Está en la página 1de 287

the disappearing god gap?

This page intentionally left blank


Religion in the 2008 Presidential Election
corwin e. smidt kevin r. den dulk bryan t. froehle james m. penning stephen v. monsma douglas l. koopman


Oxford University Press, Inc., publishes works that further Oxford Universitys objective of excellence in research, scholarship, and education. Oxford New York Auckland Cape Town Dar es Salaam Hong Kong Karachi Kuala Lumpur Madrid Melbourne Mexico City Nairobi New Delhi Shanghai Taipei Toronto With ofces in Argentina Austria Brazil Chile Czech Republic France Greece Guatemala Hungary Italy Japan Poland Portugal Singapore South Korea Switzerland Thailand Turkey Ukraine Vietnam

Copyright 2010 by Oxford University Press, Inc.

Published by Oxford University Press, Inc. 198 Madison Avenue, New York, New York 10016 Oxford is a registered trademark of Oxford University Press All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Library of Congress Cataloging-in-Publication Data The disappearing God gap? : religion in the 2008 presidential election / by Corwin E. Smidt . . . [et al.]. p. cm. Includes bibliographical references (p. ) and index. ISBN 978-0-19-973471-9; 978-0-19-973470-2 (pbk.) 1. Religion and politicsUnited StatesHistory21st century. 2. PresidentsUnited StatesElection2008. 3. United StatesReligion1960 I. Smidt, Corwin E., 1946 BL2525.D573 2010 324.973'0931dc22 2009026454

1357 9 8 6 4 2 Printed in the United States of America on acid-free paper


We want to acknowledge and express our appreciation for the generous support of the Lynde and Harry Bradley Foundation, which helped to underwrite substantially the costs of the Henry Institute National Survey on Religion and Public Life upon which much of our analysis is based. We also wish to thank those staffers of both presidential campaign organizations who willingly provided their insights and perspectives on the campaigns. And, nally, we wish to thank Ellen Hekman for her detailed and careful assistance in helping to prepare the manuscript for publication.

This page intentionally left blank


Introduction, 3 ONE Religion and Presidential Campaigns in Perspective, 18 TWO Religion and the Political Landscape in 2008, 42 THREE Religion and the 2008 Presidential Primaries, 72 FOUR Religion and the Summer Interlude, 107 FIVE Religion and the Fall Campaign, 134 SIX Religion and Election Day: Voter Mobilization in 2008, 164

SEVEN Religion and Election Day: Voting Patterns, 191 EIGHT The God Gap Revisited, 221 Notes, 234 References, 248 Index, 272



the disappearing god gap?

This page intentionally left blank


ver the past several decades, scholars have increasingly recognized that religion plays a vital role in American politics. The study of religion and politics has mushroomed from occasional analyses, largely ignored by the scholarly community, to a major subeld of study. This new scholarly attention has been especially focused on the role of religion in electoral politics. In the aftermath of the 2004 presidential election, for example, a number of different studies examined the relationship between attendance at worship services and vote choice (e.g., Campbell 2007; Rozell and Whitney 2007, chs. 25). Religiously observant voters were much more likely to vote for Republicans than Democrats. Thus, the God Gap was born, a label used to describe the tendency of those who are highly religious to vote Republican and those who are less so to vote Democratic. Or, as political commentator Michael Barone had earlier noted (Carnes 2004): Americans increasingly vote as they pray, or dont pray. By the 2008 presidential election, however, the political landscape appeared to have changed in some important ways. In the presidential primaries, Democratic candidates redoubled their efforts to appeal to religious voters; the Christian Right had become more fragmented, as longtime leaders such as Jerry Falwell and D. James Kennedy had passed from the scene; a new generation of evangelical voters appeared to be emerging, one apparently more environmentally sensitive and less reexively Republican in their preferences; and, nally, the ongoing conict in Iraq, rising oil prices, and a stagnant economy had largely pushed social issues off the table. Historically, religious afliation has been the most important factor in the relationship between religion and politics.1 Ones religious group or

faith tradition served as the basis for party afliation. Not surprisingly, different religious groups have predominated in party politics at different times, as religious groups have waxed and waned and the landscape has changed. But while the specics have changed, religion has remained a constant force in American politics. Most crucially, highly religious people were found across various political divisions. In the early 1900s, for example, a Roman Catholic and a New England Protestant could have similarly high levels of religious devotion and participation but still have vastly different political attitudes and partisan preferences. In fact, it is precisely their equally intense levels of participation in different religious traditions that may explain their political differences. Today, however, religion itself appears to be undergoing a transformation, and that transformation has implications for American electoral politics. While religious afliation was a key political factor in the past, over the last few decades a new cleavageone based on religious beliefs and values has emerged. The divide is no longer between (for example) Baptists and Episcopalians, but between religious traditionalists on one side and secularists and modernists on the other. In this religious order of American politics, membership in particular religious traditions is less relevant (Kohut et al. 2000; Green 2007). Traditionalists across religious faiths and traditions (e.g., Catholic, evangelical, Jewish) side with each other politically, while nontraditionalists, regardless of their religious afliation, nd common cause on the other side of the political divide. This book analyzes the role of religion in the 2008 presidential election in light of this apparent restructuring of religion. We provide context by assessing the role of religion in American political culture, as well as comparing the inuences of religion in the nomination process and the general election in 2008. We examine not only how voters respond to candidates on the basis of their religious characteristics, but also how candidates may seek to appeal to religious voters, foster relationships with them, and work to turn them out to vote. Our goal is to ascertain more fully the ways in which changes in religion over time relate to presidential campaigns.

The 2008 Election in Perspective

The political context of the 2008 election was both markedly different from and similar to the one held four years earlier. First, it was the rst time since 1952 that neither of the two major party nominees was a sitting president or vice president. In 2004, President George W. Bush was seeking reelection as the incumbent against his challenger, Senator John

The Disappearing God Gap?

Kerry. Given their ofce, sitting presidents seeking reelection have many weapons at their disposal, including access to free and extensive media coverage. As nonincumbents, neither McCain nor Obama had that kind of access in 2008. Second, the combination of two characteristics of American politics the two-party system and popular distrust of any concentration of power over timeaffected the electoral context differently in 2008 than in 2004. After eight years of one-party control of the presidency, there were many more voters in 2008 who were inclined to believe that it is time for a change than what is typically found after four years of an administration. Moreover, while the inclination toward periodic turnover in party power is common in American politics, the desire was amplied in 2008 by growing frustration with the war in Iraq, a third factor. Although the war was a matter of public debate during the 2004 presidential election, increasing numbers of Americans expressed discontent over the conict in Iraq in the years following Bushs reelection victory. The commitment of troops to Iraq was closely tied to the Bush administration, and President Bush ran for reelection in 2004 on the idea of staying the course in Iraq. However, the continued bombings that claimed innocent lives seemed a daily occurrence; casualties among American soldiers mounted; the Iraqi security forces did not appear to be able to assume command; and the government in Iraq did not appear to be stable. All these factors contributed to a growing dissatisfaction with existing American policy in Iraq. Fourth, partly as a result of this discontent with policy in Iraq, an important shift in political power occurred following the 2006 congressional election. Democrats won control of Congress in 2006, giving the party majorities in both houses on Capitol Hill for the rst time since 1994. This outcome suggested that the partisan tide in national politics may have turned toward the Democrats, and that the Democrats were well poised for victory in the 2008 presidential election. In addition to changing attitudes about Iraq, certain other political issues were markedly different between the 2004 and 2008 elections. Issues related to the economy were much more important as the 2008 campaign unfolded. For example, while energy policy had been discussed intermittently over the past several decades, the rapid rise in oil prices during 2007 and 2008 meant that transportation costs jumped dramatically between the two electionsincreasing the cost of the production and transportation of market goods, as well as leading to increased discontent over fuel prices. The economic standing of many families and important nancial institutions were even more adversely affected by the declining housing market and increasing numbers of home foreclosures

that led to a crisis in nancial markets and economic meltdown less than two months prior to the election. The focus of the campaign changed dramatically. Certainly, hot-button social issues (e.g., abortion and gay marriage) were still present, but their relative salience diminished substantially as economic issues moved front and center. Another important factor was that the two major candidates in the race for the 2008 Democratic presidential nomination were an African American (Barack Obama) and a woman (Hillary Clinton). While many women supported the Clinton campaign, the Obama campaign not only (eventually) solidied support among African Americans, it also attracted many young adults. Of course, none of these factors necessarily guaranteed a Democratic victory in the fall, but both campaigns mobilized new people into the political process. Indeed, participation in the Democratic primary process was at a record highonce again suggesting that the 2008 general election held great promise for the Democratic Party. Finally, a key religious movement waned in inuence in 2008. The leadership and organizational apparatus of the Christian Right, a political movement of religious conservatives long linked to support for the GOP, was clearly in disarray and decline. Many analysts still commented on the strength, if not the vitality, of the Christian Right in the 1996 or 2000 presidential elections, and clearly President Bushs reelection contributed to ongoing discussion of the relative strength of the Christian Right as well. But the Christian Coalition, historically the most powerful of the grassroots organizations associated with the Christian Right, has largely disappeared from the political scene and its organizational strength is now virtually nonexistent. In addition, many of the old leaders of the Christian Right, such as Jerry Falwell and D. James Kennedy, passed away prior to the 2008 election, creating a kind of leadership vacuum within the Christian Right. Many of the remaining leaders are aging (e.g., Pat Robertson) and lack the same sway among conservative Christian voters as in the past. On the other hand, each presidential election does not begin totally afresh. On the contrary, presidential elections are marked by a great deal of continuity, including the relative size of various voting blocs, the cultural expectations of presidential candidates, and the role of the two major parties in the election campaigns, as well as other factors. First, the relative size of major social groups in the electorate does not change markedly from one election to the next. This continuity matters because many of these social groups are linked to support for one or the other of the major political parties. For example, over the course of the past several decades, African Americans have been strong supporters of

The Disappearing God Gap?

the Democratic Party and its candidates. However, the proportion of African Americans within the total electorate changes only marginally over a four-year period. Thus, the proportion of African Americans in the electorate remained approximately the same in 2008 as it was in 2004.2 The same is true with regard to the proportion of all other major social demographic groups in the electorate, including women, Roman Catholics, and the college educated. Second, cultural values do not change signicantly in a four-year period. For example, cultural expectations related to the role of the president in American politics and the kind of qualities a president should possess remain basically the same across the span of two presidential elections. Similarly, while there were some shifts in attitudes toward religion in politics from 2004 to 2008, cultural expectations related to the role of religion in politics are generally unlikely to shift signicantly over a four-year period. Third, while the nominees of the major parties may change, the major parties that compete in presidential elections remain the same (with a few historical exceptions). The Democratic and Republican parties have long dominated the American political system. Some activists may be attracted to a party because of a particular candidate, but most activists who work for and contribute money to each party do so at least partly because of the partys historic policy emphases. Thus, the activist core of each party makes it difcult for either party to shift substantially in its policy positions. Given this continuity, voters are also largely familiar with the general emphases, priorities, and inclinations of both parties. As a result, neither major party is likely to shift substantially in their policy emphases and goals from one election to the next. Fourth, while candidates may change, the partisan identications of voters are typically stable. Research suggests that partisan identications tend to be formed relatively early in life and that, once formed, they become resistant to change. Consequently, the percentages of Democrats and Republicans in the electorate do not shift much over a short period of time. Moreover, many voters cast their ballots in line with their partisan identications, and thus the percentage of votes that the candidate of either major party is likely to receive falls within a certain range given the distribution of these partisan identications within the electorate. While these general patterns were somewhat upset in the 2008 election, the shifts in overall party identication and partisan voting were most pronounced among new voters. Older voters tended to hold their partisan loyalties. Fifth, over time, various social, economic, and religious groups develop ties to each of the political parties. Again, it is not that every member of

these particular groups necessarily votes according to these ties, but a sufcient number do so that conventional wisdom tends to associate such groups with supporting that particular party. For example, members of labor unions and Jews are linked to the Democratic Party, while business leaders and evangelical Protestants are linked to the Republican Party. These elements of continuity, among others, suggest that changes in the level of support for the two candidates between the 2004 and 2008 election were probably not as substantial as media portrayals of the 2008 election sometimes suggested. To be sure, the election was a landmark in many ways, as we will discuss. Yet the long view suggests that Obamas victory was not necessarily the result of some dramatic change in the electorate. The old political adage is that elections are won at the margins. A candidate does not have to substantially change the proportion of votes won among members of some social group in order to dramatically affect the outcome of an election. A shift of even 4 percent among women voters, for example, could result in a two-percentage-point national shift in support for the two major candidatesmoving a 51 to 49 percent result in favor of one party to a 51 to 49 percent result in favor of the opposing party. Thus, even with this substantial continuity from one election to the next, the results of elections can nevertheless change dramatically.

Perspectives on the Relationship of Religion to American Politics

What role, then, did religion play in both generating change and reinforcing continuity in the 2008 election? To answer that question, we need to address another question: in what ways does religion affect elections in general? We suggest there are at least two major ways religion may shape American electoral behavior. The rst approach focuses primarily on the mass public. Here religion is treated as an individual trait, and therefore as something that varies from individual to individual. Some members of the electorate are religiously afliated, while others are not, and those who are religiously afliated vary in terms of their specic religious faith traditions and denominations. Likewise, some members of the electorate hold particular religious belief strongly, while others may hold beliefs weakly or not at all. However, regardless of whether one analyzes religion in terms of specic religious beliefs or afliations, this approach suggests that religion functions largely as an internal mechanism that may shape ones political thinking and decision making.

The Disappearing God Gap?

The second way that religion shapes electoral politics is through political activists and elites using religion to generate candidate support and shape voter turnout. As we discuss in more detail below, different kinds of actors can engage in such activities, and various means can be employed in the effort to activate or intensify the salience of religion for political purposes. But regardless of the particular actors or methods involved, this approach focuses on external efforts to activate and mobilize religious voters.

Religion as Voter-driven Linkage

Historically, there have been two competing theoretical explanations for how religion serves as an internal mechanism in electoral politics: the ethnoreligious perspective and the theological restructuring perspective. The former emphasizes religious group afliations and draws more heavily on sociological perspectives; the latter emphasizes religious beliefs and values and draws more heavily from cognitive psychology. The ethnoreligious perspective adopts Emile Durkheims (1915) focus on religion as a social phenomenon, emphasizing the political implications of an individuals afliation with a religious group. The restructuring perspective traces its roots to Max Weber ([1930] 1992), who saw religion embodied in beliefs, which shaped political attitudes and behavior. More recently, something of a synthesis has emergeda perspective that views religion as embodying belonging (or afliation), beliefs, and behavior, with all three inuencing political life. Religious Belonging. Many pollsters, pundits, and politicians have relied implicitly on an ethnoreligious interpretation of American politics. As developed by historians, this theory identies the key religious groups as the historic denominations born in Europe and later multiplying on Americas shores. Presbyterians, Lutherans, Baptists, Episcopalians, Methodists, and myriad other Protestants combined distinct religious worldviews with other cultural attributes, such as ethnicity, race, or region. They were soon joined by other traditions, including Catholics, Jews, Eastern Orthodox, and other outside or minority religious groups at the time. All these religious groups developed their own political cultures, often in conict with neighboring groups. These cultures were fostered by religious leaders, houses of worship, and ethnic communities. In the nineteenth and early twentieth centuries American party politics involved competing alliances of ethnoreligious groups (Kleppner 1970, 1979; Jensen 1971; Formisano 1983; Swierenga 1990). For many historians, religion shaped American politics primarily through religious belonging, with partisan afliations and voting behavior

reecting political expressions of shared values derived from the voters membership in, and commitment to, ethnic and religious groups (Kleppner 1970, 35). Given a two-party system and high religious diversity, specic religious groups naturally sought like-minded allies to inuence American politics, as each religious group, no matter how large, needed allies if it wished to impact electoral politics. In the nineteenth century, a coalition of pietists (primarily Whig and Republican) faced an alliance of liturgicals (primarily Democratic), later joined by southern white Protestants as a result of the Civil War (Kleppner 1970 and 1979; Jensen 1971). By the mid-twentieth century, these coalitions had reorganized, but ethnoreligious loyalties remained at their base. Mainline Protestants3 provided both leadership and faithful voters to the Republican Party, while Catholics, Jews, black Protestants, and other religious minorities including out-groups such as southern evangelicalsconstituted the bedrock of the Democratic Party. As a result, early social science research on voting in the 1940s found substantial partisan differences pitting most Protestants against Catholics, Jews, and southern evangelicals, even with the class-based politics of the New Deal supposedly dominant (Lazarsfeld, Berelson, and Gaudet 1944; Berelson, Lazarsfeld, and McPhee 1954). Despite the historical value of the ethnoreligious model, some observers argue that it has lost relevance in contemporary politics. In particular, the underlying bases for ethnoreligious politicsthe powerful social integration within religious traditions, the social isolation of those traditions, and the strong tensions among traditions (Kleppner 1979) have largely vanished. Nevertheless, while some of the foundations of the ethnoreligious model may have dissipated, ones religious group afliation may still continue to be politically important today. First, the political behavior of certain close-knit religious groups like black Protestants, Latino Catholics, Jews, and Latter-day Saints suggests that the model may have some staying power. And second, even afliation with a church in the historic evangelical, mainline Protestant, or Catholic traditions may still matter politically, in part because Americans tend to perceive membership as elective today, allowing believers to choose a congenial religious and politicalenvironment (Green and Guth 1993). Religious Beliefs. Although many analysts still focus on religious tradition, variously dened (Manza and Brooks 1999; Steensland et al. 2000; Layman 2001; Leege et al. 2002), some sociologists argue that ethnoreligious descriptions of religious life and its implications for politics have less utility today than in the past. As ascriptive afliations4 break down and as geographic mobility has increased, Americans move more freely among


The Disappearing God Gap?

gure i.1. A county-level mapping of the 2008 presidential election. White sections are the counties that voted for McCain; gray sections are the counties that voted for Obama.

religious settings, ignoring the historic ties of doctrine, denomination, ethnicity, region, and even family (Ammerman 1997). As people re-sort themselves into congenial theological environments, religion has been restructured into two camps with opposing worldviews, fostered by competing religious institutions and leaders. As Robert Wuthnow (1988) and James Davison Hunter (1991) have argued, old religious traditions have been polarized by theological, social, and cultural conicts into a conservative, orthodox, or traditionalist faction on one side, and a liberal, progressive, or modernist one on the other. For some theorists, the growing number of secular Americans is a natural extension of the liberal or progressive sideand may even be the product of struggles over restructuring (Hout and Fischer 2002). Wuthnow explored the split primarily within religious institutions, but Hunters apocalyptic title, Culture Wars, projected the divisions into the polity, as a threat to social stability. Much of the media attention to the 2004 presidential election reinforced this perspective, suggesting that the electorate is cleaved in two, as represented by ubiquitous electoral maps that starkly distinguish red from blue states. Despite Obamas relatively strong victory, those cultural distinctions appeared to persist in the 2008 election, as suggested in the county-level map in gure i.1. Although scholarly reaction to the culture wars thesis has often focused on these purported political manifestations (Williams 1997;


Fiorina, Abrams, and Pope 2005; Nivoli and Brady 2006), Wuthnow and Hunters original formulations were rooted in theological changes that had led to the emergence of two competing, largely opposing, worldviews. The competing camps were characterized by alternative belief systems, different religious practices, and adherence to rival religious movements.5 Indeed, the identication of these competing forces probably constitutes the most valuable insight of the restructuring perspective. Although critics are rightly skeptical about extreme statements of the restructuring theory, evidence for a milder version is convincing, especially in older American religious institutions. The religious press reveals continual battles between traditionalists and modernists in almost every major Protestant body, as well as in the American Catholic Church. Although rooted in theology and practice, these struggles also produce opposing moral, social, economic, and political perspectives. To be sure, culture war theorists overstate the consequent polarization, both within religious institutions and the mass public; there are centrists in the religious wars, and moderates in the political wars. But the religious divisions they identify may well inuence politics, if only because both religious and political elites are polarized, thus shaping the cues presented to the public (Guth et al. 1997; Fiorina, Abrams, and Pope 2005). Conclusion. Over the past two decades, there has been growing evidence that a new religious order in American politics may be emerging. However, though these changes have diminished the effects of the old religious order, they have not fully eliminated them. Hence, both orders are currently operating. But the question is whether this coexistence will continue, or whether one such order will come to dominate and replace the other order. It is highly unlikely that one election can provide the complete answer to this question, but the 2008 election may well provide some clues related to the issue.

Religion as Elite-driven Linkage

Clergy may encourage the faithful to vote on Election Day; religious interest group activists may issue and distribute candidate scorecards; candidates for public ofce may appear before some religious groups but not others; political leaders may frame their messages to target certain religious groups. Clearly, political actors can engage in different kinds of activities to try to mobilize voters on Election Day, as well as to try to dissuade other voters from turning out (Leege et al. 2002). Briey, political elites may adopt four different means by which to use religion to mobilize voters. First, candidates may choose to appeal or not


The Disappearing God Gap?

to appeal to particular religious groups by posturing themselves in particular ways. Religious posturing simply refers to a candidate providing some modicum of attention to a religious group, some recognition of the particular concern(s) of such a group, or some identication with a group without necessarily embracing the group or even its political positions. For example, posturing may mean a candidate chooses to appear with certain religious leaders but not with others.6 It may mean that a candidate goes on political pilgrimages (Domke and Coe 2008, 7580) by visiting people or places that have an elevated signicance for a religious group, and, in so doing, build appreciation within the targeted group with this display of solidarity. Second, political candidates and public gures may engage in religious signaling (Domke and Coe 2008). This may involve the use of symbols that are more familiar to certain kinds of voters than others, or it may be some more direct form of communication. Regardless, the intent is to signal its intended audience that the person shares or reects certain values with them. For example, Bill Clintons campaign in the 1992 presidential election was based on a new covenant with the American people. The word covenant is a term imbued with rich religious meaning, something that religious voters would easily recognize. Similarly, through most of President George W. Bushs years in ofce, evangelical Michael Gerson served as his speechwriter, and he was well acquainted with religious language that resonated with his religious base of voters. For example, in Bushs 2003 State of the Union Address, the president evoked an easily recognized and quite famous line from an old gospel hymn when, speaking of Americas deepest problems, President Bush said, The need is great. Yet theres power, wonder-working power, in the goodness and idealism and faith of the American people.7 And candidate Obama frequently referenced an awesome God, a phrase contained in a song frequently sung within contemporary Christian services. In each of these examples, a candidate or president provides a recognition of or identication with a groupwithout conveying any specic policy positions or necessarily embracing that particular group. A third way political elites may seek to activate voters is by framing their political messages in particular ways that may appeal to religious voters. Though related to religious signaling, framing is an analytically distinct endeavor. Here the candidate chooses to discuss issues through the particular analytical lenses that reect the viewpoint of the particular group(s) from which one is seeking support. For example, the issue of gay marriage might be viewed as a moral issue or an issue of equality under the law, and candidates are free to choose whichever lens or framework they


wish to address the issue. In so doing, political candidates (as well as elected ofcials) work hard to provide cues to voters about themselves that relate to the core values of which they as members of particular groups care deeply. For many Americans, these core values are grounded in faith, with elections thereby becoming moral referendums (Domke, Shaw, and Wackman 1998). Finally, in addition to these other efforts, political elites may engage in direct efforts at mobilizing religious voters to go to the polls. For example, Barack Obama worked with ministers of African-American congregations to engage in a massive voter registration drive because he believed the black church and its members could be decisive in a tight presidential race. So, too, McCain, who has had an uneasy relationship with evangelicals, nevertheless sought to replicate the success of George W. Bush in getting out the white evangelical vote. Moreover, in addition to candidates seeking to mobilize religious voters, many religious interest groups choose to engage in the political process by distributing information to voters as a means of inuencing voting decisions and policy (e.g., Wilcox and Sigelman 2001; Guth et al. 2007).8 Quite a range of moral or religious interest groups have been active in recent elections, covering the breadth of the liberal-conservative political continuum. As we discuss in chapter 6, many new such groups were formed between the 2004 and 2008 presidential elections, particularly on the more liberal end of the political spectrum. Participation in congregational life also fosters the building of personal communication networks that are associated with political participation, and during recent presidential elections there has been a great deal of political discussion among members of religious congregations (Guth et al. 1998, 2002, 2007). Personal contacts are one of the most important and effective means of eliciting participation, and campaign organizations and political parties routinely use face-to-face and telephone contacts to persuade and mobilize voters (e.g., Eldersveld 1956; Huckfeldt and Sprague 1992; Wielhouwer and Lockerbie 1994; Gerber and Green 2000a, 2000b, 2001; McClurg 2004). Moreover, those who are most likely to be contacted are those already somewhat predisposed to participate and to vote for their partys candidates (namely, those who are registered to vote and are socially connected to their communities). The net result is that Democrats mainly contact Democratic-oriented groups, while Republicans focus on Republican-oriented groups (Rosenstone and Hansen 1993; Gershtenson 2003; Wielhouwer 2003; Panagopolous and Wielhouwer 2008). Thus, the Bush campaign and allied groups made the mobilization of evangelical Protestants a high priority in 2004 (Monson


The Disappearing God Gap?

and Oliphant 2007), while Democrats undertook similar efforts among black Protestants. After all, houses of worship provide a natural venue for organizing people for many purposes, including politics. It is clear that churches continue to be, at least for some traditions, particularly important venues through which political contacts are made and participation encouraged (see also Brewer, Kersh, and Petersen 2003).

Religion and the 2008 Presidential Election

It is sometimes said that two things one should not discuss with guests across a dinner table are religion and politics. This adage partly reects the fact that many people take both matters very seriously and, as a result, to raise either topic can invite conict. Still, religion is far more important to most Americans than politics: nearly a majority of Americans say that religion is the most important thing in their life, but almost no one makes the same claim about politics. This book analyzes the role of religion in the 2008 presidential election and the extent to which it was an important factor shaping the outcome of the election campaign. It differs from previous examinations in that, rather than focusing simply on how religion shapes voting choices on Election Day, it examines the role of religion throughout the whole course of the campaignrevealing the many ways in which religion comes into play during the election campaign. Any contention that religion played only a minor role in shaping the outcome of the 2008 election adopts far too narrow a view of presidential elections. For example, religion played a role in shaping the selection of the nominees of both political parties. Clearly Romneys Mormon faith and the relationship of Obama to his controversial pastor were matters of public discussion and contention. In addition, for the rst time in decades, Democrats made a concerted effort to reach religious voters, including evangelical Protestants. The early Democratic religious forums and the fact that each major Democratic candidate had a religious outreach staffer as part of his or her campaign organization made religion a part of the Democratic story from the beginning of the election campaign. Democratic candidates were much more prone during the 2008 campaign to talk about their religious faith and how it shaped their approach to public policy. Thus, any assessment of the role of religion in the 2008 election must take a broader view of the election than simply the voting of religious groups on Election Dayit must examine both primary and general elections in terms of how campaign teams targeted their religious messages and structured their organizations


to reach out to different religious groups. Only when the vote totals on Election Day are understood within the context of the total campaign effort can one properly interpret the role of religion in election results. This volume also provides a historical and cultural context for understanding the role that religion plays in presidential election campaigns, and it provides comparative analyses with the 2004 presidential election in order to assess the extent to which religion shaped election outcomes similarly or differently across the two elections. As we noted earlier, elections are conventionally said to be won at the margins. In other words, small changes in voting patterns from one election to the next can alter the election outcome signicantly. When alterations in power occur, attention is typically paid to the new and what has changed across the two elections, and certainly changes in religious voting patterns may have helped to account for such a change in power. However, the ip side of the fact that elections are won at the margins is that, even with signicant changes in electoral outcome, there is likely to be far more continuity than change in voting patterns across the two elections. After all, even though the outcome of the 2008 election was substantially different than that of the 2004 election, Obama won only 5 percent more votes in 2008 than Kerry did in 2004. Consequently, the basic patterns of how religion undergirded the voting results in 2004 likely continued to be largely present in 2008, with only marginal changes having occurred across the two elections that contributed to the substantial change in results. Thus, an analysis of religion in the 2008 election provides an important vantage point from which to assess both the change and the continuity evident across the two elections. Finally, one could argue that the 2008 presidential election provided the perfect storm in which to test the extent to which religion might undergird American election results. After all, prior to the banking crisis that captured public attention in late September, polling suggested it would be a very close election contest, with McCain actually enjoying a slight postconvention edge. Yet if religion is still a major factor in structuring voting decisions in an election that is centered on economic issues (e.g., the fallout of a major economic crisis, the loss of jobs, the demand for universal health carenot to mention an unpopular war) rather than social issues, then such a result would suggest the central importance of religion in shaping American election outcomes. This study, particularly the analysis of voting on Election Day, will be based largely (though not exclusively) on the Henry Institute National Survey on Religion and Public Life. This study is a national survey of 3,002 Americans conducted April 8May 10, 2008, followed by those


The Disappearing God Gap?

persons being re-interviewed following the presidential election (November 525, 2008).9 However, our analysis of the role of religion in the 2008 election, and in American presidential elections more generally, will not be conned to this sole source of data. Our analysis will draw upon other sources of information as well as a variety of other surveys conducted both prior to, and during, 2008 in order to support our contentions. These surveys typically contain a multiplicity of religious measures that enable the analyst not only to assess with great precision the religious characteristics of the respondents but also to assess the particular features of religious life that contribute most strongly to the religious divide politically.




