally uniform sample (Americans), Cohen & Hill (2006) found radical differences in the correlations between intrinsic and extrinsic religiosity among three

Religion and Religious Tolerance 31 monotheistic religious affiliations with many shared scriptures and religious values: Protestants (negative correlation), Catholics (orthogonal) and Jews (positive). In contrast, with culturally diverse samples, we have found a merely significant, but not radical difference between Christians and Buddhists in their strongly positive interrelationship of coalitional and devotional religiosity. When comparing Buddhists and Christians in factor analyses, regression slopes and scale intercorrelations, we did not always control for ethnicity. Malaysia was the most ethnically controlled sample, since almost all Christians and Buddhists were ethnic Chinese. This was also the sample that appeared to show the most striking differences between Buddhists and Christians in their intercorrelations. However, when ethnicity was controlled along with other demographics in the multiple regressions of Study 1b, individual variation in devotional and coalitional religiosity remained robust predictors of intolerance in the expected directions, and religious affiliation had no independent predictive role. It may also be argued that our failure to detect significant differences in correlations between Buddhists and Christians was related to our skewed sample sizes in each sample, with Christians and Buddhists being unequally represented in every sample. This criticism is most aptly applied to Sample 3, with only 24 Buddhists represented (the reliability of the factor analysis for Buddhists in Study 3 can be challenged on similar grounds, as noted earlier). While there were very few Buddhists in Sample 3, in Sample 1, where the sample sizes were the highest for both Buddhists and Christians, differences in intercorrelations and factor analyses were the weakest, and all differences in correlations and regression coefficients were non-significant except for the correlation

Religion and Religious Tolerance 32 between devotional and coalitional religiosity (a religious difference nominally present in the other samples in spite of the smaller sizes of some of their subsamples). Those who believe that there is more tolerant potential in Buddhism may suggest that the negative predictive power of devotional religiosity is driven by Buddhists. Those who believe that there is more tolerant potential in Christianity will assert the opposite. It is hard to argue, moreover, that the differential predictive role of devotional and coalitional religiosity is driven more by Buddhism or by Christianity, since this differential predictive role was more obvious among Christians in Samples 1 and 3, and more obvious among Buddhists in Sample 2. Generally, across the three samples, the larger subsample revealed a predictive difference between devotional and coalitional religiosity, while the smaller subsample did not or did so less significantly. This would suggest that the predictive divide between devotional and coalitional religiosity is Ndriven rather than driven by any particular religion. Finally, it may be objected that our measures are primarily based on self-reports and thus may not reflect real differences between devotional and coalitional religious processes in motivating religious intolerance. The self-reports were not simply opinion measures, however, but included self-reported prayer frequency and attendance at religious services, which were coded on a binary basis (regular vs. not) rather than a continuous basis. It is unlikely that self-reports about such concrete activities vastly distorted actual practice for prayer and religious attendance, and yet these measures showed a predictive pattern similar to the more opinion-based self-report measures. Further, the remarkable consistency in the pattern of self-reports between Christians and Buddhists in different nations is itself a significant finding. Across several different ways

Religion and Religious Tolerance 33 of measuring religious intolerance there was a consistent predictive division between devotional and coalitional religiosity in predicting that intolerance. All these considerations undermine a dismissal of our findings as based in “mere self-report.” We do consider it important for later studies to investigate causal relationships experimentally, however, and we are currently conducting two studies in that vein. In one study (Hansen, Norenzayan & Bashah, 2006), we find that under “divinity salience” (similar to the mortality salience paradigm, but asking people to imagine what would happen to them if they “encountered the Divine”) participants become either less exclusivist or less inclined to civil intolerance of other faiths. In no subsample does divinity salience increase exclusivity or civil intolerance. In another study (Hansen, Sharif & Norenzayan, 2006), scrambled sentences that, when unscrambled, prime either devotional concerns, coalitional concerns, or neither (control) reveal a significant difference between conditions such that the coalitionally-primed show the most civil intolerance (based on the measure by Sullivan et al, 1982), the devotionally primed the least, and the control primed in between. In a study of West Bank Israeli settlers (Ginges, Hansen & Norenzayan, 2006), settlers who were first asked about how often they pray were much less likely to attribute courage to Baruch Goldstein’s suicide attack against Palestinians than settlers who were first asked about synagogue attendance (those in the control condition were in between). These preliminary experimental studies suggest that the opposite predictions of self-reported devotional and coalitional religiosity in the present research are not statistical artifacts resulting from multicollinarity or some other analytical flaw.

Religion and Religious Tolerance 34 Also, some terror management-related studies on the effects of mortality salience suggest that awareness of immortality or Divine agency can eliminate or even reverse the effect of mortality salience on worldview defense, including intolerant worldview defense. For instance, while mortality salience usually inspires outgroup-derogating defense of ingroup worldview, certain aspects of worldview defense are eliminated when assuring participants of their literal immortality (Dechesne et al, 2003) and mortality salience can even lead majority Christian participants to believe more, not less, in the supernatural powers of Buddha and Shamanic spirits (Norenzayan & Hansen, 2006). Together these experimental studies corroborate the present correlational studies to suggest that a strong devotion to God, Buddha or comparable Divinity as an omnipresent all benevolent being or force may reduce reliance on ingroup-outgroup distinctions to understand the world (even if devotion tends to co-occur with the coalitional cohesion that is related to those distinctions). Internal validity Our theoretical framework sidesteps the theoretically and empirically problematic intrinsic-extrinsic dichotomy, and we have confined our analyses to measures for religious intolerance and religious exclusivity with high internal reliability, convergent and divergent validity. And while our analysis has included measures from the religion and prejudice literature, we have introduced an important family distinction between measures that are more about manifesting religious devotion and those that are more about setting the boundaries around religious (or cultural) coalitions. The theoretical perspective that emerges can guide future research into cultural psychology that is more mindful of the distinct role of religious affiliation and religiosity generally.

