Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

Transcription

1 Religious Orthodoxy in American Society: The Myth of a Monolithic Camp Author(s): Nancy J. Davis and Robert V. Robinson Source: Journal for the Scientific Study of Religion, Vol. 35, No. 3 (Sep., 1996), pp Published by: Blackwell Publishing on behalf of Society for the Scientific Study of Religion Stable URL: Accessed: 15/02/ :44 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org. Society for the Scientific Study of Religion and Blackwell Publishing are collaborating with JSTOR to digitize, preserve and extend access to Journal for the Scientific Study of Religion.

2 Religious Orthodoxy in American Society: The Myth of a Monolithic Camp NANCY J. DAVISt ROBERT V. ROBINSONt Through an analysis of national survey data, we test several hypotheses regarding the extent to which the religiously orthodox constitute a monolithic conservative front. We find that the orthodox generally are conservative on social issues of sexuality, reproduction, and schooling of children, but are moderate or liberal on gender, racial, and economic issues. There is little tendency for the orthodox to cohere around a single ideological position or, as individuals, to take stances that are consistent across a broad range of issues. The religiously orthodox are also internally divided on many social and economic issues along the lines of race, sex, class, and age. We suggest that these factors may explain the weak support that the religiously orthodox historically have given to organized political movements of the Religious Right. With vivid imagery of a "culture war," the mass media, political and religious leaders, and academics have portrayed American society as divided into hostile "camps" of religious conservatives and liberals. Each of these camps is depicted as a united front locked in battle with the other side over fundamental social concerns. In this paper we examine this imagery of monolithic camps at war through an analysis of the belief structure of one of the two sides - the religiously conservative, or orthodox. In a related paper, we have shown that the greater conservatism of this group relative to theological progressives on many issues, including sex education, abortion, birth control, nonmarital sex, homosexuality, pornography, school prayer and gender roles, does not extend to racial or economic issues (Davis and Robinson forthcoming). Here we examine in depth the orthodox pole of the religious spectrum in order to assess whether the religiously orthodox constitute the monolithic conservative camp that they have been portrayed as in the media, by leaders of quasireligious groups themselves, and, to some extent, in recent social science scholarship. Specifically, we test whether the religiously orthodox (1) hold views across a wide spectrum of issues which are conservative, inegalitarian, and opposed to government intervention in the economy, (2) exhibit the high degree of consensus on specific issues that the imagery of a moral camp implies; (3) hold internally consistent views on a range of sexuality/reproduction, gender, racial, and economic issues, and (4) are united in their stances on these issues across gender, racial and class categories. f Nancy J. Dauis is professor and chair of the Sociology and Anthropology Department at DePauw University, Greencastle, IN f Robert V. Robinson is professor of Sociology and director of the Institulte of Social Research at Indiana University, Bloomington, IN ? IJottrnal fir the Scientific Study of Religion, 1996, 35 (3):

3 230 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION THEORETICAL BACKGROUND In a seminal book published in the late 1980s, The Restructuring of American Religion, Robert Wuthnow (1988) argued that the symbolic boundaries in American religion had changed in the period since World War II. Denominational antagonism had declined only to be replaced by an internecine struggle, rooted largely in educational differences, between religious conservatives and religious liberals (see also Glazer 1987). American religion had split into two camps engaged in "symbolic warfare" (1988: 138). Each side comprised more than two fifths of the American population, leaving less than one fifth of Americans in the middle (Wuthnow 1988: 133). According to Wuthnow (1988: 114), the polarization between religious conservatives and liberals was both reflected in and further exacerbated by the growth of special purpose groups, many of them aimed at limiting or promoting government action on welfare, education, and equal rights laws. The "great divide" in American religion coincided with political divisions in the larger society between conservatives and liberals, with theological conservatives taking more conservative political stances than theological liberals on a wide range of social, political, and economic issues (1988: 132, , 239). Wuthnow (1988: 239) argued that the "political uniformity" within each side across a wide range of issues contributed to the public impression of "deep polarization between two monolithic camps," imagery drawn even more sharply by the media. James Davison Hunter (1991), in his influential book, Culture Wars, built on some of Wuthnow's arguments about the erosion of denominational antagonisms and the growth of para-church, special interest groups, but postulated a more intractable struggle between the two sides. Hunter (1991: 49) argued that the hostilities are so difficult to resolve because they are rooted in "fundamentally different conceptions of moral authority." On one side, the orthodox believe that God is the ultimate arbiter of what is right and wrong, that the revealed word of God as recorded in sacred texts is inerrant and of timeless relevancy, and that God exercises a real and active presence in everyday life. In the orthodox view, moral standards are absolute, unchanging, and universal; moral authority is external and transcendent. On the other side, progressives, who include religious modernists as well as nonbelievers and secularists, assert that humans are the ultimate arbiters of what constitutes moral action, that the morality of action is publicly contested and judgeable only in its cultural context, and that humans are responsible for their own fates (Hunter 1991: 44-45). For Hunter (1991: 42, 115) the struggles over a wide range of issues from abortion to affirmative action to what is to be taught in public schools are ultimately rooted in the conflict between orthodox and progressive moral visions. While the association of religious orthodoxy with political conservatism and of theological progressivism with political liberalism is "far from absolute," the "relationship between foundational moral commitments and social and political agendas is too strong and consistent to be viewed as coincidental" (1991: 46). Hunter (1991: 46 ) asserts that the orthodox are inclined to take conservative positions on most of the issues of the day, while progressives take liberal stances, a tendency that is "true for most Americans," but is even stronger among the leaders, spokespersons, and organizations representing the two camps. While Hunter noted that the majority of Americans "occupy a vast middle ground between the polarizing impulses" (1991: 43), he devoted the bulk of this book to describing the "culture war" between orthodox and progressive camps of Americans. Wuthnow and Hunter differed in what they viewed as the root of the religious divide (Wuthnow saw this as education-based, while Hunter viewed this as rooted in fundamentally opposed moral cosmologies), in the size of the opposing sides relative to the middle (Wuthnow assumed a small middle group while Hunter postulated a large one), and