Religion and Presidential Campaigns in Perspective

Our form of government has no sense unless it is founded in a deeply felt religious faithand I dont care what it is. President Dwight Eisenhower

lthough the u.s. constitution forbids the imposition of any formal or legal religious tests for public ofce, that has not prevented Americans from imposing informal religious tests on candidates especially candidates for the White House. Indeed, few members of unconventional religious denominationsand no avowed atheistshave been elected president. And few are likely to be elected in the near future for, although the United States is religiously pluralistic, it is indeed highly religious. Most voters throughout American history have preferred a religious presidential candidate, though they are cautious about members of religious groups that are considered out of the mainstream. This fact was not lost on either of the major party candidates in the 2008 election. As we will show later, both the McCain and Obama campaigns actively sought to convince religiously motivated voters that the candidates religious commitments were both sincere and acceptable. In comparison to other Western countries, this level of religious openness among political elites is unusual. Political leaders in most European democracies, for example, rarely speak publicly about their faith commitments, largely because the public does not expect them to do so. Even those political parties with a religious identity in their namesthe

Christian Democrats in Denmark, Finland, Germany, and the Netherlands, for exampleno longer expect their party candidates to draw explicit connections between their faith and their politics. This European difdence about religion contrasts with the major American parties and their standard-bearers. Both presidential candidates in 2008 not only worked to show the consistency between their particular policy goals and the values of various key religious groups but both also took pains to demonstrate that their positions were actually based in religious teaching and that their lives were shaped by religious faith. Why is the United States different from its closest peers? The United States has a distinctive political culture that encourages religious openness among political elites and organizations. Many Americans insist that religious believersincluding political leadershave a right to advocate for their principles and values in the political process. Indeed, for some religious believers, advocating such principles and values from a religious point of view is not merely a right but an obligation of responsible citizenship. Moreover, many citizens see religion as an indicator of a candidates character, a marker of what kind of leader he or she will be. In response, candidates often frame their political preferences or character in religious terms, which inevitably makes the religious faith of those candidates fair game for public consideration. Consequently, news media, citizen groups, and individual voters often take an interest in a candidates religious life, which may include the intensity of a candidates personal faith and the politics of a candidates house of worship. In this chapter we provide context to the role of religion in the 2008 presidential election. First, we provide a historical perspective by briey examining the role of religion in previous presidential campaigns. In doing so, we not only seek to recognize that religion has permeated American presidential elections from the founding of the Republic but also seek to understand the different ways in which religion may arise in presidential election campaigns. Second, we examine several enduring features of religion in American political culture in order to place the 2008 presidential election within a broader cultural context. By political culture, we mean a set of values and dispositions that are widely shared among citizens. Political scientists are interested in political culture because it can affect the quality of citizen interactions and the responsiveness of government to citizen needs.1 Dispositions such as trust and tolerance, for example, increase the likelihood that citizens will work with others on matters of common concernan indispensable feature of healthy and stable democracies (Putnam

Religion and Presidential Campaigns in Perspective


2000). As a result, social scientists and others concerned about good democratic practice have examined a wide range of associations and institutions that shape political culture, including family, education, friendships, work, andmost importantly for this bookreligion.

Religion and American Presidential Elections in Historical Perspective

Any detailed examination of the role of religion in presidential elections throughout American history would take us well beyond the connes of this chapter. However, we wish to provide a brief historical overview in order to differentiate between several different ways in which religion can come into play within such elections. Understanding them will help provide a context for understanding the role religion played in the 2008 election. We therefore briey review several past presidential elections and suggest that these elections can be grouped into three categories based on the role religion played in them: (1) elections where religion became an issue because one of the candidates was seen as being out of the mainstream of American religious life, (2) elections where religion became an issue because a candidates religious afliation was used to signal the moral or immoral character of that candidate, and (3) elections where religion became an issue because it was used by one or more candidates to appeal to and mobilize voters.

Opposition to Candidates of Outside Religious Groups

At the time of the founding, American religious diversity was largely within the Protestant wing of the Christian faith tradition. Over time, this diversity has expanded considerably. Despite this diversity, there has always been an evolving cultural hegemony, with some religious groups being seen as religious insiders and others as religious outsiders. In this section we consider two presidential elections where one of the candidates religion became an issue because it was seenor his opponents attempted to portray itas being out of the mainstream of American religious life. The Presidential Election of 1800. In the rst years of our new republic, a major religious controversy erupted in the presidential election of 1800, which pitted the sitting president, John Adams, against Thomas Jefferson. Jefferson, while religious and even Christian in a very broad sense, denied many of the basic teachings of Christianity, such as the divinity of Jesus Christ and the miracles recorded in the Bible. This placed him outside the


The Disappearing God Gap?

mainstream of the then prevalent Protestant Christianity. While John Adams did not subscribe completely to the tenets of traditional Protestant orthodoxy either, Adams, with his roots in the Puritan tradition of New England, was certainly less theologically adventurous than Jefferson. The resulting attacks on Jeffersons theology were intense. Alexander Hamilton referred to Jefferson as an atheist, and Adams is reported to have referred to Jefferson as an enemy of Christianity (McCullough 2001, 113). Dr. John Mason, one of the foremost preachers of the day, published a pamphlet before the election titled The Voice of Warning to Christians, in which he denounced Jefferson as a profane philosopher and an indel (Parton 1874, 5). Another pamphlet, written by a lay critic, charged that Jefferson is no Christian! He does not believe in the deluge. He does not go to church and asked whether Jefferson, who denies the truth of Christianity, and avows the pernicious folly of all religion [should] be your governor? (6). In spite of these attacksor, perhaps, in part because of themJefferson won. But this early election stands as a testimony to the fact that when one presidential candidate is perceived to be outside the religious mainstream, his (or her) faith is likely to become an issue in the campaign. The Presidential Election of 1928. Anti-Catholicism has deep roots in American culture. During the nineteenth and early twentieth centuries, members of the Protestant establishment tended to view Catholicism, with its hierarchical structure, emphasis on compliance with authority, and lack of democratic governance as being in sharp conict with American values (Massa 2003). In 1928, the Democratic Party chose Al Smith, governor of New York, to be its nominee in the presidential election. Smith was a Roman Catholic, the rst member of his church to win a major American partys presidential nomination. In view of the United States history of anti-Catholicism, it is not surprising that Al Smiths nomination generated considerable controversy across the United States. Even seemingly moderate Protestants were wary of the Smith candidacy. Peel and Donnelly (1931, 63), for example, report that a majority of Methodists and Baptists were opposed to Smith and that Presbyterian and Lutheran ministers voiced anti-Smith attitudes to their congregants. Asserted Methodist bishop Adna W. Leonard of Buffalo, No Governor can kiss the papal ring and get within gunshot of the White House (Moore 1956, 47). Fear of Smiths Catholicism was epitomized by the rumor that the pope had his bags packed and was ready to move to Washington after a Smith victory (Boller 1984, 225). Things got so bad that Republican candidate Herbert Hoover felt the need to distance himself from the religious critics, asserting that such
Religion and Presidential Campaigns in Perspective


(anti-Catholic) sentiments give violence to every instinct I possess (Peel and Donnelly 1931, 63). However, the attempts of Hoover, the Smith campaign, and such groups as the Knights of Columbus proved unable to douse the anti-Catholic ames. Matters came to a head when the Atlantic published An Open Letter to the Honorable Alfred E. Smith, challenging Smith to discuss the church-state issues raised by his Catholicism. Smith, who was certainly no theologian and preferred not to discuss the nexus between religion and politics, felt no choice but to take up the challenge. Smiths reply, printed in the Atlantic, portended later addresses of John F. Kennedy and Mitt Romney. In it, Smith asserted his belief in religious liberty, separation of church and state, and opposition to outside (papal) interference in American politics and government. Nevertheless, Smith lost the election and his Catholicism was interpreted as being a major factor in his loss. It was another thirty-two years before a major political party dared nominate another Catholic for president (John F. Kennedy).

Religion as a Sign of Moral Character

The Presidential Election of 1884. The 1884 election pitted Grover Cleveland, the Democratic candidate from New York, against James G. Blaine, the Republican candidate from Maine. It was a campaign marked by unprecedented personal attacks and mudslinging. Although Cleveland himself was a Protestant, the increasingly prominent role Catholics were playing in the Democratic Party became an issue and was used to challenge the moral character of Cleveland. His clean reputation was also soon sullied by charges that he had fathered a child out of wedlock. Equally serious were reports that the child had been removed to an orphanage and that the mother was living in an asylum. These were serious charges in Victorian America (Summers 2000, ch. 11). The Cleveland campaign responded with unusual candor, admitting the charges and pointing out that Cleveland had nancially supported his child. However, this did not prevent Blaine supporters from trumpeting the now-famous taunt, Ma, Ma, wheres my Pa? Gone to Washington, Ha, Ha, Ha (Brodsky 2000, 86). Clevelands supporters, in turn, countered with charges of political corruption on Blaines part, including claims he had sold his inuence while in Congress for large sums of cash (for which there was some persuasive evidence). Blaines supporters, however, sought to counter these charges by tapping into the virulent anti-Catholic feelings of many Protestants. The opposition of Protestants and Republicans to Catholics in the late nineteenth century was largely rooted both in different moral visions and


The Disappearing God Gap?

in more practical political and economic concerns. Protestants tended to see the growing numbers of Catholics as posing a cultural threat to such practices as refraining from the use of alcohol and gambling and as a political threat to religious liberty through an anticipated growing papal inuence over American government and a weakening of Anglo-Saxon, Protestant political hegemony. In addition, the demands of Catholics for public support of parochial schools was seen as threatening to fragment the American educational system and to drain the existing, Protestantdominated, public school system of needed revenues (Kleppner 1979, 215217). These religiously based tensions culminated in the nal week of the campaign when the Rev. Samuel Burchard, the leader of a group of New York ministers, declared at a campaign rally that we are Republicans, and dont propose to leave our party and identify with the [Democratic] party whose antecedents have been rum, Romanism, and rebellion (Summers 2000, 282). In this one phrase Burchard sought to link Cleveland with alcohol consumption, Roman Catholicism, and the southern states that had precipitated the Civil War by leaving the Union less than a quarter of a century earlier. Though Cleveland won in a very close election, the campaign still lives as a clear example of an election where religion played a prominent role by being used to attack or defend the candidates moral character. The Presidential Election of 1928. We saw earlier that the 1928 Democratic candidate, Al Smith, generated widespread controversy because of his afliation with the Catholic Church. As a result, we classied that election as one where religion played a major role because one of the candidates was from a religious faith deemed to be outside mainstream American religious life. But that election also shares characteristics of an election where religion became an issue because a candidates religious afliation was used to signal the moral or immoral character of that candidate. Al Smith was not only a Catholic but also an opponent of the prohibition of alcohol. While passage of the Eighteenth Amendment, prohibiting the manufacture and sale of alcoholic beverages, was a victory for the prohibitionist forces in 1919, it by no means ended the debate over Prohibition. The Republican Party tended to be in the dry (pro-Prohibition) camp and, before the campaign, Hoover had labeled Prohibition a great social and economic experiment, noble in motive and far reaching in purpose, arguing that it must be worked out constructively (Peel and Donnelly 1931, 225; Boller 1984). As a result of Hoovers position, dry Republicans and members of the Prohibition Party tended to endorse Hoovers views.
Religion and Presidential Campaigns in Perspective


Democrats tended to be somewhat more divided than Republicans on the issue of alcohol, although their 1928 platform contained a dry provision. However, this party platform statement did not prevent Smith from adopting a very different position from his partys declared position. In a speech in Milwaukee, Smith maintained that Prohibition was a farce, that it promoted disrespect for all law, and that it was the main cause of racketeering (Peel and Donnelly 1931, 58). Smith called for fundamental changes in the provisions for national prohibition, sparking the erce wrath of the Anti-Saloon League, the Womens Christian Temperance Union, and the Prohibition Party. Indeed, Smiths opponents labeled him Al-coholic Smith and spread the story that Smith was once so drunk in public that he had to be held up by two men. Some of the criticism of Smith took the form of a rather subtle whispering campaign in which Smiths opponents attempted to link together Smiths alleged corruption (Tammany Hall connections), opposition to Prohibition, and Catholicism, suggesting that, collectively, they demonstrated Smiths immorality and lack of tness for the White House. And it went far beyond a whispering campaign when the then-famous evangelist Billy Sunday denounced Smith supporters as damnable whiskey politicians, bootleggers, crooks, pimps and street walkers (Menendez 1977, 41). In the end, it is clear that Smith lost the election in large part because of his Catholicism and the more general linking of his religion to the corruption of big city political machines and the use of alcohol.

Religion as a Vehicle for Voter Mobilization

The 1980 election, in which Ronald Reagan of California defeated the incumbent Democratic president, Jimmy Carter, illustrates the third type of religious involvement in presidential elections: elections where religion became an issue because it was used by one or more presidential candidates to appeal to and mobilize voters. Jimmy Carter had been elected president four years earlier as a born again, evangelical Christian active in his Southern Baptist church. But in ofce he had disappointed many of his fellow evangelicals, who saw him as too liberal on both economic and cultural issues. This created an opening for the Rev. Jerry Falwell and others, who created a number of new organizations that were part of the new Religious Right.2 The most well known of these organizations was the Moral Majority, dedicated to supporting pro-life, pro-family, pro-moral, and pro-American causes and candidates (Martin 1996, 201; Diamond 1998, 6371). The Moral


The Disappearing God Gap?

Majority worked to register more evangelicals and other pro-moral voters and to inform them of the stances taken by candidates on issues of special concern to them, including abortion, pornography, homosexuality, and secular humanism (Martin 1996, 201; Diamond 1998, 67). Ronald Reagan, in his 1980 campaign, worked successfully to appeal to conservative religious voters. When he appeared in August 1980 at a Dallas meeting sponsored by the Religious Roundtable, a highly conservative group, Reagan, in an often quoted statement, said: I know this is non-partisan, so you cant endorse me, but I want you to know that I endorse you (Martin 1996, 216). Throughout the campaign Reagan appeared before similar conservative religious groups and spoke their language. Meanwhile, Jerry Falwell and other religious gures crisscrossed the country bringing their message that mixed Christianity, politics, fears of where the country was heading, and support for Reagan and other Republican candidates. And television producers were unable to resist putting Falwellwith his quick tongue, charismatic personality, and often extreme statementson their interview shows at a time when the Religious Rights organizations were few in number, and it was plausible for a single organization to claim to represent the movement as a whole (Diamond 1998, 67) . While the Moral Majority claimed to have registered four million new voters, other estimates placed it around two million voters (Lipset and Raab 1981), as thousands of recently engaged voters worked in their churches and neighborhoods to register and turn out voters for Ronald Reagan, an effort that paid off on Election Day (e.g., Guth 1983, 3537; Smidt 1988).

Religion and American Presidential Elections in Cultural Perspective

Studies of political culture often use cross-national comparisons to illumine both continuities and differences in the public role of religion across the globe (Inglehart 1990; Smidt 2003). While many patterns emerge in these analyses of political culture, the United States often stands out as an exceptional case. Already in the 1830s, Alexis de Tocqueville, the astute French social observer and author of the classic Democracy in America, noted the key civic role of religion in the United States in contrast to Europe. Today, the United States continues to defy trends in many other democratic and economically developed societies, where religious belief and behavior have waned considerably over the past century (Pew Research Center 2002; Greeley 2003; Fowler et al. 2004).
Religion and Presidential Campaigns in Perspective


These patterns are evident in table 1.1, which displays a comparison of the religious beliefs and practices of the American people with residents of Canada and nine economically developed countries in Europe. These data are drawn from a cross-national survey of religion conducted in 1998. Included in the study were residents of historically Protestant (Germany, Great Britain, the Netherlands, Norway, and Sweden) and historically Catholic (France, Italy, Poland, and Spain) countries. The results of this study clearly reveal that Americans are distinctive in their overall religiosity. Americans are much more likely to report weekly church attendance than those who reside in other Western countries; only those who live in Poland are more likely than Americans to attend church weekly. Moreover, while weekly church attendance tends to be somewhat higher in historically Catholic than Protestant countries, Americans are still much more likely than either Italians or Spaniards to attend church weekly. When compared to those Protestant countries with the highest levels of church attendance (Great Britain and the Netherlands), Americans are more than twice as likely to attend church weeklyand they are more than four or ve times more likely to do so than Germans, Norwegians, or Swedes. Such distinctiveness is not limited to worship attendance; it is also present in private prayer life and religious beliefs. Nearly half of the American people report that they pray daily; almost no other country examined
table 1.1 The Distinctiveness of American Religious Life
Believe in Attend Weekly Country Canada France Germany Great Britain Italy Netherlands Norway Poland Spain Sweden United States (%) 19 13 8 14 29 14 7 37 27 6 32 Pray Daily (%) 19 17 10 15 32 24 14 40 23 10 47 Life after Death (%) 70 52 35 14 29 60 53 77 56 51 81 Believe in Miracles (%) 55 38 50 59 72 40 38 64 46 27 79 Considers Self Very Religious (%) 11 7 8 7 14 17 9 16 14 4 26

Source: International Social Survey Program: Religion II, 1998.


The Disappearing God Gap?

comes close to this level (Poland, with two-fths praying daily, ranks second). For nearly every comparison, Americans are more than twice to nearly ve times more likely to pray daily than those in other countries. The same holds true in terms of religious beliefs. Four out of ve Americans report that they believe in life after death and that they believe in miracles, ranking rst among the eleven countries. Given these patterns, it is no surprise that Americans are also the most likely to label themselves as very religious. Clearly, then, when compared to its closest peers, religion in America is particularly vibrant and distinctive.3 For many Americans, however, religion is not only relatively robust in their private lives but also touches nearly every area of public life, including the electoral process (see, e.g., Verba, Schlozman, and Brady 1995; Putnam 2000; Smidt et al. 2008).

Church and State, Religion and Politics

Just as American religiosity is distinctive, so, too, are American beliefs and practices regarding the proper relationship of church and state (Monsma and Soper 2008). Some of that distinctiveness stems from our ongoing constitutional experiment with religious liberty (Witte 2000). In addition to the prohibition of religious tests in Article 6 of the U.S. Constitution, the First Amendment in the Bill of Rights requires that Congress make no law respecting an establishment of religion, nor prohibiting the free exercise thereof. It is no surprise that the U.S. Constitution addresses the relationship of church and state in such a prominent place. The original purpose of the Religion Clauses of the First Amendment was to ensure that the national government did not meddle in how states dealt with religion. In fact, many states persisted in formal establishments of religion well into the nineteenth century, which meant that government supported particular churches with special taxes and other advantages. By the early twentieth century, however, the U.S. Supreme Court had interpreted the Religion Clauses to apply broadly to all levels of governmentnational, state, and local. Today, most citizens believe that the Religion Clauses erect, in Thomas Jeffersons famous words, a wall of separation between church and state, even though they do not always agree on the height of that metaphorical wall or the precise reasons for its existence ( Jelen and Wilcox 1995). Throughout American history, religious believers have played a role in the debate over how far the separation of the church from the state ought to be, and sometimes in surprising ways. Some of the strongest proponents
Religion and Presidential Campaigns in Perspective


of rather strict separation of church and state have been religious believers themselves, especially those who are in the minority. The earliest examples included Baptists and Quakers in the colonial era. They argued that their own religious freedom would be jeopardized if the state supported other religious groups through special taxes or forced oaths. In other words, religion needed to be protected from the state, not the state from religion. Indeed, decades before Jefferson spoke of the metaphorical wall, Roger Williams, founder of Rhode Island and a prominent Baptist theologian of the seventeenth century, called for a hedge or wall of separation between the garden of the church and the wilderness of the world (Williams 1963).4 In the twentieth century, groups associated with another religious minorityJewsbecame some of the most prominent advocates for a separationist perspective. Such groups as the American Jewish Committee, American Jewish Congress, and the Anti-Defamation League have raised their voices against direct state support for religious schools or faith-based social services, among other church-state issues (Sorauf 1976; Ivers 1995). Of course, precolonial Baptists and modern-day Jews are not the only groups intensely focused on church and state, nor do they hold a monopoly on how much distance to place between the two. Many religious leaders and advocacy organizations take a different approach from strict separationists, insisting that, while the state ought not to fund houses of worship in their purely religious activities, it can and should be supportive of a range of faith-based activities and ideas (Brown 2002; den Dulk 2006). These accommodationists argue, for example, that religion should be treated no differently than other groups in terms of state funding when religious groups provide a service to the public good such as education or social services (for various perspectives, see Monsma 1996; Dionne and Chen 2001; Fowler et al. 2004, ch. 8). These differences notwithstanding, even when they disagree about specics, most citizensseparationist or accommodationist, religious or nonreligiousconcur that the state ought not be free to support a religious institution in any way it wishes. The vast majority of citizens agree that the Constitution imposes healthy limits on church-state interaction; the debate is about the nature and scope of those limits, not whether they ought to exist. This separation of church and state, however, does not necessarily imply the separation of religion and politics. The former focuses on how institutions of government interact with religion through public policy, while the latter focuses on how religious believers and faith-based organizations engage the political process. In fact, religion and politics are so


The Disappearing God Gap?

intertwined that even the strictest separation of church and state could never fully disentangle the two. The constitutional status of state support for religionfor example, money for religious schools or social services, displays of religious symbols on public propertyhas been the subject of a great deal of controversy. But to say that government ought to avoid direct support for religion is not the same thing as saying that religious believers or groups ought to set aside their faith when participating in public life. Moreover, even the highest wall of separation of church and state cannotand ought notprevent citizens from casting their ballot based on their religious beliefs or their religious identity. Religious tests for ofce can further illustrate the difference between church-state separation and the separation of religion and politics. Despite explicit language forbidding the practice in the U.S. Constitution, at various points in our history, both federal and state governments have barred atheists, agnostics, and members of minority religious faiths from ofce. In modern history, however, courts have roundly rejected these religious tests, arguing they involve the government in playing favorites among religious groups and therefore clearly violate the assumptions of church-state separation (see, e.g., Torcaso v. Watkins 367 US 488 [1961]). Nevertheless, even though courts have vigorously enforced the constitutional ban on religious tests, the same principle does not apply to a private citizen. Citizens are free to enter a voting booth and cast a ballot for whatever reasons they choose, including a candidates religion. Americans are much more sanguine about the religiously motivated political engagement of private citizens and groups than they are about the state supporting (or inhibiting) religion. Over three-quarters of American citizens today say that religious groups should stand up for their beliefs (Smidt et al. 2005), suggesting a high level of comfort with religion-based political mobilization. Religion also appears to play a role in the decision making of voters. In 2004, for example, nearly half of Americans said that their personal faith was an important factor in their vote choice, and in 2008 more than two-fths did so.5 It is not surprising that Americans would claim such a prominent role for religion in their political choices. Nearly three-quarters of Americans (73.8 percent) say that religion is an important part of their life, and well over half (56.2 percent) claim that it provides either quite a bit or a great deal of guidance in their day-to-day decisions.6 The difference between church-state and religion-politics separation, however, can be quite subtle and complex. Consider American attitudes toward clerical endorsements of candidates. Clergy are private citizens, so one might expect the public to support their freedom to engage in
Religion and Presidential Campaigns in Perspective


political activity. Yet as table 1.2 demonstrates, only about a quarter of American citizens believe that clergy should be permitted to endorse presidential candidates during religious servicesa percentage that is more or less consistent across all major religious traditions in the United States. Even African American Protestants, with their long history of direct political advocacy from the pulpit (Harris 1999), are generally uneasy about explicit clergy endorsement during worship services. Just over half of black Protestants say that endorsements should not be permitted, in comparison to the third who would allow them (the remainder have no opinion).7 If Americans are generally supportive of allowing religion into political participation, why are they resistant to clergy endorsements? The answer lies with the perception that clergy endorsements cross both a constitutional and cultural line and that there are both prudential and religious reasons for refraining from such endorsements from within houses of worship. From a legal and prudential point of view, endorsement during religious services might well threaten a churchs tax-exempt status, which forbids this kind of advocacy on behalf of a particular candidate. For these believers, then, the question is not whether the faithful (including clergy) can be politically active; rather, the question relates to what repercussions may be associated with violating a law that seeks to keep church and state separate and ensure that the state does not tax the church. In fact, as long as clergy do not act under the authority of a religious institution (e.g., speaking as a pastor from a pulpit), they are entirely free to endorse candidates as private citizens.8 From a cultural point of view, many believers view religious services as a place for gaining spiritual wisdom and sustenance, with direct political advocacy from clergy diminishing the power of the experience. Tocqueville (2003) once made the counterintuitive observation that both the political and spiritual inuence of clergy in the United States increases when they
table 1.2 Clergy Endorsement by Religious Tradition, Spring 2008
Religious Tradition Evangelical Protestants Mainline Protestants Black Protestants Catholics Jews Religiously Unafliated Total Disagree (%) 55 60 51 63 71 53 57 No Opinion (%) 16 12 15 12 9 19 15 Agree (%) 29 28 34 25 20 28 28

Source: Henry Institute National Survey of Religion and Public Life, April 2008.


The Disappearing God Gap?

are perceived to be the least political. In other words, the power of religion is stronger when it is not dragged down by the common perception of politics as self-interested, divisive, or even corrupt. In fact, in a survey taken just before the national party conventions in 2008, a slight majority of Americans (52 percent) agreed with the statement that churches and other houses of worship should keep out of political matters (Pew Forum 2008d). This result reected a small, but important, shift of opinion from earlier years. From 1996 to 2006, opinion on the question had been rather stablewith a majority (51 to 54 percent) holding that churches should express their views and a minority (ranging from 43 to 45 percent) contending that churches should stay out of politics. Interestingly, this reversal was largely due to a shift among conservatives, many of whom appeared to change their minds about the role of churches in political debate. The 2008 survey suggested no signicant difference between political liberals and conservatives in their response to the survey question. To be sure, some of this sentiment, especially among conservatives, was undoubtedly shaped by the widely publicized controversies surrounding the Reverends Jeremiah Wright and (to a lesser extent) John Hagee just a couple of months before the conventions. Yet these data, coupled with most Americans rejection of clergy endorsement, suggest that many Americans share a general concern about the role of religious institutions in the political process, even while they see an important role for religion in individual decision making. While it can be subtle and complex, the distinction between government support of religion and religious involvement in politics is important. The sociologist Robert Bellah (1967, 34) puts it this way: The separation of church and state has not denied the political realm a religious dimension. Although matters of personal religious belief, worship, and association are considered to be strictly private affairs, there are, at the same time, certain common elements of religious orientation that the great majority of Americans share. These have played a crucial role in the development of American institutions and still provide a religious dimension for the whole fabric of American life, including the political sphere. It is important to note that Bellah made these observations in a landmark study of presidential rhetoric, which tends to be steeped in religious symbols. Presidents speak openly in religious terms because citizens themselves are accustomed to a role for religion in their own political thinking and in the political culture as a whole.
Religion and Presidential Campaigns in Perspective


Political Culture and the Religion of Candidates

The religious identity of leaders themselves frequently becomes a campaign issue. Even some of our most celebrated presidentsThomas Jefferson and Abraham Lincoln, for examplewere widely and sometimes viciously criticized for their apparent lack of religiosity. To some extent, candidates themselves, who seek support from religious groups, have generated this level of scrutiny. Savvy candidates use cultural and religious values to marginalize their opponents supporters and mobilize their own. While this can enliven public discourse and legitimately focus public attention on the character aws of candidates, it can also produce highly contentious and polarizing campaigns. The historical record reveals an ambivalent public. On the one hand, voters clearly favor religious candidates. On the other hand, those candidates expose themselves politically if their religion is too far from the mainstream. What counts as mainstream, however, can change over time. The contrast of the 1928 and 1960 elections is a good illustration of a religious minorityin this case, Catholicismachieving sufcient cultural status to see one of its own elected to the presidency. While Al Smith, the Democratic candidate in 1928, lost in part because of public concerns about his Catholicism, a little over thirty years later John F. Kennedy won the presidency despite his Catholic heritage. This ambivalence about a presidents religion continues to the present day. Mass opinion prior to the 2008 elections indicates the continuing linkages between favorable political perceptions of candidates and perceptions of a candidates religiosity. Citizens were very unlikely to have favorable impressions of a candidate if they thought the candidate lacked sincere religious faith. Typically, at least 70 percent of Americans reported that they agreed with the statement It is important that the president should have strong religious beliefs (Smidt et al. 2005, 441; Pew Research Center 2007). This sentiment varies somewhat by party afliation, with Republican Party identiers being somewhat more likely to express such views than either independents or Democrats. Nevertheless, as can be seen from table 1.3, basically two-thirds of independents as well as two-thirds of both weak and strong Democratic Party identiers stand in agreement. Not surprisingly, there is a stronger relationship between level of church attendance and the importance of a president holding strong religious beliefs than between partisan identication and such assessments. This is due to the fact that many independents and Democrats are regular church attenders. Nevertheless, there remains a cultural expectation that the president should hold such religious beliefsas seen by the fact that


The Disappearing God Gap?

table 1.3 Importance of the President Having Strong Religious Beliefs

by Partisan Identication and Level of Church Attendance Agree (%) Partisan Identication Strong Republican Weak Republican Independent Weak Democrat Strong Democrat Church Attendance Weekly or more Once or twice a month A few times a year Seldom Never 86 73 63 52 37 3 3 4 5 3 12 24 33 43 61 100 100 100 100 101 (1213) (484) (540) (497) (284) 88 76 64 63 66 3 3 5 2 3 9 21 32 35 31 100 100 101a 100 100 (428) (352) (1131) (399) (566) Uncertain (%) Disagree (%) Total (%) (N )

Source: Religion and Public Life Survey, 2007.