Religion and Religious Tolerance 35 External validity In addition, our research was not confined to samples of mostly Christian, white or North American participants, a practice common in most psychology of religion literature. Our inclusion of Buddhists, ethnic East Asians and students from Malaysia allowed us to control for ethnicity and nationality in investigating the psychological role of religiosity and religious affiliation. The use of religions that are widely divergent in doctrinal principles, historical development, regions of origin and dominance allowed the examination of whether there is a similar relationship between coalitional cohesion variables, religious devotion variables and intolerance variables across religions. The similarity found supports the universality of the relationships, i.e. that these relationships may be psychologically wired together rather than put together by accidents of particular religious texts and traditions and particular national cultures. Buddhism and Christianity: apples and oranges or two peas in a pod? The studies have corroborated our theoretical framework that identifies two types of “religiosity” that are, to some extent, seamlessly interrelated and yet, when considered separately, are opposite predictors of intolerance. These kinds of religiosity appear to work in largely the same way among both Christians and Buddhists. Devotional and coalitional religiosity are not the only ways to conceive religiosity, nor are the measures we used for these two constructs exhaustive or free of cultural bias—the fact that both forms of religiosity are more prevalent in Christians than in Buddhists suggests that these are forms of religiosity more relevant to Christians (or, perhaps, conversion-oriented monotheists generally) than to Buddhists. However, these two kinds of religiosity travel relatively intact to Buddhism in terms of their inter-relationship and their respective

Religion and Religious Tolerance 36 independent relationships with various kinds of religious intolerance. This consistency of interrelationships across samples, religions and countries suggests that the interrelationships may be universal, reflecting natural rather than culturally contingent religious processes. While Buddhists show less of a relationship between devotional religiosity and coalitional religiosity, it is perhaps shocking to some scholars of Buddhism that these modes of religiosity are positively correlated at all among Buddhists. While it has been repeatedly found in predominantly Christian samples that devotional religiosity (including Allport’s “intrinsic” religiosity) is positively correlated with coalitional or “right wing” authoritarianism and related constructs (Duck & Hunsberger, 1999; Kahoe, 1975; Leak & Randall, 1995; Moghaddam & Vuksanovic, 1990; Watson, Sawyers & Morris, 2003), Buddhism, with its history largely unsullied by internecine warfare and wars of Buddhist conquest, might be expected to show the reverse relationship—with more Divine-oriented (Buddha-oriented or Dharma-oriented) individuals being less coalitional. Moreover, Buddha’s life and teachings--breaking away from the comforts and privilege of wealth and power to face poverty, suffering and death, telling adherents not to believe what he taught unless they found it amenable to reason and good living— seem particularly hard to reconcile with coalitional divisiveness. Many would also argue that the reverse relationship should be seen in Christians too, given that they worship someone born in a stable as Christ, rather than an uncontroversially anointed king or emperor. The robust positive relationship between devotional and coalitional religiosity among Christians is in stark contrast with the insistence by more liberal theologians that Jesus’s teachings, as well as the canonical Gospel versions of Jesus’ birth life and death,

Religion and Religious Tolerance 37 were profoundly incompatible with authoritarianism—how can authoritarianism or any coalitional orientation be compatible with worshipping someone who was poor, unmarried, outspoken, weaponless, estranged from his family; who broke and preached against what others perceived to be religious laws; who consorted with lepers, traitors, prostitutes and religious outsiders; who died from excruciating torture carried out by soldiers of the greatest civilization in the Western Hemisphere and set in motion by political and religious leaders upset by the challenge he posed to their authority? Given the teachings, character and life story of both Buddha and Jesus, it is certainly not wildly implausible to expect that both Buddhist and Christian devotional religiosity should, in theory, be inversely related to coalitional religiosity. That the relationship is instead robustly positive for both suggests that intensity of religiosity (both devotional and coalitional) may operate on principles that are not as doctrine-contingent as might be imagined. Devotional and coalitional religiosity might be tightly wound among both Christians and Buddhists not because Jesus and Buddha went out of their way to exhort their followers to bind them together, but simply because they go together psychologically, and would go together in almost any sample of people, regardless of what religious teachings had distinguished those they revered as deities or deity-like exemplars of enlightenment. All this suggests that, to some extent, comparing Buddhism and Christianity is a matter of comparing apples of different shapes and sizes, rather than comparing apples and oranges. That is, with regard to the relationships between religious devotion, coalitional cohesion, and religious intolerance, Buddhism as believed and practiced is arguably not a fundamentally different religion from Christianity as believed and

Religion and Religious Tolerance 38 practiced.

The same religious psychological variables associated with tolerance and

intolerance among Christians are associated with tolerance and intolerance among Buddhists. While different religions may be more or less intolerant among different national groups, in different cultural contexts or at different times in history, the way one holds to one’s religious orientation, more than the unique features of the religious dogma espoused, appears to be the major determinant of tolerance or intolerance. How Buddhists and Christians hold to their religious beliefs appears to matter more for determining intolerance than their affiliation per se. Cultural psychological implications The relationship found in all samples between devotional religiosity, coalitional religiosity and intolerance corroborates William James’ assertion that “the baseness so commonly charged to religion’s account are thus, almost all of them, not chargeable to religion proper, but rather to religion’s wicked practical partner, the spirit of corporate dominion. And the bigotries are most of them in their turn chargeable to religion’s wicked intellectual partner, the spirit of dogmatic dominion” (James 1982/1902, p. 337). Corporate and dogmatic dominion would both be subsumed as coalitional religiosity in our theoretical framework. James himself was inclined to venerate a very individualistic form of religion, and individualism may appear to be the best strategy for avoiding corporate and dogmatic dominion. However, Buddhism has flourished in the last centuries primarily in “collectivist” cultures and Christianity in “individualist” cultures. It is not clear that religion growing on collectivist soil is necessarily more intolerant than religion growing

Religion and Religious Tolerance 39 on individualist soil, since religious affiliation seems to be less consequential compared to personal intensity of devotional and coalitional religiosity. The paradox of religiosity and intolerance Devotional religiosity, arguably the most distinctively religious (i.e. faith-oriented and divine-oriented) aspect of religiosity, walks a predictive tightrope, closely allied both with coalitional religiosity and to varying degrees religious tolerance also, in spite of the strong inverse relationship between coalitional religiosity and tolerance. Religious devotion may embody a distinct psychological function, one related to but not subsumed under coalitional religiosity. Certainly devotion to the supernatural Divine can be groomed as a weapon of war, ideological rigidity and intolerance as the opening quotation by Dawkins claims. However, the evidence presented here suggests that devotion to the numinous and divine in one’s faith also has potential to be associated with a softening or reversal of the inclination to such intolerance, perhaps because attention to a transcendent divine figure like God or Buddha overwhelms ingroupoutgroup distinctions in much the same way that attention to a transcendent God can overwhelm self-other distinctions within a potential coalition group. One may also consider the nuanced relationship between devotional religiosity, coalitional religiosity and intolerance as analogous to the nuanced relationship between the social inclusion aspect self-esteem, narcissism (or the superiority aspect of selfesteem) and aggression.