4 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 231 in the degree of conflict they saw between the two camps (Hunter postulated more conflict, as evidenced in his more frequent use of the imagery of war). Yet they shared the assumptions that (1) there is a divide in American religion between religious conservatives and liberals and (2) this theological divide tends to coincide with divisions between political conservatives and liberals on a wide range of issues. Both Wuthnow and Hunter noted that public opinion is complex and that the religious division only "tends to" align with political divisions. Hunter (1991: 43), for example, observed that "though competing moral visions are at the heart of today's culture war, these do not always take form in coherent, clearly articulated, sharply differentiated world views." Nonetheless, the thrust of both of both Wuthnow's and Hunter's arguments and the language they used to describe the religious division imply that each of the two sides is comprised of like-minded individuals. Wuthnow (1988: 133, 256) depicted the antagonists in the religious divide as "almost separate religious communities" that have spawned civil religions that "frequently appear to be at fundamental adds with one another." Wuthnow (1988: 132, 207, 215, 239; 1989: 21-22) described the religious division as "the great divide," a "chasm," and a "ravine," sometimes used war metaphors to describe the conflict ("waging symbolic warfare," "battle,""captains of an impressive religious army"), and interpreted public opinion surveys as indicating "political uniformity" within each side. Hunter, as the title of his book conveys, used rhetoric that depicts even greater polarization and hostility between the two sides, each of which "operates within its own constellation of values, interests and assumptions " and is "worlds apart" from the other (1991: 128). He stated that his use of "the language of confrontation, battle, even war... is not merely a literary device but an apt tool to describe the way in which the many issues contested in American public discourse are being settled" (1991: 64). That, for Hunter (1991: 42, 49), the struggles over social, political, and economic issues share a common root in "fundamentally different" moral cosmologies also points to a degree of commonality and consistency of opinion within each side. "Camp"- one of the labels applied by Wuthnow (e.g., 1988: 133, 256) and Hunter (e.g., 1991: 106, 128) to the opponents in the culture conflict - is defined as "a group or body of persons acting in unanimity; esp. a group engaged in promoting or defending a given theory or doctrine" (Webster's Third New International Dictionary, unabridged 1981: 322). Finally, neither Wuthnow nor Hunter devotes much attention to whether either group is divided in its political stance along the lines of race, class, or gender.1 Wuthnow's and Hunter's arguments suggest a degree of uniformity or consensus within each of the sides of the religious division. The imagery of monolithic camps is most sharply drawn by the leaders of para-church groups, by politicians seeking votes from these "blocs," and by the mass media, as both of these authors noted (Wuthnow 1988: 239, ; Hunter 1991: 59, 96ff). That the national leaders of orthodox and progressive groups have themselves presented broad political agendas as gestalts encompassing a wide range of issues (e.g., the Moral Majority's "Agenda for the 1980s" and, more recently, the Christian Coalition's "Contract with the American Family") might lead one to expect uniformly conservative or liberal positions on each side (Pohli 1983: 532; Jelen 1990: 118). There is, moreover, no question that the mass media have fanned the flames of conflict between religious conservatives and liberals and have portrayed the former group, especially, as a monolithic voting bloc (e.g., the "Christian Right"). Yet there are several reasons to believe that this imagery of monolithic camps is overdrawn. We focus on the religiously orthodox because they have thus far had greater success than theological progressives in mounting a political movement and a national presidential campaign and because they have received greater attention from the media and scholars of religion. Ethnographic studies of Protestant Fundamentalist, Catholic Pentecostal, and Orthodox and Hasidic Jewish congregations have found considerable diversity, complexity, and incorporation of secular ideologies in the attitudes of religious conservatives (e.g.,

5 232 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION McGuire 1982; Pohli 1983; Ammerman 1987; Neitz 1987; Stacey and Gerard 1990; Davidman 1991). For instance, while ethnographies of the religiously orthodox have found many congregants who hold conservative views on the appropriate roles for women and men in the home and workplace, they also identify a significant minority who take nontraditional stances or who, for economic reasons, are unable to live out roles they see as divinely ordained (Pohli 1983; Ammerman 1987; Rose 1987; Stacey and Gerard 1990). In research based on national data for the 1972, 1980, and 1984 elections, Kiecolt and Nelsen (1988) found that conservative Protestants were actually less consistent in their beliefs and had less consensus than liberal Protestants. Analyzing national data for 1988, Jelen (1990) found little attitudinal consistency on a broad range of issues among Evangelical Protestants. One reason for this may be that Scripture does not provide an unambiguous basis for taking positions on contemporary issues, including those relating to gender, race, and economic policy. That, historically, orthodox Christianity has been associated with pro-slavery and abolition movements, with racial segregation and civil rights, with John Birch conservatism and Huey Long populism, and with anti-/era activism and Evangelical feminism suggests that a wide range of views on gender, race, and class can find support in biblical texts (Aptheker 1970; Cox 1987; Tamney, Burton and Johnson 1989; Billings 1990). Thus we expect considerable disagreement within the religiously conservative or orthodox on specific social and economic issues. On many of the concerns over which Wuthnow (1988) and Hunter (1991) see American society as divided, but especially those related to gender, racial, and economic equality, some groups are likely to benefit from conservative positions on these issues while others are hurt. For religious conservatives to advance a united front on these issues requires that the self-interest of those disadvantaged as a result of conservative policies on gender, race, and class be superceded by religious authority. Yet, as many biblical scholars have shown (e.g., Cone 1969; Gutierrez 1973), Scripture can provide a basis for rejection of conservative principles for those who feel disadvantaged by them. Thus we expect significant differences within the religiously orthodox between men and women on gender issues, between whites and people of color on racial issues, and between rich and poor on economic issues. The selfinterest on specific issues and the absence of a single scriptural line on most issues should produce considerable inconsistency of opinion across issues among religiously orthodox individuals (see also Warner 1988:297; Hart 1992). Because the religiously orthodox or conservative draw disproportionately on the disadvantaged - racial and ethnic minorities, the working class and poor, and women (e.g., Woodrum 1988:569), we question whether the attitudes of this group, as a whole, are uniformly conservative, inegalitarian, and opposed to government intervention to reduce inequalities of various forms. Wuthnow (1988:114) argues that one reason that the Religious Right first mobilized was to resist what they saw as increasing government encroachment into all areas of life. Yet government may be seen by disadvantaged groups as the primary source for redress of gender, racial, and economic injustice. Given the demographic composition of the religiously orthodox, we expect their views on gender, racial, and economic issues to be more liberal than Wuthnow's and Hunter's discussions suggest. Thus we hypothesize, based on the ambiguity of Scriptural texts, the conflicting selfinterest of the diverse groups comprising the religiously orthodox, and the tendency for the orthodox to draw adherents disproportionately from groups that would not benefit from conservative or inegalitarian government policy: H1 The religiously orthodox are not uniformly conservative on gender, racial, and economic issues. H2 The religiously orthodox exhibit little consensus of opinion on specific issues. H3 The religiously orthodox, as individuals, hold inconsistent views on sexuality/reproduction, gender, racial, and economic issues.

