A Comparative Study of Mystical Experience Among Christian, Muslim, and Hindu Students in Tamil Nadu, India

A Comparative Study of Mystical Experience Among Christian, Muslim, and Hindu Students in Tamil Nadu, India FRANCIS-VINCENT ANTHONY Department of Practical Theology Salesian Pontifical University CHRIS A. M. HERMANS Department of Empirical Practical Theology Radboud University Nijmegen CARL STERKENS Department of Empirical Study of Religion Radboud University Nijmegen Hood developed a Mysticism Scale based on the theoretical work of Stace. The scale was tested by Hood and others in a comparative perspective. Using an abridged version of Hood s Mysticism Scale, we join the debate with a study of a much larger number of Christian, Muslim, and Hindu respondents (1,920 college students) living in Tamil Nadu, India. Our empirical analysis yields a moderately reliable model of mystical experience that permits comparison between the three religious traditions. We argue for the usefulness of a comparative model of vertical mysticism that combines with the complementary common characteristics of noetic quality and ineffability. Vertical mysticism has a revelatory, ineffable character and is comparable in the experience of adherents of the Christian, Islamic, and Hindu traditions. INTRODUCTION This study focuses on the mystical experiences of college and university students living in the southern state of Tamil Nadu, India. The proportions of Hindus (88.7 percent), Muslims (5.5 percent), and Christians (5.7 percent) in the area make Tamil Nadu a suitable place for a comparative study of mystical experience among adherents of Christian, Islamic, and Hindu traditions. Taking stock of Stace s view (1961:21) that all men, or nearly all men, are in some sense or other rudimentary or unevolved mystics, we seek to analyze mysticism as a normal phenomenon, reported by healthy and functioning persons struggling to find a meaningful framework within which to live out their experience as foundational as at least what is real for them, if not in some sense as the ultimate Real (Hood et al. 1996:267). We begin with an overview of the theoretical framework of mystical experience propounded by Stace (1961), which formed the basis of Hood s Mysticism Scale. Insofar as mystical experience can be linked to personal characteristics we also consider its social location. Although the Reinert and Stifler (1993) replication of Hood s scale included some Hindu and Buddhist monks and nuns, no comparison based on religious affiliation was reported. Hood et al. (2001) did compare Christians and Muslims using the same mysticism scale. Our comparative research goes beyond these attempts by comparing Hindus with Christians and Muslims, testing for unidimensionality and scalar invariance. Our comparative study, based on responses from 1,920 students, attempts to meet a long felt need underscored by Hood et al. (2001:704). In their view, if comparison of Christianity and Islam is pertinent because both are monotheistic religions Correspondence should be addressed to Francis-Vincent Anthony, Department of Practical Theology, Salesian Pontifical University, Rome. E-mail: vincent@unisal.it Journal for the Scientific Study of Religion (2010) 49(2):264 277 C 2010 The Society for the Scientific Study of Religion

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 265 with shared historical roots, it is even more challenging to compare Christians, Muslims, Hindus, and Buddhists because the comparison might yield very different findings. THEORETICAL FRAMEWORK Hood s Mysticism Scale is based in part on the conceptual framework of mysticism propounded by Stace. Stace (1961) outlines a conceptualization of mysticism that is cross-cultural, a-historical, and unbiased by religious ideology. His conceptual framework rests on three constructs: (a) a distinction between mystical consciousness and its interpretation; (b) a distinction between the core characteristics of extrovertive and introvertive mysticism; and (c) identification of universal common characteristics (see Figure 1). Stace (1961:31) makes a distinction between mystical consciousness and its interpretation on the basis of his enquiry into Christian, Islamic, Judaic, Hindu, Buddhist, and Taoist mystical experiences. The distinction is analogous to the distinction between sense experience and its interpretation, although he concedes that we may never come across an altogether uninterpreted experience. He recognizes the directness and seemingly uninterpreted state of unity in mysticism: The unity is perceived, or directly apprehended. That is to say, it belongs to experience and not to the interpretation, in so far as it is possible to make this distinction (1961:66). Stace uses the term mystical consciousness in this context, but not always consistently; at times he equates it with mystical experience, which includes the interpretation of the state of unity. In our view this state of unity belongs to the realm of mystical consciousness and not so much to that of interpretation. Empirical research supports the claim that mystical consciousness and its interpretation can be relatively independent (cf. Hood and Williamson 2000). Mystical Figure 1 Stace s common core theory of mystical experience; item numbers of our measuring instrument are noted in parentheses see the Appendix for list of items Core characteristics (mystical consciousness) Extrovertive mysticism Introvertive mysticism Unifying quality (9, 12) Ego quality (5, 6, 11) Inner subjective quality (3) Temporal-spatial quality (2) Common characteristics (interpretation) Noetic quality (1, 7) Ineffability (8) Positive affect (10) Religious quality (4) Paradoxicality a a This ninth criterion, that there can be no logic in an experience in which there is no multiplicity (Stace 1961:270), was not measured because Hood doubted its theoretical and empirical usefulness.