Percentages over 100 due to rounding.

nearly two-thirds of those who attend church only a few times a year still agree with the statement, as do a majority of those who report that they seldom attend, and more than a third who never attend. Whats more, a sizable percentage of Americans are open to explicit candidate expressions of religiosity. At least half (50 percent in 2008, 56 percent in 2004) of Americans disagreed with the statement It makes me uncomfortable when candidates discuss their faith (Smidt et al. 2005, 441; Pew Research Center 2007), and a plurality in August 2008 (36 percent) actually wanted more expression of religious faith and prayer from the presidential candidates (Pew Forum 2008d). Of course, these statistics raise a chicken-egg question puzzle: do citizens who already favor a particular candidate simply assume that the candidate has socially desirable traits such as religiosity? Or does the religiosity of a candidate generate a citizens favorable attitude toward that candidate? Studies of the role of political parties in presidential elections suggest that the answer to both questions could be yes. A Republican is less likely than a Democrat to perceive candidates from the Democratic Party as religious, and vice versa. In polling at the beginning of the primary season, for example, only 45 percent of Republicans (compared to
Religion and Presidential Campaigns in Perspective


87 percent of Democrats) thought Hillary Clinton was religiousa gap of 42 percentage points. A similar pattern existed for all the other candidates, though the gap was less pronounced among Republican candidates (Pew Research Center 2007, 6). This suggests that partisanship shapes perceptions of religiosity; the greater the intensity of party attachment, the greater the level of religiosity attributed to a candidate from the respondents party. Yet there is some counterevidence suggesting that religion has an effect independent of partisanship in creating perceptions of candidates. For example, among those partisans who do acknowledge that candidates of opposing parties are religious, there are much higher levels of politically favorable perceptions of such opposite-party candidates (Pew Research Center 2007, 8). This suggests that religiosity does increase favorability regardless of party identication. In addition, the specic faith of a candidateevangelical, Catholic, Jewish, and so onmatters less than the simple fact that a candidate is perceived as religious. The epigraph to this chapter, in which President Eisenhower appears to suggest that religion in general is important to public life, but that the specic form does not matter, could apply to the attitudes of many voters.9 For American voters, atheism is the chief pariah. Perhaps the best evidence of this is that citizens generally say they are less likely to cast their vote for atheists than Muslimseven in the wake of the terrorist attacks of September 11, 2001, and its resultant negative effects on mass attitudes about Muslims. A Pew Forum survey in 2003 indicated that, while 38 percent of Americans would be reluctant to vote for a Muslim candidate, 52 would have the same reluctance about an atheist. More recent polls have uncovered similar patterns in public opinion, with as many as 60 percent of voters claiming that they would be less likely to vote for a candidate if he or she were an atheist (Keeter and Smith 2007). Still, perceptions of a candidates specic form of religious faith can have an impact on the candidates electoral prospects. Even though voters prefer a religious candidate to an atheist, they also appear to have a hierarchy of preferable religious groups. Hence, in the nineteenth and early twentieth centuries, Catholics were rarely even considered as potential presidential or vice presidential nominees, partly because of their perceived status as being out of the mainstream. Today there is considerably greater reluctance in the electorate to vote for a Muslim or Mormon candidate than, say, an evangelical Protestant or a Catholic (Pew Forum 2003; Keeter and Smith 2007). These perceptions, of course, are strongly associated with the size of the religious traditions; in general, the larger the tradition, the


The Disappearing God Gap?

higher it sits on the electorates hierarchy of religions. On the surface, there appear to be some interesting exceptions to this general rule. Evangelical Protestants comprise about a quarter of the population, compared to less than two percent for Mormons, Jews, atheists, and Muslims. Yet as table 1.4 shows, the percentage of American voters who reported that they would be less likely to choose a candidate if he or she were an evangelical was not far behind the percentages for these other minority faiths. Still, the effect is mitigated by the fact that a higher percentage of voters claimed they would be more likely to vote for a candidate if he or she were evangelical. Mormons and Muslims have no such advantage, with double-digit percentage point differences between those less likely and more likely to vote for them. Atheists fare even worse, with an astounding percentage point gap of well over half the voters polled. A particularly powerful picture of the independent effect of a candidates religion emerges when controlling for party and candidate qualications. Even if a party nominated someone perceived as qualied for the job, many members of that party would vote against that nominee on the basis of religious identity. This is evident in table 1.5, which examines the responses of the American people to a hypothetical question in the summer of 2008. In this Newsweek poll, it is clear just how the particular faith tradition of the partys presidential nominees might shape support for nominees from within their own party. While nearly nine in ten respondents indicated that they would vote for their partys nominee if such a person was a Jew, only three-quarters indicated they would vote for a Mormon, and less than half would do so if their partys nominee were a Muslim or an atheist.

table 1.4 Willingness to Support a Presidential Candidate by Candidates Religious

Afliation: 2007 More Likely Candidates Religious Afliation Atheist Muslim Mormon Evangelical Christian Jew Roman Catholic More Likely (%) 3 3 5 19 9 13 Less Likely (%) 61 45 25 16 11 7 Less Likely (Percentage Points) 58 42 20 3 2 6

Source: Religion and Public Life Survey, 2007.

Religion and Presidential Campaigns in Perspective


table 1.5 Willingness to Support a Qualied Presidential Candidate of Ones

Party by Candidates Religious Afliation: 2008 Candidates Religious Afliation Jew Mormon Muslim Atheist Yes, Would (%) 89 74 48 46 No, Would Not (%) 11 26 52 54 (N) (985) (971) (962) (992)

Source: Newsweek, Obama and God Survey, July 910, 2008.

Perhaps the best actual, rather than hypothetical, example in the 2008 presidential campaign of this type of hierarchy of preferences was the controversy during the early primaries over Mitt Romneys Mormon faith. In late 2007, the public perceived Romney as the most religious of all candidates for president from either major party, but that did not give him an edge.10 Despite the fact that just over half of Americans claimed they had a favorable view of the Mormon Church (formally, the Church of Jesus Christ of Latter Day Saints), a quarter responded that they would be less likely to vote for a candidate if he or she were Mormon (Keeter and Smith 2007). Romney was compelled to respond to this unease in a December 2007 speech entitled Faith in America. While he did not distance himself from his faith, he focused intensely on commonalities between his values and mainstream culture. He also argued, as had John F. Kennedy in the 1960 campaign, that religious authorities in his church would never exert an inuence on his own judgments as president, thereby seeking to assure voters that he was committed to the general principle of separation of church and state. It is instructive to compare Romney to Mike Huckabee at that point in the campaign. In late 2007, Huckabees star was rising after some strong polling results and extensive media coverage. Why? There are undoubtedly many reasons, but one argument is that Huckabees own folksy evangelicalism played better in mainstream American culture, especially in southern states. Moreover, evangelicals who attended church frequently were the most likely of all religious groups to say that Romneys Mormon faith would negatively affect their vote choice (41 percent said they would be less likely to vote for a candidate who is Mormon). One could make a similar argument about McCains pick of Sarah Palin as a running mate. Palin was widely heralded as a likeminded soul by evangelical voters, who were clearly energized by her entry into the campaign. We will discuss more fully the role of Romneys, Huckabees, and Palins religion in later chapters, but for present


The Disappearing God Gap?

purposes the contrast of these candidates illustrates how Americans want their presidents to be religiousas long as that religiosity is mainstream. Obama, too, had to face questions about the radicalism of his Christian church, as well as rumors that he was heavily inuenced by Islam in his childhood. In fact, as we explore more fully in chapter 4, Obamas life experience reects a religious eclecticism that resulted from a highly mobile and unconventional upbringing. His biological father was raised a Muslim but became an atheist later in life; his stepfather was a Muslim who nevertheless borrowed liberally from other faiths; his mother, as Obama describes her, was spiritual, but she had a healthy skepticism toward organized religion. He attended a Catholic day school for several years, as well as an Indonesian public school with mostly Muslim students and teachers. As a young man, he became a convert to Christianity and began attending Trinity United Church of Christ, a largely black congregation in Chicago. He was also heavily inuenced by his close work with Catholics as a community organizer. His membership at Trinity became the center of a political storm when controversial statements by Rev. Jeremiah Wright, Trinitys former senior pastor, were made public. Wright took three issues that are explosive by themselvesrace, religion, and politicsand combined them to criticize American politics and culture in an animated tirade from his own pulpit. Like Romney, Obama was compelled to respond in a major speech. While the episode temporarily diminished his favorability ratings among voters, his level of likely votes did not dip (Pew Research Center 2008b). Obama did resign from the church, however, as the controversy swirled for several weeks, partly because Wright reiterated his ideas several times in major news outlets.

Why Religion Shapes Perceptions of Candidates

Why does the religion of the president seem to matter to voters? One reason is institutional. The U.S. government vests power in the president as both head of administration and head of state. Unlike other countries where these two roles are split, the president both administers the government and performs many of its ceremonial and symbolic roles. American constitutional development has only reinforced these multiple roles, so that today the American presidency has become arguably the preeminent institution of American government. Quite understandably, elections for such an important position have attracted the attention of citizens with various political and policy interests, including religious ones.
Religion and Presidential Campaigns in Perspective


The institutional role of the president as head of state lends itself to the theory of a distinctively civil religion in the United States. Bellah (1967) rst articulated this idea (at least in the U.S. context) in an essay comparing the rhetoric of Presidents Kennedy and Lincoln. He argued that presidents often invoke ideas and engage in rituals that are essentially religious, even when they do not make any reference to Christianity or other traditional religious expressions. Hence, in the United States, there are beliefs (e.g., a special manifest destiny for the country, or that the United States is a city on a hill for other countries to emulate), rituals (e.g., the Fourth of July, the State of the Union address), and sacred spaces (e.g., the Washington Memorial, Gettysburg) that are part of a distinctive civil religion. Presidents benet when they are seen (metaphorically) as the high priests of this civil religion. They preside over periods of national mourning and triumph, and are often judged on the sincerity of their emotion; they are entrusted with protecting national traditions and institutions, and risk loss of support when they overlook or trivialize them.11 They must also be careful to avoid appearing too unconventional, which would be tantamount to the heresy of being unpatriotic or out of step with mainstream values. Moreover, in our religiously pluralistic culture, presidents often nd themselves strenuously avoiding any hint of favoritism toward their specic religious faith. Another reason citizens care about the religion of the president is related to public policy. Some might believe that a president who is committed to a particular religion will be more likely to agree with them on salient policy issues. For example, prolife voters might think it is more likely that a high-attending evangelical or Catholic candidate will agree with them on abortion than a candidate with lower levels of religiosity. Sometimes this is put in partisan terms: in recent elections, for example, the GOP has been widely perceived as more friendly toward religion than Democrats, including on those values issues that seem to animate many religious believers (Pew Forum 2003, 1314; Pew Forum 2007).12 The connection between policy, a candidates specic faith, and a voters perceptions can be overstated. First, both voters and candidates are rarely in agreement with everything their religious institutions teach. Second, the most salient issues to voters in a given yearthe economy, for example may not have a particularly religious cast. Finally, the correct or most prominent policy view of a religion may not be clear. Consider debates in recent decades over U.S. policy in the Middle East. For sincere religious reasons, many evangelicals insist that the United States should strongly support Israel and take a hard line against its adversaries in the region. But other evangelicals believe, also for sincere religious reasons, that we ought


The Disappearing God Gap?

to have a more balanced policy in the Middle East. Evangelical leaders share these disagreements (for a discussion, see den Dulk 2006). These cautions notwithstanding, the interconnection of religion, policy, and presidential politics can be important, especially when issues deal with morality. Many political scientists have noted that religious voters who perceive an issue in moral terms are more likely to view that issue as salient in a campaign. Moral crusades have been a trademark of American politics, dating back at least as far as the early Puritans (Fowler et al. 2004). Whether this is because Americans are a uniquely religious people or because they tend to adopt Manichean views of reality, Americans have frequently responded with enthusiasm to the dramatic appeals of moral crusaders. Thus, for example, the prohibition movement used religion and religious terminology in its campaign against drinking alcoholic beverages (demon rum, as it was described in the 1884 presidential election). Indeed, as we mentioned in the introduction, some scholars have even suggested that such values voters are at the center of a culture war that is often fought on the battleground of presidential politics. While the importance of these issues was somewhat diminished in the 2008 presidential election, we argue later that they still had a profound effect on blocs of voters who mattered at the margins. A nal reason that the religion of a presidential candidate might matter to a voter has less to do with policy and more to do with the candidates personal attributes. Political scientists have long known that voters take into account their own perceptions of the personal virtue or compassion of a candidate (see Barker, Lawrence, and Tavitz 2006). Table 1.6 reveals some of the key qualities of the candidates in the 2004 presidential election.

table 1.6 Most Important Quality of Presidential Candidates in 2004 Election

Alla (%) Cares about people Religious faith Honest/trustworthy Strong leader Intelligent Will bring change Clear stand on issues (9) (8) (11) (17) (7) (24) (17) Bush (%) 24 91 70 87 9 5 79 Kerry (%) 75 8 29 12 91 95 29 Nader (%) 1 b 1 0 0 0 0

Source: National Election Pool et al. 2004.


All respondents regardless of their particular candidate preference; bLess than 1 percent but greater

than zero.

Religion and Presidential Campaigns in Perspective


The voters religion can shape these perceptions. Not surprisingly, for example, voters who attend religious services at least once a week constitute 77.5 percent of voters who saw religious faith as the most important quality of a presidential candidate. Voters from that group were also the most likely (43.9 percent) to say that the top quality is honesty and trustworthiness.The assumption for many of these voters seems to be that personal morality has public relevance.

In a controversial 2004 interview about his religious faith, thenstate senator Barack Obama explicitly connected civil religion to religious neutrality: Alongside my own deep personal [Christian] faith, I am a follower, as well, of our civic religion. I am a big believer in the separation of church and state. I am a big believer in our constitutional structure. . . . I am a great admirer of our founding charter, and its resolve to prevent theocracies from forming, and its resolve to prevent disruptive strains of fundamentalism from taking root in this country. As Ive said before, in my own public policy, I am very suspicious of religious certainty expressing itself in politics. (Waldman 2008c) For Obama, civil religion includes beliefs about conventional religion itself. As an individual, his specic faith in Christianity is deep yet personal, and he implies that religion plays some role in shaping his politics and policy preferences. But from a public perspective, he suggests religion needs to stay within its proper bounds. Theocracies, religious certainty in the political realm, or even simply disruptive forms of traditional religion are a threat in our political culture, and hence some undened level of separation of religion and politics is warranted. Obama articulates here the widespread values of American political culture. For most Americans, religion is deep and personal. But while the personal nature of belief does not stop them from using religion to guide their day-to-day practice, they are generally suspicious when religion is used to disrupt day-to-day life or to support certain specic, concrete public policy positions. In such cases, it is seen to improperly intrude into the domain of the state. Those who use religion to argue for political change are often viewed with some suspicion. The message is that, in effect, religion ought to be a stabilizing force; it becomes problematic when it challenges the status quo.


The Disappearing God Gap?

Nevertheless, when religion is used successfully to disrupt the status quo, it frequently comes later to be accepted as an appropriate tool by which to bring about change. The Reverend Martin Luther King and the civil rights struggle of the 1960s is instructive here. King used religious beliefs and values to argue uncompromisingly for the equality of all persons and for that equality to be reected in the nations public policies. Barack Obama and almost all Americans today support the appropriateness of King doing so, in spite of Kings use of religious certainty to support public policies that disrupted the status quo. This suggests that the acceptance of religion as a basis for certain public policy stances may be rst rejected when it challenges a prevailing consensus, but that it may later come to be accepted after such change has been achieved. In the nal analysis, then, the role of religion in shaping perceptions of presidential candidates is subtle and complex. In fact, it reects the complexity of political culture itself. On the one hand, religion is a key part of identity for many Americans, and so it is inevitable that they will use religion as a guide to their own choices in presidential elections. On the other hand, many Americansindeed, many Americans with a deep religiosityare also uneasy with religion that seems too closely mixed with specic public policy positions that challenge the prevailing public policy consensus. The test for presidential candidates is to campaign in ways that appear to balance these two orientations toward religion in politics. In a religiously pluralistic culture, it is a formidable challenge.

Religion and Presidential Campaigns in Perspective



Religion and the Political Landscape in 2008

he political landscape is much like the physical landscape. Geographical terrain rarely changes markedly from one day to the next. Yes, change occurs at the margins; wind or water can erode the terrain slowly over time. But the broad contours and basic outlines typically remain the same, with day-to-day change largely undetectable to the naked eye. To be sure, some catastrophic events can permanently alter even this fundamental geological outline, but such dramatic occurrences are relatively rare. Nevertheless, in the context of such stability and constancy, even change that is incremental can appear signicant, if it is observed. So it is with the electoral landscape. No presidential election starts totally afresh from the previous contest. The long-standing partisan loyalties, political values, and policy positions held by voters provide certain parameters within which the new presidential election campaign is waged, placing important constraints on the extent to which the political landscape may change from one election to the next. Erosion does occur as international crises, domestic challenges, new political issues, and candidate choices modify the contours of the landscape. But the basic terrain does not typically shift markedly from one election to the next. Nevertheless, there is also a longstanding political adage that elections are won at the marginsin other words, small changes in the percentage of votes cast by different groups can lead to a fundamental change in the result. Thus, even minor changes can be politically signicant. Consequently, to understand the role of religion in the 2008 presidential election, we must start with the political landscape of the 2004 election.

We begin this chapter with an assessment of the relative size of different religious groups within the American electorate as well as an examination of their voting patterns in 2004.

Religious Groups and Political Thinking

It is no exaggeration to state that American electoral politics is conducted largely in terms of group politics. To forge a winning coalition, candidates for public ofce and political parties invariably think about which social and demographic groups are safely part of their base of supporters, which are rmly entrenched in the camp of their opposition, and which groups are up for grabs. While candidates and political parties seek to attract the votes of individuals, they frequently attempt to win such votes by viewing and appealing to them as members of a group. Religion constitutes a fundamental category of identity and association within American life (Warner 1993, 1059). As a social phenomenon, religion is expressed through afliation with a local church or other house of worship, a specic denomination, or a religious tradition. By virtue of their afliation with a particular social group, individuals come to share certain common experiences. Through patterns of association and interaction, as well as variation in religious teachings about the way religion is linked to politics, members of different religious groups acquire divergent political attitudes and behavior. They may experience distinctive patterns of communication, receive different kinds of information, be exposed to varying interpretations of political events, and be subject to different political recruitment and mobilization tactics. As a result, citizens respond to political stimuli differently based on their membership in different religious groups. As some have argued: [Religious] belonging can matter in politics by providing a forum in which religion can be linked to political issues, parties, candidates and activities (Kohut et al. 2000, 13). In this sense, however, religious groups are not distinctivethey simply function like other social groups.

Religious Traditions
Since the 1820s, many successful presidential campaigns have forged winning coalitions out of various enthnoreligious groups.1 During much of this time, members of the historically dominant mainline Protestant churches, such as Episcopalians, Presbyterians, and Methodists, have aligned with the Republican Party. On the other hand, religious minorities,
Religion and the Political Landscape in 2008


such as Roman Catholics, Jews, and evangelical Protestants (especially southern evangelicals), have aligned with the Democratic Party. These cleavages continued even through the New Deal of the 1930s. By the 1980s, however, these religious alignments were undergoing important changes. The relative numbers of mainline Protestants had declined, the partisan preferences of evangelical Protestants had shifted toward the GOP, and the historic support of Catholics for the Democratic Party had diminished. Black Protestants had become a crucial Democratic bloc of voters, while growing religious diversity through immigration added increasing numbers of Muslims, Hindus, Buddhists, and others to the equation, usually on the Democratic side. Nevertheless, even with these changes, many analysts continue to think in ethnocultural terms, referring to the evangelical vote, the Catholic vote, the Jewish vote, or the Muslim vote. Over the past several decades, many scholars have employed the concept of religious tradition to capture these ethnocultural categories and to classify Americans religiously, as religious traditions constitute a useful and increasingly popular conceptualization of religious belonging (Layman 2001, 60). A religious tradition comprises religious denominations, movements, and congregations that exhibit similar beliefs and behaviors and which are interrelated in some historical and organizational fashion (Kellstedt and Green 1993; Kellstedt et al. 1996). Members of a religious tradition exhibit a characteristic way of interpreting the world, based on common beliefs and practices, though not all members necessarily hold these particular beliefs or exhibit these behaviors. Religious afliation has proved to be to be a powerful predictor of political attitudes and behavior (Kellstedt and Green 1993; Kellstedt et al. 1996, 1997, 2007; Kohut et al. 2000; Layman 2001; Layman and Green, 2005; Guth et al. 2006; Smidt 2007; Green 2007; Green et al. 2007). So what are these religious traditions? Within Americas overwhelmingly Christian population, one can differentiate among evangelical, mainline, and black Protestant traditions, the Roman Catholic tradition, and the Eastern Orthodox tradition.2 And with growing religious pluralism, surveys reveal Americans of non-Christian traditions ( Judaism, Islam, Buddhism, Hinduism), although usually in numbers too small for extensive analysis. Finally, the unafliated population (Hout and Fischer 2002) can be regarded as a tradition in its own right, although it includes several types, such as unafliated believers (unattached to a church or denomination but exhibiting at least modest levels of religiosity), the nonreligious unafliated, and nally, those who describe themselves as either agnostics and atheists (Green et al. 2007; Kellstedt 2008).


The Disappearing God Gap?

For purposes of analysis in this book, we will sometimes examine those afliated with six specic religious traditions: evangelical Protestafnts, mainline Protestants, black Protestants, Roman Catholics, Jews, and those who are religiously unafliated (or secular)though at other times the analysis will also examine Hispanic3 Protestants, Hispanic Catholics, and other faiths as separate categories. While the use of the six particular traditions does not capture the full range of all religious traditions, the decision to limit the analysis to these six categories is based on several considerations. First, every researcher faces the practical matter of balancing the desire to capture the full range and complexity of American religious life with the need to present such information in a clear and manageable way: to examine all religious traditions and their subtraditions creates unmanageable tables in terms of the presentation of data, so we have opted to focus primarily on the major religious traditions within American religious life. Second, despite restricting analysis of the religious tradition variable at times to these six specic groups, we have not sacriced too much breadthas the overwhelming majority of Americans are included in the analysis in that these groups capture approximately 9094 percent of the American population, depending on the particular survey employed. Finally, scholars have given the most attention to the six traditions in previous elections, thereby providing us with points of comparison in our analysis of the 2008 campaign.

Religion and the 2004 Presidential Election

Immediately following the 2004 presidential election, there were commentators, pollsters, and political gures who claimed that religious factors had sealed George W. Bushs electoral victory over John Kerry.4 Much of this initial commentary was based on exit poll results that revealed that more voters had chosen moral values as the basis for their vote than any other factor. This claim was quickly contested, though subsequent and more rened analyses revealed that religiously motivated values voters did contribute to GOP success in the 2004 election (Campbell 2007; Rozell and Whitney 2007). On the surface, it was clear that Bush had won by substantial margins over Kerry among those who attended church regularly. The media seized on this nding and labeled it the God gap. At a deeper level, however, the results of the 2004 presidential campaign suggested that an emerging partisan religious conguration had become more solidiedone that might provide a narrow, but potentially stable, majority for future GOP presidential candidates. Certainly, the difference in voting patterns
Religion and the Political Landscape in 2008


between the religiously observant and nonobservant was one component of the larger puzzle, but the incessant focus on this single pattern tended to obscure the wider array of factors that shaped the outcome of the 2004 vote. While this focus on religious observance was a function of the medias and politicians tendency to simplify matters, it also reected differences of opinion among analysts as to what facets of religion were the most politically relevant in shaping voting decisions. These differences were reinforced by the lack of religious questions in most surveys, which prevents a more complete analysis. The religious coalitions in 2004 provide an important backdrop for the strategies of presidential aspirants in 2008, both during the primary and general election campaigns. Clearly, two factors in the political importance of any group are their relative size and their partisan proclivities. Table 2.1 reports the relative size and the voting behavior of the major religious traditions, along with the corresponding proportion that each tradition contributed to the total vote captured by each of the candidates.5 Evangelical Protestants and Roman Catholics constitute the two largest traditions, with approximately one-quarter of Americans indicating they are afliated with a religious body linked to evangelical Protestantism, while slightly less than one-quarter (21.9 percent) indicate that they are afliated with the Roman Catholic Church (when Latino Catholics are included). Mainline Protestants and the religiously unafliated are about equal in size, at around one-sixth of the total population each, with black Protestants, Jews, and others lling out the remaining religious traditions. These religious afliations clearly shape voters political choices. Evangelical Protestants voted Republican in overwhelming numbers in 2004, while religious minorities (Jews, black Protestants, and the unafliated) were staunchly Democratic. Mainline Protestants and Catholics, as a whole, were clearly swing groups, almost evenly divided between the parties. Finally, the voting patterns by religious tradition afliation reveal that the Republican and Democratic parties have very distinct religious bases. In 2004, Bush drew heavily from voters afliated with three different religious traditions (evangelical Protestants, mainline Protestants, and nonHispanic6 Roman Catholics). Members of these three religious traditions contributed nearly 80 percent of Bushs overall vote total, while black Protestants, Jews, Hispanic Protestants, Hispanic Catholics, the religiously unafliated, and those remaining in the catch-all category of other, together contributed less than 20 percent of Bushs total vote. Evangelical Protestants alone contributed approximately two-fths of all the votes Bush received in 2004, and, as a result, evangelical Protestants


The Disappearing God Gap?

table 2.1 Religious Groups and the 2004 Vote

Bush Percentage of Potential Religious Group Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalists Centrists Modernists Jews Other faiths Unafliated Voters 24.9 14.2 8.6 2.1 16.2 3.2 7.8 5.2 2.6 9.3 4.5 17.4 2.8 10.1 4.5 1.9 5.9 17.3 Percentage of Two-Party Vote 77.6 85.4 65.6 60.0 50.3 64.3 50.7 38.8 64.0 17.4 31.7 52.9 67.7 51.1 33.8 26.8 47.6 28.0 Percentage of Republican Coalition 39.8 27.8 10.0 2.0 18.4 5.5 9.0 3.9 2.6 2.6 1.9 19.8 5.8 11.3 2.7 1.4 5.3 8.1 Percentage of Democratic Coalition 12.1 5.0 5.5 1.6 18.9 3.1 9.2 6.5 1.6 13.2 4.4 18.5 1.6 11.3 5.7 4.1 5.6 21.6

Source: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron.

served as the primary religious base of the Republican coalition in 2004. Clearly, Bushs support was highly concentrated religiouslylargely drawn from those afliated with religious bodies tied to the three largest religious traditions. Kerrys votes, on the other hand, were spread far more evenly across the various religious traditions, as ve of the nine religious categories contributed more than 10 percent to Kerrys total. Roman Catholics, mainline Protestants, and the religiously unafliated each contributed relatively similar levels of support (around 20 percent each), while black Protestants and evangelical Protestants contributed lower, but relatively equal, proportions (around 12 to 13 percent each). Thus, the Democratic coalition of voters in 2004 was much more diverse religiouslydrawing support from a variety of different religious groups, with no particular religious group constituting the religious base of the coalition. As discussed in the introduction, however, those associated with a particular religious tradition can, despite their commonalities, be quite diverse
Religion and the Political Landscape in 2008


in their religious perspectives. Not all Roman Catholics or evangelicals, for example, hold the same theological beliefs or think alike religiously. Moreover, religious traditions themselves may be undergoing profound transformations. Some analysts (e.g., Wuthnow 1988; Hunter 1991) have contended that there are important changes within American religious life, which are leading to the restructuring of American religion itself. According to this theory, major divisions have been emerging within religious traditions based largely on theological differences with regard to religious authority: more orthodox believers adhere to a notion of an external, denable, and transcendent authority and accept traditional doctrines of the church, while more progressive adherents are prone to replace historic religious interpretations with new ones based on personal experience or scientic rationality (Hunter 1991, 44). Such modernists within each religious tradition are joined by the religiously unafliated, who largely see morality in much the same way. Initially, these religious divisions were largely evident with regard to such issues as abortion, feminism, and gay rights, but they soon appeared to shape other political issues as well. Not all analysts necessarily accept the presence of these divisions (Fiorina, Abrams, and Pope 2005), but others (e.g., Layman 2001; Green 2007) have found modied empirical support for them. In table 2.1 we can see the role of religious restructuring in the 2004 presidential election. We have divided the three largest religious traditions into three different categories, with traditionalists, representing the most theologically orthodox and typically the most observant members of each tradition, at one end of the continuum, while modernists anchor the other end.7 The data clearly reveal that voters within each of the three largest religious traditions varied markedly in their support for each of the two major candidates in 2004, as traditionalists within each of the three major religious traditions were far more Republican than others in their tradition, with modernists generally voting overwhelmingly Democratic. Indeed, the voting patterns of traditionalists across the three major religious traditions look much more alike than the pattern of voting within different segments of the same religious tradition. Taken together, the data in table 2.1 provide a much clearer picture of the different religious bases that exist for the Republican and Democratic candidates and serve as an important backdrop to the 2008 presidential election. While Bush received two-fths of his vote from evangelical Protestants alone, the bulk of this support was drawn from traditionalist and centrist evangelicals. When mainline Protestant and Catholic traditionalists are added to the mix, it almost attains the level of having a traditionalist majority within the GOP (49.1 percent). Of course, no


The Disappearing God Gap?