Social inclusion self-esteem and superiority-asserting

(narcissist) self-esteem are strongly positively correlated, but social inclusion self-esteem is found to negatively predict aggression while superiority self esteem and narcissism positively predict aggression (Bushman & Baumeister, 1998; Kirkpatrick, Waugh,

Religion and Religious Tolerance 40 Valencia & Weber, 2002). It is quite possible, in fact, that devotional and coalitional religiosity reflect a kind of collective social inclusion self-esteem and collective narcissism respectively. The contributions of this research While this paper is indebted to and replicates some findings from previous work in the psychology of religion (e.g. Laythe et al, 2001), the present research reaches substantially beyond past research both theoretically and empirically. These studies speak not to racial prejudice or homophobia, but specifically to religious intolerance, including support for religious violence. Religious intolerance is a form of prejudice that has had tremendous political and economic relevance throughout human history and is becoming increasingly relevant today as more people see the modern world as a religiously-colored clash of civilizations (Huntington, 1993). If all religions are not equal in their inclination to be intolerant at any one time, this may simply be because some religions nourish coalitional cohesion and religious devotion in different ways under different circumstances. Nevertheless, it is possible that some religions, due to the nature of their most accessible or reinforced doctrinal points in their most revered texts, may be chronically more inclined to one form of religiosity or other regardless of external circumstances. Classifying and comparing the frequency of coalitional vs devotional or numinous passages in the revered texts of various religions may relate to whether some religions are cross-contextually more inclined to propagate intolerance and violence (or tolerance and peace). This would be a fertile area for further research, though we suspect that natural inclinations shape the relationship between religiosity and intolerance far more than textual exhortations. To declare with despair that religious Christians can

Religion and Religious Tolerance 41 never be tolerant because the New Testament contains passages that appear to assign to hell those who fail to believe in Jesus is as preposterous as declaring that religious Jews can never be tolerant because of genocides apparently ordered by God in the Tanak or Old Testament or to declare that religious Muslims can never be tolerant because the Koran contains passages that appear to recommend the killing of non-Muslims or Muslims who convert to another religion. Does God have a religion? These studies are another illustration of how psychological science can contribute to the raging debates over religion and culture that are taking place inside and outside the academy. The present studies offer more support to a nuanced ecumenical view than to a view that excoriates all religious sentiment as inherently intolerant and violent or to a view that exalts some religious groups as uniquely inclined to be authors of tolerance and peace while indicting others as uniquely inclined to be authors of intolerance and war. Our research is most succinctly summarized by the famous statement of devout Hindu Mohandas Gandhi: “God has no religion”. An orientation towards the numinous or divine is associated with tolerance when coalitional cohesion is controlled, and this is true of religions as historically and theologically distinct as Buddhism and Christianity. To the extent people exalt the numinous or divine more than they defend the boundaries of their religion, culture or ideology, they are ready to live in tolerant peace with their neighbors. To the extent people exalt their religion, culture or ideology more than the numinous divine, their coalitional faith may potentially be, in Dawkins’ words, on a par with the hydrogen bomb.

Religion and Religious Tolerance 42 References Allport, G. (1950). The individual and his religion. New York: Macmillan. Allport, G. (1959). Religion and prejudice. Crane Review, 2, 1-10. Allport, G. W.& Kramer, B. M. (1946). Some roots of prejudice. Journal of Psychology: Interdisciplinary & Applied, 22, 9-39 Allport, G.W. & Ross, J.M. (1967). Personal religious orientation and prejudice. Journal of Personality and Social Psychology, 5, 432-443. Altemeyer, R.A. (1981). Right-wing authoritarianism. Winnipeg, Manitoba, Canada: University of Manitoba Press. Altemeyer, R.A. (1988). Enemies of freedom: Understanding right-wing authoritarianism. San Francisco: Jossey-Bass. Altemeyer, R.A. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press. Altemeyer, R. A. (1998). The other "authoritarian personality. " In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 30, pp.47-91). New York: Academic Press. Altemeyer, R.A. (1999). Right-Wing Authoritarianism [1990 version]. In J.P. Robinson, P.R. Shaver & L.S. Wrightsman (Eds.), Measures of Political Attitudes (Vol. 2), (pp. 104-106). Toronto: Academic Press. Altemeyer, R.A. (2002). Dogmatic behavior among students: Testing a new measure of dogmatism. Journal of Social Psychology, 142, 713-721. Altemeyer, R.A., & Hunsberger, B. (1992). Authoritarianism, religious fundamentalism, quest, and prejudice. International Journal for the Psychology of Religion, 2, 113133.

Religion and Religious Tolerance 43 Appleby, R. S. (2000). The Ambivalence of the Sacred: Religion, Violence and Reconciliation. Lanham: Rowen and Littlefield. Atran, S. (2002) In gods we trust: The evolutionary landscape of religion. Oxford University Press. Oxford. Atran, S. (2003). Genesis of suicide terrorism. Science, 299, 1534-1539. Atran, S. & Norenzayan, A. (2004). Religion’s evolutionary landscape: Cognition, commitment, compassion, communion. Behavioral and Brain Sciences, 27, 713770. British Broadcasting Corporation. What the World Thinks of God. Retrieved January 1, 2005 from the World Wide Web: http://news.bbc.co.uk/1/hi/programmes/wtwtgod/3475359.stm Batson, C.D., Schoenrade, P. & Ventis, L. (1993). Religion and the individual. New York: Oxford University Press. Bushman, B. J. & Baumeister, R. F. (1998). Threatened egotism, narcissism, self-esteem, and direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of Personality and Social Psychology, 75, 219-229. Cohen, A. B. & Hill, P. C. (2006). A cultural approach to religiousness and spirituality among American Catholics, Jews and Protestants. Unpublished manuscript. Cohen, A. B. & Rozin, P. (2001). Religion and the morality of mentality. Journal of Personality & Social Psychology, 81, 697-710. Cohen, A. B., Siegel, J. I. & Rozin, P. (2003). Faith versus practice: Different bases for religiosity judgments by Jews and Protestants. European Journal of Social Psychology, 33, 287-295