266 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION experience as a concept entails the distinct but not separable aspects of the consciousness of unity and its interpretation. While mystical consciousness in Stace s perspective is basically similar all over the world in different periods and religious contexts, mystical experiences can vary insofar as the interpretations of this state of unity may differ from one religious tradition to another. Stace s second distinction is between extrovertive and introvertive mystical experience, based on the direction of the stimuli that lead to the experience of unity. The essential difference between them is that extrovertive experience looks outward through the senses, while introvertive experience looks inward into the mind. Both culminate in perception of an ultimate unity what Plotinus called the One with which the perceiver experiences union or even identity. But extrovertive mystics using their physical senses perceive a multiplicity of external material objects sea, sky, houses, trees mystically transfigured so that the One or Unity shines through them. Introvertive mystics, on the other hand, by deliberately shutting off the senses and obliterating from consciousness the entire multiplicity of sensations, images, and thoughts, seek to plunge into the depths of their own egos. There, in that darkness and silence, they claim to perceive the One and be united with it not as a Unity seen through multiplicity (as in extrovertive experience), but as the wholly naked One devoid of any plurality whatever. (Stace 1961:61 62) Stace identifies two core characteristics of extrovertive and introvertive mysticism. Extrovertive mysticism is characterized by awareness of unity with the universe or the perception of all things as one (unifying quality), and by apprehension of the One as inner subjectivity or life in all things (inner subjective quality). In introvertive mysticism the experience of unity refers to a pure state, in the sense that the mystical consciousness has no substantive content and is even characterized by a loss of self (ego quality). Such consciousness of nothingness is accompanied by distortion of time and space (temporal-spatial quality). He describes it as follows: When the self is not engaged in apprehending objects it becomes aware of itself....one may also say that the mystic gets rid of the empirical ego whereupon the pure ego, normally hidden, emerges into the light. The empirical ego is the stream of consciousness. The pure ego is the unity which holds the manifold of the stream together. This undifferentiated unity is the essence of the introvertive mystical experience (1961:86 87). These descriptions of the very essence of mystical experience suggest that the extrovertive type is an incomplete kind of experience which finds its completion and fulfillment in the introvertive kind of experience. The extrovertive kind shows a partly realized tendency to unity which the introvertive kind completely realizes (Stace 1961:132). In pointing out the interconnection between extrovertive and introvertive consciousness, Stace underscores that mystics themselves generally do not distinguish between the two types, indicating that there is also a wider set of characteristics common to both extrovertive and introvertive mystical consciousness. These common characteristics of mystical experience, Stace claims, are universal in all cultures, religions, and ages. He identifies five psychological and phenomenological common characteristics that can be seen as universal factors directing the interpretation of mystical consciousness: noetic quality (perception of special knowledge or insight), ineffability (difficult to articulate), positive affect (experience of peace or bliss), religious quality (perception of sacredness or wonder), and paradoxicality (cf. Hill and Hood 1999:364; Stace 1961:31 37, 131 32). Together with Hood et al. (1996:257 58) we may sum up the three fundamental assumptions in Stace s proposal as follows: first, despite the variations in interpretation, mystical consciousness is universal and is essentially identical; second, conceptually there is a clear distinction between the introvertive and extrovertive forms of mysticism; third, although there is a set of common core characteristics, mystical experience need not always imply all characteristics, since there can be borderline cases. Stace s common core theory has been challenged by Katz (1978) and others, who favor the diversity theory and argue that no unmediated experience is possible, and that in the

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 267 extreme, language is not simply used to interpret experience but in fact constitutes experience (Katz 1978:256). Here the category consciousness may help to clarify that, while we can speak of pure consciousness without interpretation or qualification, experience always implies interpretation. The opposition between the common core theory and the diversity theory seems to be compounded by Stace s identification at times of consciousness with experience. As explained above, mystical experience can be seen as comprising both mystical consciousness (consciousness of union with reality/reality) and its interpretation. While in some cases we may speak of uninterpreted pure consciousness, mystical experience always implies a greater or lesser degree of interpretation of the consciousness of union. Research Questions EMPIRICAL RESEARCH Against the background of this conceptual framework of mystical experience, the research questions that we wish to address are: (1) What comparative model of mystical experience emerges among Christian, Muslim, and Hindu students once group-specific differences have been ascertained? (2) Are there significant differences in the levels of mystical experience between Christian, Muslim, and Hindu students? (3) Which personal (sociocultural, socioeconomic, and socioreligious) characteristics are related to the level of mystical experiences among Christian, Muslim, and Hindu college students? (4) To what extent can agreement with mystical experience among Christian, Muslim, and Hindu college students be explained by personal (sociocultural, socioeconomic, and socioreligious) characteristics? Measuring Instrument The measurement instrument is based on Hood s Mysticism Scale. This scale has been proven empirically to be acceptable in diverse cultural contexts and religious traditions (Hill and Hood 1999; Holm 1982; Hood 1975; Hood et al. 2001). Hood s Mysticism Scale comprises 32 items, half of which are formulated negatively to prevent response set. Eight common core characteristics were operationalized with four items each from Stace s concept of mystical experience. The two-component structure reported by Hood (1975) and the three-factor solution advanced by Caird (1988) and Reinert and Stifler (1993) suggest a three-factor Mysticism Scale, compatible with Stace s conceptualization: extrovertive mysticism (experience of unity with the external world) measured with 12 items; introvertive mysticism (experience of nothingness) measured with 8 items; and religious interpretation measured with 12 items (Hood, Morris, and Watson 1993; Hood and Williamson 2000). Since mystical experience is only one of many themes in a broader cross-religious study of participative and conflictive tendencies among religious traditions, we abridged the measuring instrument to 12 items. This was possible because Hood s scale of 32 items consists partly of negative formulations (16 items), which to a great extent simply reverse the positive formulations. We opted for positive formulations only in accordance with modifications in our answering format. Each dimension (extrovertive mysticism, introvertive mysticism, and religious interpretation) is represented by four items. Our choice of items is based on the interitem correlations in Hood s initial study (1975) and the results of our own pilot study. Since English is the language of education (in colleges) for all respondents, it was not necessary to translate the items. However, it should be noted that linguistic consideration can play a role in evaluating the appropriateness of an instrument in a multilingual context (Van de Vijver and Leung 1997:38 40).