Republican presidential candidate can win an election solely on the basis of support from this traditionalist core; the partys nominee still needs support from other religious and unafliated groups to win election in November. Nevertheless, no GOP candidate in 2008 could avoid some appeal to the partys traditionalist baseparticularly in seeking to capture the GOP nomination. Consequently, the problem strategically for Republican candidates in the 2008 election was that they needed to retain their religious base without repelling other voters. On the other hand, with its far more diverse religious base, Democratic candidates confronted an even more complicated strategic situation. Kerry drew support from across a wide spectrum of religious groups, including mainline Protestants, Roman Catholics, African-American Protestants, Jews, and other religious minorities. However, the largest religious voter bloc that supported Kerry was the religiously unafliated, which contributed over one-fth of his vote total. But given the generally religious nature of the American public and the fact that most Americans want their president to be a religious person (see chapter 1), many Democratic strategists believed by 2008 that the party would be at a political disadvantage were it perceived as unfriendly toward religion, as it was during the 2004 campaign. Accordingly, they thought that Democratic candidates for public ofce must nd their religious voice and that, at a minimum, the party and its candidates needed to avoid Kerrys uncertain tone when confronted with faith-related issues (Guth 2009, 120). But to seek the support of religious voters across such diverse traditions, and to do so without angering its core of secular voters and activists, would be a challenge. Indeed, some analysts and party leaders contended that such a religious turn was not necessary, as the Democratic Party could win elections by enhancing their vote among their growing natural constituencies: Latinos, specialists in technological and knowledge industries, younger voters, and those who were primarily secular in orientation or less connected to religious institutions (Judis and Teixeira 2002). Others contended that the Democrats should simply ignore the GOPs appeals to moral values and emphasize economic issues insteadparticularly among working-class voters whose economic interests were largely aligned with Democratic programs (Frank 2004). Which particular strategy Democrats should pursue and the extent to which religious values should be emphasized in the 2008 election was, in part, a decision each Democratic candidate for president needed to make particularly in the primaries. Conversely, Republican candidates needed to consider just how they might best expand their base of supporters
Religion and the Political Landscape in 2008


beyond their relatively concentrated religious base from the 2004 presidential election. The playing eld in 2008 was largely unchanged in that the various religious traditions made up about the same proportions of the electorate as they had in 2004. This is evident in table 2.2, which is based on data drawn from our national survey of 3,002 Americans, conducted in April 2008. Evangelical Protestants remain the largest religious tradition within American religious lifeconstituting a little more than a quarter of the American electorate, while Roman Catholics also compose nearly a quarter of the electorate (21.8 percent, with Hispanic Catholics included). Mainline Protestants and the religiously unafliated constitute the two other relatively large blocs of religious voters. Black Protestants form less than 10 percent of the electorate, with the remaining portion being composed of Jews and those who express a variety of other religious faiths (e.g., Mormons, Jehovah Witnesses, Muslims, Hindus, etc.), who together represent less than 10 percent of the population.
table 2.2 The Religious Landscape: Spring 2004 and Spring 2008
2004 Religious Tradition Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalists Centrists Modernists Jews Other faiths Unafliated (%) 24.9 14.2 8.6 2.1 16.2 3.2 7.8 5.2 2.6 9.3 4.5 17.4 2.8 10.1 4.5 1.9 5.9 17.3 2008 (%) 26.3 14.8 9.8 1.7 16.7 3.4 7.8 5.5 2.9 6.3 6.4 15.4 3.5 7.7 4.2 1.1 7.2 17.6

Sources: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron, MarchMay 2004 (N = 4,000); Henry Institute National Survey on Religion and Public Life, AprilMay 2008 (N = 3,002).


The Disappearing God Gap?

Religion and Partisan Identications

However, the relative size of religious traditions is not the whole picture of the political landscape. A variety of other factors contribute to the context within which the election campaign was waged. One such factor is the distribution of partisan identications as the campaign commenced. Scholars of elections and voting behavior have long noted that many Americans enter each presidential election campaign with certain persistent loyalties or standing decisions (Lazarsfeld, Berelson, and Gaudet 1944; Campbell et al. 1960; Key 1966). One school of thoughtthe socalled Michigan School of voting behaviorsuggests that these initial, though tentative, commitments to particular parties and their candidates reect a long-standing, emotional tie to a political party that typically originates early in life through political socialization.8 These identications are relatively stable orientations that color and shape political perceptions and evaluations of both issues and candidates. Simply put, those who self-identify as Democrats are likely to vote Democratic, while those who identify as Republican are likely to vote Republican. Thus, knowing the distribution of partisan identications prior to the election provides some important contextual information about the nature of the campaign and which party and which candidate likely held an advantage going in. Table 2.3 provides the percentage of self-identied Democrats and Republicans within each of the major religious traditions in spring 2008 as well as over the course of the past four presidential elections. At rst glance, the relative proportions of Democrats and Republicans dont appear to have changed much over the past sixteen years. In 1992, Democrats held a modest edge of four percentage points; by early 2008 that margin had grown slightly, to nine points. Much of this aggregate change has occurred over the past four years, as the proportion of the electorate identifying as Republican has diminished. The percentage of the electorate claiming to be Democrats, however, remained at basically the same level in 2008 as found in each of the last four presidential elections. However, beneath this largely stable pattern in partisan identications has been a shift in the way the major religious traditions have aligned themselves with the two major parties. Probably the most commonly recognized shift in partisan loyalties has been among evangelical Protestants. This group, representing, as we have noted, a little more than a quarter of the electorate, favored the Republican Party over the Democratic Party by sixteen points in 1992, but now leans Republican by twenty-eight points. In 1992, evangelical Protestants trailed mainline Protestants in
Religion and the Political Landscape in 2008


table 2.3 Major Religious Traditions and Partisan Identication: 19922008

19922008 1992 Rep (%) All Evangelical Protestant Mainline Protestant Black Protestant Roman Catholic Hispanic Catholic Jews Unafliated 37 48 50 10 38 22 18 30 41 32 32 77 43 49 45 41 41 53 49 15 39 29 29 37 Dem Rep (%) 42 33 34 80 44 52 51 40 39 51 50 12 37 25 24 32 1996 Dem Rep (%) 42 33 33 74 43 57 47 41 39 60 46 11 43 16 22 27 2000 Dem Rep (%) 43 27 41 72 45 64 70 45 33 55 35 6 38 14 19 16 2004 Dem Rep (%) 42 27 47 71 40 58 63 42 2008 Dem Rep Net Change Dem

(Percentage Points) 4 +7 15 4 0 8 +1 14 +1 5 +15 6 3 +9 +17 +1

Sources: 1992: National Survey of Religion and Politics, Bliss Institute, University of Akron (N = 4,001); 1996: National Survey of Religion and Politics, Bliss Institute, University of Akron (N = 4,034); 2000: National Survey of Religion and Politics, Bliss Institute, University of Akron (N = 6,000); 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron (N = 4,000); 2008: National Survey of Religion and Public Life, Henry Institute (N = 3,002). Rep = Republican; Dem = Democrat. Partisan leaners included with Republicans and Democrats. Independents are omitted for ease of presentation; independents equal to 100 minus the sum of each pair of numbers in a single year.

their level of Republican partisan identications, but evangelical Protestants today have replaced mainline Protestants as the religious tradition most strongly aligned with the Republican Party. In fact, 2008 marks a turning point in the partisan dispositions of mainline Protestants (who form slightly less than one-fth of the American electorate).9 Historically, mainline Protestants have been the mainstay of the Republican coalition. Even as late as 1992, mainline Protestants were heavily Republican. But in 2008, mainline Protestants were for the rst time, at least since the beginning of the New Deal, more Democratic than Republican. On the other hand, Roman Catholics have long been a major part of the Democratic coalition. In 2008, a plurality of non-Hispanic Catholics remained Democratic in their partisan identicationsbut only barely so. As a result, non-Hispanic Roman Catholics (whose total numbers approximate those of mainline Protestants) were the most likely of the large religious traditions to be up for grabs in 2008. In terms of the other religious traditions, a majority of black Protestants and Jews identied as Democrats. Black Protestants have moved slightly away from the Democratic Party since 1992, but such movement has not led them to the GOPas the percentage of both Democratic and Republican party identiers among black Protestants actually declined slightly between 1992 and 2008. Jews, on the other hand, moved much more substantially into the Democratic fold over the same period of timeas only a plurality of Jews identied as Democrats in 1992, but now more than three-fths do so. Finally, the religiously unafliated, who account for nearly one-fth of the total electorate, have moved away from the Republican Party over the past fteen years. While much of this shift away from the GOP has occurred within the past four years, there has been a steady march away from the party since 1996. In 1996, nearly two-fths of the religiously unafliated claimed to be Republicans. However, that proportion declined to one-third in 2000, and further to one-fourth in 2004, while becoming less than one-sixth in 2008. Yet even where there may appear to be relative stability in partisan loyalties within a major religious tradition, changes may be occurring among different components of that tradition. As is evident from table 2.4, there are sharp differences in partisan identications within the ranks of evangelical Protestants, mainline Protestants, and Roman Catholics, depending upon whether such religionists fall within the traditionalist, centrist, or modernist ranks of each religious tradition. In both 2004 and 2008, a majority of traditionalist evangelicals, mainline Protestants, and
Religion and the Political Landscape in 2008


table 2.4 Religion and Self-Identied Partisanship: Spring 2004 and Spring 2008
2004 Partisan Identication Republican Democrat 2008 Partisan Identication Republican Democrat

(%) All Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalists Centrists Modernists Jewish Other faiths Unafliated 38 60 70 47 43 46 59 44 39 40 11 16 42 60 38 39 22 30 27 42 27 20 34 40 41 32 40 49 45 72 64 45 25 46 53 70 42 45 33 55 65 46 45 35 57 31 28 18 6 14 38 55 39 16 19 29 16

(%) 42 27 22 30 43 47 31 50 56 54 71 58 40 22 40 63 62 41 42

Sources: National Survey of Religion and Politics, Bliss Institute, University of Akron, MarchMay 2004 (N = 4,000); Henry Institute National Survey on Religion and Public Life, AprilMay 2008 (N = 3,002). Partisan leaners included with Republicans and Democrats. Independents omitted for ease of presentation; independents equal to 100 minus the sum of each pair of numbers in a single year.

non-Hispanic Catholics identify as Republicans, while the majority of modernist mainline Protestants and Roman Catholics classify themselves as Democrats, as do a substantial proportion of modernist evangelical Protestants. When compared to their counterparts in the 2004 presidential election at roughly the same point in the campaign calendar, traditionalist evangelical Protestants remained heavily Republicanthough slightly less so than four years ago (see table 2.4); centrist evangelical Protestants, though more Republican than Democratic, were also somewhat less Republican in 2008 than in 2004. On the other hand, between 2004 and 2008,


The Disappearing God Gap?

modernist evangelical Protestants moved slightly toward both the Democratic Party and the Republican Party and somewhat away from independence. Among mainline Protestants, traditionalists remained almost as heavily Republican in 2008 as they were in 2004. However, a large shift to the Democratic Party occurred among centrist and modernist mainline Protestants. Centrist mainliners went from being Republican by four percentage points in 2004 to being Democratic by nineteen points in 2008, while Republican identication declined by more than 10 percent and Democratic identication jumped by 7 percent among modernist mainliners between 2004 and 2008. Traditionalist non-Hispanic Catholics remained strongly Republican in 2008. Modernist Catholics, who were already Democratic in 2004, became even more so in 2008, while centrist Catholics shifted away from being rather Democratic in their partisan identications in 2004 to being more evenly divided in 2008. However, partisan differences between traditionalist and modernist parts of the same religious tradition tend to be equal to, if not greater than, the political differences across the three major religious traditions overall. For example, the 20 percent difference in the percentage of Republican identiers between all evangelical Protestants and all mainline Protestants is equal to the percentage difference between traditionalist and modernist evangelical Protestants, traditionalist and modernist mainline Protestants, or between traditionalist and modernist non-Hispanic Roman Catholics. Thus, beliefs and practices are beginning to replace afliation as the primary religious basis of political cleavages. Ones religious tradition continues to shape political tendencies, but such tendencies are shaped even more by the specic kind of person one is religiously within that particular tradition.

Religion and Political Issues

In addition to their partisan afliations, voters bring different policy preferences to the campaign. Frequently, ones partisan identication shapes ones stands on issues, so that a voters issue position reects the views of party leaders and ofceholders. However, some deviate from conventional party stands on particular issues, with these deviations varying from issue to issue and from individual to individual. It is this lack of consistency across all issues that creates the opportunity for candidates and parties to focus on particular wedge issues that may divide members who identify
Religion and the Political Landscape in 2008


with the same political party. Moreover, voters frequently baseor at least justifytheir voting decisions on the particular issue stands that candidates and political parties adopt. Using wedge issues to peel off different segments of a partys core can enable a candidate to overcome a disadvantage based on partisan identications. Religion can meaningfully shape public policy attitudes in at least two ways.10 First, religion may shape the specic policy positions of voters. Here religion shapes the direction of opinion on an issue. For example, is a citizen for or against government funding for embryonic stem cell research? Does the voter support maintaining troop levels or withdrawing troops from Iraq? Second, religion may shape the policy agendas of voters. Here religion shapes the salience of an issue for a voter and the likelihood that the person will act on that position. Using the examples of stem cell research and Iraq, the relevant questions are comparative: is funding for stem cell research more or less important to a voter than troop levels in Iraq? And if so, is the citizen likely to act (through his or her vote) on this relative assessment of the issues?

Issue Positions
Economic Issues. Even though political cultures differ about governments role in addressing various social and economic problems, most religious faiths command their adherents to show charity toward the poor and destitute. However, such faiths may differ as to whether such charity and aid should be provided through the institutions of civil society or through the state. Typically today, liberalism on issues of economic policy tends to be associated with greater support for the role of the state in providing aid and services to the disadvantaged, while conservatism tends to favor means outside the state in addressing such needs. With regard to religious afliation and positions on economic issues, several general patterns seem to be fairly well establishednamely, that Jews and black Protestants hold very liberal positions on a wide range of economic and social justice questions (including welfare spending, national health insurance, and tax policy), while evangelical Protestants hold more conservative positions on such matters, with Catholics and mainline Protestants adopting positions more in the middle (Smidt 2001; Wald and Calhoun-Brown 2007). These general patterns are fairly robust across time and across issues (Wilson 2009). Table 2.5 examines the relationship between religious tradition and the respondents positions on various economic issues (some in 2004 and others in 2008 or for both years, depending upon when such questions


The Disappearing God Gap?

were asked).11 If responses truly reect real positions on these issues and not simply random responses to the questions posed, then one would anticipate that patterns in 2004 should continue in 2008. Of course, where there are fewer respondents falling within a category (e.g., Jews), one might expect more instability of responses across time, as changes in only a few respondents may have dramatic effects. Finally, depending upon the wording of the questions, the actual percentages expressing agreement or disagreement may change, but the particular patterns of which religious traditions tend to stand most in agreement or opposition should prevail across the two elections. Three questions found in table 2.5 explore the role of government in addressing various social welfare issues. Note the strong language used in the statement: The government should spend more to ght hunger
table 2.5 Economic Issue Positions by Religious Tradition (Percentage Agreeing)
Evan. Issue Year Prot. Main. Prot. Black Prot. Rom. Cath. Jews Relig. Unafl.

The government should spend more to ght hunger and poverty even if it means higher taxes on the middle class. 2004 2008 43 43 52 51 52 65 51 52 65 66 56 59

Minorities need governmental assistance to obtain their rightful place in America. 2004 32 35 58 42 56 40

The economically disadvantaged need governmental assistance to obtain their rightful place in America. 2004 57 54 62 57 72 62

Strict rules to protect the environment are necessary even if they cost jobs or result in higher prices. 2004 2008 52 42 59 52 40 52 57 52 67 74 55 65

Free trade is good for the economy even if it means the loss of some U.S. jobs. 2004 2008 34 30 32 33 16 21 31 30 48 61 27 34

Sources: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron, MarchMay 2004; Henry Institute National Survey of Religion and Public Life, AprilMay 2008.

Religion and the Political Landscape in 2008


and poverty even if it means higher taxes on the middle class. In order to agree, respondents not only had to agree that more should be spent to help the needy but that such spending should increase even if it meant that there would be increased taxes on the middle class which, presumably for most respondents, meant even if my own taxes were to increase. In addition, it is important to note that both the words hunger and poverty are included in the question, and, as a result, those who might favor having the government address one such issue rather than the other will be more ambivalent with regard to how to respond to the question. Several clear patterns emerge in table 2.5. First, there are important differences in the political stands adopted by members of different religious traditions, and these differences are consistent with previous ndings that Jews and black Protestants hold very liberal positions on a wide range of economic and social justice questions, while evangelical Protestants hold more conservative positions on such matters, with Catholics and mainline Protestants falling in between. On this particular question, a little more than half of mainline Protestants and Roman Catholics stand in agreement with the statement, while nearly two-thirds of Jews (and black Protestants in 2008) but only a little more than twofths of evangelical Protestants do so. Though the percentages change, a similar pattern prevails when respondents were asked in 2004 whether they agreed with the statement that minorities need governmental assistance to obtain their rightful place in America. Those who were most likely to agree with the statement were members of minority groups, as a majority of black Protestants and Jews agreed with the statement, while evangelical and mainline Protestants were the least likely to agree the statement. However, when the same question was posed in terms of whether the economically disadvantaged rather than minorities need government assistance, then the percentage differences across the religious groups shrink considerably, with a majority of members within each of the religious traditions expressing agreement with the statement. In fact, the percentage of evangelical Protestants who express agreement with the statement exceeds the percentage of mainline Protestants who agree, is equivalent to the percentage of Roman Catholics who agree, and nearly reaches the same percentage found among black Protestants. Clearly, the particular group to be aided by government assistance makes a big difference. Two other issues dealing with economic matters are examined in table 2.5namely, environmental protection and free trade. With regard to issues we have examined thus far, which relate more directly to government


The Disappearing God Gap?

services and aid to minorities (or to the poor), black Protestants and Jews stand largely in agreement with one another, while evangelical Protestants are typically standing at the other end of the distribution. But black Protestants and Jews generally part ways and stand largely opposed to each other on the environment and free trade. When asked whether they agreed with the statement that Strict rules to protect the environment are necessary even if they cost jobs or result in higher prices, Jews were the most likely to agree (with more than twothirds agreeing in both 2004 and 2008), while black Protestants were the least likely to agree with the statement in 2004 and were among the least likely in 2008. However, despite this difference between Jews and black Protestants, a majority of nearly every religious group examined (ve of the six religious traditions in both 2004 and 2008) expressed agreement with the need for environmental protection. Thus, whatever variation is evident across members of different religious traditions, such differences largely occur within the framework of majority support. On the other hand, Jews are the only religious group where a majority (or nearly a majority) of its members expressed agreement with the statement that Free trade is good for the economy even if it means the loss of some U.S. jobs. Black Protestants were the least likely to express agreement with the statement, as only one-fth agreed in 2008 (an even smaller percentage did so in 2004). Except for Jews, slightly less than onethird of the members of the various religious traditions stood in agreement with the free trade statement. Obviously, support for free trade is not common among Americans as a whole, and its level of support does not vary much from one religious group to anotherexcept for the differences noted between Jews and all other religious traditions. Finally, it is important to note that, on economic issues alone, particular issues could drive a wedge between what might seem to be natural allies. The fact that black Protestants and Jews can stand together on one economic issue but then stand at polar positions on another economic issue indicates the presence of such wedge issues. It also reveals that advancing social issues is not necessarily the only means by which to insert a wedge among voters united by economic concernsas such unity on economic matters is dependent on the particular economic issues under consideration. Cultural Issues. It was not long ago that conventional wisdom held that economic issues were the driving mechanism of American politics. Certainly, the formation and persistence of the New Deal coalition suggested such an interpretation. Moreover, with the end of World War II and the economic prosperity that followed, many scholars contended that American politics had come to be dened by a fundamental consensusan
Religion and the Political Landscape in 2008


end of ideologyand, as a result, political leaders now simply needed to make incremental and rened technical decisionsrather than choosing between different ends, goals, or values. After all, if politics reects nothing more than people wishing to advance economically, then the goal of politics is the same for all. During the 1960s, however, various social issues (e.g., prayer in public schools, civil rights, busing) began to surface in American politics, and soon others emerged as well (e.g., abortion, gay rights). At the same time, the salience of such social issues rose dramatically (Scammon and Wattenberg 1970), and has only increased since (Jelen 2009). These issues raise fundamental questions about the communitys goals and therefore counter the idea of a postwar consensus. Indeed, what makes these kinds of issues particularly contentious is that they are not merely technical matters of economic regulations but call into account basic beliefs about morality, identity, and lifestyle. In fact, some analysts (e.g., Hunter 1991, 1994) have gone so far as to suggest that lifestyle issues superseded economic divisions as the principal axis of political conict in the United States, labeling the divide a culture war. Table 2.6 shows the percentage of respondents reporting agreement with various statements related to cultural issues according to religious afliation.12 Several noteworthy patterns emerge. First, there is a large degree of consensus on two issueslarge majorities favored prayer at public school graduation ceremonies and the posting of the Ten Commandments and other religious symbols in public buildings. Jews are the only religious group in which a majority stand opposed to graduation prayers and the public posting of religious symbols, though the religiously unafliated split roughly in half on such issues. Evangelical Protestants were the most likely to favor these practices, while black Protestants were also notable for their strong support.13 On the other hand, there was as much disagreement on abortion, samesex marriage, gay rights, and public funding of faith-based social services as there was agreement in favor of graduation prayers and the public posting of religious symbols. Evangelical Protestants stand most opposed to allowing abortion, as less than one-third expressed agreement with the statement on that issue. Jews overwhelmingly expressed agreement with such a statement, as did three-quarters of the religiously unafliated. About three-fths of black Protestants and mainline Protestants support abortion under such conditions, while Roman Catholics were evenly divided on the matter. However, when the focus shifts from abortion to that of gay rights, some important changes occur in the pattern of support. First, black Protestants join evangelical Protestants in their relatively low level of


The Disappearing God Gap?

table 2.6 Cultural Issue Positions by Religious Tradition (Percentage Agreeing)

Evan. Issue Year Prot. Main. Prot. Black Prot. Rom. Cath. Jews Relig. Unafl.

Private citizens should be able to provide prayers at public school graduation ceremonies. 2008 89 73 80 73 29 52

Local communities should be allowed to post the Ten Commandments in public buildings. 2004 2008 84 86 67 72 67 81 65 69 34 37 45 55

Abortion should be legal and solely up to the woman to decide. 2008 30 65 65 50 83 74

Gays and lesbians should be permitted to marry legally. 2008 19 50 37 42 68 64

Homosexuals should have the same rights as other Americans. 2004 43 59 39 63 82 73

Public funding should be available to churches and houses of worship to provide social services. 2004 58 47 62 51 37 35

Sources: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron, MarchMay 2004; Henry Institute National Survey of Religion and Public Life, AprilMay 2008.

support for gay rights generally and gay marriage more specically (although the specic level of opposition is greater to gay marriage than to gay rights more generally). Jews and the religiously unafliated are the most supportive of these matters (with approximately two-thirds expressing support for gay marriage and even greater proportions for gay rights more generally). A majority of mainline Protestants and Roman Catholics tend to agree that Homosexuals should have the same rights as other Americans, but fewer of those afliated with these traditions support gay marriage. Overall, on matters related to gay rights, mainline Protestants and Roman Catholics fall between evangelical and black Protestants on the one side and Jews and the religiously unafliated on the other. Finally, with regard to faith-based initiatives, both evangelical Protestants and black Protestants favor public funding of social services
Religion and the Political Landscape in 2008


provided by churches and other houses of worship. Approximately threefths of those with such religious afliations support such funding, while approximately half of Roman Catholics and mainline Protestants do so. Jews and the religiously unafliated stand in greatest opposition, as only a little more than one-third of each group expresses agreement with the statement on the issue. Once again, we can see potential wedge issues. For example, black Protestants tend to stand in greater agreement with their Jewish and secular Democratic coalition partners on issues related to abortion, but these Democratic religious groups dramatically part ways on gay rights, the posting of religious symbols in public places, and the organized expression of religion in public ceremonies. Foreign Policy. In both the 2004 and the 2008 presidential campaigns, the major foreign policy issues centered around the war on terror and the war in Iraq. Table 2.7 examines the issue positions on various matters of foreign policy expressed by those afliated with different religious traditions. First, by the advent of the 2008 presidential election, only a minority of Americans thought that The U.S. did the right thing in taking military action in Iraq. In fact, in terms of religious tradition afliations, only evangelical Protestants exhibited majority support for this statement. Mainline Protestants ranked next, though less than half agreed. Black Protestants were the least likely to express agreement with the statement. Only a quarter of the religiously unafliated agreed, while a little more than one-third of Catholics and Jews did so. In 2004, when respondents were posed a question about whether it was necessary, given the threat of terrorism, for the United States to be able to take preemptive military action against other countries, a majority of Americans regardless of their religious tradition afliation expressed agreement with such a statement. However, evangelical Protestants were the most likely to express such agreement, while black Protestants, Jews, and the religiously unafliated were the least likely to do so. Nevertheless, most Americans in 2004 continued to reject some return to isolationism in American foreign policy, as only a minority of those across all six religious traditions expressed agreement with the statement that The U.S. should mind its own business internationally. However, while black Protestants and the religiously unafliated were the most likely to express agreement, Jews were the least likely to do so. Only one third of evangelical Protestants, mainline Protestants, and Catholics expressed agreement with the statement. There is, however, one foreign policy issue on which Jews and evangelical Protestants tend to stand in relative agreementnamely, the need to


The Disappearing God Gap?

table 2.7 Foreign Policy Issue Positions by Religious Tradition (Percentage

Agreeing) Evan. Issue Year Prot. Main. Prot. Black Prot. Rom. Cath. Jews Relig. Unafl.

The U.S. did the right thing in taking military action in Iraq. 2008 55 41 20 36 37 25

Given the threat of terrorism, the U.S. must be able to take preemptive military action against other countries. 2004 72 62 54 62 57 57

The U.S. should mind its own business internationally and let other countries get along as best they can on their own. 2004 34 34 43 36 17 44

The U.S. should support Israel over the Palestinians in the Middle East. 2004 53 33 24 30 75 21

Source: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron, MarchMay 2004; Henry Institute National Survey of Religion and Public Life, AprilMay 2008.

support Israel over the Palestinians. In fact, these were the only two religious traditions in which a majority of their members supported such a statement. Only a third or less of those afliated with the other religious traditions were willing to express such agreement, with only one-quarter of black Protestants and one-fth of the religiously unafliated doing so. In sum, the way in which those afliated with different religious traditions take stands on matters of public policy varies across economic, cultural, and foreign policy issues. Overall, evangelical Protestants tend to take relatively more conservative positions, while black Protestants, Jews, and the religiously unafliated tend to take more liberal positions. Mainline Protestants and Roman Catholics generally fall more in the middle on most issues. However, on certain issues, these general patterns do not necessarily prevail. As a result, such issues may serve as important wedge issues in the campaign. Whether or not they do so is partially dependent on whether campaign strategists choose to use them in such a fashion. Religious Traditionalism and Issue Stands. How do these differences play out in terms of the traditionalism of ones religious beliefs? Table 2.8
Religion and the Political Landscape in 2008


table 2.8 Issue Positions by Religious Traditionalism: Spring 2008 (Percentage

Agreeing) Evan. Prot. Issue Trad. Cent. Mod. Main. Prot. Trad. Cent. Mod. Rom. Cath. Trad. Cent. Mod.