Religion and Religious Tolerance 44 Dawkins, R. (1989). The selfish gene. Oxford: Oxford University Press. Dawkins, R. (2006). The God delusion. London: Bantam Press. Dechesne, M., Pyszczynski, T., & Arndt, J. (2003). Literal and symbolic immortality: The effect of evidence of literal immortality on self-esteem striving in response to mortality salience. Journal of Personality & Social Psychology, 84, 722-737. Dennett, D. (2006). Breaking the spell. New York: Allen Lane. Derrida, J. (1998). Faith and knowledge: The two sources of ‘religion’ at the limits of reason alone. In J. Derrida and G. Vattimo (Eds.), Religion (pp. 1-78). Stanford: Stanford University Press. Donahue, M.J. (1985). Intrinsic and extrinsic religiousness: Review and meta-analysis., Journal of Personality & Social Psychology, 48, 400-419. Duck R., & Hunsberger, B. (1999). Religious orientation and prejudice: The role of religious proscription, right-wing authoritarianism and social desirability. International Journal for the Psychology of Religion, 9, 157-179. Fiorito, B. & Ryan, K. (1998, August). Development and preliminary reliability of a new measure of religiosity: Religiousness and spirituality questionnaires. Poster session presented at the annual meeting of the American Psychological Association, San Francisco, CA. Fisher, R.D., Derison, D., & Polley, C.F. (1994). Religiousness, religious orientation, and attitudes towards gays and lesbians. Journal of Applied Social Psychology, 24, 614-630.

Religion and Religious Tolerance 45 Freedom House. Religious freedom by religious background. Retrieved August 25, 2005 from the World Wide Web: http://www.freedomhouse.org/religion/publications/rfiw/fig2.htm Ginges, J., Hansen, I. G. & Norenzayan, A. (2006). Religion and support for combative martyrdom. Unpublished manuscript. Griffin, G.A., Gorsuch, R.L. & Davis, A.-L. (1987). A cross-cultural investigation of religious orientation, social norms, and prejudice. Journal for the Scientific Study of Religion, 26, 358-365. Guthrie, S. (2001). Why gods? A cognitive theory. In J. Andresen (Ed.), Religion in mind: Cognitive perspectives on religious belief, ritual and experience (pp. 94111). Cambridge: Cambridge University Press. Hanh, T. H. (1995). Living Buddha Living Christ. New York: Riverhead Books. Hansen, C. (1992). A Daoist Theory of Chinese Thought. Oxford: Oxford University Press. Hansen, I. G. & Norenzyan, A. (in press). Between Yang and Yin and Heaven and Hell: Untangling the complex relationship between religion and intolerance. In P. McNamara, (Ed.), Where God and Science Meet: How Brain and Evolutionary Studies Alter Our Understanding of Religion. Greenwood Press/Praeger. Hansen, I.G. & Norenzayan, A. (2006). Religion and war, religion and peace: Evidence for a nuanced relationship. Poster session presented at the annual meeting of the Society for Personality and Social Psychology, Palm Springs, CA. Hansen, I.G., Norenzayan, A. & Bashah, E. (2006). Terrorized into tolerance? Mortality and divinity salience can attenuate religious bigotry. Unpublished manuscript.

Religion and Religious Tolerance 46 Hansen, I.G., Norenzayan, A. & Sharif, A. (2006). An experimental manipulation of devotional and coalitional religiosity on religious intolerance. Unpublished raw data. Hoge, D.R. (1972). A validated intrinsic religious motivation scale. Journal for the Scientific Study of Religion, 11, 369-376. Hoge, D.R. & Carroll, J.W., (1973). Christian beliefs, nonreligious factors, and antiSemitism. Social Forces, 53, 581-594. Hoge, D.R. & Carroll, J.W. (1975). Religiosity and prejudice in northern and southern churches. Journal for the Scientific Study of Religion, 12, 181-197. Hunsberger, B. (1995). Religion and prejudice: The role of fundamentalism, quest, and right-wing authoritarianism. Journal of Social Issues, 51, 113-129. Hunsberger, B. (1996). Religious fundamentalism, right-wing authoritarianism, and hostility toward homosexuals in non-Christian religious groups. International Journal for the Psychology of Religion, 6, 39-49. Hunt, M. O. (2002). Religion, race/ethnicity, and beliefs about poverty. Social Science Quarterly, 83, 810-831 Huntington, S. (1993). The clash of civilizations? Foreign Affairs, 72, 22-49. Jacobsen, M. & Bruun, O. (2000). Human Rights and Asian Values: Contesting National Identities and Cultural Representations in Asia. Richmond: Curzon. James, W. (1982/1902). The Varieties of Religious Experience. New York: Penguin Books. Juergensmeyer, M. (2003). Terror in the Mind of God: The Global Rise of Religious Violence. Berkeley: University of California Press.

Religion and Religious Tolerance 47 Kahoe, R.D. (1975). Authoritarianism and religion: Relationships of F-scale items to intrinsic and extrinsic religious orientations. JSAS Catalog of Selected Documents in Psychology, 5¸284-285. Kahoe, R.D. & Meadow, M.J. (1977, October). Religious orientations dimensions: Individual and institutional interrelations. Paper presented at the annual convention of the Society for the Scientific Study of Religion, Chicago. [163] Kimball, C. (2002). When Religion Becomes Evil. San Francisco: HarperSan Francisco. Kirkpatrick, L. (1993). Fundamentalism, Christian orthodoxy, and intrinsic religious orientation as predictors of discriminatory attitudes. Journal for the Scientific Study of Religion, 32, 256-268. Kirkpatrick, L. (1999). Toward an evolutionary psychology of religion and personality. Journal of Personality, 67, 921-952. Kirkpatrick, L.A. & Hood, R.W. (1990). Intrinsic-extrinsic religious orientation: The boon or bane of contemporary psychology of religion? Journal for the Scientific Study of Religion, 29, 442-462. Kirkpatrick, L. A., Waugh, C. E., & Valencia, A. (2002). The functional domain specificity of self-esteem and the differential prediction of aggression. Journal of Personality and Social Psychology, 82, 756-767. Laythe, B., Finkel, D., Kirkpatrick, L. A. (2001). Predicting prejudice from religious fundamentalism and right-wing authoritarianism: A multiple-regression approach. Journal for the Scientific Study of Religion, 40, 1-10.

