STRADDLING THE SACRED AND SECULAR: CREATING A SPIRITUAL EXPERIENCE THROUGH BELLY DANCE

Abstract

Few studies examine spiritual experiences through seemingly secular activities. Who has spiritual experiences while belly dancing? How does belly dance become spiritual? Using surveys and interviews with 77 belly dancers in the United States, this study shows that belly dance is spiritual for people who consider spirituality important in their lives, have belly danced for many years, rarely attend worship services, and are not affiliated with Judeo-Christian religions. Belly dance becomes spiritual when dancers “let go” and experience various connections. The dance itself and the environment in which one dances facilitate spiritual experiences. Implications for spirituality are discussed.

I am grateful to Scott A. Desmond, Rachel L. Einwohner, Carolyn A. Kapinus, and Kat Lebo [Shuvani] for numerous conversations about various topics discussed in this article and for reading earlier drafts. I appreciate the helpful comments from two anonymous reviewers and the journal's editor. I also acknowledge Ashley Brooke Baker, Daniel R. Flowers, and Amanda Zimmerman for their very helpful research assistance. This project would not be possible without the participation of the 77 subjects who graciously took time out of their extremely busy schedules to talk with me. I very much appreciate their candor and openness. They are referenced in this article by pseudonyms. Any and all errors are my own. This research was partially supported by a New Faculty Research Grant from Ball State University. An earlier version of this article was presented at the 2007 annual meeting of the Society for the Scientific Study of Religion in Tampa Bay, Florida.

Notes

¹Much debate surrounds the proper name for this dance. It originated in the Middle East and North Africa, collectively known as “the Orient,” so it is sometimes called “Oriental dance.” In the Middle East, it is known as “Raqs Sharqi” or “Raks el Sharki,” which in English translates into “dance of the East” or “Middle Eastern dance.” “Belly dance” derives from the French term “Danse du Ventre” (abdominal dance), used to describe the performances at the 1893 Chicago World's Fair (Carlton 1994; Monty 1986). Because it is the term most familiar to the American public, the dance discussed in this article includes American adaptations (American Tribal Style, American Cabaret, and tribal fusion), and this article focuses on American dancers, I use the term “belly dance.”

²Crosby (2000) distributed a written survey at a dance seminar in Dallas and through the Med-dance listserve, resulting in a sample of 45 respondents. This a 12% response rate, which is larger than the less than 10% response rate for another comparable survey (Sherezzah 1998). It is difficult to estimate the response rate for this study because of the diverse recruiting methods and multiple waves used. Most people who expressed interest in the study participated. I am satisfied the sample represents a range of people with a variety of belly dance experiences and demographic characteristics, with the possible exception of religious affiliation, which is indicative of the region from which the sample was drawn.

³Very few respondents disagreed with the survey statement. Several selected the middle category of “neutral.” When asked in the interview to explain their answer, almost all of the people who selected “neutral” stated that they did not think belly dance was spiritual; therefore, creating a dichotomous “yes” or “no” variable from both the survey and interviews is a more accurate representation of the data.

^a Because religious affiliation and the dependent variable yield a 2 × 2 contingency table, a Fisher's Exact Test is a more appropriate test of significance than a chi-square test to analyze these two variables. The Fisher's Exact Test tests the probability of the observed values occurring by chance. The interpretation of the p value is comparable to interpreting the p value of a Chi-square test. If the p value is .05, in 95% of the cases, the observed values result from a strong relationship between the two variables and significantly differ from what they would be if they occurred by chance. The Fisher's Exact Test assumes a one-tailed version of the test (Siegel 1956). Although SPSS calculates a 2-tailed test, the one-tailed results are reported.

⁴There are two limitations to using a chi-square analysis for these data. First, chi-square assumes a probability sample, which was not used in this study. Second, due to the small sample size, one cell in both variables “years belly dancing” and “service attendance,” and two cells in the variable “spirituality in life,” have an expected value of less than five. An expected value less than five means the difference between the expected and observed frequencies in that cell may artificially raise the chi-square value and falsely suggest a significant relationship (Siegel 1956). However, these limitations are not overly problematic. Cochran (1954) suggests the chi-square test should not be used if more than 20% of the expected frequencies are smaller than five or if any expected frequency is smaller than one. In this study, even though 33% of the expected frequencies for the importance of spirituality in one's life are below five, no expected frequency falls below one for this or any other variable. Only 16% of the expected frequencies for the other two variables are smaller than five. Also, the qualitative data show the same general relationships that are suggested by the chi-square analysis. Future research may wish to conduct similar statistical analyses on a representative and larger sample than the one used in this study.

⁵I defined the religious affiliation variable in several ways. I first used all religions represented in the sample (Protestant, Catholic, Jewish, Hindu, Pagan, Other, None) but this yielded too many categories with very small numbers. I also collapsed the categories into the following combinations: Protestant vs. everyone else, Christian vs. other religion vs. no religious affiliation, and Christians vs. everyone else. Christians vs. everyone else and Protestant vs, everyone else yielded a Fisher's Exact Test one-tailed p value of .06. Christian vs. other vs. no religion yielded a p value of .1. These results, along with the qualitative data, suggest a general pattern that those affiliated with Judeo-Christian religions tend not to consider belly dance spiritual, even though the quantitative results hover around the accepted cut-off for statistical significance (p < .05).