268 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION We should point out that the items we used from Hood s original scale (form D) represented the eight common core characteristics of mystical experience (see the Appendix for list of items): ego quality (items 5, 6, 11), temporal-spatial quality (item 2), unifying quality (items 9, 12), inner subjective quality (item 3), noetic quality (items 1, 7), ineffability (item 8), positive affect (item 10), and religious quality (item 4). Hood used a four-point Likert-type response ranging from +2 ( this description is definitely true of my own experience/s ) to 2 ( this description is definitely not true of my own experience/s ), with no midpoint but with the additional option, I cannot decide. The original format of the item (e.g., I have had an experience in which a new view of reality was revealed to me ) seemed weighted in favor of a positive response and rather indirect in its approach to respondents mystical experience. That is, it focused less on the experience and more on the description of it. In Hood s formulation the respondents had to choose between This description is probably (or definitely) true (or not true) of my own experience/s. We opted for a more direct formulation of items and responses. For example, our item 7 reads: Did you ever have an experience in which a new view of reality was revealed to you? The respondents could answer on a four-point Likert scale: Certainly no, Probably no, Probably yes, Certainly yes. We have included three categories of background variables in our research: sociocultural characteristics (age, gender, language, urbanization, and field of specialization at college/university), socioeconomic characteristics (caste, educational level of mother, educational level of father), and socioreligious characteristics (agents who have influenced the religiosity of the respondents, namely, parents, relatives, friends, religious community, teachers/professors, and media). All measuring instruments for the background variables have been constructed by the authors. Age is measured in three categories: (1) 17 19 years, (2) 20 22 years, and (3) 23 26 years. Gender is coded (1) male and (2) female. Language is coded (1) the respondent s mother tongue is Tamil or (2) not Tamil. Urbanization is measured in three categories: (1) village, (2) town, and (3) city. Field of specialization is measured in two broad categories: (1) Art & Social Sciences and (2) Natural Sciences. Caste, a characteristic element of the Indian society, is measured in terms of (1) Forward Caste, (2) Backward Castes, (3) Most Backward Castes, (4) Scheduled Caste, (5) Scheduled Tribes, and (6) Other Caste. 1 Educational level of mother and of father is measured in three categories: (1) no formal education or primary education, (2) middle or high school, and (3) secondary school, undergraduate level or higher level education. The influence of socioreligious agents (namely, parents, relatives, friends, religious community, teachers/professors, and media-persons linked to one s religion) on the understanding and practice of one s religion is measured with a Likert scale ranging from (1) very unfavorable to (4) very favorable. Sampling and Data Collection The sampling and data collection procedure has already been described in previous articles (Anthony, Hermans, and Sterkens 2005, 2007) as part of a wider research project on interreligious 1 The Indian Constitution (Articles 341 & 342) identifies ST, originally considered outcastes (comprising over 24 percent of the Indian population), and Scheduled Castes (SC) as groups who have historically suffered oppression and denial of equal opportunity. The government proposes to reserve a certain percentage of jobs in the public sector for ST (7.5 percent) and SC (15 percent). Later on, reservations were introduced for Other Backward Castes (Constitution Article 340). The detailed division of castes and the percentage of job reservations can vary according to states. In the state of Tamil Nadu, a further distinction was made between Backward Castes (BC) and Most Backward Castes (MBC) in 1971. In our research we followed the classification used by the Tamil Nadu government: ST are traditionally identified with adivasi (original indigenes and outcastes); SC are identified with dalit (untouchables). MBC and BC are consequently the better end of the disadvantaged groups. Those who do not belong to these categories and cannot claim reservations are said to belong to the Forward Castes (FC). We also included the category Other Castes (OC), since there may be respondents who do not associate themselves with these categories.