Religion and Religious Tolerance 48 Leak, G.K. & Randall, B.A. (1995). Clarification of the link between right-wing authoritarianism and religiousness: the role of religious maturity. Journal for the Scientific Study of Religion, 34, 245-252. Lee, R. (1997). Sacred Tensions. Columbia, S.C.: University of South Carolina Press. Manji, I. (2003). The Trouble with Islam, A Wake-Up Call for Honesty and Change. Toronto: Random House Canada. Martel, Y. (2001). Life of Pi. Orlando, Florida: Harcourt. Menand, L. (2001). The Metaphysical Club: A Story of Ideas in America. New York: Farrar, Straus and Giroux. Moghaddam, F.M. & V. Vuksanovic. (1990). Attitudes and behavior toward human rights across different contexts: The role of right-wing authoritarianism, political ideology, and religiosity. International Journal of Psychology, 25, 455-474. Morris, P. (1996). Community beyond tradition. In P. Heelas, S. Lash, & P. Morris (Eds.), Detraditionalization: Critical reflections on authority and identity (pp. 222–249). Cambridge, England: Blackwell. Navarette, C. D. & Fessler, D. M. T. (2005). Normative bias and adaptive challenge: a relational approach to coalitional psychology and a critique of terror management theory. Evolutionary Psychology, 3, 297-325. Nelson-Pallmeyer, J. (2003). Is Religion Killing Us?: Violence in the Bible and the Quran. Harrisburg: Trinity Press International. Neuberg, S.L. & Newsom, J.T. (1993). Personal need for structure: Individual differences in the desire for simpler structure. Journal of Personality & Social Psychology, 65, 113-131.

Religion and Religious Tolerance 49 Nietzsche, F. (1895/1968). The Anti-Christ. (R. J. Hollingdale, Trans). New York: Penguin Books. Norenzayan, A. & Hansen, I. G. (in press). Supernatural belief in the face of death. Personality and Social Psychology Bulletin. Rokeach, M. (1960). The open and closed mind: investigations into the nature of belief systems and personality systems. Oxford, England: Basic Books. Rowatt, W. C. & Franklin, L. M. (2004). Christian Orthodoxy, Religious Fundamentalism, and Right-Wing Authoritarianism as Predictors of Implicit Racial Prejudice. International Journal for the Psychology of Religion, 14, 125138. Sanchez-Burks, J. (2002). Protestant relational ideology and (in)attention to relational cues in work settings. Journal of Personality and Social Psychology, 83, 919929. Saroglou, V., Delpierre, V., & Dernelle, R. (2004). Values and religiosity: A metaanalysis of studies using Schwartz' s model. Personality & Individual Differences, 37, 721-734. Smith, H. (1991/1958). The World’s Religions: Our Great Wisdom Traditions. New York: HarperCollins. Sullivan, J.L., Pierson, J.E. & Marcus, G.E. (1982). Political tolerance and American democracy. Chicago: University of Chicago Press. Tix, A. P., & Frazier, P. A. (2005). Mediation and Moderation of the Relationship Between Intrinsic Religiousness and Mental Health. Personality & Social Psychology Bulletin, 31, 295-306.

Religion and Religious Tolerance 50 Trimble, D.E. (1997). The Religious Orientation Scale: Review and meta-analysis of social desirability effects. Educational & Psychological Measurement, 57, 970986. Tweed, T.A. (2001). The American encounter with Buddhism, 1844-1912: Victorian culture & the limits of dissent. Chapel Hill, NC: The University of North Carolina Press. Watson P., Sawyers P., & Morris, R.J. (2003). Reanalysis within a Christian ideological surround: Relationships of intrinsic religious orientation with fundamentalism and right-wing authoritarianism. Journal of Psychology & Theology, 31, 315-328.

Religion and Religious Tolerance 51 Appendix A All variables were coded in binary terms since there was not a clear ordinal structure among the potential answers to some items. Participants who gave “don’t know” responses to any item (or pair of items in the case of “sacrificial scapegoating”) were not analyzed in relation to that item. Predictors: prayer frequency, religious attendance, exclusivity Question Answers Would you say you pray… Regularly only occasionally, at times of crisis only occasionally, at special religious events at religious festivals during the course of a normal year never don' t know I regularly attend an organized religious service agree disagree don' t know My God (Beliefs) is the only true God (Beliefs) agree disagree don’t know Dependent measure: scapegoating Question Answers I blame people of other religions for much of the Agree trouble in this world Disagree don’t know Question I would die for my God (beliefs).

Dependent measure: sacrificial scapegoating Answers Question Answers agree I blame people of other agree religions for much of the disagree trouble in this world don’t know disagree agree disagree don' t know Don' t know agree disagree don' t know

1. BCLR: Binary coding for logistic regression 2. NA: not analyzed

BCLR1 1 0 0 0 0 NA2 1 0 NA 1 0 NA BCLR 1 0 NA BCLR 1 0 13 0 0 0 13 0 NA

Religion and Religious Tolerance 52 3. agreeing with one statement while answering “don’t know” to the other was also coded as sacrificial scapegoating

Religion and Religious Tolerance 53 Appendix B Faith Devotion Scale [adapted from Intrinsic Religious Motivation (Hoge, 1972)] 1. My religious belief involves all of my life. 2. Although I am a religious person, I refuse to let religious considerations influence my everyday activities.* 3. One should seek Divine guidance when making every important decision. 4. In my life I experience the presence of the Divine. 5. Although I believe in my religion, I feel there are many more important things in life.* 6. My faith sometimes restricts my actions. 7. Nothing is as important to me as serving the Divine as best I know how. 8. It doesn’t matter so much what I believe as long as I lead a moral life.* 9. I try hard to carry my religion over into all my other dealings in life. 10. My religious beliefs are what really lie behind my whole approach to life. * Reversed items