6 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 233 H4 The religiously orthodox are divided by sex, race, and class in their attitudes. Specifically, men take more conservative positions than women on gender relations; whites take more conservative positions than people of color on racial issues; and middle-class and affluent people take more conservative positions than the working class and poor on economic issues. DATA AND METHODS Data In testing these hypotheses we analyze data from the NORC's General Social Survey for This is a national sample of the noninstitutionalized population of the United States, 18 years of age or older. The 1991 GSS had 1517 respondents, of which 1359 or 89.6% also completed a questionnaire for the International Social Survey Program module on religion; 158 individuals did not complete the ISSP questionnaire. Our sample consists of the respondents to the ISSP module. Logistic regression analyses of the social characteristics (sex, race/ethnicity, occupational prestige, income, business ownership, education, region, and religious denomination) of those who completed versus did not complete the ISSP questionnaire showed that African Americans, Indians (native Americans) and residents of the South are slightly overrepresented in the ISSP subsample of the GSS.2 As suggested by the ISSP, we do not weight the ISSP subsample. Measures Religious orthodoxy versus progressivism is measured by a three-item index. Respondents were asked: (a) "Which one of these statements comes closest to describing your feelings about the Bible?" (1) The Bible is an ancient book of fables, legends, history and moral teachings recorded by man. (3) The Bible is the inspired word of God but not everything should be taken literally, word for word. (5) The Bible is the actual word of God and it is to be taken literally, word for word. (b) "How much do you agree or disagree with each of the following: The course of our lives is decided by God"? (1) Strongly disagree. (2) Disagree. (3) Neither agree nor disagree. (4) Agree. (5) Strongly agree. (c) "How much do you agree or disagree with each of the following statements: Right and wrong should be based on God's laws"? (1) Strongly disagree. (2) Disagree. (3) Neither agree nor disagree. (4) Agree. (5) Strongly agree. The three items are summed to form an index that ranges from 3 to 15. (The first item is scored from 1 to 5 so that it will carry the same weight as the other two items). High scores indicate religious orthodoxy or conservatism. Factor analysis of these items yielded a single factor with an eigenvalue of Cronbach's alpha, a measure of reliability, is.68 for the three items. All analyses are conducted on individuals whom we classified as religiously orthodox on the basis of their scores on this index. Since the scale is a continuum, there is no clear boundary separating the religiously orthodox from other respondents. To ensure that we analyze only those respondents who conform most closely to the beliefs underlying religious orthodoxy - not the ambivalent, "muddled middle" (Hunter 1994:106) that one would

7 234 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION expect to be positioned between the extremes - we include in our sample of religious conservatives only those who were further than one standard deviation (2.979) above the mean (9.967), that is, those who scored 13 or higher. This yields a sample of 297 respondents or 21.8% of the ISSP sample.3 Hunter (1991: 159) estimated that approximately 20% of Americans hold strongly to each of the two moral visions, while Wuthnow (1988:133) classified 41% of a sample of Americans as "religious conservatives" (with 18% "very conservative"). Since the highest score among the 30 Jewish respondents in the ISSP sample is 12, there are no Jewish respondents in the orthodox subsample.4 Our index of religious orthodoxy more closely measures Hunter's conceptualization than that of Wuthnow. In his analyses of survey data, Wuthnow (1988:343n) uses a measure of religious conservatism that allows individuals to place themselves on a six-point scale of conservatism/liberalism.5 We prefer our index for several reasons. First, there is nothing in Wuthnow's discussion of religious conservatism and liberalism to suggest that this must be self-assessed. Our measure is in accord with the general spirit of Wuthnow's discussion,6 even though we choose to operationalize religious conservatism/liberalism differently. Second, we prefer a measure which classifies individuals based on their beliefs rather than what they perceive themselves to be because the former places individuals along a continuum in relation to other individuals in the sample while the latter places each person on a scale that is unique to that individual. Third, and most importantly, we prefer a measure that does not use the terms "liberal" and "conservative," which for most Americans are probably more associated with political labels than they are with religious identities, and which may thus inflate the correlation between religious conservatism or liberalism and political conservatism or liberalism. Independent variables are operationalized as follows: Age is measured in years. Race/ethnicity is measured by a dummy variable series consisting of African American, Latino, and other minority, with "non-latino" white as the omitted reference category. Sex is a dummy variable labeled Female, with male as the reference category. Denomination is measured by a dummy variable series consisting of Catholic and none/other, with Protestant as the reference category. Education is measured in years of schooling. Income refers to family income before taxes in 21 categories, with midpoints representing each. Occupational prestige is a scale developed for NORC; individuals without an occupation are coded to the mean for employed persons, so that in regression analyses this variable indicates effects among employed persons only. Ownership is a dummy variable indicating whether the respondent or spouse owned a business versus both worked for someone else (reference category). Rural is a 10-category ordinal variable ranging from large central city (over 250,000 people) to open country. South is a dummy variable indicating residence in the South versus all other regions (reference category).7 Dependent variables are described briefly in the text below; wordings of specific questions are given in Davis and Smith (1994). All variables are coded so that the high pole represents the conservative position. Information on many of the dependent variables was collected from randomly selected subsamples of two thirds or one half of the GSS respondents. Because these subsamples overlap and differ from item to item, it is not possible to form indexes of similar questions on social, gender, racial, and economic issues. RESULTS Wuthnow (1988:132, , 239) and Hunter (1991:46, 96, 115, 263-4) argued that the religiously orthodox/conservative tend to take politically conservative positions on a broad spectrum of issues (see also Klatch 1992). Wuthnow maintained that the political involvement of religious conservatives grew partly in reaction against the expansion of