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 269 participation and conflict. A selective stratified sample, taking into account students gender, religious affiliation, area of residence, and educational level, was drawn from 16 colleges and Madras University. In order to assure an appropriate gender balance, eight women s colleges were selected, the remaining eight being principally for men. Madras University represents a fully fledged co-educational system. The demographic characteristics of the 1,920 respondents reveal the adequacy of our sampling procedure. The respondents represent the genders rather equally: 55.6 percent are women and 44.4 percent men. As for religious affiliation, 45.3 percent are Christians (28.1 percent Catholics, 12.8 percent Protestants, 4.4 percent from other Christian denominations), 41.1 percent are Hindus, and 13.3 percent Muslims, with a few (.4 percent) Jains or Buddhists. As the study focused on Christians, Hindus, and Muslims, students belonging to other religions were not included in our analysis. The percentages of students according to religious affiliation do not reflect the ratio in the Tamil Nadu population. The smaller proportion of Muslims in our sample is due to the fact that they rarely attend colleges other than their own, which are also fewer in number. Almost all the respondents (98.4 percent) are between 17 and 25 years of age. A vast majority (86 percent) are undergraduates, 65.6 percent of these in the final or third year of their studies. Our study focused on third- or final-year undergraduates, since some of them may not continue to postgraduate levels. Only 13.7 percent of our respondents are postgraduate students and a few (.3 percent) are enrolled in advanced studies (M.Phil. and Ph.D.). Procedure of Data Analysis As discussed more extensively elsewhere (Harkness, van de Vijver, and Mohler 2003; cf. Sterkens and Anthony 2008; Van de Vijver, Van Hemert, and Poortinga 2008; Van de Vijver and Leung 1997), the comparative nature of our research necessitates an appropriate data analysis procedure. Measurement invariance must be established when different groups are compared or when the same group is compared across time. In order to establish measurement invariance we use two procedures. In the first step we assess the dimensionality of the mystical experience scale via factor analysis. In the second step we test whether the scale can be used to compare the mystical experience of Christians, Muslims, and Hindus. Dimensionality assessment involves an exploration of the factor structure for the pooled sample of all three religious groups. After that, the factor structure is evaluated for each subgroup (Christians, Muslims, and Hindus). In this evaluation the factor structure of each subgroup meets the following criteria (eigenvalue >.1; factor loadings >.45; commonality >.20), otherwise the items are removed from the scale. Finally, we establish the factor structure of our comparative scale within the whole sample using only the items that met our criteria. Comparability is assessed as suggested by Meredith (1993). We first assess whether the measurement models are the same for all groups using confirmatory factor analysis by testing whether the factor loadings are the same across groups. We then compare the relationships of one factor with other factors. Finally, we test whether the item-intercepts are the same across the groups and compare the latent means (or composite scores). To evaluate the adequacy of the models we have used the CHI2, the RMSEA, and the GFI. We then make a cross-religious comparison of the scores on the common factor mystical experience using a Scheffé test. We relate the scores of mystical experience to background characteristics of the students and describe the social location of mysticism in relation to sociocultural, socioeconomic, and socioreligious student characteristics. Finally, we conduct a multivariate regression analysis.

270 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION Table 1: Factor analysis (PAF, oblimin rotation), commonalities (h 2 ), percentage of explained variance, and reliability (Cronbach s alpha) of comparable mystical experiences among Christian, Muslim, and Hindu students as a whole F1 h 2 7. Did you ever have an experience in which a new view of reality was revealed.56.32 to you? 8. Did you ever have an experience that cannot be expressed in words?.49.24 11. Did you ever have an experience in which something greater than yourself.49.24 seemed to absorb you? 6. Did you ever have an experience in which your own self seemed to merge.48.23 into something greater? Cronbach s alpha.58 Number of valid cases 1,869 Explained variance = 25.8%. F1 = Vertical mysticism (mystical union with a higher reality). 1 N = 1,920. Results of Empirical Analysis Research Question 1 What comparative model of mystical experience emerges among Christian, Muslim, and Hindu students once group-specific differences have been ascertained? When free factor analysis (step one) was run on items manifesting strong correlations (6, 7, 8, 11, 12), they yielded a single factor. We found that item 12 ( Did you ever have an experience in which you realized the oneness of yourself with all things? ) yielded low commonality for Christians and Muslims. The factor loading for Hindus, by contrast, was quite high (.51). In Stace s conceptual framework item 12 represents one of the core characteristics of extrovertive consciousness unifying quality. This suggests that the comparative model underrepresents the experience of mystical union with all reality (perception of oneness) in the case of Hindus. We shall discuss this difference in more detail in the final section. Table 1 presents results from the principal axis factoring (oblimin rotation method) of the four remaining items for the entire group of students. Analysis of the data with LISREL 8.51 (Jöreskog and Sörbom 1998) resulted in an acceptable model fit for the test of scalar invariance (X 2 [df = 18] = 38.75; RMSEA =. 043; GFI =.99). This means that we can assume that the measurement model is equivalent for the three different religious groups (Christian, Muslim, and Hindu), and allows meaningful comparison between these groups. Of the four items, two represent the complementary (active and passive) aspects of mystical union implying loss of self. On the one hand, an experience in which one s own self seems to merge into something greater (item 6); on the other hand, an experience in which something greater than oneself seems to absorb one (item 11). In the three factor analysis performed by Hood, Morris, and Watson (1993:1177) item 6 emerged as part of extrovertive mysticism. In Hood s original scale (form D) items 6 and 11 represented ego quality (loss of self), conceptualized as central to introvertive mysticism (Hood 1975:31). In our comparable factor analysis these two items point to one of the core dimensions of introvertive mysticism, namely, loss of self in the experience of union with a higher reality. The other two items (7 and 8) of our comparative model stand for the common characteristics of noetic aspect and ineffability. In Stace s theoretical framework ineffability (item 8) an experience that cannot be expressed in words is one of the common characteristics, but in Hood s