Religion and Religious Tolerance 54 Appendix C Religious Exclusivity Scale 1. Only one religion is true; the other religions are false. 2. All religions teach essentially the same Divine values.* 3. Having faith is very important, but what particular faith you have is less important.* 4. Some religions are closer to the truth than others. 5. The Divine is revealed to us through different religions.* 6. Prayer can only be effective if you are praying to the right Divine Name. 7. The Divine answers prayers regardless of the Name by which the prayer calls on the Divine.* 8. Most religions have some truth to them, but there isn’t any one religion that can claim it is the only true faith.* 9. Many of the major religions have similar moral teachings; you can be a good person no matter what religion you profess.* 10. A deeply-felt religious belief can have a very positive impact on your life, but there is no one religion that is the best of all.* 11. Only one religious faith leads to salvation; all the others fall short. 12. It is not enough to believe in a religion that is partly true, because there is one religion that is absolutely true. 13. Although there is prayer and meditation in all religions, only if you pray according to the one true religion will your prayer do good. 14. If you pray sincerely, your prayer can do good, no matter what religion you believe in.* 15. Only one religion can lead your soul to final safety and peace. 16. There is only one religion that understands the source of all goodness and evil in this world; without believing in that religion, you cannot truly do good or escape evil. 17. If you say that all spiritual paths are equally true, you have said something false, because there is only one true path. 18. If you are kind and loving, you will be close to salvation no matter what religion you believe or practice.* * Reversed items

Religion and Religious Tolerance 55 Appendix D: Descriptive summary of the passage adapted from Martel (2002) The passage includes four short paragraphs describing how Piscine came to consider himself simultaneously Hindu, Christian and Muslim, as well as one passage describing a profound religious experience which he describes as follows: “Every element lived in a harmonious relation with its neighbor, and all was related. I felt like the center of a small circle coinciding with the center of a much larger circle. The individual soul met the world soul: Atman met Allah. The presence of God is the finest of rewards.” [The more richly-written version of this is on pages 62 and 63 in the original].The following day, Piscine is walking with his parents (described as not religious) in the park and is simultaneously (though coincidentally) approached by the three religious authorities he sees regularly: the Hindu pandit, Christian priest, and Muslim Imam. After arguing at first over whether Piscine is Hindu, Christian or Muslim, these authorities begin insulting each other’s religions, as in the following exchange: IMAM: Piscine, can this be true? Hindus and Christians believe in many gods. There is only one God. PANDIT: And Muslims have many wives. PRIEST: Piscine, there is salvation only in Jesus. PANDIT: Nonsense! Christians know nothing about religion. IMAM: Yes, Christians refuse to recognize the last and holiest messenger, peace be upon him. PRIEST: The holiest messenger was Jesus Christ the Son of God, and if you don’t accept him as your Lord and Savior you will burn in hell. IMAM: I’m afraid you are mistaken—it is you who will burn in hell.

Religion and Religious Tolerance 56 PANDIT: You will both burn in each other’s hells. The real question is, why is Piscine involved with these foreign religions? [Some of this

exchange (the exchange on hell) was

invented for this study. The rest can be found in Martel (2001) on pages 67-68]

The three authorities then demand that Piscine choose one of the three faiths. Piscine responds as follows: “Mahatma Gandhi said, ‘All religions are true.’ I just want to love God” (see Martel, 2002, p. 69). Piscine then says the following about Gandhi (invented for this study): “There is something rude about interrupting an interfaith battle with the name of a man who died begging the religions around him to bring the people he loved together, not to tear them apart.” The three wise men then pull away with “stiff, grudging smiles on their faces” (p. 69) and Piscine’s father takes the family out for ice cream.

Religion and Religious Tolerance 57 Appendix E Devotion to divine scale 1. An important part of my religious practice is to feel the presence of the Divine in everything I do. 2. I have no interest in feeling a sense of awe about ‘the truth and greatness of the Divine’.* 3. An important aspect of my prayer experience is to feel a deep sense of connectedness with the Divine. 4. Fundamental to my religious experience is an absolute trust in the Divine. 5. It is not important for me to feel an intimate personal relationship with the Divine.* 6. For me, an important part of religious practice is experiencing the power and majesty of the Divine. 7. I never try to achieve the feeling of oneness with the Divine.* 8. I do not have much faith in ‘Divine will’.* 9. Part of my being religious involves a constant search for deeper truths behind the words of my sacred texts. 10. I try to avoid centering my thoughts and will around Divine principles.* 11. An essential aspect of my religious practice is to try to understand Divine law. 12. I almost never think about the Divine during the day.* 13. It is not very important for me to search for knowledge about what the Divine would have me do.* 14. I rarely or never think about ways to improve myself to be more attuned to Divine law.* 15. I ask religious questions in order to better understand my relationship to the Divine. 16. My search for Divine guidance is an important religious activity. * Reversed item

Religion and Religious Tolerance 58 Appendix F Antipathy towards the Characters (Religious Antipathy) 1. Refer to the following scale to indicate how you feel about each character in the story. 1. 2. 3. 4. 5. 6. 7. 8. 9. Extremely Very Generally Slightly Neutral Slightly Generally Very Extremely negative negative negative negative positive positive positive positive Piscine Patel The pandit The priest The imam Piscine' s parents

1 1 1 1 1

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

2. How wicked (evil, bad) and how righteous (good) do you think each character is in the story? [1 = extremely wicked; 9 = extremely righteous] 3. State how likely it is that each character will go to heaven or hell. If you do not believe in heaven and hell, check here ____ and leave the question blank. [1 = definitely hell; 9 = definitely heaven; 2 and 8 = “very probably hell/heaven”; 3 and 7 = “more probably hell/heaven”; 4 and 6 = “slightly more probably hell/heaven”] 4. Do you think that each character will have good or bad experiences after they die? If you are not sure, mark neutral. If you do not believe that people have experiences after they die, check here ____ and leave this question blank. [1 = totally bad (experience after death); 9 = totally good experience after death; 2 and 8 = “very bad/good”; 3 and 7 = “generally bad/good”; 4 and 6 = “slightly bad/good”] 5. If you ruled the world and could easily punish people who did not do as you told them, whom would you forbid from talking about their beliefs to your friends and family and to others in your religious group? [1 = definitely forbidden; 9 = definitely not forbidden; 2 and 8 = “very probably”; 3 and 7 = “more probably”; 4 and 6 = “slightly more probably”] Feelings about Hypothetical Killing of the Characters (Aggressive Antipathy) 6. How sad or joyous do you think the Divine would feel at the killing of each character—in other words, how much do you think each character deserved to be killed? Assume that the Divine is not joyous to see a righteous person killed. Please pay close attention to the scale. If you do not believe that the Divine “feels” emotions, then please check here ____ and leave the question blank.