Abortion should be legal and solely up to the woman to decide. 21 35 52 43 57 84 20 50 83

Gays and lesbians should be permitted to marry legally. 9 26 42 28 49 78 13 43 67

Strict rules to protect the environment are necessary. 38 43 47 38 64 74 31 46 74

The government should spend more to ght hunger and poverty. 40 44 54 35 51 61 31 50 69

Free trade is good for the economy. 34 27 26 36 30 41 32 27 35

The U.S. did the right thing in taking military action in Iraq. 66 48 51 55 38 33 66 34 27

Source: Henry Institute National Survey of Religion and Public Life, AprilMay 2008.

examines differences among traditionalists, centrists, and modernists within each of the three largest religious traditions (i.e., evangelical Protestants, mainline Protestants, and Roman Catholics) on each of the cultural, economic, and foreign policy issues we looked at earlier. When we examine them in this fashion, we nd that the traditionalists from these three largest Christian traditions generally (and not surprisingly) adopt the more conservative stance on such issues, while the modernists within each tradition adopt the most liberal views. This is true for each of the issues examined except free trade, where there is little variation and little consistency across such categories within each tradition. However, with regard to all other issues examined, the percentage differences between traditionalists and modernists within the same religious tradition are typically far greater than the differences observed across the broader religious tradition categories.


The Disappearing God Gap?

Clearly, then, both religious afliation and religious traditionalism matter in shaping the issue stands of the American electorate. Consequently, those candidates entering presidential primaries initially, and then later the presidential nominees themselves, need to determine whether to cast their appeals more toward members of specic religious traditions (e.g., toward evangelical Protestants or toward Catholics) or more toward those groups that transcend such religious traditions (e.g., religious traditionalists or religious centrists). On the one hand, it is far easier to identify and contact members of, create liaisons with, appear before, and nd spokespersons for members of particular religious traditions such as Catholics or mainline Protestants than it is to do the same for religious traditionalists or religious modernists. However, any candidate who wishes to target his or her appeals more effectively would want to know how best to reach the intended audiences within such religious traditions, as not all members of a religious tradition will be equally receptive to a particular message.

Issue Agendas
For many American voters, religion inuences not only the particular stands they adopt on policy proposals but also the relative importance they place on such issues. All laws impose someones values on the rest of society, because they specify either directly or indirectly that certain forms of behavior are preferred and enforced over other forms. While not all values necessarily derive from religion, it frequently serves as the basis of value formation for many Americans. Thus, both the political agenda and the policy stands that many Americans adopt may have a religious basis. At the close of 2007, and prior to the Iowa caucuses and the New Hampshire primary, voters were divided on what was the most important issue confronting the nation, and those afliated with different religious traditions tended to differ in their assessment. A Gallup poll conducted in December 2007 asked an open-ended question as to what constituted the most important issue confronting the nation at that time. Table 2.9 examines various responses according to the respondents religious afliation. As is evident from the table, foreign policy issues dominated the concerns of the voters just prior to 2008, with a third of the respondents identifying such matters as the most important issue. Economic issues ranked second, with a quarter identifying such issues as the most important, while one-fth of the voters cited what may be labeled social or cultural issues.
Religion and the Political Landscape in 2008


table 2.9 Most Important Issue by Religious Tradition

Evan. Nature of Political Issue Economic issues Social welfare issues Social/cultural issues Foreign policy Environment Social and political institutions Total (N) 5 100 (261) 10 101 (209) 4 99 (70) 4 100 (242) 21 100 (14) 12 101 (101) Prot. (%) 24 13 28 29 1 Main. Prot. (%) 20 12 21 34 4 Black Prot. (%) 14 17 20 44 0 Roman Catholic (%) 30 13 18 34 1 Jews (%) 7 29 7 29 7 Religiously Unafliated (%) 32 12 6 37 2

Source: Gallup News Service Survey #40, December 2007.

Those afliated with different religious traditions had somewhat different priorities. Evangelical Protestants were rather equally divided in terms of the relative emphasis placed on economic issues, social issues, and foreign policy issues, with approximately one-quarter of evangelical Protestants citing each set of issues as being the most important. However, evangelical Protestants were the most likely of the various religious traditions examined to cite some type of social or cultural issue as being the most important. Mainline Protestants stressed a variety of different issues. More than one-third of mainline Protestants cited foreign policy as being the most important issue, while one-fth of mainliners noted economic issues, while another one-fth reported social issues as most important. However, one-tenth of mainline Protestants cited social welfare issues, and another one-tenth noted issues related to American social and political institutions as most important. One-fth of black Protestants cited cultural or social issues as their most important issue. And while a substantial number of black Protestants cited economic issues and social welfare issues, even these two types combined could not eclipse the percentage of black Protestants who cited foreign policy as being most important. In fact, of all the religious traditions examined in table 2.9, black Protestants were the most likely to cite foreign policy issues as the most important ones confronting America. Roman Catholics were divided into four distinct segments. Nearly one-fth of Catholics cited social issues as the most important, and more


The Disappearing God Gap?

than 10 percent cited social welfare issues. However, two sets of issues dominated the concerns of Catholics, as nearly one-third cited economic issues and another third named foreign policy. At the close of 2007, Jews were the least likely to cite economic issues as being the most important issues confronting America. Instead they tended to emphasize three different kinds of issues: social welfare, foreign policy, and issues that were generally related to social and political institutions. In fact, Jews were the most likely of all religious traditions to cite social welfare issues, and among the least likely to stress cultural issues as being the most important. The religiously unafliated or the more secular segment of the American people were also not moved by cultural issues; rather, they were primarily concerned with economic and foreign policy, each of which was named by over one-third of them as the most important issues confronting the nation at the end of 2007. Certainly, it would be a distortion to suggest that certain religious groups were concerned about only one particular type of issue, as those afliated with different religious traditions tended to cite with some frequency three or four types of issues as being the most important. Still, the tendency of different religious traditions to emphasize distinctive types of issues is also evident in table 2.9, as evangelical Protestants were the most likely to cite cultural issues, black Protestants to cite foreign policy issues, Jews to cite social welfare issues, and Roman Catholics and secular Americans to cite economic issues. Under such circumstances, different types of issues can be stressed for different types of religious audiences, providing opportunities to mobilize different kinds of voters based on different types of issues. But as we will see later, economic issues became progressively more important during the summer of 2008 and, then, following the banking crisis of late September 2008, dominated the discussion during the closing weeks of the campaign. Not surprisingly, different types of issues can be stressed when one enjoys a certain level of economic security and material satisfaction. It is harder, however, to be concerned about issues related to abortion or gay marriage when you have just lost your job and cannot nd suitable employment, or when you are about to lose your home because you have defaulted on your mortgage. Under circumstances of deep economic uncertainty and disruption, those afliated with different religious traditions are likely to become much more alike in terms of their issue concerns and priorities. As a result, it becomes harder to utilize different issues to mobilize different segments of the electorate, and it reduces the opportunities to employ wedge issues to ones partisan advantage.
Religion and the Political Landscape in 2008


Democratic Lessons Learned in 2006

As we have noted throughout this book, scholars and political practitioners alike have observed a growing God gap in American politics from the 1980s to the early 2000sa gap that seemed to come to fullest fruition in the 2004 presidential election (e.g., Green and Silk 2003; Olson and Green 2006; Campbell 2007). The idea of the gap is simply that religious behavior predicts political behavior. More specically, religiosity is associated with vote choice and partisanship. The more religious one is (as measured, for example, by religious attendance), the more likely one is to vote Republican or claim to be a Republican; the less religious, the more likely to vote for or identify with the Democratic Party. Hence the gap had been widely perceived as a special problem for the Democrats (Smith and Craighill 2006). Some argue that religious association with the Republicans was largely the result of the Republican Partys own outreach in the late 1970s, culminating in Ronald Reagans victory in the 1980 presidential election. But recent research suggests that this interpretation does not tell the full story. Catholics and evangelicals were moving as much away from the Democratic Party in the early 1970s as toward the Republicans later in the decade, partly due to the widespread perception that the Democrats had developed an agenda on reproductive rights and other issues that was anathema to traditional family values. As a result, the Democrats could no longer count on Catholics, who increasingly became swing voters, and also lost decisively the vote of southern evangelicals (Layman 2001; Bolce and DeMaio 2002). After many decades of walking in the religious wilderness, Democrats reinvigorated their religious outreach in the midterm elections in 2006. On the one hand, in the aggregate the 2006 elections were a resounding victory for the Democrats, who wrested back control of both the House of Representatives and Senate. The victory suggested opportunities for the Democrats in the 2008 presidential election, especially over the issues of the economy and the war in Iraq (as we have noted in this chapter). On the other hand, despite their overall victory, Democrats achieved only marginal progress among the most religious voters in 2006, as they largely either persisted in their support of the GOP or simply stayed home (Keeter 2006). Still, not all was lost, as some inroads among religious voters had been madesufcient to encourage the Democratic Party not to abandon its initial efforts at religious outreach. The Democratic Party had already laid some groundwork for faithbased outreach before the 2006 elections. In 2005, the Democratic National Committee created the Faith in Action Initiative, which sought


The Disappearing God Gap?

ways to broaden the faith-based issue agenda and thereby draw in religious voters who were disenchanted with the party. The initiative included annual conferences and a religious advisory board with representatives from every major tradition. Party leaders in the House and Senate also created working groups focused on the partys religion problem, which not only helped hone a faith-based message but also developed competent operatives who would prove useful in the 2008 elections. Also noteworthy was the fact that in 2006 Common Good Strategies (CGS), a two-person consulting rm formed in 2005, consisting of two young evangelicals, Mara Vanderslice and Eric Sapp, sought to help Democratic candidates talk about religious faith.14 Their rm was contracted to work in seven campaigns in the 2006 elections: two U.S. Senate races (Ohio and Pennsylvania), three gubernatorial races (Michigan, Ohio, and Kansas), one congressional race in North Carolina, and the attorney general race in Kansas. In the end, CGSs candidates won all seven races, in part by targeting religious communities. While each race and candidate was different, CSGs candidates tended to employ several common themes: following Christs example by serving others, promoting the common good, protecting the environment as Gods creation, [and] alleviating poverty (Rosen 2007), with the net result being that CSGs candidates did 10 percentage points better than the Democratic national average among white15 evangelicals and Catholics who regularly attend church (Gibbs and Duffy 2007; Belkin 2008). In addition, the party invested in several candidates with religious bona des in the 2006 elections, particularly the recruitment of Robert Casey, a pro-life Catholic who ultimately defeated incumbent Republican Rick Santorum for a Pennsylvania Senate seat. This new religious strategy did arouse considerable opposition among some secular Democratic activists, but it nevertheless helped the Democrats to recapture Congress (Rosin 2007; Guth 2009, 120), and it provided party leaders with a glimpse of a future strategy. Despite the potential of alienating the secular base, the DNC and Democratic congressional committees began to see religious outreach as a risk worth taking. The social, religious, and political context in the wake of the 2006 elections also signaled a favorable environment for Democrats. New religious leaders were emerging, particularly in the evangelical world, and they brought with them attention to a new set of policy concerns. To be sure, a host of Catholic, mainline, and Jewish leaders were already laboring in liberal politics and therefore expected to be supportive. But other emerging leaders were even more heartening to Democratsand perhaps disconcerting to Republicans. Some relatively young evangelical leaders were as
Religion and the Political Landscape in 2008


conservative on family issues as older religious elites, but they were also not focused as narrowly on those issues. Pastor Rick Warren, for example, whose Saddleback Church hosted a key candidate forum during the 2008 primaries, was opposed to abortion rights and same sex marriage, but he had also been an outspoken advocate for greater attention to issues ranging from poverty to HIV/AIDS in sub-Saharan Africa. Richard Cizik, the former head of the government affairs ofce of the politically conservative National Association of Evangelicals, insisted with other moderate and conservative leaders that their constituents take a greater interest in the environment (see, e.g., National Association of Evangelicals 2004). But while Cizik sought to prod evangelicals to exhibit greater care for the environment, he nevertheless later lost his job foramong other factors his measured support of same sex civil union protections. Intraparty politics and trends in party afliation also suggested the time was right for Democrats to reach out to religious voters. The religious vote in the 2006 elections notwithstanding, various pieces of evidence suggested softening religious support for the Republican Party. Two of the early front-runners for the Republican nominationJohn McCain and Rudy Guilianiwere perceived as having little to no religious bona des. And the faith commitment of the other, more religious nomineethe Mormon Mitt Romneywas problematic. (Mike Huckabee had not yet emerged as a serious candidate in late 2006 or early 2007.) At the popular level, a small yet signicant percentage of evangelical voters, most notably those younger than thirty, were no longer identifying with the GOP (Gibbs and Duffy 2007). And as we have seen in this chapter, the key morality issues that seemed to animate religious voters did not appear as prominent as in past elections. Democrats saw an opening to recast some of their strongest issues in religious terms. The front-running candidates on the Democratic side took early steps to take advantage of this religious and political environment. Unlike John Kerry, the Clinton, Edwards, and Obama campaigns each had a robust faith outreach team. Already by late 2006, Clinton had hired Burns Strider to spearhead her religion efforts. Strider, an evangelical and former Southern Baptist missionary, had previously served as the chief advisor on faith and politics for House Democrats (Bolton 2006). Strider was known for his weekly roundups on Faith, Family, and Values, all terms with great symbolic resonance among rank-and-le religious conservatives. Obama turned to his former aide Joshua DuBois, a young Pentecostal pastor; DuBois set out to organize faith forums across the country (including a half dozen in New Hampshire, the important primary state, in early 2007). Later, following the 2008 presidential election, DuBois would


The Disappearing God Gap?

be named as the head of the White House Ofce on Faith-Based and Neighborhood Partnerships. It is also worth noting that the candidates received a great deal of cajoling and aid from outside their own campaigns. Just before and after the 2006 elections, several pundits and public intellectuals published hopeful books on the role religion could play in center-left politics (Press 2005; Sullivan 2008a; Dionne 2008a). Key center-left think tanks such as the Center for American Progress convened religious leaders to discuss concerns and strategy (Bolton 2006). Prominent religious voices were actively courted by Democratic leaders and activists. A good illustration is Jim Wallis of Sojourners, a left-leaning organization with roots in a kind of social gospel evangelicalism. Wallis had been equally relentless in his criticisms of Republican policies and past outreach of Democrats to religious voters, as evidenced in his 2006 book, Gods Politics: Why the Right Gets It Wrong and Left Doesnt Get It. He sought to help the Democrats get religion through numerous consultations, as well as spearheading a conference on Faith, Values, and Poverty in June 2007. By early 2008, his various efforts with Democratic leaders had apparently convinced him that they were moving in the proper direction; his next book was titled The Great Awakening: Reviving Faith and Politics in a PostReligious Right World. In sum, the combination of the political and social environment and the existing party groundwork led each campaign to develop faith outreach. Every bloc matters in tight elections, and surely 2000 and 2004 had suggested that 2008 might be yet another battle down to the wire. Democrats simply could not afford to ignore religious voters. But they also had to work out ways to reframe issues so that they could attract those voters without alienating the long-standing secular base of the party. The change had to be incremental yet signicant.

Religion and the Political Landscape in 2008



Religion and the 2008 Presidential Primaries

ampaigns for president of the United States almost never end. Within days after a new president is chosen, speculation focuses on who might seek the same ofce four years later. Potential challengers, careful to not look too ambitious, quietly take steps to sound out advisors and seek out early contributors and organizers, particularly among each partys core of support and especially in the early primary and caucus states like Iowa and New Hampshire. These activities become more intense and public after the midterm elections, as a great deal of planning is necessary before the rst straw polls, debates, and other campaign events, which now crowd into the summer of the year before the election. The jockeying for the 2008 presidential nods for both parties was typical. Work intensied right after the 2006 midterm elections, in which Republicans lost big and gave up control of both the U.S. House and Senate (Feldmann 2006; Nagourney 2006). While the timetable of the two major party primary races was typical, the role of religion in them was not. As we noted near the end of the last chapter, the early Republican eld was dominated by two candidates, John McCain and Rudolph Giuliani, neither of whom have ever been closely associated with the partys active religious base (Balz and Cohen 2006). The leading Democratic contenders, Hillary Clinton, Barack Obama, and John Edwards, on the other hand, were all intentionally increasing their religious proles and making formal and informal connections to religious groups (Bolton 2006). Moreover, there are important differences between presidential primaries and the general election. Certainly one important difference is

that the basis on which voters choose candidates differs greatly between the presidential primaries and the general election. In the general election, political party identication plays a very important role in shaping voter choice. But in primary elections, voters choose among candidates from the same party. Moreover, candidates from the same party typically do not differ as much on issues as do candidates from different parties, and, as a result, theories about what motivates vote choice in general elections are not easily transferable to the primary context (Bradberry 2009, 10).

Religion and the Early Stages of the Campaign

Months prior to the rst head-to-head contests in Iowa and New Hampshire, the various announced (and even unannounced) candidates start to compete in a variety of ways that begin to shape public perceptions of their candidacy. Journalists, in particular, monitor various aspects of the process to inform the public, both through print and the electronic media, about the changing dynamics of the relative strength of the candidate within the party, the organizational capacity of the candidate, and the nancial health of the candidates campaign organization. In this initial stage of the presidential selection process, candidates seek to attract campaign contributions to support their candidacy, solicit the endorsement of inuential elites within the party and within groups typically associated with the party (e.g., the head of a major labor union), and attend various summits or party gatherings in which straw polls may be conducted that test the attractiveness of the various candidates with different segments of voters within the party.

Democratic Candidates
After Republicans lost control of Congress in the midterm election of November 2006, the race for the 2008 Democratic presidential nomination began in earnest. In early February 2007, the Democratic National Committee held an audition of sorts for ten Democrats who had put themselves forward as potential presidents (Nagourney and Heealy 2007). At that point, the war in Iraq was the dominant issue, and Democratic candidates were competing with each other over who would end it better and more quickly. The eld included four U.S. senatorsJoe Biden of Delaware, Hillary Clinton of New York, Christopher Dodd of Connecticut, and Barack Obama of Illinois. They were joined by four other major
Religion and the 2008 Presidential Primaries


guresformer North Carolina senator John Edwards, former Iowa governor Tom Vilsack, sitting New Mexico governor Bill Richardson, and Ohio congressman Dennis Kucinich. A major issue in judging the viability of candidates has always been fund-raising prowess. The rst real test of that came in early April, when contribution reports from the rst quarter of 2007 became public. In that period, Clinton and Obama raised more than $20 million each and Edwards took in more than $12 million. These three quickly became dened as the top tier of candidates (Kornblut and Mosk 2007), while the others were quickly labeled as also-rans, and media attention narrowed accordingly. Several Democratic presidential contenders for 2008 had a faith or church strategy, but each of the top-tier Democratic candidates was particularly focused on religious outreach (Democratic Rivals 2007). All three had hired religious outreach advisors early in their campaigns.1 From a personal perspective, each of these presidential aspirants also had public claims to a sincere faith (Rosen 2007; Fields 2007). In October 2006 Time magazine had published a long piece by Obama entitled My Spiritual Journey, a long excerpt from one of his two already-published memoirs, along with a photo essay promoting him as the possible next president (Scott 2008). Anonymous friends of Hillary Clinton claimed that religion has always been important to her, that she attended prayer group meetings while rst lady, and that she joined a Senate prayer group shortly after winning election in 2000. It would be wrong, these friends noted, to discount her expressions of faith as cynical maneuverings for a 2008 presidential run (Bolton 2006). John Edwards, too, early expressed his religious bona des, claiming that since 1996, when his teenage son died in an auto accident, his faith had come roaring back from a time of spiritual dryness. Since then, he claimed, My faith informs everything I think and do (Kennedy 2007). Based on the lessons learned in the 2006 congressional elections, the Democratic Party wanted to signal to religious voters that the party was not going to ignore them. Consequently, these three major Democratic candidates were showcased in early June 2007 in a special candidate forum on the campus of Washington D.C.s George Washington University that was particularly targeted toward religious issues and the religious background of the candidates. The forum was sponsored by the evangelical group Sojourners and broadcast by CNN. In an evening that the Washington Post called unprecedented, all three candidates described how faith inuenced both their politics and their personal lives (Bacon 2007). Answering questions from the moderator and a group of ministers and


The Disappearing God Gap?

religious leaders familiar with Democratic politics, none of the candidates offered answers that strayed from party orthodoxy on key issues. But there were also questions and answers on the rare territory, for Democrats, of the candidates religious beliefs and practices. The generally favorable coverage of the event led to expectations that Democrats would use more religious language and symbolism in their pursuit of moderate Catholic and mainline Protestant voters (Silk 2007). A favorable summary of the event in The Christian Century, the agship magazine of mainline Protestantism, stated that the old stereotype of Democrats being hostile to religious faith and uncomfortable in front of evangelicals was being chisel(ed) away, with candidates giving Americans a glimpse of their soulful side (Democratic Rivals 2007). The Christian Science Monitor noted the irony: in 2008, the top Democratic contenders may be more comfortable elding questions on religion than todays top Republicans (Feldmann 2007). Clinton appeared to be winning early. By the end of 2007, Clinton held a substantial lead in superdelegates, and she was leading in the national polls with about 40 percent of likely Democratic voters (Thee 2007; Nagourney and Thee 2007). In national polls, Obama and Edwards were below Clinton but far above the rest of the pack. The picture differed slightly in the early selection states of Iowa and New Hampshire, as the top three were generally more tightly packed and farther above the rest of the eld (Thee 2007; Nagourney and Thee 2007).

Republican Candidates
Like the Democrats, many Republican presidential aspirants had made clear their intentions by early 2007. Former Wisconsin governor Tommy Thompson, former Virginia governor Jim Gilmore, and sitting Kansas senator Sam Brownback were in, and former Massachusetts governor Mitt Romney was forming an exploratory committee. Three U.S. congressmenRon Paul of Texas, Duncan Hunter of California, and Tom Tancredo of Coloradowere also candidates. The eld was rounded out by former Arkansas governor Mike Huckabee, Senator John McCain of Arizona, and former New York City mayor Rudy Giuliani. Very early polls typically placed McCain at the top of the pack in terms of public opinion, with Giuliani close and sometimes leading (Its Clinton and McCain 2007). Romney had signicant nancial resources of his own, and he was also generally placed by pundits and GOP insiders in the top tier with McCain and Giuliani, while the others were typically relegated to second-tier, if not extreme long-shot, status (Canellos 2006).
Religion and the 2008 Presidential Primaries


Traditionally, organized conservative religious groups have had a great deal of inuence in the GOP nomination process. The Christian Coalition, for example, is widely credited with helping Republicans win control of the House and Senate in 1994, largely through the strength of its state-by-state mobilization efforts. With President George W. Bushs election in 2000, the conventional wisdom was that the major Religious Right leadership groups, such as Focus on the Family and the Family Research Council, had one of their own in the White House. Although the relationship between the Bush White House and top Religious Right organizations was more complex and conicted than usually reported (Rozell and Whitney 2007), religious conservatives had recently exerted strong inuence on internal Republican politics at the level of presidential nominations. This is why the early portion of the 2008 Republican race was so extraordinary. None of the three top-tier candidates was a comfortable t for these traditional religious groups. McCain was very quiet about his faith. Moreover, in his 2000 campaign for the presidency, he had strongly criticized the leading gures and institutions of the Religious Right,2 and at the start of the 2008 campaign he seemed insufciently enthusiastic about key issues on the agenda of politically active social conservatives. Still, in recent years, McCain had been seeking to make amends with the Religious Right. Mitt Romney had adopted nearly all the policy positions in which religious conservatives had an interest, but his Mormon faith was considered outside the Christian tradition by most Christian denominationsa major stumbling block for many Republicans (Helman 2006; Nagourney and Goodstein 2007). Giuliani, the lone or co-frontrunner in early polls, was apparently only a nominal Catholic, and he held views on abortion and gay rights that were opposite those of most religious groups involved in GOP politics (Page 2007a). Even more worrisome to older leaders of the Religious Right was that Giuliani was, at the time, narrowly ahead of McCain among self-identied evangelical Republican voters, although they suspected that Giulianis support would fall once his positions and personal life (two of his marriages had ended problematically) became better known (Blunt 2007). The GOP long shot with the most credible religious credentials, Mike Huckabee, was problematic for other reasons. As a former pastor, Huckabee had the clearest and most genuine religious credentials, but as governor of Arkansas he displayed a propensity for compromising with the left on taxes and spendingtoo easily so for many Republicans (Dinan 2007a). As with the Democrats, formal Republican events began early. The GOPs rst major candidate unveiling was in early May 2007 at the Ronald


The Disappearing God Gap?

Reagan Presidential Library in southern California. In this ten-candidate debate, key differences on moral issues came out clearly. Huckabee, Brownback, Tancredo, and McCain all noted some concerns with teaching evolutionary theory alone in schools, whereas the six others did not (Nagourney and Santora 2007). Only one candidate, Giuliani, supported abortion rights to any extent, and even he seemed to move rightward in declaring his intention, if elected president, to support conservative federal judges with strict constructionist viewsa position usually at odds with the reasoning behind Roe v. Wade. Giuliani even asserted he would accept Roes repeal, although would not push for it (Dinan 2007b). On federally funded embryonic stem cell research, another hot-button issue of some religious conservatives, only Romney of the top-tier candidates clearly opposed it (Balz and Shear 2007). The outcome of the debate was mixed. The religiously problematic Mormon Romney was most in line with the issues of concern to most religious conservatives, whereas the ostensibly more electable Giuliani and McCain, and even the more comfortably orthodox Huckabee and Brownback, had clear, if individualized, sets of policy departures from various powerful portions of the party (First GOP Debate 2007). A later, relatively broadly participatory event for Republicans was the Iowa straw poll, held as a fund-raising and publicity stunt in Ames on August 11, 2007. Romney decided early to expend a great deal of time, effort, and money there to establish his credibility and organizational prowess. In response, the other front-runners McCain and Giuliani essentially pulled out rather than compete with Romney, who was reported to have spent up to $5 million in the state before the straw poll to inuence its outcome (Shear and MacGillis 2007a). With Romney all but assured of rst place, the Iowa straw poll became a place to see who the most likely challengers, other than McCain and Giuliani, might be. Brownback and Thompson had toiled long in Iowa, and Huckabee had dabbled late. As expected, Romney won easily with about 32 percent of the straw voters. Brownback took a disappointing third place with 15 percent of the vote, suggesting he was not the voters choice as a Christian alternative to Romney (Nagourney and Zeleny 2007). Tommy Thompson fared so poorly that he pulled out a few days later. The better-than-expected prize went to Mike Huckabee, who took second place with 18 percent, an impressive showing given his small investment of time and money. He had a quiet campaign that was seemingly backed by homeschool proponents, a movement especially prominent among evangelicals in Iowa (Slevin and Bacon 2007). Together, the
Religion and the 2008 Presidential Primaries


Huckabee and Brownback share of the vote was one percentage point larger than the share won alone by the heavy-spending Romney. While missed by most commentators, the numbers were instructive: the two most openly religious candidates had, despite long odds and little spending, shown that, while conservative Christians active in GOP politics had not settled on any one candidate, a large share of them seemed uncomfortable with Romney. Brownback was personally disappointed in his Iowa showing and began to campaign indifferently, but Huckabee enjoyed a boost in popularity and donations from his good showing (Cook 2007). Through September and early October the GOP race stayed unsettled. Opinion polls kept Giuliani at or near the top in GOP voter preferences nationwide, usually at around 30 percent. McCain and Romney were close together and about ten points back, but McCain seemed to be losing support and nancial backing on a daily basis. Former U.S. senator Fred Thompson of Tennessee joined the race in September, with polls before and immediately after his announcement usually giving him a share roughly equal to Romney and McCain (Page 2007b). The religious factor provides the greatest explanatory power in understanding the ups and downs of the fall GOP campaigns. The traditional religious leaders active in the party were concerned about Giulianis front-runner status, and seemed a bit perplexed and disturbed that selfidentied evangelical Republican voters either did not know or did not care about the nonconservative elements of Giulianis positions and lifestyle. Even as these religious leaders argued Giulianis poll numbers would drop as voters learned more about him, they seemed unable to make that happen quickly enough, and worked among themselves to rally around an alternative candidate (Blunt 2007). Some even contemplated more drastic measures, raising the possibility of running a third party candidate should Giuliani be the GOP nominee (Hallow 2007a). However, for many of these politically active religious leaders, each of the major alternatives to Giuliani had important weaknesses. McCain had always had trouble with core religious activists, his campaign seemed to be oundering, and his friend Fred Thompsonwho had entered the race when McCains campaign seemed especially weakwas drawing new attention (Hallow 2007b). Although Thompson was not particularly religious himself, his rationale seemed to be that he was an acceptable alternative to Christian Right leaders who were very strongly against Giuliani and uncomfortable with both Romney and Huckabee (Kranish 2007). Romney had sufcient funds, the right stand on issues, and many other substantive and stylistic advantages. But his Mormonism remained a very large problem among such leaders and their followers. Huckabee was the


The Disappearing God Gap?

perfect theological t, but he failed on nonreligious issues. Seen by some as too nave in foreign policy and insufciently supportive of the Bush administration in the Middle East, he had also criticized much of the GOP establishment, including religious leaders, for being out of touch with the economic concerns of ordinary voters, emphasizing his humble economic roots and Main Street (as opposed to Wall Street) values. Finally, he was perceived to be insufciently tough as a campaigner, matching up poorly at the time against anticipated Democratic nominee Hillary Clinton (Bailey 2007). The mid-October 2007 Value Voters Summit illustrated the complexity of the situation. The summit was a two-day extravaganza hosted by the Family Research Council (FRC) and attended by about 2,500 conservative religious activists. Most of the GOP candidates personally addressed the group. If the group could reach a consensus on a candidate, observers inside and outside the religious movement agreed, it would be a signicant plus for that candidate on the road to the nomination (Luo 2007a). That would not happen, however. A key part of this summit was a complex presidential preference straw poll. Voters could vote in the straw poll by two methods. Any member of FRCs Action group could vote, whether or not they attended the conference. Romneys campaign had been actively encouraging people to join this group and vote on-line, and approximately 3,500 new people had joined in the past few months, many presumably at Romneys urging. Of the 2,500 people at the summit, only 1,500 voted in the poll at all, and only about 1,000 of those voted at the summit itself. The results announced at the summit were telling. Overall, nearly 5,800 votes were cast, and Romney barely edged out Huckabee with 27.6 percent to 27.2 percent of the overall vote. But of the votes cast in person at the summit, Huckabee was the runaway winner, with over 50 percent, and Romney a distant second with 10 percent (Luo 2007b). Giuliani and McCain nished eighth and ninth, respectively, with only handfuls of votes (Eichel 2007). To anyone watching closely, there was a growing ssure between leaders of religiously conservative groups and the more spontaneous and independent grassroots voters. While Romney had the organizational wherewithal and foresight to be competitive among religious Republican voters, the immediate enthusiasm seemed more strongly behind Huckabee. Romney seemed to be doing well with some D.C.-based leaders of the organized religious groups, perhaps because he was uniformly conservative on economic and foreign policy issues and had the nancial resources to run a top-level campaign. But Huckabee seemed to have strong grassroots support and was holding his own, if not making real gains, among the rank and le (Luo 2007c).
Religion and the 2008 Presidential Primaries


His more populist economic views gave him strength in drawing lowerincome religious voters to his candidacy (Buzz for a Baptist 2007). Given the impasse among conservative religious leaders and followers, the race was wide open (Kranish 2007). Romney tried to address his Mormon problem on December 6 with a speech at the George H. W. Bush Library in Texas, one that gained generally favorable reviews among the leaders of religious groups. In it, Romney had a difcult task. On the one hand, he had to present himself as a deeply and openly religious person to appeal to a similar key voting bloc in Republican primaries. On the other hand, he had to cast himself as a strong believer tolerant of other beliefs, a president who would not impose his particular views on others (Luo 2007d). For most leaders of the Christian Right organizations, Romney did well (Balz 2007). Still, from the perspective of the mass public, Romneys prospects were unsettled. While Romney was perceived to be the most religious candidate (with more than one-fth perceiving him to be very religious), more than one-half of the American people responded that they did not know enough about his religiosity to evaluate him (see table 3.1).

