Re ' 'ously Based Politics: Religious Elites and the Public*

Transcription

1 Re ' 'ously Based Politics: Religious Elites and the Public* DANIEL V.A. OLSON, Indiana University at South Bend JACKSON W. CARROLL, Hartford Seminary Abstract Will religion become a major source of political conflict in America? Wuthnow describes a growing conservative-liberal rift in American religion that threatens to fan the flames of political conflict. Warner agrees that there is a split in religion butfeels it will foster little political conflict because the two sides pursue nonopposing agendas. Among religious elites (seminary faculty) we find two political agendas (liberal and conservative), both linked to religion but only partly opposed to each other. Within the general public we find that the same two agendas are uncorrelated and only the conservative social-moral agenda is linked to religion. While important, religion will not soon become the major axis of political conflict in America. Since the late 1970s many observers of the American political scene have been disturbed by the apparent injection of religiously based values and goals into public policy debates (Rothenberg & Newport 1984). Much of this concern centers on the appearance of New Christian Right (NCR) organizations formed to pursue conservative moral and economie goals by activating the approximately 50 million evangelicals in the country and in particular the fundamentalist wing of this community. The airn from the beginning was to mobilize a group of people who had traditionally avoided polities because they saw it as a dirty, corrupt business... by convincing these people that political involvement was a God-given responsibility. (Zwier 1982, quoted in Wilcox 1989) While some liberals may fear the creation of a strong conservative coalition with the power to obstruct the pursuit of liberal political goals, more concern * We thank the Pew Charitable Trusts for supporting this work through a planning grant to study seminary cultures. Thanks to the Center for Social and Religious Research, Hartford Seminary, and the Indiana University Summer Faculty Fellowship program for support to the primary author during various stages of this project. Thanks to John Green, Larry lannaceone, Edward Lehman, R. Stephen Warner, and two anonymous reviewers for comments on earlier drafts of the article. Special thanks to Larry lannaecone for help in developing the final data set. Direct all correspondence to Daniel V.A. Olson, Department of Sociology, Indiana University at South Bend, 1700 Mishawaka Ave., P.O. Box 7111, South Bend, IN or BITNET: DOLSON@IUBACS. The University of North Carolina Press Social Forces, March 1992, 70(3):

2 766 / Social Forces 70:3, March 1992 centers on the potential for NCR organizations to link religion and polities, to use religion as the basis for political mobilization. At worst the New Christian Right is seen as the American equivalent of Islamic fundamentalism that threatens to undermine liberal democratie traditions of political tolerance (Lawrence 1989). Less dramatically, some fear that as public issues become religiously and morally charged, pragmatic solutions based on compromise will become increasingly difficult. At the verg least, such a scenario suggests that America may be headed into an era of increasing political and cultural conflict. But is this where America is headed? Are fears of religiously based political conflict really justified? Wuthnow (1988, 1989a) suggests that such conflict will increase, but Warner (1988, 1989) implies that religion will play only a minor role in future political conflicts. In this article we examine the basis of both Wuthnow's and Warner's assertions. First, we explore the connections between religion and polities among religious elites (seminary faculty), where we expect such connections to be most apparent (Converse 1964). Then we look for similar connections within the general public. Among elites, we find some support for Wuthnow and streng support for Warner. Within the general public, our findings strongly support Warner's analysis and suggest that in the 1990s, religiously based political conflict will be limited to social-moral issues (abortion, sexual morality, civil liberties, women's roles). Religion is unlikely to fuel political conflict over most other political issues, including economicwelfare and social-spending issues. Two Views of the Religious Divide Wuthnow (1988, 1989a, 1989b) argues that since the 1960s there has been a growing split between religious conservatives and religious liberals, a division that crosscuts the older barriers separating Protestants, Catholics, and Jews. He suggests, in fact, that American religion is being "restructured." As the conservative-iiberal divide widens, the importance of denominations is declining to the point that today many conservative Catholics may feel more in common with conservative Presbyterians than with liberal Catholics. Wuthnow recognizes that the current liberal-conservative split has roots in divisions that began within American Protestantism at least one hundred years ago, the divisions between those who supported the social gospel and the fundamentalists. The social-gospel view is the belief that "Christianity is not fulfilled and not true to its origins and history unless and until it immerses itself in alleviating the world's miseries and social pathologies" Qohnstone 1975:239). The fundamentalists took their name from the 1910 pamphlet The Fundamentals that codified their orthodoxy and clarified their opposition to liberal theology and the social-gospel movement. Despite many changes, the heirs of these two traditions live on within American Protestantism today and are identified by scholars with a variety of labels, e.g., private and public Protestants (Marty 1970), this-worldly and other-worldly (Roozen, McKinney & Carroll 1984), evangelicals and liberals (Warher 1988), or evangelical and mainline (Roof & McKinney 1987).

3 Religiously Based Politics / 767 According to Wuthnow, religious conservatives and liberals (including many secularists) disagree not only about religion but also about the role of government in public life. He argues that many of the most hotly debated issues of the past several decades (e.g., civil rights, womeri s rights, homosexual rights, military and social spending, abortion, pornography, school prayer, the teaching of creationism) arise out of differences rooted ultimately in their two ways of viewing God and the world. These include a conservative public theology [that] champions strong traditional morality, strong national defenses and a heady brew of free enterprise. It stakes its authority in a literal interpretation of the Bible and claims to return the nation to biblical ideals on which it was founded. [Opposed to this is] a liberal public theology [that] argues for a more relativistic code of personal morality, a cooperative multilateral spirit in foreign relations and strong government initiatives capable of infusing norms of social justice into the capitalist mind. It also stakes its authority on biblical tradition, but often interprets this legacy as cultural wisdom rather than divine revelation. (Wuthnow 1989b) For Wuthnow, the two religious camps underlie an important ideological and political split in American culture, one that could have quite divisive consequences for America's future. Warner's (1988,1989) views differ from Wuthnow's in two ways that make religiously based political conflict less likely. First, Warner suggests that despite the media attention given in the 1980s to symbolic debates between groups such as the Moral Majority and the American Civil Liberties Union, the liberal and conservative traditions are not as opposed as they initially appear. He suggests that it is theology, not politics, that unites conservatives. With the exception of social-moral issues, the political views of conservatives are diverse and vary in much the same ways as those of the general public. On the other hand, Wanner lees religious liberals as united not by "doctrine but... [by] an optimistic, socially responsible attitude. What holds the liberal constituency together, in other words, is less religion, narrowly conceived, than worldly morality" (Warner 1988:56). Thus, while the Gore values of the two traditions differ, they are not directly opposed. According to Warner, evangelicals rally round a theological flag, whereas liberals unite around a political and social justice banner. Warner is not arguing that the political unity of liberals has no theological roots, only that the political and social justice outcomes of actions are more visibly the basis of liberal consensus than are the potentially diverse theological positions used to justify these actions. Wanner (1988) acknowledges that theological conservatives hold more conservative views on social-moral issues such as "abortion, homosexuality, and prayer versus sex education in the public schools" (298). Inasmuch as these are also political issues, Wanner implies that religious conservatives, like religious liberals, have a distinctive political agenda. Nevertheless, it is conservative theology, not social-moral issues, that forms the basis of conservative unity and identity. Warner's second disagreement with Wuthnow concerns the size and strength of the liberal religious constituency. In a review of Wuthnow's (1988) book, Warner (1989) suggests that religious liberals make up a fairly small