8 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 235 government intervention in "welfare, education, equal rights legislation and other kinds of regulation" (1988:114; see also Hunter 1991: ). We hypothesized under H1, however, that because the Bible contains multiple messages and layers of meaning and because the religiously orthodox draw disproportionately on the disadvantaged of society, they may not hold conservative, anti-statist positions on gender, racial, and economic issues. Table 1 shows the distribution of religiously conservative respondents on 24 issues, including school prayer, abortion, nonmarital sex, birth control, pornography, gender relations, race, and economic inequality. Responses have been classified into three categories - liberal, moderate (including "don't known"), and conservative by grouping together weak and strong responses (e.g., "somewhat agree" and "strongly agree"). As Wuthnow and Hunter would expect, the bulk of orthodox/conservative religionists take conservative positions on school prayer, abortion, premarital and extramarital sex, homosexuality, and pornography.8 Yet, surprisingly, three quarters of the religiously orthodox favor sex education in public schools, and they are evenly split on whether the birth control pill should (47%) or should not (49%) be made available to teenagers Even on the question of abortion on demand, nearly one quarter (24%) take the liberal position. TABLE 1 DISTRIBUTION OF ATTITUDES ON CULTURAL, GENDER, RACIAL AND ECONOMIC ISSUES, FOR THE RELIGIOUSLY ORTHODOX Percentage STANDARD Issue LIBERAL MODERATE CONSERVATIVE TOTAL DEVIATION N School prayer in public schools Sex education in public schools Pill to teens Abortion on demand Premarital sex Extramarital sex Homosexuality Pornography law Woman's work hurts family Husband=work, wife=home Home vs. country Women in politics Woman President Government spending on blacks Government help blacks Black President School busing Racial intermarriage Government reduce income gap Government provide jobs for all Government pay medical care Government help the poor Spending on welfare Spending on Social Security Weak and strong responses (e.g., strongly agree and somewhat agree) were combined to compute the percentages in the liberal or conservative categories. Middle responses and "don't know" were combined to form the moderate category. The orthodox have more varied feelings on issues of gender than on the sexuality/reproduction issues above. Just under one half of the orthodox sample believe that family life suffers when a woman has a full-time job and that the husband should work and the wife should look after the home and family, but the other half take moderate or liberal

9 236 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION positions on these two items. Thirty percent of religious conservatives feel that women should run their own homes and leave running the country to men; 35% agree that men are better suited than women to politics, and 14% would not vote for a qualified woman of their party for President. Yet this leaves two thirds or more of the orthodox taking liberal or moderate positions on these gender issues. Nor are religious conservatives uniformly conservative on questions of racial and economic inequality. On only one issue - the busing of school children - is there a clear conservative majority (59%), and even this issue may reflect belief in parental or neighborhood control over schools as much as attitudes on race. Conservative religionists lean toward the conservative pole on the questions of welfare spending and special treatment by the government to improve the condition of blacks. Yet, on the remaining eight questions concerning racial and economic justice, the orthodox tend to take liberal positions. In general, the stances taken by orthodox religionists on gender, racial, and economic inequality are not those that Wuthnow's and Hunter's arguments would lead us to expect. At the least, the orthodox are far less opposed to government efforts to combat racial inequality, provide jobs and health care for all citizens, and reduce the gap between rich and poor than the leadership of organized para-church groups such as the Moral Majority and the Christian Coalition appear to be. In a related paper, we found that the religiously orthodox were relatively no more conservative than theological progressives on questions of racial and economic justice (Davis and Robinson forthcoming). We also found that the religiously orthodox draw disproportionately on African Americans, Latinos, people with low incomes and less prestigious jobs, and women, which may help to explain the liberal/moderate stance of many religious conservatives on questions of gender, racial, and economic justice. Consistent with H2, the standard deviations of the items in Table 1 suggest considerable disagreement among orthodox religionists on specific issues. For three-category items, the standard deviation has a lower limit of 0.0 (unanimity) and approaches 1.0 (maximum disagreement) as the sample size increases.9 The standard deviations in Table 1 are skewed toward the high end of the possible range, that is, toward disagreement. Moderate standard deviations (.4 to.6) are found on some questions - sex outside marriage, homosexuality, pornography, and spending on Social Security. Yet there is nearly as much disagreement as possible (>.9) on many issues, including women's role in the home and in politics, whether birth control should be made available to teenagers, whether premarital sex is wrong, whether school children should be bussed, and whether government should reduce the gap between rich and poor and provide jobs to those who want them. The standard deviations in Table 1 do not indicate the political uniformity or consensus of opinion that Wuthnow's and Hunter's use of such terms as "sides,""factions" and "camps" implies. We hypothesized (in H3) that religiously conservative individuals will hold inconsistent beliefs from one specific issue to the next. Kiecolt and Nelson (1988: 51) define political belief system consistency as "the extent to which individuals consistently take conservative or liberal positions on issues." They measure this by the correlation of selfassessed political liberalism-conservatism with attitudes toward social issues. In Table 2 we show the correlations of a similar scale (seven points ranging from "extremely liberal" to "extremely conservative") with the 24 issues considered above. The correlations indicate the extent to which individuals hold positions on these issues that are consistent with their own perceived liberalism-conservatism. Of the 24 correlations, only five suggest such consistency - those for opposition to sex education in public schools, opposition to making birth control available to teens age if their parents do not approve, belief that sexual relations before marriage are always wrong, and beliefs that too much money is being spent on