Perceptions of Candidate Religiosity and Favorability Ratings

At this early stage in the campaign process, the role of religion in shaping public opinion about these candidates was somewhat mixed. On the one hand, voters tended to know relatively little about the religious characteristics of most candidates. But on the other hand, their evaluations about the religiosity of such candidates shaped their favorability assessments of them. Table 3.1 shows voters assessments of the perceived religiosity of each of the three major Democratic candidates.3 Note that many Americans reported that they did not know much about the religiosity of most presidential candidates at this point in the campaign. In part, this lack of knowledge about the religiosity of presidential candidates is a function of the fact that Americans do not know very much about most presidential candidates at this point in time. Still, the one exception is noteworthy: despite her highly visible public presence for nearly two decades, nearly onequarter of the American people reported that they did not know enough about Hillary Clintons religious life to assess her level of religiosity. Of course, if a candidates religiosity did not matter in public evaluations of the candidate, then such perceptions of a candidates religiosity might matter little. But for better or worse, public perceptions of a candidates religiosity shape public evaluations of that candidate. This is evident in table 3.2.


The Disappearing God Gap?

table 3.1 Perceived Religiosity of Candidates Prior to the Iowa Caucuses

Perceived Religiosity of Candidate Not too/ Very Religious (%) Democratic candidates Hillary Clinton John Edwards Barack Obama Republican candidates Rudy Giuliani John McCain Mitt Romney 9 11 22 41 40 21 15 9 5 35 39 53 100 99 101 12 18 15 41 37 36 24 10 10 22 35 40 99 100 101a Fairly Religious (%) Not at All Religious (%) Dont Know (%) Total (%)

Source: Religion and Public Life Survey, August 118, 2007.


Percentages over 100 due to rounding.

Table 3.2 examines whether the respondent reports either very favorable or mostly favorable evaluations4 of the three top candidates in the two major parties by perceptions of their religiosity, while controlling for the partisan identication of the respondent. Since those who identify with a particular party are prone to have more favorable views of the candidates of their parties, it is important to control for such partisan identications. Doing so allows one to ascertain whether evaluations of a particular candidate are more a function of partisan identications or perceptions of religiosity. Clearly partisan identications help to shape evaluations of candidates. At each level of perceived religiosity, Democratic partisan identiers give higher evaluations of their partys candidates than do those who do not identify as Democrats, and the same patterns holds true with regard to Republican candidates, when comparing Republican party identiers and those who do not identify as Republicans. For example, among those who perceive a candidate to be very religious, there is typically about a fteen- to twenty-point difference in the percentage expressing positive evaluations of that candidate between those who identify with the party of that particular candidate and those who do not. However, while the level of approval varies by partisan identication, the pattern is consistent. Regardless of whether or not one identies with the party of the candidates being evaluated, the greater the perception of
Religion and the 2008 Presidential Primaries


table 3.2 Favorable View of Candidate by Perceived Religiosity Controlling for Party
Identication Perceived Religiosity of Candidate Very Very/Mostly Favorable View Democratic candidates Hillary Clinton Democrats Others John Edwards Democrats Others Barack Obama Democrats Others Republican candidates Rudy Giuliani Republicans Others John McCain Republicans Others Mitt Romney Republicans Others 88 51 79 48 72 20 91 70 76 60 46 41 88 72 92 66 64 36 89 73 89 64 42 26 89 77 83 58 52 19 95 78 90 61 74 15 Religious (%) Fairly Religious (%) Not too/Not at All Religious (%)

Source: Pew Religious and Public Life Survey, August 118, 2007.

being religious, the higher the percentage of respondents who report favorable evaluations of the candidate. As perceptions of the religiosity of the candidate declines, the percentages of those who report favorable assessments of the candidate also decline. And it is when the candidate is not perceived to be very religious that the effects of partisan identications on candidate evaluations tend to be the greatest, as the difference in positive evaluations of such candidates between those who identify with the party and those who do not tend to be much greater at the low end of perceived religiosity. Thus, these data suggest that it is to the candidates political advantage to be perceived as being religious, as such perceptions are related to greater positive evaluations of the candidate, and that this


The Disappearing God Gap?

advantage works to its greatest advantage among those who do not identify with the candidates political party.

The Iowa and New Hampshire Contests

Iowa and New Hampshire have exerted early, and many argue disproportionate, inuence on the presidential selection process. Iowas caucuses provide a test for a candidates organizational skills in a rural and smalltown midwestern market. Supposedly this quintessential yover state tests a candidates appeal to the broad middle of America. On the other hand, New Hampshires New England location and tiny primary electorate supposedly make it an ideal testing ground for door-to-door retail politics. The argument for the Granite State is that its citizens can detect lack of sincerity or competence with so much up-close interaction. Defenders of the existing system claim Iowa and New Hampshire, particularly in concert, effectively winnow down primary elds to the best remaining contenders. But whatever may be the arguments for these early contests, each state provides an opportunity for religious groups (among others) to generate momentum for their favored candidates. As a result, the religious composition of each state helped to shape its election outcome in important ways.

Iowa selects presidential delegates in hundreds of local caucuses, where support is determined by each candidates ability to turn out many supporters at a particular place on one particular evening in the winter. In every state, churches are a means of social afliation and networking, but this is particularly true in Iowa, where approximately 79 percent of people (compared to 68 percent nationally) afliate with either a Protestant or Catholic church (Pew Forum 2008c). Successful presidential candidates learn to work the Iowa church- and faith-afliated networks strongly; such afliates include church-related schools and an increasingly popular homeschool network. The history of both political parties includes church-related Iowa caucus surprises. Born-again outsider Jimmy Carter in 1976 saw the opportunity Iowa presented, and he successfully organized an impressive turnout in the caucuses to launch his successful presidential campaign. Evangelical television personality Pat Robertson placed second to Bob Dole in 1988 (both placed ahead of sitting vice president George H. W. Bush).
Religion and the 2008 Presidential Primaries


Iowa Republicans. Republican presidential candidates, in particular, nd religious networking indispensible in Iowa. White Protestants are extremely inuential in the state GOP, a disproportionate share of the partys caucus-goers identify themselves as born again or evangelical (Shear and MacGillis 2007b), and state party ofcials typically work closely with members of pro-life and other socially conservative organizations. Romney had been particularly active in the more politicized religious networks, and while his Mormonism continued to trouble Iowa voters, he seemed to be having some success. Huckabee had an easier time in explaining his faith to Iowa voters, but a harder time getting into the established Iowa networks. He built his strategy more among the newly emerging homeschool network in Iowa, along with other elements of the faith community most resistant to Romney (Slevin and Bacon 2007). Media reports detected a late and mostly spontaneous burst for Huckabee and, indeed, he won with a surprisingly large share of the vote: 34 percent (Nagourney 2008a). Romney appeared to fall in the last few days, and he came in second with 25 percent. Essentially tied for the GOPs third spot were Fred Thompson and McCain. Giuliani trailed far behind at 4 percent after a very weak effort. Huckabees strongest supporters were religious and conservative voters, who turned out disproportionally in the 2008 caucuses. Nearly half (45 percent) of GOP caucus-goers described themselves as very conservative, and 35 percent of these voters supported Huckabee compared to 23 percent for Romney and 22 percent for Thompson (Pew Forum 2008a). Self-identied evangelical Christians constituted 60 percent of caucusgoers, and they went overwhelmingly for Huckabee (46 percent compared to 19 percent for his closest challenger, Romney), with Huckabee taking 56 percent of those voters who said that it mattered a great deal that a candidate shares your religious belief (Pew Forum 2008a). Iowa showed Huckabees strength among religious conservatives, the traditional core constituency of the Republican Party, as well as Romneys very weak attachment among themat least as long as there was a viable alternative in the race. Iowa Democrats. As with Republicans, religion plays a more signicant role in the lives of Iowa Democrats and their evaluation of presidential candidates than in many other states (George 2007). Each of the leading Democratic contenders for the White House made it a point to discuss their faith journeys, to talk about how their beliefs inuenced their policies, and to reach out to religious communities. But Obama, in particular, made religion a signature part of his campaign through his own public appearances in places where Democrats rarely venture, scheduling faith


The Disappearing God Gap?

forums and choosing to employ faith-based voter mobilization (Gorski 2007). Still, prior to the Iowa caucuses, Hillary Clinton was considered the prohibitive favorite for the nomination, and some thought John Edwardss populist prounion rhetoric would do him well. Barack Obama, however, pulled off the surprising rst of his many victories, winning with 38 percent of the vote, with Edwards and Clinton trailing together at 30 and 29 percent, respectively (Nagourney 2008a). Though the delegate count was signicantly closer, the media focused on the overall vote totals in this rst contest. Most media reports noted that the chief component of Obamas coalition was younger voters. Only a few, however, noted that Obama did well in more religious parts of the state, by using strong moral language reminiscent of the civil rights era in describing his campaigns purpose (Harper 2008) and by organizing faith forums in hundreds of local congregations across the state (Cummings 2008). With this mix of youthful and religious enthusiasm at his back, Obama immediately moved to the top of the prospective nominee list. Donors and commentators were surprised by Obamas strength, and discussion of a protracted primary battle arose. The eld also quickly narrowed somewhat, as Biden and Dodd withdrew from the race in short order (Cooper 2008a).

New Hampshire
The New Hampshire primary on January 8 quickly followed on the heels of the Iowa caucuses, with both Republican and Democratic races affected by the Iowa results. For Democrats, Clinton was no longer the obvious favorite, and Obama seemed the most likely to overtake her. Thus, Clinton needed to stop Obamas momentum in New Hampshire. For Republicans, Huckabees surprising Iowa strength showed the weaknesses of its front-runners, Romney and McCain, with the core of the party, and New Hampshire became even more important to the Romney and McCain campaigns. Fortunately for them, New Hampshire is no Iowa in religious terms. The Granite State is both more Catholic and secular, and far less evangelical Protestant. Twenty-ve percent of New Hampshire citizens have no religious afliation, ten percentage points more than in Iowa (Pew Forum 2008b). Similarly, only 11 percent of New Hampshire citizens call themselves evangelical, fewer than half the percentage of Iowans, whereas Catholics make up 40 percent of New Hampshire residents but only 22
Religion and the 2008 Presidential Primaries


percent of Iowans (Pew Forum 2008b). And the political mix with religion is quite different, as New Hampshire faith-related organizations tend to be very small, low-budget, grassroots groups that hold little political power at the state level. Thus, New Hampshire seemingly provided a far better electoral environment than did Iowa for both the McCain and Romney campaigns. New Hampshire Democrats. In the aftermath of his surprising Iowa win, Obama fueled speculation that he might ride a wave of momentum through New Hampshire and then onward, fairly quickly to the nomination. Indeed, his Iowa win was reected immediately in New Hampshire polls, where his lead quickly reached ten percentage points (Balz 2008a). That was his high point, however. Clinton fought doggedly and, although attacked by both Obama and Edwards in a televised January 5 debate, she showed her strength in retail politics. The race turned quickly tighter and personal, as she asserted that only she could bring substantive change, mocking Obamas claim as a change agent to be merely rhetorical: Making change is not about what you believe; its not about a speech you make. Its about working hard (Healy and Zeleny 2008a). Voters seemed to rally toward Clinton, as she won a surprising threepoint victory over Obama in the popular vote (Helman and Milligan 2008). Exit poll results, however, revealed some interesting distinctions between the Democratic candidates along religious lines. Obama won 45 to 29 percent in the large category of New Hampshire Democratic primary voters who claimed no religious afliation; the numbers were reversed among Catholic voters in favor of Clinton, and the two major candidates virtually tied among Protestants. Obama won a plurality of both the most frequent and least frequent worshipperswinning by a 38 to 31 margin among those who attended religious services once a week or more and by a similar margin among those who never attended. Clinton won by ve percentage points among those in the other two categories those who attended between a few times a month and a few times a year (Pew Forum 2008b). New Hampshire Republicans. New Hampshire was far more favorable ground for McCain than Iowa. McCain had conducted countless town hall meetings in the Granite State throughout the fall and winter, banking on it to establish him as a top-tier candidate, just as his victory in 2000 had done eight years earlier. In fact, McCain seemed to be basing his entire campaign comeback on doing well in New Hampshire, a goal that became even more important after his distant third-place nish in Iowa. Huckabees religious identity was less a threat to his opponents in New Hampshire, as New Hampshires Republican voters were far less


The Disappearing God Gap?

evangelical than Iowas and made up only 20 percent of the primary electorate (Bacon 2008). Moreover, Huckabee had not put much time into New Hampshire, focusing before it on Iowa and after it on the more religious southern states whose primaries would follow later in January. Still, Huckabee was worth watching in terms of just how much support he might draw away from his chief competitor for religious voters, Mitt Romney, and how this would advance McCain. Polling just before the 2008 New Hampshire primary showed McCain with a slight edge over Romney, with Giuliani and Huckabee far back. In the end, McCain earned 37 percent of the vote, a ve-point margin over Romney. Huckabee beat Giuliani for third, 12 to 9 percent, a positive nish for the former and another disappointment for the latter (Helman and Milligan 2008). In exit poll results, Huckabee actually won by several points among those who attended worship services more than once a week, while McCain won every other category of church attenders. Huckabee also won a plurality of white evangelical Protestants, while McCain won among voters of every other religious category (Pew Forum 2008b).

The Remaining January Primaries

The rst two contests had already shaped both Republican and Democratic races. The remainder of an awkward January schedule would further dene the contests, almost settling the Republican nomination and dening the battle lines for what would turn out to be the longest possible Democratic contest. Once again these contests were held on different days spread across different regions of the country. And given the different religious composition of these different states, different candidates chose to focus their campaign efforts on winning the contest in different states. There were clear and open contests in Nevada and South Carolina, but Michigan and Florida were both problematic, especially for Democrats, as the latter two states had staged early primaries in violation of national party guidelines, leading eventually to a battle over whether to count their delegates.

McCain had earned a strong bounce from his New Hampshire win; a Washington Post/ABC News poll now had him at about 30 percent support, far ahead of Huckabee (20 percent), Romney (19 percent), and Giuliani
Religion and the 2008 Presidential Primaries


(15 percent) (Balz and Cohen 2008). But with a bounce that seemingly stemmed from just one contest, McCains lead seemed far from certain. No one seriously challenged Romney in the January 19 Nevada caucuses, as many voters were adherents of, or at least comfortable with, his Mormon faith. His organizational ability in the caucus environment, particularly among fellow Mormons, showed itself as he drew over half the support on caucus night. He easily led the pack, as he received nine out of ten Mormon votes (Milligan 2008). The only mild surprise was Ron Pauls second place showing, beating out McCainwho was focused on the South Carolina primary held on the same day and on the Florida primary scheduled ten days later. The three more important GOP contests were primariesthe Michigan primary on January 15, South Carolina four days later, and the Florida primary on January 29. Michigan loomed largest for the top three, particularly for McCain and Romney. Both had high expectations in this industrial state. Romneys father, George, had been an auto executive who became a very popular GOP governor of Michigan in the 1960s, and McCain had won the 2000 presidential primary over George W. Bush, drawing strongly on independent voters in the states open primary system (Santora and Nagourney 2008). Huckabee was hoping to stay a close third behind the top two and remain viable for the South Carolina primary, relying on pockets of support where conservative religious groups were strong (Bacon and Eilperin 2008). Romney had, perhaps, the most at stake. He had just lost New Hampshire, and he had ties so deep in Michigan that he announced his presidential candidacy there rather than in Massachusetts, where he was governor. Michigan was hard fought. The dominant issue was the economy, as the state was in a recession closely connected to the faltering domestic auto industry. The state had the nations highest unemployment rate (7.4 percent on the primary date) and was continuing to lose jobs from car companies and other parts of its heavy manufacturing employment base. McCains straight talk got him into some trouble in the state. Burnishing his economic conservatism, McCain sharply attacked Romneys tax record as Massachusetts governor and touted his own position in favor of maintaining the Bush tax cuts. But on job retention, McCain may have been too frank when he declared that there are some jobs that arent coming back to Michigan (Santora and Nagourney 2008). Romney saw McCains statement on jobs as an opportunity to separate himself from McCain; he called the Arizona senator pessimistic and declared, based on his family heritage, that he had the automobile industry in my blood and veins (Weisman 2008). Romneys strategy succeeded that day, as he won 39 to


The Disappearing God Gap?

30 percent over McCain, with Huckabee third at 16 percent (Broder 2008a). With Michigan bringing no surprise and Nevada written off by all but Romney, South Carolina was the next important contest. In 2000, George W. Bush had nally put away John McCain in the state, relying on controversial tactics, friends, and rumors that rallied the states strongly evangelical voter base against McCain (Eichel 2008). As a result, McCain had worked hard in 2008 to cultivate some evangelical support. Others in the primary avoided the state altogether; Romney focused on Mormonfriendly Nevada, while Giuliani was postponing his rst full-scale effort for Floridaa far more religiously diverse state whose primary was ten days later (Luo and Cooper 2008). Huckabee put a great deal of resources into South Carolina, as he needed to win or do very well there for his campaign to be perceived as viable. Polls indicated that this was possible. McCain and Huckabee were at the top of most South Carolina polls, with Romney and Fred Thompson each garnering about one-half the support of the top two (Balz and Eilperin 2008). In the last few days of South Carolina campaigning, Thompson abetted McCain, intentionally or unintentionally, as the Tennessean attacked Huckabees tax and budget policies when he was Arkansas governor, arguing they were insufciently conservative (Cooper and Luo 2008). In the end, McCain won the state, beating Huckabee by three percentage points; Thompson placed third and withdrew the next day (Nagourney 2008c). Florida emerged ten days later as the test of the remaining Republican big threeMcCain, Romney, and Huckabeeagainst Giuliani, who at last was making an all-out effort (Nagourney 2008b). Campaigning in Florida with virtually no competition before the South Carolina contest, Giuliani did well in early polls, but following the South Carolina contest his numbers fell. A few days out from the Florida vote, McCain and Romney were at the top of Florida polls with about 30 percent each, and Giuliani was nearly tied with Huckabee at around 15 percent each (Current Polls 2008). Florida turned out to be Giulianis rst, and last, stand: McCain won 36 to 31 over Romney, with Giuliani a distant third with about 15 percent, barely above Huckabee (Cooper and Thee 2008). The former New York City mayor quickly withdrew and endorsed John McCain. Thus, during the January contests, Huckabee used his faith-based organizing in Iowa to propel him from obscurity, and he was able to use South Carolina, the other key January contest with a strongly religious GOP base, to keep himself in the running. Romney had strongly courted
Religion and the 2008 Presidential Primaries


this same base, but despite his nancial advantages and socially conservative positions on issues dear to that base, was never quite able to close the deal with them, probably because of his Mormon faith. And the seemingly strong Giuliani, the most secular of the GOP candidates, never could gain a foothold in a series of Republican electorates across different states whose electorates were strongly inuenced by religious messages and faith-based policy positions. So as the February primaries and the Super Tuesday (February 5) in particular loomed, the GOP eld had narrowed to three major candidates, plus pesky Ron Paul. John McCain was a clear, but not-quite-sure, leader, over Romney; Huckabee was a doggedly persistent third candidate with a fairly strong base among the religiously inclined Republican primary voters.

After the New Hampshire vote, Democratic momentum returned to Clinton. January remained a treacherous month for the Democrats. Both Michigan and Florida had violated party rules by moving their primaries earlier in the year, and national Democratic leaders were urging all Democratic presidential candidates to boycott the states. Most did avoid active campaigning in those two states, and competed openly only in Nevada and South Carolina. The exceptions were Clinton and Dodd in Michigan, leading to the peculiar situation of Clinton outpolling a category of uncommitted primary voters 56 to 39 percent. Whether this was a delegate victory, however, remained unresolved, since the national Democratic Party ruled that Michigans 128 pledged delegates and 29 superdelegates would not count in the nominating contest unless the primary were delayed (Slevin 2007; Issenberg 2008a). The delegates ultimate fate would be decided later. In a different state with no such technicalities, Clintons New Hampshire momentum carried to Nevada, as she won the caucus 51 to 45 percent, despite Obamas endorsement by hotel unions inuential in the large cities (Milligan 2008). Obama did run strongly in rural areas throughout the state, and actually beat Clinton in the delegate count, thirteen to twelve. Edwardss support, however, had collapsed, and he became increasingly irrelevant to the race, except perhaps as a future endorser of one of the two major candidates (Murray and Kornblut 2008). Race rst rose openly as an issue in the South Carolina Democratic primaries. While Hillary Clinton campaigned in some Super Tuesday states that were voting before South Carolina, husband Bill Clinton served


The Disappearing God Gap?

as her major surrogate in South Carolina (Balz and Murray 2008a). During his presidency, Clinton had enjoyed strong African-American support, and South Carolinas large African-American population might be expected to give his wife some share of their vote based on that history. But the Clinton campaign, behind in state polls, apparently took a calculated risk to inject race as a divisive issue, gambling that the risk in South Carolina would be outweighed by reward in later states (Healy 2008a). Obama had a far less overt racial strategy in South Carolina. Rather, his main tactic was a strong dose of religion, with a great deal of work organizing in both African-American churches and more racially diverse mainline Protestant congregations. Perhaps given uncertainty about his religious faith (rumors that he was a Muslim circulated widely), Obama campaigned openly as a Christian (see gure 3.1), and he employed a new style of faith-based organizing, going into white-dominated churches and hosting nondenominational faith forums all across the state (Cummings 2008). Election-day vote totals may have revealed something of a backlash against Clintons tactics, as Obama won by a larger-than expected two-toone margin over Clinton, 55 to 27 percent. Regional favorite son John Edwards trailed far behind with only 18 percent. Edwards seemed at an end, and Obama emerged as the clear frontrunner (Balz, Kornblut, and Murray 2008).

gure 3.1
Religion and the 2008 Presidential Primaries


But Obamas momentum slowed with Clintons win ten days later in Florida, the other unsanctioned January primary. Clinton won 50 to 33 percent, without either candidate campaigning (Broder 2008b). More important than the Florida primary itself was Edwardss suspension of his candidacy one day later. He did not immediately endorse either remaining major candidate, athough both candidates vowed to pick up his central theme of ghting poverty (Balz and Kornblut 2008a). Thus, heading into Super Tuesday on February 5, neither Clinton nor Obama had a clear advantage. In a few days, twenty-three states and territories and roughly 1,700 delegates would be at stake, and they would bring more media attention than any presidential primary day in American history (Saul 2008).

The Super Tuesday Primaries

The Tuesday on which most states hold their presidential primary elections is known as Super Tuesday. Traditionally, Super Tuesday has had as many as ten or so states with simultaneous elections. In 2008, however, many states chose to move their primary dates up as early as party rules allowed. In all, twenty-three states had Democratic Party contests that day, and twenty-one (most of them overlapping) had Republican primaries or caucuses. Spread across the nation, from California to New York, with every region of the nation substantially represented, it was close to constituting the rst-ever national primary for both parties. A sweep or near sweep by either Clinton or Obama would clinch it; but each candidate would rather ght to a draw than let the other win. For Republicans, it looked like the day McCain would wrap it up. By Super Tuesday, leading Republican ofcials were rallying around McCain. Giuliani had just endorsed him, and on January 31, big state governors Arnold Schwarzenegger of California and Rick Perry of Texas did the same (Eilperin and Shear 2008). The Romney campaign, especially, was showing strain as party establishment types started to coalesce elsewhere, while Huckabee maintained a loyal cohort that was preventing Romney from shoring up the religious vote (Kornblut and Shear 2008). The vote was mostly predictable on the Republican side. McCain won California and 173 of its delegates, as well as his home state of Arizona, and the states of Connecticut, Delaware, Illinois, Missouri, New Jersey, New York, and Oklahoma, most of them big. Romney won his home state of Massachusetts, and the smaller states of Alaska, Colorado, Minnesota, Montana, North Dakota, and Utah. Huckabee did better than expected,


The Disappearing God Gap?

winning his home state of Arkansas but more surprisingly the Souths Alabama, Georgia, Tennessee, and West Virginia. With his big state victories, McCain won by far the most delegates (Balz and Kornblut 2008b). A closer look at GOP results show that the core evangelical vote was split. Christianity Today reported that evangelicals had divided their support nearly evenly on Super Tuesday between the top three [Republican] candidates, as exit polls showed that Huckabee had won 34 percent of these voters, Romney 31 percent, and McCain 29 percent (Hansen and Pulliam 2008). In delegates, McCain had more than 700, with 1,191 needed to win. Romney was second in delegates with about 290, and Huckabees tally climbed above 200 (Curl 2008a). As the likely nominee, McCain began to directly court the more conservative supporters of Huckabee and Romney (Feldmann 2008), even though older evangelical leaders resisted. In fact, Focus on the Familys James Dobson even stated that he would not vote in November if McCain were the nominee, a path that would doom McCain if many other evangelicals were to follow suit (Hansen and Pulliam 2008). Obama had once again surged in the polls as well as in the size of crowds he attracted following his South Carolina win (Kornblut and Shear 2008). People noticed. Senator Ted Kennedy of Massachusetts, a lion in the Senate and the party, endorsed Obama on January 26 in a move that infuriated the Clinton campaign (Zeleny and Hulse 2008). With Edwards out, Obama and Clinton met in a one-on-one debate for the rst time on January 31. They were surprisingly civil, looking like they wanted to move beyond the racially charged language and tactics of South Carolina and other recent clashes (Zeleny and Healy 2008). With a total of nearly 1,700 Democratic delegates at stake, Super Tuesday was a huge prize for both. At the same time, each candidate seemed to be focusing on a different array of states that, if such activity predicted voting, made it likely that the night would be inconclusive. And so it was. Both Obama and Clinton could, and did, plausibly claim victory. Obama won thirteen mostly smaller states, and Clinton was victorious in ten, including the big prizes of New York, New Jersey, California, and even Kennedys own Massachusetts (Balz and Kornblut 2008b). The delegate race was also close, with only thirteen delegates separating each candidates Super Tuesday totalsObama with 847 and Clinton with 834 (Murray and Mosk 2008). African-American voters continued their South Carolina trend and overwhelmingly supported Obama, and he was also doing well among more highly educated voters, and younger college-age and collegebound Democrats. Clintons demographic strengths were also becoming
Religion and the 2008 Presidential Primaries


clearer. She was relatively stronger among white women, older voters, Roman Catholics, and the less educated, working-class Democrats, with Latinos at this stage being more strongly for Clinton than Obama (Balz and Kornblut 2008b) The main conclusion of the Democrats Super Tuesday results was that the Democratic race was not over. Few observers or participants, including apparently Clinton and her campaign team, had predicted the race would continue beyond February 5. While the Super Tuesday results suggested to the campaigns that the nomination process would continue for at least another month, they couldnt foresee some of the issues that would shape the remaining contests, including unexpected controversies of faith and politics.