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 271 Table 2: Vertical mysticism scale; reliability of the comparative model, percentages of explained variance, and number of valid cases for Christian, Muslim, and Hindu students considered separately Cronbach s Alpha % of Explained Variance Valid Cases Christians.56 24.4 852 Muslims.57 25.6 237 Hindus.58 25.9 773 empirical study it turns out to be a component of introvertive mysticism. In a three-factor solution (Hood et al. 2001:694, 702) based on data from Iranian Muslims, the characteristic of ineffability fits the interpretation and introvertive factors equally well. In the case of U.S. Americans, however, ineffability is indicative of introvertive mysticism. The significance of ineffability seems to differ according to religiocultural setting. In Caird s study (1988:125) with a small sample of 115 respondents, the two-factor solution confirmed Stace s view of ineffability rather than Hood s. The two-factor solution in the Reinert and Stifler (1993:387) study also confirms ineffability as part of the common interpretive categories. We conclude, therefore, that the interpretive category of ineffability has a particular affinity with introvertive mysticism. Item 7 ( an experience in which a new view of reality was revealed ) has the highest loading (.56) on the factor. In Hood s three-factor solution items representing the noetic aspect form part of the interpretive category. This accords with Stace s understanding of the noetic aspect as a common characteristic that can qualify both introvertive and extrovertive mysticism. Two items in our comparative model (items 6 and 11) represent the complementary aspects of loss of self in the union with a greater reality, which is the core of introvertive mystical consciousness. The other two complementary interpretive items (items 7 and 8) indicate the noetic quality of this union (leading to perception of new insight) and its ineffable nature (experience of union being difficult to articulate). The comparative model of mystical experience thus represents a revelatory and ineffable experience of union with a greater or higher reality. Hence we label it experience of mystical union with a higher reality or vertical mysticism. The reliability (α) and percentage of variance explained by the comparative model of mystical experience are moderate when all respondents are taken together (Table 1), and remain almost unaltered when each of the three religious groups is examined separately (Table 2). Overall, the reliabilities are low, likely due to the limited number of items employed. Research Question 2 Are there significant differences in the levels of mystical experiences between Christian, Muslim, and Hindu students? As shown in Table 3, the experience of vertical mysticism reported by Hindus tends slightly toward ambivalence (mean 2.89), whereas Christians and Muslims affirm it as probably yes (means 3.13 and 3.05, respectively). Hindus with their lesser involvement are found to differ significantly from both Christians and Muslims (Scheffé s test: F-value: 27.81; sign. <.000). We would have expected Hindus to manifest a higher level of involvement. While mysticism is a key component of Hinduism, being both its source and center, in Christianity and Islam it is only a minor strand (Stace 1961:342 43). The lesser involvement indicated by Hindus tendency toward ambivalence suggests that vertical mysticism may not be as important for them as for Christians and Muslims. At the same time, there is evidence that the comparative model underrepresents mystical union with all reality, thus inadequately capturing the Hindus experience.

272 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION Table 3: Descriptive statistics (mean, SD) of all items in the vertical mysticism scale comparable among Christian, Muslim, and Hindu students Item 7 Item 11 Item 8 Item 6 All Items Mean SD Mean SD Mean SD Mean SD Mean SD Christian students (n = 852) 3.02.92 3.23.89 3.29.95 2.96.95 3.13.61 Muslims Students (n = 237) 3.05.89 3.08 1.00 3.33.92 2.75 1.00 3.05.63 Hindu students (n = 773) 2.80.98 2.95.99 3.10 1.04 2.71 1.04 2.89.67 Scale: 1 = certainly no; 2 = probably no; 3 = probably yes; 4 = certainly yes. See the Appendix for the formulation of the items. Research Question 3 Which personal (sociocultural, socioeconomic, and socioreligious) characteristics are related to the level of mystical experiences among Christian, Muslim, and Hindu college students? This question concerns the social location of mysticism (see Table 4). Among Christian students, female students report a higher level of vertical mysticism (r =.12). Four socioreligious characteristics are also associated with vertical mysticism: perceived favorable influence of friends, religious community, teachers/professors, and media. The association between the influence of teachers/professors on the religiosity of Christians and vertical mysticism is strongest (r =.19). Among the Muslim students, only the socioreligious characteristics are associated with vertical mysticism: influence of relatives, religious community, teachers/professors, and media. The strongest associations are with regard to the favorable influence of relatives (r =.23) and of media (r =.20). Table 4: Social location of vertical mysticism among Christian, Muslim, and Hindu students Christians Muslims Hindus Sociocultural characteristics Age Gender.12.10 Language.09 Urbanization Field of specialization Socioeconomic characteristics Caste Education mother Education father Socioreligious characteristics Parents.08 Relatives.23.10 Friends.12.08 Religious community.12.15.14 Teachers/professors.19.16.18 Media.14.20.13 Note: Correlations (eta for the nominal variables (gender and language); Pearson s r for the ordinal variables) between vertical mysticism on the one hand and some personal characteristics on the other hand. All correlations are significant at p <.00 level ( )orp <.05 level ( ).