10 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 237 welfare and on efforts to improve the condition of blacks. The average correlation of attitudes on specific issues with political liberalism-conservatism is only.10. TABLE 2 ZERO-ORDER CORRELATIONS BETWEEN ISSUES AND LIBERAL/CONSERVATIVE ORIENTATION AND AVERAGE CORRELATION OF EACH ISSUE WITH EVERY OTHER ISSUE, FOR THE RELIGIOUSLY ORTHODOX CORRELATION WITH MEAN CORRELATION WITH Issue LIBERALISM/CONSERVATISM EVERY OTHER ISSUE School prayer in public schools.134 (195).056 (162) Sex education in public schools.192* (197).134 (164) Pill to teens.249* (197).135 (164) Abortion on demand.051 (199).054 (138) Premarital sex.148* (297).141 (224) Extramarital sex.070 (297).078 (224) Homosexuality.109 (297).055 (224) Pornography law.123 (198).124 (146) Woman's work hurts family.060 (297).159 (224) Husband=work, wife=home.108 (297).152 (224) Home vs. country.133 (197).131 (164) Women in politics.012 (197).154 (164) Woman President (197).104 (164) Government spending on blacks.162* (154).174 (124) Government help blacks.116 (198).144 (146) Black President.016 (197).088 (164) School busing.069 (197).062 (164) Racial intermarriage.061 (199).118 (138) Government reduce income gap.115 (297).048 (224) Government provide jobs for all.108 (297).068 (224) Government pay medical care.063 (198).127 (146) Government help the poor.112 (198).180 (146) Spending on welfare.181* (154).126 (124) Spending on Social Security.033 (297).031 (224) *p <.05. Number of cases on which correlation with liberalism/conservatism is computed is given in parentheses (col. 2). The mean number of cases on which correlations of each item with every other item are based is given in parentheses (col. 4). Significance tests cannot be computed for the mean correlation of each item with every other item. Our measure of self-assessed political liberalism-conservatism suffers from the same problem that we identified with Wuthnow's self-assessed measure of religious liberalismconservatism, namely, that subjective measures place individuals along a scale that is unique to each individual. Thus we report results for a second indicator of belief consistency - the average correlation of each attitude with the 23 other attitudes (Table 2). These correlations range from.03 to.18 and average only.11. This modest average correlation is all the more remarkable because the 24 items include multiple indicator s of similar attitudes (e.g., those related to sexuality, gender roles, racial inequality, and economic inequality) which should inflate inter-item correlations. Regardless of which indicator of attitude consistency is used, the orthodox exhibit little ideological consistency across a broad spectrum of attitudes. Another sense in which we hypothesized (in H4) that the religiously orthodox would not constitute a united, monolithic, conservative front is that the self-interest of sexes, races, and classes creates internal divisions, leading some of the religiously conservative to take politically conservative positions and others to take liberal positions. In Tables 3 to 6, we regress sexuality, gender, racial, and economic attitudes on the social characteristics of the religiously orthodox.10 We did not hypothesize any effect of self-interest on attitudes

11 TABLE 3 REGRESSION COEFFICIENTS FOR THE RELATIONSHIP BETWEEN ATTITUDES TOWARD S AND REPRODUCTIVE RIGHTS AND SOCIAL CHARACTERISTICS, FOR THE RELIGIO Independent School Sex Pill to Abortion Premarital Extr Variable Prayer Education Teens on Demand Sex Catholic None/other Female (.078) (.192).019 (.085).037 (.207) (.230).296 (.563) (.088) (.215).187* (.199) (.488).498* African American (.057) (.062) (.168) (.064) (.146) * Latino Other minority Occupational prestige Income (in $1000s) Ownership (.075) -.379* (.110).137 (.136).001 (.003).002 (.002) (.081).023 (.119).328* (.147) (.003).005* (.002).019 (.219) (.324).016 (.398) (.008).013* (.005).152 (.084).053 (.124) (.152) (.003) (.002) (.190).052 (.281).356 (.345).005 (.007).003 (.004).256 Education (.099) (.107) (.289) (.111) (.251) Age Rural (.011).000 (.002) (.011).005* (.002).043* (.031).005 (.005).115* (.012) (.002) (.027).012* (.004).094 South Constant (.020) (.055) (.021) (.060) (.058) (.162) (.022) (.062) (.050) (.140) R N * p <.05. Standard errors in parentheses. The reference category is Protestant white males who do not own a business and are living outs

12 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 239 toward schooling, sexuality, and reproduction. Nonetheless, we find that women are more opposed than men to abortion, premarital and extramarital sex, and pornography (see Table 3), perhaps because sex may be more of a resource for women, and, as such, women may have more of an interest than men in restricting sex to marriage and increasing the cost of non marital sex by limiting abortion. African Americans are more liberal than whites on abortion and sex outside marriage, and less opposed to pornography, while Latinos and other minorities are more opposed than whites to pornography. Poor people are more favorable than those with higher incomes to sex education in public schools and to making birth control available to teens, perhaps because teenage pregnancy is higher among the poor and its consequences are experienced more directly. As expected, women are significantly more likely than men to disagree that family life suffers if a women takes a full-time job, that husbands should work and wives should look after the home, and that women should take care of running their homes and leave running the country to men (see Table 4). Women are also more liberal than men in support for government efforts to improve the living standards of blacks and to reduce income inequality and in feeling that too little is being spent on welfare (see Tables 5 and 6). Since women's wages are on average lower than men's and since welfare programs are directed at TABLE 4 REGRESSION COEFFICIENTS FOR THE RELATIONSHIP BETWEEN GENDER ATTITUDES AND SOCIAL CHARACTERISTICS, FOR THE RELIGIOUSLY ORTHODOX Independent Women's Work Husband=Work, Home vs. Women in Woman Variable Hurts Family Wife=Home Country Politics President Catholic (.210) (.212) (.091) (.098) (.074) None/other (.515) (.520) (.223) (.239) (.182) Female -.464* -.748* -.184* (.153) (.155) (.066) (.071) (.054) African American (.200) (.202) (.087) (.093) (.071) Latino (.296) (.299) (.128) (.138) (.105) Other minority *.379* (.364) (.368) (.157) (.169) (.129) Occupational prestige (.007) (.007) (.003) (.003) (.003) Income (in $1000s) (.004) (.004) (.002) (.002) (.002) Ownership (.265) (.267) (.114) (.123) (.094) Education -.084* -.095* -.036* -.033* (.028) (.028) (.012) (.013) (.010) Age.011*.016*.007* (.004) (.004) (.002) (.002) (.002) Rural (.053) (.054) (.023) (.025) (.019) South (.047) (.149) (.064) (.069) (.052) Constant R N *p <.05. Standard errors in parentheses. The reference category is Protestant white males who do not own a business and are living outside the South.