4 768 / Social Forces 70:3, March 1992 segment of the American public. In addition, Warner (1988) argues along with others (e.g., Hadden 1969, Roof & McKinney 1987) that compared with conservatives, religious liberals give relatively weak support to their churches and are more likely to disagree with their religious leaders on political issues. As we shall see below, Wuthnow's and Warner's analyses are not totally contradictory, but they have quite different implications for political conflict. 1f Wuthnow is right, both religious traditions have the capacity to politically mobilize the public and are directly opposed. With religion and politics becoming aligned and polarized along a single liberal-conservative dimension, religiously based political conflict appears likely. If Warner is right, both religious traditions are linked to political agendas, but these agendas direct their followers in different and unrelated directions. What is critical to one camp is peripheral to the other. Moreover, only the conservative camp has a constituency large enough to strongly promote its political agenda. Religious Elites: Data and Variables We first examine the religious and political views of seminary faculty. We do so for two reasons. First, if religion and politics are aligned along a single conservative-liberal dimension, as Wuthnow suggests, this alignment should be most evident among religious elites, where ideological constraint is greatest (Converse 1964). Second, seminary faculty have great potential influence over the religious views of others. In this country, they are the primary interpreters, guardians, and gatekeepers of religious traditions. These faculty train religious professionals, who, in turn, enter positions of similar influence. The linkage of beliefs forged by theological elites can have an important influence on those in the general public. If elites strongly link political platforms to religious positions, their followers may be more inclined to do so. If no such linkages exist among elites, they are unlikely to exist within the general public. The data we examine were collected by Ladd and Ferree of the Roper Center in 1981 and were designed to assess the political and religious beliefs of American theological faculty. In late 1981 and early 1982 they mailed 2,000 questionnaires (each containing 208 items) to a sample of faculty at seminaries and theological schools in the U.S. Because of the nature of the theological items, faculty at Jewish institutions were not included. The list of schools was obtained from the Association of Theological Schools (AT5) in the United States and Canada and from the United States Catholic Conference. However, the list also includes schools that are not officially affiliated with either AIS or the Catholic church. Most of these are Bible colleges and training schools without academie accreditation and tend to be theologically conservative. Ladd and Ferree randomly sampled 2,000 people from lists of faculty at each institution. Since their interest was in the views of those who train religieus professionals, the sample does not include faculty in undergraduate departments of religion. After two separate mailings, 1,112 completed questionnaires were received. A list of the questionnaire items and a detailed description of the sampling methods are found in Ladd and Ferree (1982).

5 Religiously Based Politics / 769 Since the Ladd-Ferree data indicate the institutional affiliation of each respondent, we were able to classify respondents using a typology of seminaries1 used with success in several earlier studies (e.g., Carroll & Wheeler 1987). This typology first separates Protestant from Catholic seminaries and then cross-classifies Protestant seminaries according to whether they are evangelical or mainline and whether they are independent or affiliated with a denomination. In previous analyses we found that faculty at independent seminaries tend to take more extreme positions (both conservative and liberal) than faculty at denominationally affiliated seminaries. Denominational ties appear to have a moderating effect on seminaries. For Protestant schools affiliated with a denomination, we classified each school, with a few exceptions, 2 according to the general mainline or evangelical tendencies of the denomination. Protestant schools not affiliated with a denomination were classified as evangelical or mainline based on our personal knowledge of these institutions. Since all the Catholic schools are affiliated with the Catholic church or one of its orders, our typology has live categories: (1) evangelical independent, (2) evangelical denominational, 3) mainline independent, (4) mainline denominational, and, (5) Roman Catholic. We were unable to classify 38 respondents, most of whom failed to indicate their institutional affiliation.4 In order to simplify the presentation of our results, we added respondent scores for similar items to create several scales. First, we created a conservative theology scale based on the sum of the standardized scores from seven theological items (see Appendix for individual items, Cronbach's a=.79). Next we created a liberal peace-and-justice scale by adding the standardized scores of 18 items dealing with defense issues, welfare issues, and issues relating to the current distribution and potential redistribution of wealth and power in the U.S. (see Appendix for individual items, Cronbach's a-.91). Finally, we created a conservative sexual-morality scale by adding the standardized scores from six items dealing with sexual morality and abortion (see Appendix for individual items, Cronbach's a-.819). Religious Elites: Alignment of Agendas Wuthnow suggests that religion and politics are becoming polarized along a single liberal-conservative dimension. If so, this should be truest among religious elites. More specifically, if Wuthnow is right, we would expect the conservative-theological scale to have a strong positive correlation with conservative sexual morality and a strong negative correlation with liberal peace-and-justice issues. Further, conservative sexual morality should be negatively correlated with liberal peace-and-justice issues. Among elites, the simple correlation coefficients support Wuthnow's claims. Not only is theological conservatism related to sexual conservatism (Pearson's r=.643, p<.0001, N-1112), but it also has a strong negative correlation with liberal peace-and-justice issues (r=-.510, p<.0001). The conservative sexualmorality and liberal peace-and-justice scales are negatively correlated (r--.446, p<.0001). The two political agendas, conservative sexual morality and liberal peace and justice, are both related to religion, and while not exact opposites, they have a strong negative correlation.