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 273 Among Hindu students, female respondents report a higher level of vertical mysticism (r =.10) as do those who speak the Tamil language (r =.09). In the case of Hindus all socioreligious characteristics are significantly connected with vertical mysticism. The strongest association refers to the perceived favorable influence of teachers/professors on the religiosity of Hindu students (r =.18). Research Question 4 To what extent can agreement with mystical experience among Christian, Muslim, and Hindu college students be explained by personal (sociocultural, socioeconomic, and socioreligious) characteristics? The results of the regression analysis are rather modest. For the Christian and Hindu students, only the influence of the media on their religiosity is a significant predictor of vertical mysticism. The level of explained variance (.02) in both groups is too low to be theoretically relevant. No other personal characteristic is associated with vertical mysticism. However, for Muslim students, two predictors explain 8 percent of the variance (R 2.08; Adj. R 2.07). The influence of relatives (β.20) and of persons in the media on religiosity is significantly associated with vertical mysticism (β.16). The positive influence of relatives is the strongest predictor, suggesting the importance of primary religious socialization (i.e., in the extended family). No other characteristic (sociocultural or socioeconomic) predicts a higher level of agreement with vertical mysticism. Vertical mysticism in the case of Muslim students is influenced by religious models close to them (relatives) and in the media. SUMMARY AND DISCUSSION The comparative model of vertical mysticism or mystical union with a higher reality emerging from our analysis is significant for many reasons. In the first place, it is the outcome of a procedure of measurement invariance that has not been used before in conjunction with the Mysticism Scale. Second, as far as comparison between religious groups is concerned, the only study has been that of Hood et al. (2001), which included 188 U.S. Americans (mostly Christians) and 185 Iranian Muslims, with separate factor analyses for each group. In our research 1,920 respondents belonging to the Christian, Islamic, and Hindu traditions form a single sample and are compared substantially. Third, from the perspective of Stace s common core theory, as will be seen below, the comparative model offers new insights and prospects. In our comparative model of vertical mysticism, one of the core characteristics of introvertive mysticism, namely, sense of loss of self, combines with the complementary common characteristics of noetic quality and ineffability. Vertical mysticism has a revelatory, ineffable character and is comparable in the experience of adherents of the Christian, Islamic, and Hindu traditions. In fact, Stace s aim was to identify such comparable common core elements. However, this model seems to represent only one type of mystical experience, namely, mystical union with a higher reality or vertical mysticism. In a way, vertical movement in mysticism characterizes Semitic religions. It is alluded to, for example, in the classic of Christian mysticism, The ascent of Mount Carmel by St. John of the Cross (1542 1591), and in more humanistic terms in The ascent of Mount Ventoux by Francesco Petrarca (1304 1374), the first modern scholar and man of letters (Robinson 1898). Christian and Muslim students report a significantly higher level of vertical mystical experience than Hindu students. Socioreligious agents (relatives and persons in the media linked to their religion) influence the level of vertical mysticism among Muslims. On the other hand, Hindus involvement in vertical mysticism tends toward ambivalence. Moreover, the comparative model of mystical experience underrepresents mystical union with a

274 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION wider reality, one of the core characteristics of extrovertive consciousness (perception of oneness). This is significant as in the Hindu tradition introvertive search for the ultimate ground of one s being leads to discovery of Atman and extrovertive search for the ground of all reality leads to discovery of Brahman. The search culminates in the realization that Atman is identical with Brahman (Ayam Atma Brahma, as stated in Mandukya Upanishad, 2). Hindu mystical traditions recognize that the ground of one s being is identical with the ground of all reality. Given that this ultimate ground, Brahman, is consciousness (Prajnanam Brahma, as stated in Aitareya Upanishad, III, v, 3), in Hindu tradition mystical experience is pure consciousness. In the Vedantic tradition, Sankara (788 820), with his philosophy of advaita (nondualism), and Ramanuja (1017 1118), with his visishthadvaita (qualified nondualism), are leading exponents of the nature of consciousness. For Sankara and the advaita school, consciousness is pure light shining by itself without any subject-object distinction, whereas for Ramanuja and the bhakti (devotional) tradition consciousness is a relationship of illumination. Notwithstanding the differences, Sankara and Ramanuja assume consciousness to be the essential nature of spirit, namely, of Brahman and of the individual soul, and on this basis they interpret the nature of the phenomenal world (Chethimattam 1996:34, 96, 1971:54). In the experience of recent mystics like Ramana Maharishi (1879 1950) of Tamil Nadu and Sri Ramakrishna (1834 1886) of Bengal the difference between the two philosophical traditions is perceived to be basically one of vantage points (Easwaran 1988:30). In relation to the external phenomenal world, advaitic and visishthadvaitic consciousness, by underscoring the radical interdependence of all reality, tend to evoke a monistic or pantheistic mystical experience. As mentioned already, our comparative model of mystical experience was found to underrepresent this mystical union with a wider reality in the case of Hindus. This points to the possibility of horizontal mysticism (union of self with a wider reality) as distinct from vertical mysticism (mystical union with a higher reality), which is more compatible with the Christian and Islamic traditions. Vertical mysticism may be more dominant in the Semitic religions (Judaism, Christianity, and Islam). Such mysticism generally has a revelatory component and is not easy to express (ineffable). In the context of these religions, mysticism