The PostSuper Tuesday Primaries

Romney quickly dropped out after Super Tuesday. At a Conservative Political Action Conference meeting in Washington, D.C., on February 7, Romney announced he would end his campaign, observing that McCain was the presumptive nominee (Helman and Issenberg 2008). Mike Huckabee had a different response, vowing to ght on for an indeterminate time in order to represent more directly the social conservatives in the party. And some early February victories suggested the core of his support continued to be quite substantial. On February 9 he won Kansas (handily) and Louisiana (by a slim margin), though that same day he lost to McCain in the more easily controlled caucus state of Washington (Issenberg 2008b; Lengell 2008). But Huckabees candidacy was eventually a losing cause. In the socalled Potomac primary on February 12, McCain swept Virginia, Maryland, and the District of Columbia, taking all 116 delegates and making it even more unlikely that Huckabee could win the nomination. On February 14, Romney ofcially endorsed McCain, released his nearly 280 convention delegates, and urged them to support McCain (Bacon and Shear 2008). If all or most of these delegates did so, McCain would have almost enough for the nomination with the prospects of victory in big states like Ohio and Texas wrapping it up. A few days after Romneys action, McCain was endorsed by former President George H. W. Bush (Cooper 2008b). Even so, Huckabee vowed to stay in the race, and McCain was unwilling to directly urge him out (Issenberg 2008b). But it was a last gasp. On March 4, McCain swept the primaries in Ohio, Rhode Island, Texas, and Vermont and nally obtained enough delegates to


The Disappearing God Gap?

mathematically eliminate Huckabee (Nagourney 2008d). Huckabee nally withdrew, endorsed McCain, allowing the victor to shift his campaign focus fully toward the general election (Shear and Slevin 2008). Huckabees candidacy highlighted for McCain his greatest strategic task for the immediate future as well as the fall campaign: to shore up his weakness among religiously motivated social conservatives, particularly at the grass roots. Those voters were attracted to Huckabees evangelicalism and his ability to connect faith with the issues of the day. McCain knew he had to devise a strategy for winning over that key bloc of Republican voters. However, the way forward for the Democrats after Super Tuesday was far more contentious. It became clear that the tie on Super Tuesday left Obama better positioned than Clinton for the upcoming contests in February, as he had more money and more favorable state demographics moving forward (Healy 2008b). Clinton had announced the day after Super Tuesday that she had personally loaned her campaign $5 million in January. Her core supporters said it was a sign of her personal commitment; most others interpreted it as a nancial red ag (Murray and Mosk 2008). On the campaign trail, the rst weeks in February were good for Obama. In the span of eleven days between February 9 and 19 Obama won all eleven contests and increased his pledged delegate lead over Clinton from more than 100 to about 160, roughly 1,200 to 1,040. He also was gaining in unpledged superdelegates, although Clinton held a lead of about 50 among the 430 declared superdelegates (Cooper and Wheaton 2008). The Clinton team was hoping March would be better, looking ahead to the March 4 primaries of four states with 370 delegatesOhio, Rhode Island, Texas (which had a combination of primary and caucus), and Vermont. Clinton seemed favored in Ohio with its more white, Catholic, and blue-collar Democratic electorate. Wins in the two major states would probably turn the tide toward Clinton again; a split or dual losses would almost certainly spell defeat for her campaign (Luo and Zeleny 2008). In the last week of February, Clintons campaign seemed to be back on its feet as the press was starting to scrutinize Obama more closely (Kurtz 2008). Chicago-based early Obama backer Tony Rezko went on trial in a political corruption case in Chicago, and that spilled over in some negative coverage for Obama (McIntire and Drew 2008). Controversy also erupted when the Canadian press reported that a key Obama economic advisor had privately offered assurances that Obamas public rhetoric against free trade, particularly the North American Free Trade Agreement,
Religion and the 2008 Presidential Primaries


should not be taken seriously (Memmott 2008). Obama lost momentum, and with it Clinton scored convincing wins in Ohio and Rhode Island. Texas, with its confusing two-step system of a primary vote followed by party caucuses, provided a mixed result: Clinton narrowly won the Texas primary and narrowly lost in the Texas caucuses (Balz and Murray 2008b). Clinton nevertheless claimed a strong comeback, pointing out that she was stronger in large states that counted for more Electoral College votes in November. Obama responded with reminders that he still led the delegate count, the ofcial way to gauge rst place. After two more victories in the second week of March, Obama was clearly looking like the eventual nominee (Harwood 2008). Weaknesses showed, however, as Obamas strong support among minorities and younger voters was not matched in support among working-class Catholics and women (Nagourney and Zeleny 2008a). Thus, while Obama was winning most elements of the party, there was an emerging Clinton Democrat demographic that might be up for grabs in the general election (Balz 2008b).

Religion during the Primary Campaign Lull

After the early March contests, the Democratic race entered a six-week lull, with no elections until April 22, an unprecedented halftime of active campaigning without the discipline of an immediate contest. In this interim, both campaigns escalated their attacks in campaign events, paid media, and daily conference calls with reporters and supporters, with the tenor of the campaign becoming increasingly rancorous and vitriolic (Kiely 2008). The central controversy during this time was religious, the resurfacing of Obamas relationship with the Reverend Jeremiah Wright, the former pastor of his home Trinity United Church of Christ congregation in Chicago. On March 14, clips of controversial sermons that had been preached by Rev. Wright surfaced on YouTube and received heavy viewing there and constant replay on the television news (Rother and Luo 2008). Among other things, Wright was videotaped as saying, God damn America for treating our citizens as less than human. God damn America for as long as she acts like she is God and she is supreme. Obamas opponents inside and outside the Democratic Party used the stark visual to question Obamas values and associations, which still seemed vague to most voters (Welch 2008). Obama had had a long and close association with Rev. Wright and Trinity United Church of Christ; indeed, the candidate had often openly


The Disappearing God Gap?

acknowledged his debt and friendship in the past. This close association had not been a political liability in Obamas local and state campaign elections, but on a larger stage and against more aggressive opponents, the relationship had become more problematic. Rev. Wright and the church advanced a form of black liberation theology, a theological perspective that frequently voiced the concerns of oppressed classes of people in stark economic and political terms. In liberation theology, economic and political marginalization is dened as un-Christian oppression, with Jesus being on the side of the oppressed and promising a this-world liberation. These themes of liberation theology, along with the expressions of Rev. Wright, seemed radical and dangerous to persons holding more conventional religious and political values in America, with the stark visual presentation of these values in the Wright video representing a clear threat to Obamas effort to forge a campaign image of moderation and coolness. Understanding the threat, Obama responded quickly and comprehensively. Four days after the clips surfaced, Obama delivered his More Perfect Union speech, an attempt to respond to the immediate situation with Rev. Wrightas well as to place it within the broader context of race and religion in the United States. He denounced Wrights particular remarks, while refusing to condemn the pastor or disassociate personally from someone whom he had publicly acknowledged as important to him and his family (Zeleny 2008a). Generally, the national media and political elites gave favorable reviews to Obamas response. For example, The New York Times editorialized that Senator Barack Obama, who has not faced such tests of character this year, faced one on Tuesday. It is hard to imagine how he could have handled it better (Mr. Obamas Prole 2008). The storm seemed to have settled down. But soon another Obama faux pas would arise. On April 11, a blogger on the Hufngton Post posted an audio recording of Obamas remarks at an April 6 private San Francisco fund-raiser in which he discussed Pennsylvania voters in disparaging terms (Bacon and Murray 2008). He is heard saying: You go into these small towns in Pennsylvania and, like a lot of small towns in the Midwest, the jobs have been gone now for 25 years and nothings replaced them. . . . And its not surprising then they get bitter, they cling to guns or religion or antipathy to people who arent like them or anti-immigrant sentiment or anti-trade sentiment as a way to explain their frustrations. Seeing another opportunity, Clinton pounced on those remarks and called them evidence of Obamas elitism and distaste for the traditional
Religion and the 2008 Presidential Primaries


values of many voters, particularly those in the upcoming primary state of Pennsylvania (Zeleny 2008b). Obama defended and tried to contextualize his comments. But without disavowing the general thrust that voters were being misled by traditional politicians, he eventually conceded that he didnt say it as well as I should have (Seelye and Zeleny 2008). On Sunday April 13, Senators Clinton and Obama participated in a Compassion Forum at Messiah College, a Christian liberal arts college in Grantham, Pennsylvania; Senator John McCain declined the invitation, citing a schedule conict. The televised event was moderated by CNNs Campbell Brown and Newsweeks Jon Meacham. Each candidate was on stage, separately, for approximately thirty minutes, before a bipartisan audience of faith leaders from around the country.5 Clinton sat for questions rst, speaking about her views on such varied topics as when life begins, end-of-life issues, the place of religion in the public square, climate change, possible boycott of the Beijing Olympics over the Tibet question, and why a loving God would allow suffering; Obama, who took the stage after Clinton, faced many of the same questionsplus some new ones. In her remarks, Clinton allowed that the potential for life begins at conception and that abortion must be safe and rare. Obama indicated that it was important to acknowledge that there is a moral dimension to abortion, which I think that all too often those of us who are pro-choice have not talked about or tried to tap down (Sinderbrand 2008). However, in an April 16 Pennsylvania debate between Clinton and Obama in Philadelphia, Obama was on the defensive over some of his recent slips (Kornblut and Balz 2008). As a result, in the days leading up to the voting Clinton seemed to be gaining. Indeed, she scored a convincing win in the Pennsylvania Democratic primary, beating Obama 55 to 45 percent, and gaining a handful more delegates than Obama (Kornblut and Murray 2008). With few large states remaining, Clinton had little chance to overtake Obamas advantage in pledged delegates. As a result, her longshot strategy was to push for the full counting of Florida and Michigan delegates and to convince enough unpledged superdelegates that Obama was unelectable because of his weakness in many large states, particularly blue-collar states such as Ohio and Pennsylvania (Balz 2008c). Meanwhile, McCains attempts to gain stronger conservative and openly religious support were controversial and inconclusive (Merida 2008). In early March, McCain attempted to mend fences with movement conservatives connected to the Council for National Policy, an umbrella organization of approximately four hundred politically active leaders of economic, foreign policy, and social conservative interest groups (Hallow


The Disappearing God Gap?

2008a). The council meets sporadically and prefers to maintain a low prole, though its members see themselves as important players in conservative politics. In a presentation to the group on March 7, McCain appeared to do reasonably well on economic and foreign policy issues, but left most members dissatised with his lack of enthusiasm for socially conservative issues (Hallow 2008b). To fully win their support, some council sources suggested that McCain needed to pick a more truly conservative running mate, with Huckabee mentioned prominently and positively (Wall 2008). From then through April, McCain seemed to be making some progress in consolidating a larger share of conservative voters (Curl 2008b). Slowly, the 2000 and 2004 Bush coalition was coming together for McCain, although later and with less enthusiasm. A special target was Catholic voters who might be approachable given the strong pro-choice record of the Democratic Party and, if Obama were the eventual nominee, his relative weakness among Catholics against Clinton (Slater 2008a). It is no surprise, then, that media coverage of the campaign during this period of the 2008 presidential election was full of stories of possible shifts in partisan loyalties among religious votersparticularly among younger voters within their ranks. But just how much had really changed in terms of the religious terrain and the way in which religion shapes political attitudes and preferences? This issue is addressed, in part, in table 3.3, which examines responses to a national survey conducted in April 2008.6 During the primary campaign, afliation with different religious traditions continued to shape the voting preferences of the American electorate, but for some religious groups the choice of the particular Democratic candidate paired with McCain was an additional factor. Evangelical Protestants continued to be the most Republican in their voting preferences; regardless of whether the Democratic candidate was Clinton or Obama, a little more than one-half of evangelicals surveyed indicated that they would vote for McCain, while about one-fth indicated their support for the Democratic candidate. Mainline Protestants, on the other hand, were rather evenly divided in their support for McCain and the Democratic candidate, whether Clinton or Obama, though Obama fared a little better than Clinton in head-tohead competition with McCain. Not surprisingly, black Protestants were far more supportive of Obama than Clinton, though their expressed support for McCain did not shift whether Clinton or Obama was the candidate. Roman Catholics, on the other hand, were far more drawn to Hillary Clinton than Barack Obama in the early months of the campaign; by April, Catholics were evenly
Religion and the 2008 Presidential Primaries


table 3.3 McCain versus Democratic Opponents by Religious Tradition: April 2008
Evan. Prot. McCain versus Clinton Clinton McCain Someone else Sit out Undecided Total McCain versus Obama Obama McCain Someone else Sit out Undecided Total 18 52 12 11 8 101 38 35 7 11 9 100 82 7 7 3 1 100 31 31 10 14 13 99 41 29 6 0 24 100 52 19 7 16 6 100 21 54 9 11 5 100 37 40 13 7 3 100 32 7 40 17 5 101a 45 30 8 10 7 100 65 18 0 6 12 101 44 18 19 11 8 100 Main. Prot. Black Prot. Rom. Cath. Jews Relig. Unafl.

Source: Henry Institute National Survey on Religion and Public Life, AprilMay 2008.

Percentages over 100 due to rounding.

divided in their support for McCain and Obama, but they were about 50 percent more likely to support Clinton than McCain. Jews also were more supportive of Clinton than Obama, though both candidates garnered far greater support among Jews than did McCain. And nally, the religious unafliated were far more drawn to Obama than Clinton, but regardless of McCains opponent, less than one-fth of the religiously unafliated supported McCain. Still, while McCain was seemingly retaining the support of many evangelical Protestants (though the intensity of their support for McCain in 2008 may have been weaker than it had been for Bush in April 2004), his religious outreach to the evangelical community was not without its difculties. In late May, McCain was forced to reject the endorsements of two relatively prominent evangelical ministers who had backed him in late February when the GOP nomination was still somewhat in doubt namely, Rev. John Hagee of San Antonio, Texas, and Rev. Rod Parsley of Columbus, Ohio. Each had made remarksthen circulating on the Internetthat were causing controversy and offense (Banerjee and Luo 2008). Hagee, a well-known pastor among southern evangelicals, is a premillennial dispensationalist, a Christian tradition that takes a literalist


The Disappearing God Gap?

approach to biblical prophecy and is particularly attentive and sympathetic to modern Israel, believing it is closely connected to the ancient biblical people of Israel. It is a controversial theological view not widely shared, at least with such intensity and focus, within the Christian community more broadly nor even necessarily within the evangelical community more narrowly. In any case, a watchdog group critical of Hagee had released a tape of his remarks about Gods role in the Holocaust that raised questions about the pastors political views and, by afliation, the views and political acuity of McCain. Parsley is a more broadly known pastor with generally less controversial views. But he, too, had delivered a widely circulated sermon that made him appear to claim that God wanted the United States to destroy Islam. Uncomfortable with the details of these pastors theological views and without language to explain them away, McCain rejected their earlier endorsements (Banerjee and Luo 2008). The rejection appeared to largely settle the matter with the broad public, and to the media it seemed an equivalency of sorts with Obamas own pastor troubles (Goodstein 2008a; Boorstein 2008). Underneath the radar, however, the timing and style of McCains repudiations reected the candidates clumsiness with religious views and gures. The endorsements offended some people of faith, but the rejection of them may have alienated othersboth those who agreed with the views of Hagee and Parsley and those who may have disagreed with the pastors but saw the episodes as an example of McCains ineptitude with the deeply religious community (Kindy 2008; Luo 2008). Moreover, during the spring months of 2008, the McCain campaign still confronted a general sentiment within the electorate that it might be a time for a change. Most voters typically cast their ballot for candidates of the same political party from one election to the next. However, among the 10 to 15 percent of the electorate who reported that they might swing their ballot from a candidate of one party in 2004 to a candidate of another party in 2008, a much higher percentage typically indicated that they would swing their vote from Bush in 2004 to the Democratic candidate in 2008 than indicated that they intended to swing their vote from Kerry in 2004 to McCain in 2008. This is evident in table 3.4, which examines such reported intentions of swing voting by the religious tradition with which the respondent is afliated. While there are some differences that arise depending on whether Clinton or Obama was the Democratic nominee, the likely swing to Obama or Clinton was overall much greater than the likely swing to McCain. More than twice as many evangelical Protestants, mainline
Religion and the 2008 Presidential Primaries


table 3.4 Swing Voting by Religion Tradition: April 2008

Evan. Prot. (%) McCain versus Clintona Consistent Republican Swing to Clinton Swing to McCain Consistent Democrat Total McCain versus Obamaa Consistent Republican Swing to Obama Swing to McCain Consistent Democrat Total McCain versus Democrat candidates combined Consistent Republican Swing to Clinton/Obama Swing to McCain Consistent Democrat Total 74 8 3 15 100 50 10 5 36 101 10 9 2 80 101 46 8 6 40 100 33 0 5 62 100 31 8 3 58 100 73 11 3 14 100 48 13 6 34 100 15 11 0 74 101 49 9 6 36 100 42 0 0 58 100 28 9 1 62 100 76 6 2 17 100 51 8 4 37 100 5 5 5 86 101b 43 7 7 43 100 22 0 11 67 101 34 6 7 52 100 Main. Prot. (%) Black Prot. (%) Rom. Cath. (%) Jews (%) Relig. Unafl. (%)

Source: Henry Institute National Survey on Religion and Public Life, AprilMay 2008.

Asked of half the sample surveyed; bPercentages over 100 due to rounding.

Protestants, and the religiously unafliated indicated that they intended to swing to the Democratic candidate than indicated that they intended to swing to McCain. And while the ratio was not as great for Roman Catholics, a greater percentage of Roman Catholics also indicated in April 2008 that they intended to switch to the Democratic candidate than switch to McCain in the fall election.

The Final Democratic Contests

On the Democratic side, the religious issue did not go away. Obama appeared to be further weakened in the days after the Pennsylvania primary with another round of controversial remarks by Rev. Wright (Murray and Slevin 2008). Wright had scheduled several media appearances for


The Disappearing God Gap?

the last few days of April, amplifying on his previous comments and starting another round of stories. This time Obama acted swiftly, and on April 29 denounced Rev. Wrights remarks and separated himself from them as far as he could (Zeleny and Nagourney 2008). The latest ap might have been expected to harm Obama again and to diminish his showing in the North Carolina and Indiana primaries on May 6, but that did not seem the case; Obama won the North Carolina primary by fourteen percentage points and lost the Indiana primary by only two (Helman and Williams 2008). Thus, after the May 6 primaries, Obama led by 168 pledged delegates with only 217 pledged delegates left to choose in the few remaining Democratic contests (Healy and Zeleny 2008b). While the numbers were close, Obama clearly had momentum, with his well-applauded response to the Wright and other controversies. He quickly assumed for himself the stance of the inevitable nominee, giving an election night speech that addressed the general election challenge against Republicans and McCain instead of discussing the ofcially undecided Democratic primary. An increasing number of Democratic superdelegates seemed to agree (Bellantoni 2008). Four days later, Obamas superdelegate total exceeded that of Clintons, as some undeclared superdelegates endorsed him, and a few others switched to him from their prior Clinton pledges (Broder 2008c). Clinton still did not concede, and looked to the unsettled Michigan and Florida issues as the last possible avenue to make signicant gains. But on May 31, the DNC rules committee accepted the Michigan state partys proposal to distribute delegates 69 to 59 in Clintons favor. The DNC allocated Floridas delegation and gave a forty-seat advantage to Clinton. But both Michigan and Floridas delegates would get half-votes each, diluting Clintons gains (Balz 2008d). After the DNC decisions, Obama remained signicantly ahead in pledged delegates with a lead of about 140 before the nal three primaries in Puerto Rico, South Dakota, and Montana the rst days of June. Clinton won two of these three, but Obama simultaneously announced sixty new superdelegate endorsements to more than offset her gains (Sabar 2008). On June 3 Obama had nally moved over the magic number of 2,117 delegates for a majority. By early that evening, major news organizations had announced that Obama had clinched the Democratic nomination. Obama simultaneously claimed the status of the presumptive nominee in a speech in St. Paul, Minnesota, the site of that summers Republican National Convention (Balz and Kornblut 2008c). While Clinton did not concede that evening, two days later she did announce she would formally congratulate Obama on his victory (Nagourney and Leibovich 2008).
Religion and the 2008 Presidential Primaries


The paths of McCain and Obama to their party nominations were each difcult in their own ways. In the early stages of both races, neither was a sure thing. McCain was very far behind in late summer 2007, a status Obama never had to endure. Obama impressed early, but struggled later on to close the deal against a very tough Hillary Clinton. Both broke some new religious-related ground in their successful efforts. For McCain, it was his signicant and ongoing discomfort with the language and leaders of traditional Republican-afliated religious organizations. Obamas new ground was his high comfort level with issues of faith and religion. The contests had started with the observation that the top Democrats were all comfortable with faith talk and faith-related issues, whereas McCain was mostly uncomfortable with the same. And at the end of the primary season, one could see evidence in both data and narrative of a new religious dynamic, different from 2000 or 2004, with Democrats making observable headway. In the aftermath of the 2004 presidential election, much was made of the religious divide in the election. This election cleavage was captured, in part, by the relationship between attendance at worship services and vote choicenamely, as voters became more observant religiously, they became less likely to vote Democratic. Then, following the election, a series of polls further revealed that only a small minority of Americans thought the Democratic Party was friendly to religion. In June 2008, however, many were speculating that the landscape was changing (Goldman and Chipman 2008; Luo 2008; Gilgoff 2008b). The Republicans had nominated John McCain who, for all his heroic national service and long-standing if sometimes unpredictable Republicanism, seemed almost adrift on religious matters. This was highly unusual, for since 1980 Republican presidential nominees had brought, or at least learned quickly to develop, openly religious language, behaviors, and ties to religious groups. McCain was not in that tradition, as he was very private with his faith. Though he was at least a sporadic churchgoer, he rarely was photographed or videotaped attending religious services. He rarely initiated conversation about his faith, nor did he use opportunities to talk about his faith or employ religious language or allusions in public. His highest loyalties were to the nation and, by implication, this nations God, but he seemed far more a patriot than a pietist. He seemed uncomfortable with the leaders of the traditional religious groups active in internal Republican Party politics, and he had warred with many of them in the past. In 2008 he was more cordial and tried to woo them to his side both


The Disappearing God Gap?

during and after the contested primary season. Nevertheless, McCains interaction with the traditional religious groups and most of the religious gures in the preprimary and primary seasons was nearly always awkward and, at times, embarrassing. Barack Obama, on the other hand, was openly faithful, particularly in his rhetoric. He was not afraid to discuss the role of religion in his public and private life, and even mentioned a personal relationship with Jesus Christ, the trademark language of evangelical faith. While Hillary Clinton and John Edwards used religious language in a more traditional social gospel sense, Obama added a dimension to the gospels social demands that was both more aspirational and more personal. Obama deftly handled the specic and very serious religious controversies in the primary season, limiting the damage of his remarks about Pennsylvania voters, and even turning the threat posed by Rev. Wrights tirades into a winning opportunity to reach above and beyond older forms of racism and race relations. Part of the reason for Obamas felicity with religion, and McCains difculty, was the different standard of religious gatekeepers in both parties. In the GOP, religious groups are long-standing, well-organized, tradition-bound, and (at least perceived to be) politically powerful. McCain, and any other candidate, had to meet carefully dened theological and political expectations to win the approval of this old, and some would say aging and declining, leadership. Mindful of their rocky past, the media pounced on any divergence between McCains words and the established orthodoxy and found frequent examples of the Arizonans religious aws. On the other hand, faith and politics is largely uncharted territory for national Democratic candidates. The Democrats in 2008, particularly Obama, had an open eld to dene and rene what it meant to be sincerely religious and a member of their party. The media had few if any established authorities with which to compare or contrast those views, and thus even major controversies such as Rev. Wright and bitter Pennsylvania voters created no concrete clash between established practices and a candidates misstep. A largely religiously illiterate media gave Obama a pass on his pronouncements, and pointed out every critique of McCains. As such, the reporting suggested a further large shift of religious voters away from Republican McCain and toward Democrat Obama. At the grass roots and with specic hot-button issues, however, the contrast of Obama and McCains religiosity probably had less traction. Obama knew the religious language, but his interpretation of his faiths application to public policy differed little from every major tenet of liberal
Religion and the 2008 Presidential Primaries


Democratic orthodoxyparticularly on hot-button social issues. Obama was strongly pro-choice and would take no action to restrict the spread of gay marriage initiatives. His support for faith-based initiatives was clouded by his willingness to restrict the types of religious groups eligible for participation based on their support for nondiscrimination in hiringa key demand that translated into protection for sexual orientation, an inviolable provision in Democratic Party politics. On broadly moral issues (not strictly religious), Obama was a dove in foreign policy and a liberal on domestic taxing and spending. McCain, on the other hand, although religiously inarticulate, had essentially the right position on abortion and gay marriage and was moving toward the conservatives even on embryonic stem cell research. He was for lower taxes and less federal spending, and appeared as a strong proAmerican patriot who, although dangerously close to a civil religion for some, mirrored such sentiments among much of the organized and politically active conservative religious community. This closer look at the grass roots suggests the scope and limits of any religious shift from Republican to Democrat in the general election. Shifts across all religious categories would be at the margins, reecting general opinion shifts more than closely faith-related ones. Bigger shifts might occur in subgroups of religious voters who might hold liberal views on social issues or who, if socially conservative, might put those issues at very low priority in making a voter choice. These smaller groups might possibly shift strongly toward the Democrats, but because of their relative small size the overall effect would be marginal. The ve months of the general election campaign, and the results of election night, would provide more denitive answers.


The Disappearing God Gap?


Religion and the Summer Interlude

y the early months of 2008, following the initial series of primary elections, the eld of presidential candidates had been winnowed down. The presumptive nominee of the Republican Party was largely determined by mid-March, but the close contest between Hillary Clinton and Barack Obama made such a determination impossible for the Democrats until mid-June.1 Weeks remained before delegates to the Democratic and Republican National Conventions were to gather in Denver, August 2528, and in St. Paul, September 14, respectively, followed by the traditional Labor Day kickoff to the general election campaign. Hence the summer months are an interlude period between the primary and general election phases of the presidential campaign. The conventional political wisdom is that candidates must move to the ideological extremes during the primaries to secure the nomination from their partys activist base, but that, immediately after securing the nomination, they must move toward the center in order to win in the general election. As a result, during the summer months, both candidates used their time to rene some of their stated positions, to position themselves more competitively, and to prepare for the national conventions. Perhaps most importantly, presumptive nominees must settle on their vice presidential running mates prior to the national conventions.

Religion and the Presidential Nominees

From a religious point of view, both Obama and McCain contrasted sharply with their parties previous nominees. In 2004, the Democrat nominee John Kerry, a Catholic, largely avoided discussion of [his] faith

throughout the campaign (Wilgoren and Keller 2004). Meanwhile, the Republican candidate, George W. Bush, rarely missed an opportunity to speak to a religious audience or to highlight his Christian faith. It is hardly an exaggeration to say that the nominees of the two parties in 2008 had done a complete role reversal: Democrat Barack Obama was comfortable speaking of his own religious faith and insisted it was wrong for Democrats and progressives to avoid religious concerns; John McCain shied away from addressing religious audiences and was hesitant in speaking about his personal religious faith.