7 7 7 7 7

8 8 8 8 8

9 9 9 9 9

Religion and Religious Tolerance 59

7. How sad or joyous would you feel at the killing of each character? [1 = extremely joyous (to see the person killed); 9 = extremely sad (to see the person killed); 2 and 8 = “very joyous/sad”; 3 and 7 = “joyous/sad”; 4 and 6 = “slightly joyous/sad”] Willingness to Violate the Civil Rights of the Characters (Civil Intolerance) [based on Sullivan, Pierson & Marcus (1982)] People who believe and practice 3 religions at once Hindus Christians Muslims Nonreligious people

1 1 1 1 1

2 2 2 2 2

3 3 3 3 3

4 4 4 4 4

5 5 5 5 5

6 6 6 6 6

1. 2. 3. 4. 5.

___________ should be BANNED from being a political leader of your country. ________ should be ALLOWED to teach in public schools. * If ___________ form groups, these groups should be OUTLAWED. ___________ should be ALLOWED to make a speech in your town or city.* __________ should have their PHONE CONVERSATIONS SECRETLY LISTENED TO by your government. 6. _________ should be ALLOWED to hold rallies in your town or city.* [1 = totally disagree 9 = totally agree; 2 and 8 = “strongly disagree/agree”; 3 and 7 = “disagree/agree”; 4 and 6 = “slightly disagree/agree”] * Reversed item Support for Moral/Religious Killing (Religious Aggression) 1. The Divine rejoices in the killing of the wicked—glad that they are taken off the earth—but the Divine is sad at the killing of the righteous, because their loss is a great loss. 2. It is good when a wicked person dies. (It is a pity if they go to hell, but the Divine is better served if the wicked are gone). 3. It is good to kill a wicked person. 4. The Divine sometimes calls on the righteous to kill the wicked. [1 = totally untrue; 9 = totally true; 2 and 8 = “very untrue/true”; 3 and 7 = “generally untrue/true”; 4 and 6 = “slightly untrue/true”]

7 7 7 7 7

8 8 8 8 8

9 9 9 9 9

Religion and Religious Tolerance 60 Author Note Ian G. Hansen and Ara Norenzayan, Department of Psychology, University of British Columbia, Vancouver, Canada. This research has been supported by a Social Sciences and Humanities Research Council of Canada grant to the second author (410-2004-0197). We thank Peter Suedfeld and Del Paulhus for comments on an earlier version of this paper. We also wish to thank the following people who helped to conduct these studies or otherwise made them possible: Abdul-Lateef Abdullah, Rohani Abdullah, Juhaidi Yean Abdullah, Tessa Amundson, Su' ad Awab, Eleanor Chow, Benjamin Gibbons, Sophia Khan, Neewern Khoo, Kim Koh, Susan Lin, Michael Mackay, Norzaliana Mat Nuri, Tee Ridhuan, Suellen Teh, Claudia Vexelman, Stephanie Young, and all participants. Address correspondence to Ian Hansen, Department of Psychology, University of British Columbia, 2136 West Mall, Vancouver, BC, Canada V6T 1Z4, email: [email protected].

Religion and Religious Tolerance 61

Footnotes 1

Allport did not contrast personally-felt religious devotion with coalitional religiosity per se, but rather with a kind of instrumental

religiosity (Allport, 1950; Allport & Ross, 1967). This contrast between personally felt devotion and instrumental religiosity (known in the literature as intrinsic and extrinsic religiosity) has been critiqued as unhelpful to illuminating prejudice, however (Kirkpatrick & Hood, 1990), and partly this is because of the strong empirical correlation between personally-felt religious devotion and more coalitional forms of religiosity like authoritarianism and fundamentalism (see Batson, Schoenrade & Ventis, 1993 for a discussion). 2

For each country, a different means of sampling was used, sometimes based on several large cities, sometimes based on the whole

populations, sometimes using random digit dialing in countries with extensive telephone systems, sometimes using face to face interviews. In some cases, representativeness was aided by weighting, though weights are not considered in this study in order to simplify analysis and because they did not affect the pattern of results. For a complete listing of the specific sampling methods used for each country, please contact the authors. 3

For 225 out of 254 Buddhists in the sample, “prayer” was translated into Korean as

(gi-do), which is not specifically

Christian prayer but can denote prayer in all religions including Buddhism. Gi-do also can connote meditation. Only 21% of Buddhists indicated that they never prayed/meditated, suggesting that for the vast majority of Buddhists, prayer/meditation is a legitimate devotional activity.

Religion and Religious Tolerance 62

4

We used Altemeyer’s measure of dogmatism rather than Rokeach’s (1960) scale because the latter contained several items that were

dated and we also expected it would not translate well into Malaysian culture. 5

Due to a word processing error, items 7 and 8 from Altemeyer’s (1996) DOG scale were not presented to participants. Internal

reliability remained good, however. 6

Altemeyer (1996) notes that Quest religiosity (Batson & Ventis 1982), while often a negative predictor of prejudice, is also a highly

unstable construct, with internal reliability dropping as low as .17, and suffering from other measurement problems (e.g. lack of reversed items) even in more reliable versions.

Religion and Religious Tolerance 63

Table 1

Factor analyses of religiosity and intolerance variables, 254 Buddhists and 4710 Christians (Sample 1)

Buddhists

Christians

F1

F1

Prayer

.50

.72

Religious attendance

.68

.73

Exclusivity

.47

.42

Sacrificial Scapegoating

.17

.21

41.0

45.9

% variance explained

Devotional religiosity

Coalitional religiosity Religious intolerance

Religion and Religious Tolerance 64 Table 2 Factor analyses of religiosity and intolerance variables, 81 Buddhists in Malaysia (Sample 2)

Buddhists F1

F2

Religious devotion

.82

.17

Authoritarianism

.56

-.06

Exclusivity

.34

-.60

Fundamentalism

.54

-.44

Dogmatism

.66

-.28

Anti-pluralism

-.07

-.67

% variance

55.1

16.7

Devotional religiosity

Coalitional religiosity

Religious intolerance

Religion and Religious Tolerance 65 Table 3 Factor analyses of religiosity and intolerance variables, 48 Christians in Malaysia (Sample 2)

Christians F1 Religious devotion

.72

Authoritarianism

.34

Exclusivity

.80

Fundamentalism

.96

Dogmatism

.82

Anti-pluralism

.72

% variance

62.0

Devotional religiosity

Coalitional religiosity

Religious intolerance

Religion and Religious Tolerance 66 Table 4 Factor analyses of religiosity and intolerance variables, 23 Buddhists in Canada (Sample 3)