13 240 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION single mothers and their children, self-interest may figure in these attitudes. Women's more liberal attitudes toward government support for disadvantaged groups may also reflect a generalized identification with underdogs, a result of their own disadvantaged position. African Americans (and occasionally Latinos) are significantly more likely than whites to support government efforts to reduce racial and economic inequality (see Tables 5 and 6). The results are much less conclusive with regard to whether poor and working-class respondents are more liberal than those who are better off on economic issues. Those in high prestige jobs are actually more likely than those in less prestigious occupations to favor government efforts to provide jobs for all. The disinterest of those in less prestigious occupations in government jobs provision may reflect the sense that the jobs created would be dead-end ones paying low wages. Consistent with our expectation, however, people with low incomes are more likely than those with higher incomes to favor government efforts to help the poor and to support increased welfare spending. The well-educated are less likely than less educated people to support government efforts to equalize incomes, help the poor, and provide welfare.11 TABLE 5 REGRESSION COEFFICIENTS FOR THE RELATIONSHIP BETWEEN RACIAL ATTITUDES AND SOCIAL CHARACTERISTICS, FOR THE RELIGIOUSLY ORTHODOX Independent Gov't Spending Government Black School Law Against Variable on Blacks Help Blacks President Busing Intermarriage Catholic (.159) (.243) (.073) (.099) (.083) None/other (.389) (.596) (.178) (.242) (.204) Female * (.116) (.178) (.053) (.072) (.061) African American -.790* * -.187* -.203* -.297* (.151) (.232) (.069) (.094) (.079) Latino * * (.224) (.343) (.102) (.139) (.117) Other minority (.275) (.422) (.126) (.171) (.144) Occupational prestige *.002 (.005) (.008) (.003) (.003) (.003) Income (in $1000s) * (.003) (.005) (.002) (.002) (.002) Ownership (.200) (.306) (.091) (.125) (.104) Education * (.021) (.033) (.010) (.013) (.011) Age * *.000 (.003) (.005) (.001) (.002) (.002) Rural (.040) (.061) (.018) (.025) (.021) South (.111) (.171) (.051) (.070) (.059) Constant R N *p <.05. Standard errors in parentheses. The reference category is Protestant white males who do not own a business and are living outside the South.

14 TABLE 6 REGRESSION COEFFICIENTS FOR THE RELATIONSHIP BET1WEEN ECONOMI AND SOCIALCHARACTERISTICS, FOR THE RELIGIOUSLY ORTHOD Independent Gov't Reduce Gov't Provide Gov't Pay Gov't Help Sp Variable Income Gap Jobs for All Medical Care the Poor Catholic None/other Female African American Latino (.178) (.436) -.342* (.130) -.395* (.170) (.183).286 (.448) (.134) -.798* (.175) (.261) (.640) (.191) -.608* (.249).188 (.245).188 (.600) (.179) -.828* (.234).078 Other minority (.251) (.258) (.369) (.346) Occupational prestige (.308) (.317) -.014* (.453).005 (.425) Income (in $1000s) (.006) (.006) (.009) (.008) * Ownership (.004).063 (.004) (.006) (.005) Education (.224) (.231) (.329) (.309).051*.066* * Age Rural South (.023) (.004).031 (.045) -.344* (.125) (.025).004 (.004).016 (.046) (.129) (.035).005 (.005) (.066).171 (.184) (.033).009 (.005).060 (.062).319 (.172) Constant R N * p <.05. Standard erlrors in parentheses. The reference category is Protestant white males who do not own a business and arle living ou

15 242 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION In addition to gender, race, and class differences in the attitudes of the religiously orthodox, there is also some tendency for younger people to hold more liberal views than older people on sexuality/reproduction, gender, and racial issues (the one exception is with regard to government efforts to help blacks). This may represent a conservatizing effect of age or it may be a harbinger of greater liberalism among the religiously orthodox as more liberal younger cohorts replace older ones. In general, self-interest appears to affect the positions of groups on nearly all of the issues on which any political movement of the religiously orthodox might take a stand. Yet the modest percentages explained in the variance in attitudes on these issues (see the R2s in Tables 3 to 6) suggest that much of the difference of opinion among the religiously orthodox is not structured along the lines of variables included in these models, but may reflect different scriptural interpretations or other characteristics of individuals not included in the regression models (see, e.g., Cox 1995). Denomination has no effect on attitudes toward any of the 24 issues considered here, as Wuthnow (1988) and Hunter (1991) argue. CONCLUSIONS In two widely cited books that sparked a debate in academic and political circles alike on the political implications of religious divisions, Wuthnow (1988) and Hunter (1991) depicted America as divided into hostile religious camps in conflict with each other over social issues such as school prayer, sexuality, reproduction, as well as on issues relating to the family, gender roles, and racial and economic justice. For Hunter, whose conceptualization we have most closely followed and tested in this paper, the religiously orthodox - those who see God as the ultimate moral arbiter of right and wrong, who regard the Bible as the revealed word of God that is timeless and inerrant, and who see God as a guiding presence in their daily lives - are assumed to be politically conservative on these issues, while theological progressives - those who believe that humans should be the judge of right and wrong, that morality is situated in a context and developed by human reason, and that humans determine their own fates - are assumed to take liberal political stances. Through an analysis of national survey data, we tested several hypotheses regarding the extent to which one of these groups - the religiously orthodox - constitutes a united, politically conservative front. To ensure that we were not simply capturing possibly the ambivalent, muddled sentiments of those in the middle, we defined religious orthodoxy very narrowly. We analyzed only those respondents whose views on religion correspond most closely to Hunter's (1991) definition of religious orthodoxy, a group comprising just over one fifth of a national sample of Americans. We examined four specific senses in which conservative religionists might constitute a united front. First, we considered whether they are uniformly conservative across a broad range of sexuality/reproduction, gender, racial, and economic issues. While as both Wuthnow and Hunter expect, religious conservatives do tend to be overwhelmingly conservative on prayer in public schools, abortion, sex outside marriage, and homosexuality, they tend toward the liberal pole in their stances on many gender, racial, and economic issues and on sex education in public schools. Their views indicate far more support for government intervention to improve the conditions of racial minorities, provide jobs for all citizens, help people with their medical bills, and reduce the gap between rich and poor than Wuthnow's and Hunter's analyses would suggest. Second, we examined whether the religiously orthodox exhibit consensus on specific issues. The standard deviations of responses on issues suggested little consensus; for many issues the level of disagreement approached the maximum that the questions could show. Third, we considered whether religious conservatives are consistent in their attitudes across a broad range of issues. In other words, if individuals are conservative on abortion,