entails union of the self with the higher reality of God, who sometimes reveals himself and his intentions in an ineffable manner. By contrast, Oriental religions like Hinduism and Buddhism tend to emphasize horizontal mysticism in their introvertive and extrovertive search for the ultimate. It leads to union of self with a wider reality, an experience of the radical interdependence of all reality in a nondualistic sense. The final phase of such mysticism is attained in total loss of self or total realization of self in pure consciousness. Loss of self implies emergence of the ultimate Reality as pure consciousness. In the Semitic religions, there is a strong inclination to see this Reality as beyond or above the world as we experience it. This Reality is One, absolute, and the ultimate ground of the self. The truth of the self is embedded in this Reality. In analogy with a distinction made in philosophy of mind (Searle 1985), we could speak of a self-to-god direction of fit. 2 The truth of the self is decided by the fit of the self to the nature of the absolute Reality. The loss of self implies a transformation of the self according to this divine Reality. The self should mirror this Reality, which is beyond anything in the world (including the self). Vertical mysticism is characterized by a self-to-god direction of fit. Horizontal mysticism on the contrary can be seen as characterized by a God-to-self direction of 2 Searle (1985) distinguishes between a mind-to-world direction of fit, and a world-to-mind direction of fit. Examples of the first are beliefs. The truth of beliefs about the world is decided by the fit of beliefs to the state of affairs of the world. A belief is satisfied when it depicts the world. A desire is an example of a world-to-mind direction of fit. The idea is that usually a desire is not yet realized, that is, it does not refer to a state of affairs in the world. Instead, the world is transferred in such a way that it fits the desire.

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 275 fit. Horizontal mysticism is characterized by the perception of all things as one, that is, by the idea that the divine Reality is realized in all things (including the self). This implies the transformation of the divine Reality in order to fit the reality of all things. This implies not so much a loss of self, but a self that becomes One with this Reality. The perception of all things as one is the result of a process of transformation in which the divine Reality is realized in all things (including the self). By using the concept of direction of fit we suggest that the movement in vertical mysticism is different from horizontal mysticism. In vertical mysticism, the divine Reality transcends or is beyond the world as we experience it. Therefore, a loss of self is implied because the truth of the self is in God. In horizontal mysticism, the divine Reality is immanent, that is, within everything we experience. Therefore, the self should be transformed in conformity with this all pervading divine Reality, or brought to God-realization. The theoretical distinction between vertical and horizontal mysticism could be a fruitful path for further comparative research into religious mysticism and might be more fruitful than introvertive and extrovertive mysticism, as suggested by Stace. The qualifications introvert and extrovert refer more to the type of search one engages in rather than to the experience of the reality/reality as such. In other words, mystical experience resulting from introvertive and/or extrovertive search is perhaps better understood as vertical mysticism and as horizontal mysticism. This sheds light on the unresolved debate between the common core theory of Stace and the diversity theory of Katz. Is there a common structure of mystical experience for all religions (cf. the common core theory of Stace) or do religions differ in the type of mystical experience (cf. diversity theory of Katz)? If Stace is correct both vertical and horizontal types of mysticism are present within all religions, and if Katz is right religions will differ with regard to specific types of mystical experience. We find support for the diversity theory of Katz. In the process of constructing measurement invariance, we had to remove items that can be regarded as belonging to a horizontal model of mystical experience (i.e., oneness with all reality). The items did not seem to belong to the mindset of our Christian and Muslim students. In the case of Christians this may be due to fear of syncretistic fusion with monistic or pantheistic tendencies. For example, in its Letter to the bishops of the Catholic Church on some aspects of Christian meditation the Congregation for the Doctrine of Faith (1990: n. 12) warns of monistic or pantheistic syncretism in the use of Eastern methods of meditation. The problem with regard to pantheistic tendency is also present in Islamic theology (Ventura 2000). Obviously further research into horizontal mysticism is needed among adherents of Oriental religions in order to arrive at a full fledged, empirically tested common core theory or diversity theory. In conclusion, this study underscores the need for further comparative research. The four-item scale of vertical mysticism has a low level of reliability. Further research with more scale items on both dimensions (i.e., noetic quality and loss of self) is needed to improve the level of reliability of the vertical mysticism scale. Research is also needed to study changes across age groups and whether the change is the same for members of different religious groups. Theoretically, the distinction between vertical and horizontal mysticism looks promising for further theory building. Our findings also point to the limitations of Hood s Mysticism Scale (already foreseen by him) when it comes to comparing totally heterogeneous religions such as Semitic and Eastern religions (Hinduism and Buddhism). Finally, we think that the major theoretical problem with regard to mysticism still stands: Is there a core structure of mystical experience or does it imply diversity? In order to answer this fundamental question, we need more comparative research among members of different religions in different geographical contexts. REFERENCES Anthony, Francis-Vincent, Chris A. M. Hermans, and Carl Sterkens. 2005. Interpreting religious pluralism: Comparative research among Christian, Muslim and Hindu students in Tamil Nadu, India. Journal of Empirical Theology 18(2):154 86.