Buddhists F1

F2

Devotion to divine

.22

.71

Intrinsic religiosity

.33

.73

Authoritarianism

.63

-.04

Exclusivity

.91

.06

Fundamentalism

.92

.15

Dogmatism

.59

.14

Religious antipathy

.78

.17

Civil intolerance

.79

.03

Aggressive antipathy

.90

-.23

Religious violence

.30

-.36

% variance

52.8

15.9

Devotional religiosity

Coalitional religiosity

Religious intolerance

Religion and Religious Tolerance 67 Table 5 Factor analyses of religiosity and intolerance variables, 71 Christians in Canada (Study 3)

Christians F1

F2

Devotion to divine

.76

.53

Intrinsic religiosity

.92

.26

Authoritarianism

.63

-.33

Exclusivity

.86

-.05

Fundamentalism

.95

-.06

Dogmatism

.70

-.13

Religious antipathy

.61

-.48

Civil intolerance

.27

-.64

Aggressive antipathy

.00

-.59

Religious violence

-.07

-.58

% variance

48.3

21.7

Devotional religiosity

Coalitional religiosity

Religious intolerance

Religion and Religious Tolerance 68

Table 6 Regression slopes for devotional and coalitional religiosity among Buddhists and Christians (all samples) Sample Sample 1 [BBC multination survey]

Criterion

Intolerance Devotional (Scapegoating) Religiosity Coalitional Religiosity Devotional Intolerance Religiosity (Sacrificial Scapegoating) Coalitional Religiosity Intolerance (Antipluralism)

Sample 2 [Malaysia] Intolerance (Religious antipathy)

Sample 3 [Canada]

Predictors

Intolerance (4 index composite)

Devotional Religiosity Coalitional Religiosity Devotional Religiosity Coalitional Religiosity Devotional Religiosity Coalitional Religiosity

B (SE) [254 Buddhists] -.44 (.46)

B (SE) [4710 Christians] -.44 (.08)***

.06 (.53)

.21 (.11)*

-.21 (1.01)

.03 (.10)

2.23 (1.23)†

1.57 (.14)***

(S ) [81 Buddhists] -.29 (.12)*

(S ) [48 Christians] -.06 (.15)

.62 (.12)***

.73 (.15)***

-.21 (.12)†

-.14 (.13)

.65 (.12)***

.88 (.13)***

(S ) [24 Buddhists] -.13 (.14) a

(S ) [71 Christians] -.52 (.13) a ***

.89 (.14)***

.84 (.13)***

† p < .1 * p < .05 ** p < .01 *** p < .001 a Difference between Buddhist and Christian regression slopes is significant at p < .05

Religion and Religious Tolerance 69 Table 7: Conceptual summary of devotional and coalitional religiosity predicting intolerance among Buddhists and Christians for all measures of intolerance Predictor a

Devotional religiosity

Coalitional religiosity

Criterion Scapegoating (Sample 1) Sacrificial Scapegoating (Sample 1) Anti-pluralism (Sample 2) Religious antipathy (Sample 2) 4 index composite of intolerance (Sample 3) Scapegoating (Sample 1) Sacrificial Scapegoating (Sample 1) Anti-pluralism (Sample 2) Religious antipathy (Sample 2) 4 index composite of intolerance (Sample 3)

Buddhists Neg. (N)

Christians Neg. (S)

Neg. (N)

Pos. (N)

Neg. (S)

Neg. (N)

Neg. (S)

Neg. (N)

Neg. (N)

Neg. (S)

Pos. (N)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

Pos. (S)

a. Where an orthogonal result is nominally negative or positive it is listed as “Neg. (N)” or “Pos. (N)”; where a result is at least marginally significant it is listed as “Neg. (S)” or “Pos. (S)”.

Religion and Religious Tolerance 70

Table 8: Construct intercorrelations, by study and religion Construct intercorrelations Study 1: Buddhists Devotional Religiosity (DR) (N = 254) Coalitional Religiosity (CR) Religious Intolerance: Scapegoating (RI [S]) Religious Intolerance: Sacrificial Scapegoating (RI [SS]) Study 1: Christians Devotional Religiosity (DR) (N = Coalitional Religiosity (CR) 4170) Religious Intolerance: Scapegoating (RI [S]) Religious Intolerance: Sacrificial Scapegoating (RI [SS]) Study 2: Buddhists Devotional Religiosity (DR) Coalitional Religiosity (CR) (N = 81) Religious Intolerance: Anti-pluralism (RI [AP]) Religious Intolerance: Religious antipathy (RI [RA]) Study 2: Christians Devotional Religiosity (DR) Coalitional Religiosity (CR) (N = 48) Religious Intolerance: Anti-pluralism (RI [AP]) Religious Intolerance: Religious antipathy (RI [RA]) Study 3: Buddhists Devotional Religiosity (DR) (N = 24) Coalitional Religiosity (CR) Religious Intolerance: (RI) Study 3: Christians Devotional Religiosity (DR) (N = 71) Coalitional Religiosity (CR)

DR --

CR .37*** d (.45)

RI [S] -.06 -.02 --

DR --

CR .54*** d (.45)

RI [S] -.08*** -.02 --

DR --

CR .57*** (.83)

RI [AP] .07b .46*** a --

DR --

CR .70*** (.81)

RI [AP] .44*** b .70*** a --

DR (.80)

CR .47* (.86)

RI .31 .84*** c (.75)

DR (.89)

CR .69*** ( .88)

RI .04 .48*** c

RI [SS] .04 .13* --RI [SS] .11*** .20*** --RI [RA] .16a .52*** c --RI [RA] .47*** a .79*** c ---

Religion and Religious Tolerance 71

Religious Intolerance: (RI)

(.70)

Note: RA = religious attendance, E = exclusivity, SS = sacrificial scapegoating, S = scapegoating, DR = devotional religiosity, CR = coalitional religiosity, RI = religious intolerance. Minimum N for Buddhists correlations: N = 237; minimum N for Christians: N = 4472 a. Difference between Buddhist and Christian correlations significant at p < .1 b. Difference between Buddhist and Christian correlations significant at p < .05 c. Difference between Buddhist and Christian correlations significant at p < .01 d. Difference between Buddhist and Christian correlations significant at p