16 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 243 are they also conservative on sex education, racial equality, and welfare spending. We found little consistency across issues among orthodox religionists. Fourth, we considered whether self-interest along gender, racial, and class lines impedes the consensus needed for the religiously orthodox to mobilize as a united political movement. We found significant differences based on sex, race, and economic position within religious conservatives on many of today's most fiercely contested issues - abortion, pornography, proper roles of women and men, government efforts to help African Americans, and welfare spending. For example, while orthodox blacks and whites share common positions on some issues, there are also important differences between the races on many other issues, casting doubt on the likely success of recent efforts by groups such as the Christian Coalition to reach out to religiously conservative blacks for support on a range of conservative causes (Glazer 1987: 252; Wilcox 1990: 43). That young people among the orthodox tend to be more liberal on some issues than older people also hints at problems that religiously based, right-wing political movements may have in mounting a united political campaign. Thus, contrary to the portrayal of the religiously orthodox in the popular media, by the leaders of political movements, and in some recent scholarship, they are not a united conservative front. Instead, even Americans who are at the extreme of religious orthodoxy, the group one might most expect to be ideologically consistent, are moderate or even liberal on many issues, show little political uniformity on specific concerns, are divided on many of today's most contested issues along lines of race, sex, class, and age, and are as individuals, inconsistent in their opinions across issues. In view of the political indirection, ideological inconsistency, and disunity we have found among the religiously orthodox, we must ask whether the culture war they are said to be engaged in with theological progressives exists mainly in the minds of media pundits, leaders of political movements, and academics. We also question the widespread use of the political labels, "the Religious or Christian Right" as synonymous with religious orthodoxy or traditionalism. These labels, which many media analysts would have applied to all members of our sample based on their religious orthodoxy, fit only a small minority of traditional religionists. Our portrayal of the religiously orthodox, which is in accord in many respects with both ethnographic (McGuire 1982; Pohli 1983; Ammerman 1987; Neitz 1987; Stacey and Gerard 1990; Davidman 1991) and quantitative studies (Kiecolt and Nelsen 1988; Jelen 1990), helps to explain why such low levels of support have been found among the religiously orthodox for organized para-religious movements such as the Moral Majority or for the political campaigns of conservative Christian candidates such as Pat Robertson (Gallup 1982; Buell and Sigelman 1985; Wilcox 1987; 1989; 1990). Our findings suggest that the representation of the religiously orthodox as a monolithic camp locked in struggle with a liberal-secularist enemy is simply false. NOTES An earlier version of this paper was presented at the annual meetings of the Society for the Scientific Study of Religion held in St. Louis, in October Correspondence concerning the paper should be directed to Nancy J. Davis, Department of Sociology and Anthropology, 331 Asbury Hall, DePauw University, Greencastle, IN 46135; ndavis@depauw.edu. The authors are listed alphabetically. We wish to thank Ted Jelen and the anonymous reviewers of JSSR for their helpful comments. 1 Since the late 1980s and early 1990s when Wuthnow and Hunter wrote the influential books considered here, they have each written books analyzing a specific issue based on surveys and in-depth interviews with Americans. In his 1988 book, The Restructuring of American Religion, Wuthnow (1988: ) argued on theoretical grounds and cited survey evidence that religious conservatives were conservative on economic issues. In God and