276 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION. 2007. Religious practice and religious socialization: Comparative research among Christian, Muslim and Hindu students in Tamil Nadu, India. Journal of Empirical Theology 20(1):100 28. Caird, Dale. 1988. The structure of Hood s mysticism scale: A factor-analysis study. Journal for the Scientific Study of Religion 27(1):122 27. Chethimattam, John B. 1971. Consciousness and reality. An Indian approach to metaphysics. Tenbury Wells: Fowler Wright.. 1996. Experience and philosophy. Bangalore: Dharmaram. Congregation for the Doctrine of Faith. 1990. Letter to the bishops of the Catholic Church on some aspects of Christian meditation. London: Catholic Truth Society. Easwaran, Eknath. 1988. The Upanishads. Translated with a general introduction. London: Arkana. Harkness, Janet A., Fons J. R. van de Vijver, and Peter Ph. Mohler. 2003. Cross-cultural survey methods (Wiley Series in Survey Methodology). Hoboken, NJ: John Wiley & Sons. Hill, Peter C. and Ralph W. Hood, Jr. 1999. Measures of religiosity. Birmingham, AL: Religious Education Press. Holm, Nils G. 1982. Mysticism and intense experiences. Journal for the Scientific Study of Religion 21(3):268 76. Hood, Ralph W. Jr. 1975. The construction and preliminary validation of a measure of reported mystical experience. Journal for the Scientific Study of Religion 14(1):29 41. Hood, Ralph W. Jr., Nima Ghorbani, Paul J. Watson, Ahad Framarz Ghramaleki, Mark N. Bing, H. Kristl Davison, Ronald J. Morris, and W. Paul Williamson. 2001. Dimensions of the mysticism scale: Confirming the three-factor structure in the United States and Iran. Journal for the Scientific Study of Religion 40(4):691 705. Hood, Ralph W. Jr., Ronald J. Morris, and Paul J. Watson. 1993. Further factor analysis of Hood s mysticism scale. Psychological Reports 73:1176 78. Hood, Ralph W. Jr., Bernard Spilka, Bruce Hunsberger, and Richard L. Gorsuch. 1996. The psychology of religion: An empirical approach. New York: Guilford. Hood, Ralph W. Jr. and W. Paul Williamson. 2000. An empirical test of the unity thesis: The structure of mystical descriptors in various faith samples. Journal of Psychology and Christianity 19(3):222 44. Jöreskog, Karl G. and Dag Sörbom. 1998. LISREL R 8 user s reference guide. Chicago, IL: Scientific Software International. Katz, Steven T. 1978. Mysticism and philosophical analysis. New York: Oxford University Press. Meredith, William. 1993. Measurement invariance, factor analysis, and factorial invariance. Psychometrika 58(4): 525 43. Reinert, Duane F. and Kenneth R. Stifler. 1993. Hood s mysticism scale revisited: A factor- analytic replication. Journal for the Scientific Study of Religion 32(4):383 88. Robinson, James Harvey. 1898. Francesco Petrarca: The first modern scholar and man of letters, pp. 307 20. New York: Putnam. Searle, John R. 1985. Expression and meaning: Studies in the theory of speech acts. Cambridge: Cambridge University Press. Stace, Walter Terence. 1961. Mysticism and philosophy. London: Macmillan. Sterkens, Carl and Francis-Vincent Anthony. 2008. A comparative study of religiocentrism among Christian, Muslim and Hindu students in Tamil Nadu, India. Journal of Empirical Theology 21(1):32 67. Van de Vijver, Fons and Kwok Leung. 1997. Methods and data analysis for cross-cultural research (Cross-cultural Psychology Series). Thousand Oaks, CA: Sage. Van de Vijver, Fons J. R., Dianne A. Van Hemert, and Ype H. Poortinga. 2008. Multilevel analysis of individuals and cultures. New York: Lawrence Erlbaum. Ventura, Alberto. 2000. Il crocifisso dell islam. Al-Hallaj, storia di un martire del IX secolo. Brescia, Italy: Morcelliana. APPENDIX 3 MEASURING INSTRUMENT 1: MYSTICAL EXPERIENCE In this part, you are invited to indicate whether you have had the following types of experiences. Your answer can be one of four: certainly no (1), probably no (2), probably yes (3), or certainly yes (4). Extrovertive mysticism 3. (8) Did you ever have an experience in which everything seemed to be alive? 3 Numbers in parentheses refer to the corresponding items in Hood s Mysticism Scale.

A COMPARATIVE STUDY OF MYSTICAL EXPERIENCE 277 6. (24) Did you ever have an experience in which your own self seemed to merge into something greater? 9. (19) Did you ever have an experience in which you felt that everything in the world was part of the same whole? 12. (12) Did you ever have an experience in which you realized the oneness of yourself with all things? Introvertive mysticism 2. (11) Did you ever have an experience in which you lost the sense of time and space? 5. (4) Did you ever have an experience in which everything seemed to disappear from your mind until you were conscious only of a void (emptiness)? 8. (32) Did you ever have an experience that cannot be expressed in words? 11. (3) Did you ever have an experience in which something greater than yourself seemed to absorb you? Religious interpretation 1. (25) Did you ever have an experience that left you with a feeling of wonder? 4. (20) Did you ever have an experience that you knew to be sacred? 7. (13) Did you ever have an experience in which a new view of reality was revealed to you? 10. (5) Did you ever experience a sense of profound joy?