6 770 / Social Forces 70:3, March 1992 FIGURE 1: Means and Standard Deviations by Seminary Type (Standardized Scales)a Theology -d- Cn new "A^ "Ml GO Peace and AAI Justice ""a cn Sexual "M I.Iorality "An 1 I I i I i More Liberal Positions > CA-Catholic, MI-Mainline Independent, MD-Mainline Denominational, ED-Evangelical Denominational, EI-Evangelical Independent. If Wuthnow's analysis is correct, we would also expect the evangelical seminary faculty to hold more conservative positions on both sexual-morality and peace-and-justice issues. Figure 1 shows that this is true. In diagrammatic form, it shows the means and standard deviations of all three scales for each seminary type. The means for each seminary type are indicated by the two letter abbreviations. The lines extend one standard deviation to the left and right of the mean value. Thus, short lines indicate small standard deviations and high consensus, and long Tines reflect large standard deviations and greater disagreement within a seminary type. Consistent with the correlations discussed above, there are strong differences in all three scales between seminary faculty in different types of schools. In analysis of variance tests not shown here, differences between the live seminary types all yield F statistics that are significant at probabilities below The differences between seminary types on conservative theology are large (g2z.4518) but not surprising, since conservative theology played a major role in our classification of schools. As expected, evaipgelical faculty are the most conservative and mainline faculty the most liberal. While unsurprising, the strength of the evangelical/mainline differences lends validity to our classification scheme.

7 Religiously Based Politics / 771 Consistent with Wuthnow's analysis and the correlations discussed above, Figure 1 also shows strong differences between seminary types on both the liberal peace-and-justice scale (q ) and the conservative sexual-morality scale (1 2_.2973). Evangelicals are more conservative on both scales compared with mainline or Catholic faculty. Most analyses, including those of Wuthnow and Warner, say relatively little about the position of Catholics in the conservative-liberal religious divide. Figure 1 shows that on conservative theology the Catholic faculty score near the middle between evangelicals and mainline faculty. While this finding is consistent with our findings for Catholic laity, discussed below, we are cautious in interpreting this result, since, as the Appendix shows, the theology scale reflects issues separating conservative and liberal Protestants, but does not include items that might better differentiate Catholics, such as views on papal authority. Greeley (1989) notes that inerrant views of the Bible are not a requirement of faith for Catholics. Figure 1 shows that on liberal peace-and-justice issues, Catholics are close to the liberal positions of mainline respondents. We have greater faith in this finding because it is consistent with other observations concerning Catholic elites (e.g., Varacalli 1983), a long history of Catholic support for social legislation (Blake 1972), and recent positions taken by the American Catholic bishops. The near-average position of Catholics on sexual morality is somewhat surprising, since they take quite conservative positions on abortion. However, the abortion item is only one of six used to construct the sexual-morality scale. Catholics are more liberal on other items. The results discussed so far appear to support Wuthnow, at least among elites. However, as is clear below, the simple correlation coefficients and the mean values for seminary types teil only part of the story. Warner's claim that the two sides of the religious divide focus on different issues is also true. Religious Elites: Core and Peripheral Agendas Warner (1988) contends that the core values of the two religious parties are not in direct opposition. Evangelicals unite first around theology and secondarily around social-moral issues, while liberals unite mostly around political and social justice concerns. What is central to one side is peripheral to the other. To test this contention we seek to identify theological and political issues that are central versus those that are relatively peripheral to the belief systems of different seminary types. We assume that one measure of issue centrality is the level of consensus that an issue generates within a religious tradition. Cate keeping and social control within seminaries limit disagreement on issues that are considered fundamental to the self-conception of an organization, especially organizations, such as seminaries, devoted to the maintenance and promotion of particular ideas. These differences in levels of consensus can be seen in Figure 1, where larger standard deviations reflect greater disagreement and smaller standard deviations reflect greater consensus among faculty within a seminary type.

8 772 / Social Forces 70:3, March 1992 Thus, one sees, as Warher predicts, that on conservative theology and sexual morality, evangelical faculty are not only more conservative but also more uniformly conservative. This is especially true for independent evangelicals. Nonevangelical faculty exhibit greater differences among themselves. On theological conservatism the standard deviation for mainline independents is more than six times greater than for evangelical independents. In general, seminary groups with more liberal mean scores on theological conservatism show greater internal theological diversity. Not only are the differences in levels of consensus visually striking, but they are also statistically significant. An F test comparing the variances of evangelical and mainline faculty on the sexual-morality scale yields an F of 2.80 (df=386 and 482), implying a probability of less than.0001 that evangelical and mainline respondents share the same variance (Walker & Lev 1953). A similar test comparing the variances of evangelical and mainline respondents on the theology scale yields an F of 4.12 (p<.0001, df=386 and 482). The Jack of theological consensus among mainline faculty may partly result from the nature of the items in the theology scale, which tend to focus on issues central to conservative theology. However, our own experience with such seminariess tends to support Warner's claim that theological orientations are less central to the identity of mainline seminaries than the practical consequences of theological positions for action in the world. Moreover, among mainline faculty, theological diversity and tolerance for such diversity may be a more central value than any particular theological orientation, since such diversity gives students broader exposure to a variety of theological positions. Figure 1 shows a very different pattern for peace-and-justice issues. Mainline and Catholic respondents show the greatest internal consensus, whereas responses among evangelicals are quite diverse. Once again the results are statistically significant. The relevant F test yields a less-than.0001 probability of equal variances for evangelical and mainline respondents (F=4.79).6 These findings strongly support Warner's analysis. Among the seminary types, evangelicals show the most consensus on theology and sexual morality, but the orthodoxy that constrains these beliefs apparently exercises little constraint upon peace-and-justice positions. Evangelical faculty can hold a variety of positions on these issues. The pattern differs for mainline faculty. Figure 1 suggests that a different type of "orthodoxy" functions in mainline and Catholic seminaries, one that appears to constrain the positions of mainline faculty to a fairly narrow range at the liberal end of the peace-and-justice scale. Our experience with such seminaries suggests that this orthodoxy is often, though not always, less explicit than the theological orthodoxy of evangelical seminaries (which is sometimes enforced by making faculty sign statements of faith when they sign their contracts). Though less deliberate, this peace-and-justice orthodoxy may be as effective as official administrative policy in limiting faculty diversity at mainline seminaries. Consistent with Warner, evangelical faculty appear united around conservative theology and sexual morality while mainline and Catholic faculty are united on peace-and-justice issues. Although it is true, as Wuthnow suggests, that evangelicals are, on average, more conservative on all these issues, the Gore

9 Religiously Based Politics / 773 FIGURE 2: Liberal and Conservative Political Issues by Seminary Type (Standardized Scales)` 1 A1 MArtVJNE PROTESTANT ' 0 a ci -1 0 cri iv J -3 Ç4A EVANGEL)CAL w =, t -A Conservative Sexual Morality --> ' Ellipses are based on regression to include approximately 67.5% of the respondents in each seminary type. peace-and-justice issues of more liberal faculty appear to be peripheral issues for conservatives. Similarly, conservative theology and sexual morality appear to be peripheral issues for more liberal faculty. This pattern of results is more evident in Figure 2, which shows a scatterplot7 of the relationship between sexual-morality issues (the horizontal axis) and peace-and-justice issues (the vertical axis). We selected these two scales for Figure 2 sine Wuthnow and Warner both suggest that these issues lie at the core of the political agendas supported by religious conservatives and religious liberals.8 Each of the three ellipses in Figure 2 is drawn to inciude approximately 67.5% of the cases in each of three seminary types evangelical, mainline, and Catholic. As noted above, the two scales are negatively correlated (r=-.446), a pattere consistent with Wuthnow's view. But as Figure 2 shows, this "alignment" is more complicated than the simple correlation implies. The mass of the plotted points does not follow a straight line through the center of the graph. It follows an inverted L-shaped curve that is greatly constrained by the two orthodoxies.