17 244 JOURNAL FOR THE SCIENTIFIC STUDY OF RELIGION Mammon in America, Wuthnow (1994) finds little correspondence between church attendance and economic attitudes and rarely analyzes the effects of the conservative/liberal division in religion that was the focus of his earlier book. While Hunter's (1991) Culture Wars focused mainly on the rhetoric of leaders and organizations of the extreme poles in the "culture war," his 1994 book, Before the Shooting Begins, examines the ambiguity in the attitudes of Americans in the middle on the question of abortion and refers to the "conservative, moderate, and progressivist wings" rather than only to the polar categories of this religious division (1994: 185). 2 The logistic regression coefficient for African Americans is.708 (standard error =.277), for Indians, (.437) and for residents of the South,.888 (.180). 3 As a check on our findings based on this definition of the religiously orthodox, we conducted all analyses on four other subsamples of respondents: (1) those who scored 12 and above on the scale, (2) those who scored 14 and above, (3) those who were Protestant and identified themselves as "fundamentalists," and (4) those who were Protestant or Catholic and reported a "born-again" experience. These analyses yielded similar results to those reported in the text (details available on request from the authors). 4 Since the first item in the scale refers to "the Bible," we were concerned as to how Jewish respondents interpreted this, whether the three items scaled for Jews, and whether this might have resulted in no Jews exceeding our cutoff for the religiously orthodox. The item on the Bible appears to have been interpreted by Jewish respondents as referring to the original books of the Bible or the Torah. This item correlates similarly with the other two items for Jews (.55 with the second item and.40 with the third) as it did for Protestants, Catholics, and those who had no religious affiliation. Had Jewish respondents leaning toward the orthodox pole interpreted the item on the Bible as referring to both the Old and New Testaments and rejected its literalness for this reason, this item would have been negatively correlated with the other items in the orthodoxy scale. Factor analysis showed that for Jews all three items load on a single factor with an eigenvalue of Cronbach's alpha for the scale is actually slightly higher for Jews,.73, than for the sample as a whole, Respondents to a 1984 national survey were asked, "Where would you place yourself on this scale in terms of your religious views?" Respondents were handed a card with a six-point scale ranging from conservative (1) to liberal (6). Wuthnow (1988: 343n) treated as "religious conservatives" respondents who selected 1 or 2 and as "religious liberals" those who selected 5 or 6. 6 Wuthnow (1988: 215) states that: a conservative Christian who regards the Bible as a divine book to be taken as literally true, word for word, is likely to think there is good reason for saying that liberal Christians do not show proper respect for the Bible... In return, liberal Christians who believe that faith in the Bible must be tempered with reason, personal experience, and historical knowledge have some cause to regard conservatives as narrow-minded. 7 Southern states, as classified by NORC, include South Atlantic states(delaware, Maryland, West Virginia, Virginia, North Carolina, South Carolina, Georgia, Florida, and the District of Columbia), East South Central states (Kentucky, Tennessee, Alabama, and Mississippi), and West South Central states (Arkansas, Oklahoma, Louisiana, and Texas). While some of these states may be misclassified, the data do not allow exclusion of specific states. 8 What constitutes the liberal stance on pornography is not altogether clear since some liberals oppose pornography on grounds that it is degrading or injurious to women. 9 The upper limit is approximate because the equation used in computing the standard deviation depends on the number of cases minus one (N -1). The actual upper limit for 200 cases is 1.003, for 300 cases is 1.002, and for 1,000,000 cases is For these regressions the dependent variables are not collapsed into three categories (as they are in Table 1), in order to most accurately estimate the effects of independent variables. 11 Elsewhere (Robinson and Bell 1978; Davis and Robinson 1991), we showed that, in addition to the liberalizing or 'enlightening" effect of education that Wuthnow (1988: 163) expects, education has a "reproductive" effect on some attitudes toward inequality in that education stresses and rewards individual effort as a means toward advancement, which may reduce support for government intervention as a solution to group disadvantage. REFERENCES Ammerman, N Bible believers: Fundamentalists in the modern world. New Brunswick: Rutgers University Press.

18 RELIGIOUS ORTHODOXY IN AMERICAN SOCIETY: THE MYTH OF A MONOLITHI CAMP 245 Aptheker, H The urgency of Marxist-Christian dialogue: A pragmatic argument for reconciliation. New York: Harper & Row. Buell, E., Jr. and L. Sigelman An army that meets every Sunday?: Popular support for the Moral Majority in Social Science Quarterly 66: Cone, J. H Black theology and black power. New York: Seabury Press. Cox, H Fundamentalism as an ideology. In Piety and politics: Evangelicals and Fundamentalists confront the world, edited by R. J. Neuhaus and M. Cromartie, Washington, DC: Ethics and Public Policy Center The warring visions of the Religious Right. The Atlantic Monthly. November , 64-66, Davidman, L Tradition in a rootless world: Women turn to Orthodox Judaism. Berkeley: University of California Press. Davis, J. A. and T. W. Smith General Social Surveys, : Cumulative Codebook. Chicago: National Opinion Research Center. Davis, N. J. and R. V. Robinson Men's and women's consciousness of gender inequality: Austria, West Germany, Great Britain, and the United States. American Sociological Review 56: Forthcoming. Are the rumors of war exaggerated?: Religious orthodoxy and moral progressivism in America. American Journal of Sociology. Glazer, N Fundamentalists: A defensive offensive. In Piety and Politics, edited by R. J. Neuhaus and M. Cromartie, Washington, DC: Ethics and Public Policy Center. Gutierrez, G A theology of liberation. Maryknoll, NY: Orbis Books. Hart, S What does the Lord require?: How American Christians think about economic justice. New York: Oxford. Hunter, J. D Culture wars: The struggle to define America. New York: Basic Books Before the shooting begins: Searching for democracy in America's culture war. New York: Free Press. Jelen, T Religious belief and attitude constraint. Journal for the Scientific Study of Religion 29: _ The political mobilization of religious beliefs. New York: Praeger. Kiecolt, K. J. and H. M. Nelsen The structuring of political attitudes among liberal and conservative Protestants. Journal for the Scientific Study of Religion 27: Klatch, R Complexities of conservatism: How conservatives understand the world. In America at Century's End, edited by A. Wolfe, Berkeley: University of California Press. McGuire, M. B Pentecostal Catholics: Power, charisma, and order in a religious movement. Philadelphia: Temple University Press. Neitz, M. J Charisma and community: A study of religious commitment within the charismatic renewal. New Brunswick, NJ: Transaction Books. Pohli, C. V Church closets and back doors: A feminist view of Moral Majority women. Feminist Studies 9: Robinson, R. V. and W. Bell Equality, success and social justice in England and the United States. American Sociological Review 43: Rose, S. D Women warriors: The negotiation of gender in a charismatic community. Sociological Analysis 48: Stacey, J. and S. E. Gerard We are not doormats: The influence of feminism on contemporary evangelicals in the United States. In Uncertain terms: Negotiating gender in American culture, edited by F. Ginsburg and A. Tsing, Boston: Beacon Press. Tamney, J. B., R. Burton, and S. D. Johnson Fundamentalism and economic restructuring. In Religion and political behavior in the United states, edited by T. G. Jelen, New York: Praeger. Warner, R. S New wine in old wineskins: Evangelicals and liberals in a small-town church. Berkeley: University of California Press. Wilcox, C Popular support for the Moral Majority in 1980: A second look. Social Science Quarterly 15: _ Evangelicals and the Moral Majority. Journal for the Scientific Study of Religion 28: Blacks and the New Christian Right: Support for the Moral Majority and Pat Robertson among Washington, D.C. blacks. Review of Religious Research 32: Woodrum., E Determinants of moral attitudes. Journal for the Scientific Study of Religion 27: Wuthnow, R The restructuring of American religion: Society and faith since World War II. Princeton, NJ: Princeton University Press The struggle for America's soul: Evangelicals, liberals, and secularism. Grand Rapids, MI: William B. Eerdmans God and Mammonz in America. New York: Free Press.