10 774 / Social Forces 70:3, March 1992 Evangelical faculty are constrained to hold conservative views on sexual morality but are free to hold diverse views on peace and justice. Mainline and Catholic faculty are constrained to hold liberal positions on peace and justice but vary greatly in their views on sexual morality (with Catholics being more conservative as a group). Figures 1 and 2 show that the elites of both religious camps are each in strong agreement concerning a political agenda, but the agendas of the two are different. What is central to one camp is peripheral to the other. While this suggests that each side could mobilize their followers in support of their own political agenda, it also makel it less likely that either side could mobilize their followers to oppose the agenda of the other. Even among elites, where ideological constraint should be most evident, both sides hold quite diverse positions regarding the agenda of the other side. The political issues of the other camp are not well connected to local religious viewpoints. 1f conservative and liberal elites fail to unite in opposition to each other's agenda, one hardly expects to find such opposition among the laity. The Public: Two Independent Agendas Our findings for elites suggest the likely directions that political mobilization might take among both religious conservatives and religious liberals. But conflict is unlikely if the connections that religious elites make between religion and politics are not made by the laity. Thus, we also look for these connections within the general public. We find, contrary to Wuthnow and consistent with Warner, that within the general public the social-moral and peace-and-justice agendas are uncorrelated. Moreover, only the social-moral agenda is correlated with theological orientation. Peace-and-justice issues have little connection to religion among liberals or conservatives within the general public. The data come from the 1988 General Social Survey (GSS) (Davis & Smith 1988), a survey of the adult American population (N-1,481) that includes a special supplement on religion. Using these data we sought to construct scales that resembled those used in our analysis of elites. While the theological items contained in the General Social Survey are not the same as those used by Ladd and Ferree, they are perhaps better measures of the theological divide described by Wuthnow (1988) and Warher (1988). Wuthnow describes this division as two different ways of onderstanding religious truth. Liberals are more inclined to view religious truth as socially and culturally relative and to read the Bible symbolically rather than literally. Because they view the Bible as open to multiple interpretations and as a source of cultural wisdom more than as absolute truth, they are less inclined to view it and the proclamations of the church as determinative when making decisions. In contrast, conservatives are more likely to interpret the Bible literally and to see religious truth in absolutistic terms, true for all times and places, and open to only one correct interpretation. Thus, they are more inclined to accept the authority of the Bible and the church as a guide for individual decision making. We attempted to capture the spirit of these differences in our theologicalconservatism scale created by adding the standardized values of nine theological

11 Religiously Based Politics / 775 and religious items (Cronbach's a-.8756, see Appendix for individual items). High values on this scale reflect theologically conservative views of religious authority (the Bible and one's church). We selected a variety of political items from the 1988 GSS questionnaire that seemed to us to correspond to the two agendas found among seminary faculty. We also included items that are linked to religious positions in Wuthnow's descriptions of the liberal-conservative divide. The liberal peace-and-justice scale is based on eleven items (Cronbach's a=.6626, see Appendix) reflecting support for greater government spending and action on issues related to welfare, defense, racial minorities, education, economic equality, and redistribution of wealth. The conservative social-moral scale is based on nineteen items (Cronbach's a-.8263, see Appendix) related to conservative positions on sexual morality, abortion, women's roles, and civil liberties. Contrary to Wuthnow and consistent with Warner, within the general public the two agendas are not aligned along a single conservative-liberal continuum. In fact, they are almost entirely independent (r=-.049). While the large number of cases (1,481) makel this correlation statistically significant at the.029 level of probability (one-tailed test), it is substantively insignificant, with only two-tenths of 1% of the variations in the scales being in common (R2=.0024). Support for or opposition to one agenda has no bearing on support for or opposition to the other agenda. We are not alone in observing the independence of these agendas. Fleishman (1988), following Lipset (1981), finds, using confirmatory factor analysis on GSS data, that instead of a single conservative-liberal continuum, there are two separate, orthogonally related conservative-liberal dimensions underlying the political attitudes of Americans: an economic-welfare and an individual-liberties dimension. The connections to Fleishman's analysis are especially strong, since many of the same GSS variables that form his two dimensions are variables we used to construct our peace-and-justice and social-moral scales 9 In addition to predicting a negative correlation between the political agendas of religious conservatives and liberals, Wuthnow predicts that both political agendas will be strongly correlated with theological orientation. Contrary to Wuthnow, we find that within the general public only one of the agendas is correlated with theological orientation. The liberal peace-andjustice scale has a correlation of (not significant at p<.05, N=1,481) with conservative theology. Only the conservative social-moral agenda is related to theology (r-.304, p<.001, N-1,481). For the public, Wuthnow's claim of linkage between religion and politics holds true only for conservative social-moral issues. Many theological conservatives could be, and indeed already have been, politically mobilized by appeals to their conservative religious beliefs (a point conceded by Warner [1988]). At the same time, our findings suggest that theological orientation cannot currently be used effectively to mobilize religious liberals or conservatives to support or oppose the liberal peace-and-justice agenda.

12 776 / Social Forces 70:3, March 1992 The Public: Religious Constituencies Warner's second point of opposition to Wuthnow's analysis concerns the constituency for religiously based political conflict. Whereas Wuthnow (1988) describes two fairly evenly matched constituencies on either side of the religious divide, Warher (1989) claims that only the conservative constituency has the size and commitment necessary for successful political mobilization. In this part of our analysis, we examine the potential religious constituencies for each political agenda. We find, consistent with Warner, that the conservative social-moral agenda has a much larger religious constituency than the liberal peace-andjustice agenda. In seeking the religious constituencies for the two political agendas, we widened our search to include not only people sharing conservative and liberal theological orientations, but also people who identify themselves with conservative and liberal religious groups (despite Wuthnow's claim that theology is now more relevant than denomination)'o Thus, in Table 1 we use multiple regression to identify which religious variables best predict support for the political agendas. To lessen the possibility of spurious correlations between the religious and political variables, we simultaneously control for level of religious involvement 11 and several social background characteristics' 2 (sex, age, years of education, race, and family income) 13 We created dummy variables for six religious groups: white evangelicals,' 4 black evangelicals,' mainline Protestants, Catholics, those who said they did not identify with any religious tradition (religious "nones"), and those who identify with non-christian religions. Due to insufficient numbers, we did not create separate categories for Jews, Muslims, or other non-christians 16 We included those with no religious preference because Wuthnow and others (e.g., Connecticut Mutual Life 1981; Roof & McKinney 1987) suggest that secularists also support the liberal social justice agenda. Although we created a dummy variable for Catholic affiliation, we omit this variable hom our regressions in order to prevent multicollinearity. We found in other analyses that Catholics are moderates on nearly all the political items used to construct the social justice and social-moral scales (see also Jelen 1990). Out of 1,46917 GSS respondents in 1988, 25% (360) are white evangelicals, 10%18 (145) are black evangelicals, 22% (327) are mainline Protestant, 26% (384) are Catholic, and 8% (118) list no religious preference. Turning first to the liberal social justice agenda, we find, as we did earlier, that support for this agenda is unrelated to theological orientation (see Table 1). However, affiliation with some religious bodies is related to support for this agenda. Contrary to the expectation associated with their label as evangelicals, black evangelicals are the strongest supporters of the liberal social justice agenda. Non-Christians (almost half of whom are jews) also support this liberal agenda. Contrary to Wuthnow, who suggests that liberal Protestants and secularists are strong supporters of the peace-and-justice agenda, we find that neither 19 of these groups is supportive. Among the public both groups are neutral, art

13 Religiously Based Politics / 777 TABLE 1: Standardized Regression Betas, Liberal and Conservative Political Agendas on Religious and Social Background Predictors for the General Public Liberal Conservative Peace-and- Socialjustice Moral Independent variables Scale Scale Religious-group dummy variables: White evangelicals -.124***.017 Black evangelicals.174*** Liberal protestants Non-Christians.103*** No religious affiliation ** Catholics* Conservative theology *** Religious involvement * Social background Age -.113***.047 Sex (female).100*** -.052* Race (nonwhite).123***.079* Education (years).064* -.226*** Family income * R (N - 1,481) The Catholic dummy variable is not included in the regression equation although Catholic respondents are included in the analysis. Prior analysis shows Catholics to be moderates on both political agendas (see also Jelen 1990). *p<.05 **p<.01 ***p<.001 observation consistent with Hadden (1969), who contends that liberal Protestant laity are more conservative than their leaders on most political issues. White evangelicals are the only opponents of the peace-and-justice agenda. However, in other regressions (not shown here), we find that this opposition is limited to defense issues and government aid for racial minorities. On other economic and social welfare spending issues, they differ little from the general population. Turning to the conservative social-moral agenda, we find, again as we did earlier, that this agenda is strongly related to theological orientation even after introducing controls for denomination and social background. This agenda is unrelated to most of the religious-group dummy variables. When controls for social background and theological orientation are used, only those with no

14 778 / Social Forces 70:3, March 1992 religious affiliation show a tendency to support the conservative social-moral agenda based on their lack of religious-group affiliation, a finding that is also inconsistent with Wuthnow's portrayal of this group. In contrast to Wuthnow and consistent with Warner, the liberal peace-andjustice agenda does not appear to have a broad religiously based constituency. Liberal theological orientations are uncorrelated with support for this agenda. Support comes only from non-christians and black evangelicals. While significant, these groups together make up only 18% of the population. Consistent with Warner, there is a broad religiously based constituency for the conservative social-moral agenda. This support comes mostly from theological conservatives regardless of denomination. Such persons are much more numerous than theological liberals. The distribution of the religious-authority scale is skewed so that in the standardized version of this scale the fiftieth percentile (the median) falls at.131 (where positive values indicate conservatism). About 55% of the sample are above the mean. Looking at some of the individual items in the scale, one finds that 64% say they "know God really exists and... have no doubts about it." On this same item only 5% say they do not believe in God or do not know whether God exists. Fully 34% choose the most conservative of three possible views of the Bible, saying the "Bible is the actual word of God and is to be taken literally, word for word." These analyses lead us to conclude with Warner that the religiously conservative constituency is larger and more supportive of its political agenda than the religiously liberal constituency. We do not argue that the social justice agenda lacks supporters, only that for most of the general public (as opposed to religious elites) this support has no direct relation to religion.20 Prospects for Religiously Based Politica! Conflict Is the U.S. about to be engulfed in religiously based political conflict? We think that over the short term such conflict will be limited to the conservative socialmoral agenda and that this conflict will be far less severe than the scenario portrayed by Wuthnow. The public does not appear ready to split apart into two warring camps opposed on both religious and political issues. Even among religious elites, where we expect to find the highest levels of ideological constraint and where the political agendas of conservatives and liberals appear to be aligned with religion, neither side appears unified in its opposition to the agenda of the other. Within the general public the two agendas are not at all aligned and only the conservative agenda has much connection to religion. Moreover, the two lides have unequally matched religious constituencies, with the potential conservative constituency being much larger than the liberal constituency. The most likely short-term scenario is that religiously based political conflict will continue to be limited to debate over the conservative social-moral agenda. Outside of this agenda, attempts to politically mobilize the public on the basis of religion are likely to fail. Most of the general public fails to connect other political issues with religious belief.

15 Religiously Based Politics / 779 This finding helps to explain why New Christian Right organizations have been ineffective in mobilizing support for conservative foreign policy or economic agendas (lannaccone 1992) and why Jerry Falwell's audience share for his Old Time Gospel Hour television show dropped when his message became more political (Warner 1988). Few religious conservatives tie these issues to matters of faith. Wuthnow's claim that the conservative agenda includes "a heady brew of free enterprise" (1989b) is true only for some elites, not for the general public (lannaccone 1992). While religiously based conflict over social-moral issues is likely to continue, if Warner is right, it may be far less severe than the potential suggested by the large numbers of religious conservatives in our analysis. Warner argues that conservatives focus first on theology and only secondarily on political issues. lf true, many conservatives who scored high on the social-moral conservatism scale (because they agree, for example, that premarital sex is wrong) might nevertheless be reluctant to become politically involved in support of this agenda. If politics are secondary, it could be quite difficult to mobilize theological conservatives to support the conservative social-moral agenda. Research on New Christian Right organizations suggests that this is true. While a majority of the general public do, in fact, agree with many of the conservative social-moral positions taken by groups such as the Moral Majority (Simpson 1983), actual support for such organizations is low. In a survey of Southern Baptist clergy (a quite conservative denomination in which one would expect very strong support among clergy elites), Guth (1983) found that only 3.3% were members of such groups and only 43% identified themselves as Moral Majority sympathizers. Wilcox (1989), using a national sample of evangelicals, found that only 24% were supporters of the Moral Majority. Buell and Sigelman (1985) found that only 11% of American voters in 1980 indicated that they were sympathetic to, or supporters of, the Moral Majority, a finding consistent with other research (e.g., Shupe & Stacey 1983). Still other research shows that New Christian Right organizations have had little impact on voting behavior either in national elections (Lipset & Raab 1981, Johnson & Tamney 1982; Johnson, Tamney & Burton 1989) or in local congressional elections (Johnson, Tamney & Burton 1990). In fact, Johnson and Tamney (1985) and Simpson (1985) argue that religious conservatives tend to vote for Democrats on the basis of economic issues. Liberal Protestants gave Reagan more votes in 1980 and 1984 than conservative Protestants (Simpson 1985). In our own analyses, not shown here, we find that within the general public, conservative theology has no correlation with Republican party identification. Thus, we expect that during the next few years, religiously based conflict over the conservative social-moral agenda is likely to continue, but at levels far below the potential suggested by Wuthnow's analysis. But what about the longer-term future? Perhaps Wuthnow's (1988) religious "restructuring" is a slow process that has only just begun. Full development of Wuthnow's scenario requires that three more pieces fall into place. First, religious elites who support the peace-and-justice agenda (mostly liberal Protestants and Catholics) will have to more effectively convince their followers of the connections between their beliefs and these issues. Second, there needs to be a large increase in the numbers of religiously liberal and

16 780 / Social Forces 70:3, March 1992 secular Americans who would support the liberal agenda. Wuthnow (1988) argues that higher education recruits persons into these categories, but more recruiting is necessary. Third, the two political agendas will have to become aligned along a common conservative-liberal dimension, a process that has not even begun within the general public and that is only partially complete among religious elites. Considered separately, each of these events would involve a major change in both religion and U.S. politics. Taken together, Wuthnow's scenario seems unlikely. Though religious divisions will continue to fuel debate over abortion, women s roles, sexual morality, and civil liberties, such divisions will probably not affect most other political issues. While important, religion is unlikely to become the primary axis of political division in the U.S. Notes 1. The main reason we classify respondents by their institutional rather than by their denominational affiliation one or by self-classification is that belief systems do not persist well as free-floating phenomena. They usually are linked to particular social and organizational settings that promote them (Stark & Bainbridge 1985). Our impression is that the local cultures of individual seminaries have an important selective and shaping influence on the beliefs of both faculty and students, one that usually leads to a within-seminary homogeneity of beliefs concerning issues most central to local seminary culture. 2.There area few schools affiliated with mainline denominations but known for their generally evangelical theological stance (e.g., the University of Dubuque Theological Seminary [Presbyterian] and the Eastem Baptist Theological Seminary [American Baptist]). We classified such schools as evangelical. 3.A list of these seminaries and their classification is available from the primary author. 4.An additional 11 respondents worked for seminaries supported by peace church denominations such as the Mennonites and the Church of the Brethren. In an earlier analysis (Olson & Carroll 1989), we found that the peace church respondents differ in important ways from the groups studied here, but because of their small numbers we have excluded them from the present analysis. 5.One of us, Jackson Carroll, is employed by a mainline Protestant seminary. 6.In order to make variances in the peace-and-justice scale comparable to those in the theology scale (which is constructed entirely from dichotomous yes/no items), we dichotomized all of the political items before constructing the political-liberalism scale. The results presented here thus appear somewhat more dramatic than the results for the political-liberalism scale based on nondichotomized items. Yet, even without dichotomizing the political items, the variance for mainline respondents is still more than twice as large as for evangelicals (F-2.09, p<.0001, df-386 and 482). 7.A small random jitter has been introduced to prevent the overprinting of respondents with identical scale scores. The apparent vertical lines formed by points plotted near the left side are mathematical artifacts resulting from the smaller number of items (7) used to construct the theological-liberalism scale. 8.The social-moral issues discussed by both Wuthnow and Warher are broader than the sexual morality issues incorporated in our scale. However, as discussed below, sexual morality is highly correlated with other social-moral issues. 9. Fleishman s economic-welfare dimension includes subsets of variables covering welfare spending, urban spending, government aid, and economic conservatism. His individualliberties dimension includes variable subsets for defense, sexual mores, two abortion factors, racial discrimination, and three civil-liberties factors related to atheists, communists, and homosexuals. The major difference is that defense spending is part of his individual-liberties

17 Religiously Based Polities / 781 dimension, whereas we used defense spending as a component of our peace-and-justice scale. The similarity of Fleishman's two dimensions to the political agendas of religious elites suggests either that his two dimensions have their origins in religious divisions or that the conservative-liberal split in religion is rooted in historical circumstances that also gave birth to Fleishman's two political dimensions. 10.Wuthnow (1988) emphasizes that the religious alignment of polities is more with theological orientation than with denominational affiliation. In his "restructured" view of American religion, denominational differences are diminishing and are being crosscut by the theological differences between conservatives and liberals, now found, according to Wuthnow, in most every major denomination. 11.The religious-involvement scale is based on six GSS items (Cronbach's a-.7906) whose mnemonic names (as listed in Davis & Smith 1988) are shown in parentheses. The respondents were asked how frequently they attended religious services (ATTEND), whether they were a member of a church or synagogue (CHURHMEM), whether they prayed out loud before family meals (SAYGRACE), and how often they read the Bible (READWORD) and prayed (PRAY). The regressions in Table 1 and separate factor analyses suggest that while the religious-practice and religious-authority scales are positively correlated, they reflect two separate phenomena with differing effects on political attitudes. 12.The background variables used in Table 1 include the GSS variables, with the following mnemonic labels described in the GSS codebook (Davis & Smith 1988): AGE, EDUC, SEX, RACE, INCOME We do not report multiple regressions or controls for social background in our analyses of seminary faculty since there are comparatively few social background differences among the elites. 14.We divided Protestants into evangelical and mainline groupings based on respondents' denominational preferences, using the NORC classification variable FUND (Smith 1986,1990). Mainline Protestants are those classified by NORC as either "moderate" or "liberal." This approach was used because of the small number of "moderate" Protestants and the Jack of many differences between "moderate" and "liberal" Protestants for the variables under examination. 15.We divided evangelicals on the basis of race since blacks make up a significant portion of evangelicals, have a historical tradition distinct from white evangelicals, and hold different views on a broad range of issues. Note that some blacks are not evangelical Protestants and are included in other categories based on denominational affiliation. 16.The 5% in this residual group are quite heterogeneous in their religious background but we include them in regressions since their political positions tend to be liberal (consistent with the fact that 30 of these 70 respondents are Jews) and because we wanted to include all segments of the public. 17.We could not classify 12 Protestant respondents since they had missing values for the GSS variable FUND. 18.Note also that 20 nonwhite evangelicals whose race is recorded as "other" were also included as black evangelicals. This inflates the size of the black evangelical group somewhat. 19.Even without controls for social background and theological conservatism, those with no religious affiliation are only very weakly in favor of the social justice agenda. 20.Wuthnow (1988) argues that higher education is playing a major role in widening the religious divide by recruiting many to the side of religious liberalism and secularism. The regressions in Table 1 show that education has a strong negative relationship with the conservative social-moral agenda and a minor positive effect on the liberal peace-and-justice agenda. For the purposes of this article we do not consider these direct effects of education as a "religious" basis of political conflict. (We save our analysis of the indirect effects of education on polities via religion for, another article.) 21.However, Simpson (1985) notes that Reagan received large majorities from those within and outside conservative Protestantism who share the positions of NCR organizations on moral and social issues. According to Simpson, Reagan s victories resulted from the combined support of these social-moral conservatives and traditional Republicans.

18 782 / Social Forces 70:3, March 1992 APPENDIX All the scales described below were constructed by adding the standardized values from component items. In the case of missing values for individual items, we substituted a value of 0 (the mean value) before performing addition. Scales Used for Seminary Faculty The numbers in parentheses or brackets indicate the original values assigned to the responses for each item. The responses are presented in the same order as in the questionnau+e. Conservative Theology (a=.79) For each of the teachings below, please indicate if you personally believe in them (possible responses are [2] Yes and [1] No): the existence of God, the divinity of Jesus, immortal life, final judgment, a place of eternal torment, the inerrancy of the Bible in everything it says, the Bible's infallibility in matters of faith and morals. Liberal Peace and justice (a-.91) A variety of areas in which public funds are expended are listed below. For each one, please indicate whether, given available resources, you think we are spending: (3) too much money on it, (1) too little money, or (2) about the right amount? a. The military, armaments and defense. The defense-spending debate remains a vigorous one. Do you believe that the fiscal 1982 level of military spending in tenns of actual purchasing power, controlling for inflation: (3) should be reduced, (2) kept as is, or (1) increased? Do you think the Reagan administration's plans for a military buildup: (3) go too far, (1) not far enough, or (2) are just about right? For each of the following, please indicate whether you think the US. should: (3) increase its cominitment, (2) keep its commitment at roughly the current level, or (1) reduce its commitment. a. Selling arms and giving military aid to countries that are against communism. (Scoring reversed for coding.) b. Negotiating with the Soviet Union to obtain arms control. Do you think the use of nuclear weapons is always: (3) morally wrong, (1) always morally permissible, or (2) dependent on the circumstances? How would you characterize the competition between the United States and the Soviet Union: fundamentally a struggle in power politics or fundamentally a moral struggle? Which is the greater problem in the world today: (2) repressive regimes aligned with the U.S., or (1) communist expansion? A variety of areas in which public funds are expended are listed below. For each one, please indicate whether, given available resources, you think we are (1) spending too much money on it, (3) too little money, or (2) about the right amount? a. Solving the problems of the big cities. b.foreign aid. c. Welfare.

19 Religiously Based Politics / 783 APPENDIX (continued) It is important to reduce the tule government plays in American life. (1) strongly agree, (2) agree, (3) disagree, (4) strongly disagree. For each of the following, please indicate whether you think the U.S. should: (3) increase its commitment, (2) keep its commitment at roughly the current level, or (3) reduce its commitment. a. Giving economic aid to poor countries to improve their standards of living. Economic growth is a better way to improve the lot of the poor than redistributing existing wealth: (1) strongly agree, (2) agree, (3) disagree, (4) strongly disagree. Overall, for each of the dimensions below, please indicate where you think the quality of government in the U.S. falls on a continuum, where given the human element, "1" represents the best possible situation, and "10" the worst possible situation. a. The responsiveness of government to social needs. In the US. today, would you rate the following as: (1) fully satisfactory, (2) generally satisfactory but needing amendment, (3) requiring major amendment, or (4) fundamentally deficient? a. The distribution of income. b. The distribution of power. Here are two ways to deal with inequality; which one do you prefer? Equality of opportunity, giving each person an equal chance for a good education and to develop his or her ability. Or equality of results, giving each person a relatively equal share of income and status regardless of education or ability. (Seven-point scale, with "1" labeled "equality of opportunity," "4" labeled "in between," and "7" labeled "equality of result.". ) Consernative Sexual Morality (cv.82) 1f a married woman does not want any more children, do you think abortion is: (1) moral, (3) immoral, or (2) morally neutral? In general, do you think homosexual relations in private between consenting adults are: (1) moral, (3) immoral, or (2) morally neutral? Do you think premarital sex is: (1) moral, (3) immoral, or (2) morally neutral? How about extramarital sex? (1) moral, (3) immoral, or (2) morally neutral? Do you think divorce is: (1) moral, (3) immoral, or (2) morally neutral? How about the remarriage of divorced persons while their ex-spouses are still living? (1) moral, (3) immoral, or (2) morally neutral? Scales Based on the General Social Survey Because of space considerations and fairly wide availability of the NORC GSS codebook our descriptions of scales based on the General Social Survey refer, in parentheses, to the NORC mnemonic names for items listed in Davis and Smith (1988).

20 784 / Social Forces 70:3, March 1992 APPENDIX (continued) Theological Conservatism (a=.88) The GSS items used in this scale involve whether respondents they view the Bible literally or otherwise (BIBLE); whether they believe in God without doubt (GOD); whether their faith is free of doubts (MYFATTH); whether it is important for a good Christian or Jew to go to church or synagogue regularly (GOCHURCH), to believe in God without doubts (BELIEVE), and to follow faithfully the teachings of one's church or synagogue (FOLLOW); whether when making important life decisions the Bible (DECBIBLE) and the teachings of one's church or synagogue (DECCHURH) are important factors in reaching a decision; and whether those who violate God's rules must be punished (PUNSIN). Liberal Peace and Justice (a=.66) These GSS items include respondent's views on federal programs designed to equalize income inequalities (EQWLTH) and help the poor (HELPPOOR and HELPNOT), the sick (HELPSICK), and minorities (HELPBLK). In addition, it includes a series of variables created by first combining two variables with slightly different wording that were asked to different subsets of the 1988 sample (see Davis & Smith 1988). These include items related to federal spending on defense (NATARMS and NATARMSY), welfare (NFARE and NFAREY), the environment (NATENVIR and NATENVIY), education (NATEDUC and NATEDUCY), foreign aid (NATAID and NATAIDY), big cities (NATCITY and NATCITYY), and assistance to blacks (NATRACE and NATRACEY). Conservative Social-Moral Issues (a=.83) These GSS items include five civil liberties items that concern whether atheists (SPKATH), racists (SPKRAC), communists (SPKCOM), militarists (SPKMIL), and homosexuals (SPK- HOMO) should be allowed to speak in the respondent's local community; three items focusing on women's roles (FEWORK, FEHOME, and FEPRES); seven items involving whether respondents they think a woman should be able to obtain a legal abortion for various reasons (ABDEFECT, ABNOMORE, ABHLTH, ABPOOR, ABRAPE, ABSINGLE, and ABANY); and four items concerning whether they disapprove of sexual relations between persons in their early teens (TEENSEX), and homosexual (HOMOSEX), premarital (PREMARSX), and extramarital (XMARSEX) sexual relations. References Blake, Nelson Manfred A History of American Life and Thought. 2d ed. McGraw-HilL Buell, Emmet, jr., and Lee Sigelman "An Army That Meets Every Sunday? Popular Support for the Moral Majority in 1980." Social Science Quarterly 66: Carroll, Jackson W., and Barbara G. Wheeler Study of Doctor of Ministry Programs: Summary. Hartford Seminary Center for Social and Religious Research. Connecticut Mutual Life Insurance Company The Connecticut Mutual Life Report on American Values in the '80s: The Impact of Belief. Research and Forecasts. Converse, Philip E "The Nature of Belief Systems in Mass Publics." In Ideology and Discontent, edited by David E. Apter. Free Press. Davis, James Allan, and Tom W. Smith General Social Survey [MRDF] and cumulative codebook National Opinion Research Center [producer], The Roper Center for Public Opinion Research [distributor].

21 Religiously Based Politics / 785 Fleishman, John A "Attitude Organization in the General Public: Evidence for a Bidimensional Structure." Social Forces 67: Greeley, Andrew M Religious Change in America. Harvard University Press. Guth, James L "Southern Baptist Clergy: Vanguard of the Christian Right?" Pp in The New Christian Right: Mobilization and Legitimation, edited by Robert C. Liebman and Robert Wuthnow. Aldine. Hadden, Jeffrey K The Gathering Storm in the Churches. Doubleday. Iannaccone, Laurence R "Heirs to the Protestant Ethic? The Economics of American Fundamentalists." In Remaking Politics, Militancy, and Economies, edited by Martin E. Marty. University of Chicago Press. Delen, Ted G "Religious Belief and Attitude Constraint " Journal for the Scienhjic Study of Religion 29: Johnson, Stephen D., and Joseph B. Tamney "The Christian Right and the 1980 Presidential Election." Journal for the Scientific Study of Religion 21: _ "The Christian Right and the 1984 Presidential Election." Review of Religious Research 27: Johnson, Stephen D., Joseph B. Tamney, and Ronald Burton "Pat Robertson: Who Supported His Candidacy for President?" Journal for the Scientific Study of Religion 28: _ "Factors Influencing Vote for a Christian Right Candidate." Review of Religious Research 31: Johnstone, Ronald L Religion and Society in Interaction: The Sociology of Religion. Prentice-Hall. Ladd, Everett Carll, and G. Donald Ferree, Jr "The Politics of American Theology Faculty." This World 1: Lawrence, Bruce B Defenders of God: The Fundamentalist Revolt Against the Modern Age. Harper & Row. Lipset, Seymour Martin Political Man. Free Press. Lipset, Seymour Martin, and Earl Raab "The Election and the Evangelicals." Commentary 71: Marty, Martin E Righteous Empire: The Protestant Experience in America. Dia! Press. Olson, Daniel V.A., and Jackson W. Carroll "Theological and Political Orthodoxy among American Theological Faculty." Paper presented at the annual meeting of the Society for Scientific Study of Religion, Salt Lake City. Roof, Wade Clark, and William McKinney American Mainline Religion: Its Changing Shape and Future. Rutgers University Press. Roozen, David A., William McKinney, and Jackson W. Carroll Varieties of Religious Presence: Mission in Public Lift. Pilgrim Press. Rothenberg, Stuart, and Frank Newport The Evangelical Voter: Religion and Politica in America. Institute for Government and Politics of the Free Congress Research & Education Foundation. Shupe, Anson, and William Stacey "The Moral Majority Constituency." Pp in The New Christian Right: Mobilization and Legitimation, edited by Robert C. Liebman and Robert Wuthnow. Aldine. Simpson, John H "Moral Issues and Status Politica." Pp in The New Christian Right: Mobilization and Legitimation, edited by Robert C. Liebman and Robert Wuthnow. Aldine "Social Moral Issues and Recent Presidential Elections." Review of Religious Research 27: Smith, Tom W "Classifying Protestant Denominations." GSS Technical Report no. 67. National Opinion Research Center "Classifying Protestant Denominations." Review of Religious Research 31: Stark, Rodney, and William Sims Bainbridge The Future of Religion: Secularization, Revival, and Cult Formation. University of California Press.

22 786 / Social Forces 70:3, March 1992 Varacalli, Joseph A Toward the Establishment of Liberal Catholicism in America. University Press of America. Walker, Helen M., and Joseph Lev Statistical Inference. Holt, Rinehart & Winston. Warner, R. Stephen New Wine in Old Wineskins: Evangelicals and Liberale in a Small-Town Church. University of California Press. _ Review of The Restructuring of American Religion: Society and Faith Since World War II, by Robert Wuthnow. American Journal of Sociology 94: Wilcox, Clyde "Evangelicals and the Moral Majority." Journal for the Scienhfic Study of Religion 28: Wuthnow, Robert The Restructuring of American Religion: Society and Faith Since World War 11. Princeton University Press. _.1989a. The Struggle for America's Soul: Evangelicals, Liberals, and Secularism. W.B. Eerdmans. _.1989b. "One Nation Under Whose God?" Hartford Courant. 12 March. Zwier, Robert Born Again Politics: The New Christian Right in America. Inter-Varsity Press.