The Religious Faith of Obama

Barack Obamas basic biography is well known. His Kenyan father had been raised Muslim but later became an atheist. His mother was a white American who had mainline Protestant roots but might more accurately be termed a nonpracticing Christian. Obama was born in Hawaii; after his mother and father divorced and his mother remarried, he lived for a while in his stepfathers home country of Indonesia, where he attended a Muslim school for two years and a Catholic school for two years. During this time, he was also, in effect, homeschooled by his mother, who saw the schools he attended as inferior (Obama [1995] 2004). Later, as a teenager, he lived with his grandparents in Honolulu and attended an elite, nonreligious prep school. As he later reected: I was not raised in a particularly religious household (Obama 2006). It was during his days as a young college graduate working as a community organizer in the economically depressed, largely African-American south side of Chicago that Obama became a committed Christian believer. He has described that experience in these words: You need to embrace Christ precisely because you have sins to wash awaybecause you are human and need an ally in this difcult journey. It was because of these newfound understandings that I was nally able to walk down the aisle of Trinity United Church of Christ on 95th Street in the Southside of Chicago one day and afrm my Christian faith. . . . [K]neeling beneath that cross on the South Side, I felt that I heard Gods spirit beckoning me. I submitted myself to His will, and dedicated myself to discovering His truth. (2006) Similarly, in a summer 2008 forum on faith and the presidency, Obama began by speaking in personal terms: As a starting point, it means I believe inthat Jesus Christ died for my sins, and that I am redeemed


The Disappearing God Gap?

through him. That is a source of strength and sustenance on a daily basis. Yes, I know that I dont walk alone. But he then quickly switched to more societal terms: But what it also means, I think, is a sense of obligation to embrace not just words, but through deeds, the expectations, I think, that God has for us. And that means thinking about the least of these (Saddleback Presidential Candidates Forum 2008). It is noteworthy that what brought Obama to this point was not some crisis or tragedy in his personal life. Rather, it was his growing recognition that Christianity was speaking to the suffering he saw all around him, giving hope to those without hope and courage to work for change to those in despair. This theme comes up repeatedly in Obamas references to his faith. In 2006 he stated that he believed and still believe[s] in the power of the African-American religious tradition to spur social change (Obama 2006). In a 2008 speech in which he addressed some inammatory remarks in sermons by his pastor at that time, Jeremiah Wright, he described the church he joined as a church that serves the community by doing Gods work here on Earthby housing the homeless, ministering to the needy, providing day care services and scholarships and prison ministries, and reaching out to those suffering from HIV/AIDS (Obama 2008a). In his memoir, Dreams from My Father, Obama referred to the desire to give oneself up to a God that could somehow put a oor on despair (Obama [1995] 2004, 294). And in a speech about race, Obama (2008a) provided a basic summation of the public role of his religious faith: In the end, then, what is called for is nothing more, and nothing less, than what all the worlds great religions demandthat we do unto others as we would have them do unto us. Let us be our brothers keeper, Scripture tells us. Let us be our sisters keeper. Let us nd that common stake we all have in one another, and let our politics reect that spirit as well. In short, Obama was convinced that religion could offer hope to the hopeless and motivation to work for social change to those who otherwise might live only for themselves. From this Obama concluded that political leadersand progressive Democrats in particularhad every right and motivation to bring religious beliefs and values into their public pronouncements. He bemoaned the fact that many Democrats still refused to assume a religious posture and frame timely issues in religious terms. As he stated in a 2006 speech: The discomfort of some progressives with any hint of religion has often prevented us from effectively addressing issues in moral
Religion and the Summer Interlude


terms. . . . [I]f we scrub language of all religious content, we forfeit the imagery and terminology through which millions of Americans understand both their personal morality and social justice. Imagine Lincolns Second Inaugural Address without reference to the judgments of the Lord. Or Kings I have a Dream speech without references to all Gods children. Their summoning of a higher truth helped inspire what had seemed impossible, and [helped] move the nation to embrace a common destiny. In these religious beliefs and values one can see the strong inuence of what has been termed black liberation theologya theology that permeated Trinity United Church of Christ, where Obama rst professed his Christian faith and where he was a member for twenty years. This church membership and Trinitys pastor, the Reverend Jeremiah Wright, as we noted in chapter 3, became a campaign issue when some of Wrights very inammatory sermons surfaced early in 2008. But the theology taught there clearly inuenced Obamas religious views. Black liberation theology originated in the 1960s and is usually associated with James H. Cone, an African-American and professor of systematic theology at Union Theological Seminary in New York. He once dened black liberation theology largely in terms of its seeing God primarily as concerned with the poor and the weak in society (Gross 2008).2 In the same interview he later criticized those churches that see the gospel primarily as going to heaven when I die. And the gospel is not primarily that at all. The gospel is what happens to you now, in this world; and so I wanted the black churches to see that they cant preach the gospel unless they preach about justice and peace for all people (Gross 2008). This theology is still richly reected on the website of Trinity United Church of Christ. Its declared motto is: Unashamedly Black and Unapologetically Christian. It goes on to explain: Our roots in the Black religious experience and tradition are deep, lasting and permanent. . . . God has superintended our pilgrimage through the days of slavery, the days of segregation, and the long night of racism. It is God who gives us the strength and courage to continuously address injustice as a people, and as a congregation. We constantly afrm our trust in God through cultural expression of a Black worship service and ministries which address the Black Community. (Trinity United Church of Christ)


The Disappearing God Gap?

As a young community organizer struggling to empower a community deeply mired in poverty and often ignored by the political power structures, Barack Obama found this message powerfully attractive. Obamas campaign website listed eight key principles related to his religious faith that clearly reveals how Obamas religious faith interacted with his political philosophy. In these principles the emphasis is on how religion can form the basis for social change and a better society, a stance reective of Obamas own biography. These eight principles are: (1) God is constantly present in our lives, and this presence is a source of hope. (2) Progressives should boldly approach matters of faith and values. (3) As Joshua built on the work of Moses, leaders of todaythe Joshua Generationmust build off the foundation of previous generations to move our nation forward. (4) Faith should not be used as a wedge to divide. (5) The separation of church and state is critical and has caused our democracy and religious practices to thrive. (6) We are a nation of many faiths and of those with no faith at all. The religious practices of all must be respected. (7) Faith is a source of action for justice. (8) Government alone cannot solve all of our problemswe have an individual responsibility to be our brothers keeper and our sisters keeper. Note, however, the subtle ways in which these principles call attention to certain key facets of religious experience, reecting what we called elite posturing in the introduction. The second principle explicitly associates religiously inspired change with progressive politics, but the last recognizes the deep-seated emphasis in American political culture on individual responsibility beyond governmentespecially, one assumes, through institutions such as the church, as suggested in the campaigns evocative use of biblical language. The third principle clearly invites younger religionists to take leadership in new ways. And these points make clear Obamas commitment to religious pluralism, his recognition of our being a nation of many faiths, and the value of church-state separation.

The Religious Faith of John McCain

The basic biography of John McCain is also well known. He grew up as a military brat, with a distinguished military heritage: both his grandfather and father were U.S. Navy admirals. He graduated from the U.S.
Religion and the Summer Interlude


Naval Academy, although he was anything but a model midshipman. After graduation he trained to become a naval aviator, where he later confessed he liked to y, but not much more than I liked to have a good time. . . . I drove a Corvette, dated a lot, spent all my free hours at bars and beach parties, and generally misused my good health and youth (McCain 1999, 153). McCain nevertheless successfully completed his naval air training and was assigned, rst, to the aircraft carrier USS Intrepid and later, the USS Enterprise and Forrestal. While on a bombing mission off the Forrestal in 1967, McCains plane was shot down over North Vietnam. A life-changing saga began. When he ejected from his plane, he broke both of his arms and right knee. He was quickly captured and his ve and a half years in a POW prison began. After four days without medical treatment, he was near death. Only when the North Vietnamese discovered that their prisoner was the son of a U.S. Navy admiral did they take him to a hospital. Even then the North Vietnamese withheld medical treatment in an attempt to obtain information and admissions of guilt from this prestigious prisoner. To this day McCain does not have full use of one of his arms and his right leg. In time, McCain recovered from his injuries only to face additional harsh interrogations and torture. At one point he heroically refused release, since military rules called for the rst captured to be the rst released. After his release McCain continued to serve in the Navy until he retired in 1981. He settled in Arizona, where in 1982 he was elected to the House of Representatives and in 1986 to the United States Senate. He has served in the Senate for twenty-two years, where he developed a reputation as a moderate and even as a maverick voice within the Republican Party. Religiously, McCain was raised as an Episcopalian and now attends a Southern Baptist church in Phoenix (he is not formally a member). But it is difcult to dig deeper into the spiritual dimensions of his life. He is reluctant to discuss his faith. His time in a North Vietnamese prison was a profoundly life-changing experience, yet McCain remains somewhat of an enigma religiously. In his 1999 autobiography that covered his years in prison, one nds only eeting references to his religious faith and how it gave him strength and courage to face unbelievably difcult challenges. At one point he writes about having prayed frequently and he tells of copying passages from the Bible one Christmas Eve so he could read the story of the birth of Christ at a service with his fellow prisoners. Yet one is struck by the lack of religious references at a time when one would imagine faith would dramatically come to the fore. This is not to say, however, that religion was absent from the experience: A fellow prisoner of war, George Bud Day, has recounted how McCain led and preached at church


The Disappearing God Gap?

services, saying, He was a very good preacher, much to my surprise. He could remember all of the liturgy from the Episcopal services . . . word for word (Banks 2008). It is instructive that we learn about this aspect of McCains time as a POW not in his own memoir but from the testimony of a fellow prisoner. McCain is clearly a professed Christian, but equally clearly he is not a person comfortable with openly discussing his personal religious beliefs. When asked to describe his faith at a public forum in the summer of 2008, McCain answered: It means Im saved and forgiven. Were talking about the world. Our faith encompasses not just the United States of America but the world. Can I tell you another story real quick? (Saddleback Presidential Candidates Forum 2008). He then launched into the story of a North Vietnamese prison guard who, when McCain was being held as a POW, had one night loosened the ropes that were painfully binding his arms. Later, on Christmas Eve, in an exercise area that same guard came up to McCain and drew with his foot a cross on the ground. Understandably, McCain found this gesture to be a deeply moving, spiritual experience. And while this is a touching story, it does little to explain the nature of McCains faith or how it informs his thinking about politics or public policy, as was the intent of the forum. McCain seldom made religious references in his campaign appearances and included no mentions about his faith on his campaigns website. The closest thing one could nd there about religion was a statement on Human Dignity and the Sanctity of Life, where McCain stated: To sacrice for a cause greater than yourself, and to sacrice your life to the eminence of that cause, is the noblest activity of all. The reason for this reticence to discuss his faith is puzzling, especially given the precedent of recent Republican candidates doing so and the fact that Obama was doing so. Taylor Grifn, a McCain campaign spokesperson, explained it this way: To Sen. McCain, faith is a private matter. He believes that politicians or leaders shouldnt be judged on their religious beliefs but rather they should be judged on their preparedness to do the job (Banks 2008). But this really does not answer the question. Instead, it implies that ones religious beliefs have nothing to do with being prepared to be president, a position many Americans of religious faith would dispute, including candidate Obama given his emphasis on religion as a motivation for reaching out to and caring for others. Perhaps a better answer revolves around the tendency of McCain to see his religious values and beliefs, on the one hand, and values such as courage, love of country, service, and loyalty to ones family and comrades, on the other hand, as drawing from the same well. A statement found on his
Religion and the Summer Interlude


campaign website captured this sense of values well: John McCain is the son and grandson of military ofcers. He served as a Navy pilot, honored to live in the company of heroes as a prisoner of war in Vietnam, and is a beloved husband and father. Senator McCain has enjoyed the quiet blessings found in serving others. As noted earlier, there are few references to God or faith in his autobiography that detailed his experiences in a North Vietnamese prison, and none in the nal pages of the book where he sums up his experiences and what sustained him through terrible times. At one point, in a poignant passage, McCain recalls that his father had recounted to him how, when his father ( John McCains grandfather) was near the end of life, he had said to him: Son, there is no greater thing than to die for the principlesfor the country and the principles that you believe in. McCain then continues: And down through the years, I had remembered a dying mans legacy to his son, and when I needed it most, I had found my freedom abiding in it (McCain 1999, 349). It is signicant that McCain reports that, when in great need under torture and the deprivations of a prison camp, he had found his freedom in remembering his grandfathers words to his father about sacricing for ones county. This, we believe, is a key to understanding John McCain as a religious person. His deepest valuesindeed his religious valuesspeak of service to ones country, of courage, of sacrice, and of honor. This we would label a God and country religious perspective. Service to God and service to ones country almost seem to merge into a single quasi-religious (and quasi-secular) code of honor and love of country.

The Religious Challenges Confronting the Candidates

Both Obama and McCain campaigned for the presidency as Christians. For Obama this largely meant speaking freely of his Christian faith and using it as a basis for helping and motivating those in need and despair; for McCain this meant holding ones faith in private while living a life of service to ones country and courage in the face of lifes challenges. But whatever their personal views

of faith and public life, the political context posed some distinct religious challenges that were largely outside their control. At the end of the primary season, each candidate enjoyed several advantages and disadvantages. As individuals, each candidate confronted different public perceptions related to their religious faith and their religiosity. As candidates of their respective parties, both McCain and Obama had to deal with perceptions related to the friendliness of their political parties toward religion more generally.


The Disappearing God Gap?

table 4.1

Perceptions of Obamas Religiosity (July 2008) Percentage

As far as you know, is this true or false? Obama used a Quran for swearing in to the U.S. Senate. True False Dont know/refused to answer Obama attended an Islamic school in Indonesia during his youth. True False Dont know/refused to answer Obama was raised as a Muslim. True False Dont know/refused to answer Obama is a practicing Muslim today. True False Dont know/refused to answer
Source: Newsweek Poll, July 910, 2008.

12 49 39 39 23 38 26 51 23 12 69 19

From the start of his candidacy for president, Obama had to address allegations that he was a Muslim or that he had received education in a Muslim madrassa while he lived in Indonesia (Bacon 2007). These claims were exacerbated by Internet rumors to the effect that, if he were elected president, he would take his oath of ofce using the Quran instead of the Bible. In part to counter such allegations, Obama actively campaigned publicly as a Christian (as shown in the previous chapter), including using campaign literature with the words COMMITTED CHRISTIAN in large letters, indicating that if elected president, he would be guided by his Christian faith and stating that I believe in the power of prayer. Still, in the months before the convention, the question remained as to just how effective such efforts were in seeking to convince the American public that he was a committed Christian, as any link to the Islamic faith might well serve to depress support for his candidacy. Table 4.1 analyzes how Americans perceived Obamas religious faith in the summer of 2008. These poll results provided both good and bad news for Obama. On the one hand, there was still a sizable proportion of the American people who thought Obama continued to have some ties to Islam, and there was an even larger proportion of people who did not
Religion and the Summer Interlude


know whether Obama had any Islamic tiesthereby creating an opening for damaging misinformation to be circulated by his political opponents. For example, as the table illustrates, more than one-tenth of the American people in July 2008 thought Obama had used the Quran when he was sworn in to the U.S. Senate (with nearly two-fths responding that they did not know for sure whether such a statement was true or not). More than one-quarter of the electorate continued to think that Obama had been raised a Muslim (with nearly one-quarter responding that they did not really know about the accuracy of the statement). While only a little more than one-tenth of the American people thought that Obama continued to be a practicing Muslim, nearly one-fth reported being uncertain about whether or not this was the case. As a result, nearly one-third of the American people continued to be uncertain of, or misinformed about, Obamas religious faith only four months prior to Election Day. On the other hand, the poll also contained some positive news for Obama. By the summer of 2008, American voters were more likely to report that they thought religious faith played a more important role in Barack Obamas than in John McCains life. This is important because, as we showed in chapter 1, voter perceptions of high levels of candidate religiosity are positively related to candidate preferences. These patterns are evident in table 4.2. Whereas nearly two-thirds of the American people responded that they thought religion played an important role in Obamas life, less than half responded similarly when asked about John McCain. Moreover, these perceptions were associated with candidate preferences, as a majority of those who thought religion played an important role in the life of a particular candidate gave their support to that candidate, while a majority of those who reported that they thought religion did not play an important role in a candidates life supported his opponent. Moreover, not only did a greater proportion of Americans perceive religion to hold a more important role in Obamas life than McCains, but Obama seemed to receive a greater bounce from such perceptions than did McCain. This is also evident in table 4.2. Note, for example, that while six in ten of those who reported that they thought religious faith was important in Obamas life indicated that they were inclined to vote for Obama, only a little over half of those who thought the same of McCain indicated an intention to vote for McCain. Nevertheless, while many voters viewed Obamas personal religiosity favorably, he was still a party candidate, and the Democratic Party continued to be viewed as relatively unfriendly toward religion. When asked in July 2008, the majority of Americans (58 percent) responded that


The Disappearing God Gap?

table 4.2

Candidate Preference by Perceptions of the Religiosity of the Candidates Candidate Preference Percentage of Electorate McCain (%) Undecided (%) Obama (%) Total (%)

Do you think religious faith plays an important role in Barack Obamas life or not? Yes No Dont know/refused Total Do you think religious faith plays an important role in John McCains life or not? Yes No Dont know/refused Total 49 28 23 100 53 39 33 5 10 12 42 51 55 100 100 100 63 23 14 100 37 65 48 4 11 21 60 25 31 101a 101 100

Source: Newsweek Poll, July 910, 2008.


Percentages over 100 due to rounding.

the Republican Party was friendly toward religion, while only a little more than one-third (37 percent) responded similarly about the Democratic Party.3 But the efforts of the Democratic Party to appear to be more friendly toward religion seemingly were bearing fruit, as in 2006 only 29 percent of the American people reported that they perceived the Democratic Party as being friendly toward religion (in the Pew Center Religion and Politics Survey in 2006). To the extent that the God gap is dened by partisanship, the gap was consequential yet shrinking. Though such perceptions of the two parties might work to Obamas disadvantage and McCains advantage, what is less clear is how the two parties are viewed in comparison to each other on this matter and how candidate choices are tied to these different patterns of perceptions. To bring greater clarity to these comparisons and linkages, table 4.3 presents all the different combinations possible, as well as how respondents within those particular categories reported their candidate preferences.
Religion and the Summer Interlude


table 4.3

Candidate Preference by Perceptions of the Friendliness of the Political Parties toward Religion Candidate Preference

Perceived Friendliness of Party toward Religion Dem friendly, Rep unfriendly Rep friendly, Dem unfriendly Dem friendly, Rep neutral Rep friendly, Dem neutral Dem unfriendly, Rep neutral Rep unfriendly, Dem neutral Both neutral Both friendly Both unfriendly Total

Percentage of Electorate 3 12 7 19 7 3 20 27 2 100 McCain (%) 5 89 34 43 95 0 45 30 63 Undecided (%) 9 6 5 3 4 4 10 3 10 Obama (%) 86 5 61 55 1 96 45 67 28

Source: Newsweek Poll, July 910, 2008. Dem = Democratic Party; Rep = Republican Party.

Most Americans in July 2008 viewed the two political parties as either being both friendly toward religion or both neutral toward religion; only a small percentage viewed both parties as unfriendly toward religion. On the other hand, nearly one-fth viewed the Republican Party as friendly and the Democratic Party as neutral toward religion, while more than one-tenth viewed the GOP as friendly and the Democratic Party as unfriendly toward religion. The remainder expressed different combinations of perceptions (e.g., one party unfriendly, the other party neutral). Why these perceptions matter is that these different combinations of perceptions are clearly tied to expressions of candidate preferences. Among the part of the electorate who perceived the Democratic Party as unfriendly toward religion and at the same time viewed the Republican Party as neutral toward religion, almost all reported that they were inclined to vote for McCain; conversely, among the group who had the opposite point of view, nearly all supported Obama. A similar pattern was evident among those who perceived one party as friendly toward religion and the opposing party as unfriendly. Thus, among the group who believed the GOP to be friendly toward religion while at the same time perceiving the Democratic Party as unfriendly toward religion, nine in ten indicated their intention to vote for McCain; on the other hand, of the respondents who saw the opposite relationship between religion and the two parties,


The Disappearing God Gap?

nearly the same proportion reported that they were inclined to vote for Obama. Thus, candidates linked to parties perceived to be unfriendly toward religion are highly unlikely to receive the votes of those who hold such perceptions. However, a candidate does not have to have the party be perceived as friendly toward religion to garner votesonly that the party be perceived, at a minimum, as neutral toward religion. Overall, Obama beneted when the Democratic Party was viewed as being neutral, if not friendly, toward religion. For example, among the part of the population who viewed the GOP as friendly toward religion and the Democratic Party as neutral toward religion, Obama won a majority of support. Nevertheless, it is better for a candidate to have ones party be seen as being friendly toward religion, as among the more than one-quarter of Americans who viewed both parties as friendly toward religion, two-thirds reported that they intended to vote for Obama.

Religion and the Summer Months of the Campaign

John McCain had secured the GOP nomination by early March, while Obama did not wrap up the Democratic nomination until early Juneat the very end of the presidential primary process. The summer interlude in the campaign season gives the nominees of the two parties an opportunity to catch their breath. For a few weeks at least, the strenuousness of the primary campaign, with its continual traveling from state to state, even from town to town and from city to city within a state, along with its constant campaign appearances, talks, interviews, and even debates, is now behind the candidate. Moreover, during the summer months, the publics attention to the campaign wanes, as children are out of school and vacations begin. As a result, as we noted, the summer months of the campaign are typically used by the candidates to rene their positions and position themselves as they prepare for their conventions, particularly in terms of the choice of running mate for the fall campaign.

Religious Consultations
From the start of his campaign, the Obama organization had a religious affairs team, and during the primary campaign, it conducted weekly conference calls with prominent African-American religious leaders, evangelical and values voters activists, Catholic religious leaders, and Jewish religious and civic leaders (Sweet 2008). The campaign organization also
Religion and the Summer Interlude


had a National Catholic Advisory Board, along with a separate staff that worked on Jewish and Middle Eastern affairs. In addition to these campaign staffers and activists, Barack Obama convened in early June a group of inuential Christian leaders to help shape the kind of conversations he wished to have with voters during the general election campaign. The group was not gathered with an eye toward possible endorsement; indeed, not all the participants were Obama supporters (Sullivan 2008b). Instead, it was assembled to hear their critique of the Obama campaign in his quest for independent, middle-class, and GOP votes as well as to let the group know that his door was openin the campaign and in the White House, if he is elected (Sweet 2008). The invited guests numbered about thirty people. While it included mainline Protestant, black Protestant, and Roman Catholic leaders from across the country, the majority of attendees were white Evangelical leaders (Sullivan 2008b). Included in the group were Cameron Strang, the founder and editor of Relevant, the Christian magazine and related website that caters to young evangelicals; Max Lucado, a pastor and best-selling author of numerous Christian books; Paul Corts, the president of the Council for Christian Colleges and Universities; Catholic scholar Douglas Kmiec, a Pepperdine University law professor who served in the Reagan and rst Bush administrations; Bishop T. D. Jakes, the black pastor of a Dallas megachurch; Rev. Stephen Thurston, president of the National Baptist Convention of America; Rev. T. DeWitt Smith, president of the Progressive National Baptist Convention; and African Methodist Episcopal church leader Bishop Phillip Robert Cousin Sr. (Sweet 2008). Whether or not Obama won new political converts, the event was a clear opportunity for the campaign to signal its desire to reach beyond the conventional Democratic blocs. Following the meeting, Joshua DuBois, Obamas chief campaign staffer for religious affairs, issued a statement that noted explicitly the participation of several prominent evangelicals, while not singling out any other religious tradition. He also suggested that attendees not only discussed policy issues but also came together in conversation and prayer (Sweet 2008). The point of the meeting, he suggested, was outreach, not necessarily political mobilization. As we discuss in the introduction, this effort is a kind of posturing. By posturing, we do not mean to suggest that the efforts of the Obama campaign were insincere and merely strategic, but that the campaign was recognizing and identifying with a particular group without necessarily endorsing that group or even expecting an endorsement in return.


The Disappearing God Gap?

The Saddleback Forum

Another opportunity for posturing was the appearance of both presumptive nominees at a forum held at the Saddleback Church, in Lake Forest, California, on August 16. Rev. Rick Warren, pastor of the 22,000member evangelical megachurch and author of the best-selling book The Purpose-Driven Life, accomplished what had never been done previously in any American presidential electionnamely, to get the two candidates on the same stage prior to their national party conventions. Moreover, it was the rst time that an event featuring the two major presidential contenders had been moderated by a pastor or held in a church. Not only was it the rst joint appearance of the two presumptive nominees, it constituted one of only four joint appearances of the campaignwith the others being the three debates that were planned for later in the fall. Both candidates had been friends with Warren prior to their run for president, and he invited them to participate in Saddlebacks Civil Forum on the Presidency (Warren 2008; Rutenberg 2008).4 By candidate request, Warren, who is both theologically and socially conservative, asked all the questions rather than a panel or audience members doing so (Warren 2008).5 The event was heavily publicized in advance, and, as a result, the national media had sufcient time to speculate on what challenges each candidate faced. The day before the forum, Good Morning America noted that Obamas challenges entailed false rumors hes a Muslim, his controversial former pastor and his liberal views on abortion and other issues, while noting that the challenge for McCain was that he has not always seemed comfortable talking about his faith. The two-hour event was conducted in a nondebate format, with each candidate taking the stage separately for about an hour to respond to Warrens questions. In order to avoid bias, identical questions were asked of both candidates, though the follow-up questions differed, based in part on the answers provided by the candidates. Each interview was divided into four segments of equal time, each with a particular theme: stewardship, leadership, worldview, and Americas role in the world. Warren said the goal of the civil forum was to restore civility in our civil discourse (Chan 2008) and that he wanted to provide a different context and climate for the candidates to let the faith community and the entire electorate know who they are and how they would leadsharing not just what is in their heads, but also in their hearts (Warren 2008). In response to Warrens question about which of the sitting Supreme Court justices they would not have appointed, Obama noted Justices Thomas, Scalia, and Roberts while McCain cited Justices Ginsberg,
Religion and the Summer Interlude


Breyer, Souter, and Stevens. When asked to name an instance in which their thinking had changed over the past ten years, Obama cited the 1996 welfare reform bill passed by the GOP-led Congress and signed by President Clinton, which he opposed at the time believing it would have disastrous results, but which he now believes has been a largely successful reform of the social welfare system. When asked to dene marriage, both candidates responded that they believed marriage to be a union between a man and a woman. Obama elaborated: For me as a Christian, it is also a sacred union. Gods in the mix. However, Obama indicated his opposition to a constitutional amendment dening marriage as a union between a man and a woman, indicating that such a question has traditionally been left up to the states to dene. McCain indicated that the courts were wrong in approving gay marriages, but also noted somewhat vaguely (perhaps referring to some type of civil unions) that such a perspective doesnt mean people cant enter into legal agreements. Both candidates responded somewhat ippantly when asked certain questions. When asked, At what point does a baby get human rights in your view? Senator Obama noted that answering such a question was above my pay grade. By contrast, McCains reply to the question was direct and straightforward: At the moment of conception. At another point, each candidate was asked to dene rich. Obama rst responded Well, if you got book sales of $25 million . . . , likely referring to Mr. Warrens book sales, though Obama, too, has made millions from his books. But then he noted that under his tax plan, families making more than $250,000 a year would have to pay what he labeled a modest increase in taxes. When McCain was asked to dene rich, he rst noted that being rich and being happy were not synonymous, and then attempted to catalog the criteria by which to dene rich (specically he mentioned a home, a good job and education and the ability to hand to our children a more prosperous and safer world than the one that we inherited). McCain then stated that he wanted everyone to be rich, indicating that he did not believe in either class warfare or the redistribution of wealth. But nally at the end, he said, So I think if youre just talking about income, how about $5 million? This nal response, coupled with a news story several days earlier that McCain owned seven homes, likely left the impression that McCain was out of touch with the economic plight of most Americans. While no published polling results followed the Saddleback Civic Forum, there were some important developments and outcomes linked to the event. Not surprisingly, some groups, such as Americans United


The Disappearing God Gap?

for Separation of Church and State, protested the event as violating the principle of separation of church and state, while some third party candidates protested their exclusion from the forum. Neither criticism appeared to carry much weight with the public. Still, the Saddleback Civil Forum on the Presidency drew intense, though brief, media coverage. According to a study conducted by the Pew Research Centers Project for Excellence in Journalism and the Pew Forum on Religion and Public Life that focused on the coverage of religion in the campaign, the event attracted 10 percent of all campaign coverage of that week.6 By the end of the month, the forum no longer constituted a major press topic. Nevertheless, coverage of the event accounted for 11 percent of all religion-focused campaign coverage in the general election. And though the candidates were asked a variety of questions, it was their response to the question about the origin of life that garnered most of the media analysis (Pew Forum 2008d). Despite the fact that the Saddleback Forum was not a debate, much of the media coverage following the forum delved into various assessments of whether Obama or McCain had won the event. The general assessment among pundits was that McCain had gained more than Obama (Pew Forum 2008e). Not only did he position himself as somewhat less stiff about religion than was popularly assumed, but he also framed some key issuesabortion and marriage chief among themin ways that solidied support among key constituents. Following the event, for example, the National Right to Life organization issued a press release that led with the statement McCain scores a perfect 10. Yet the question of who won is more complex than this analysis suggests. First, McCains framing on other issues left him open to being the champion of the rich, while Obama solidied his position among those who favored greater income redistribution. Moreover, it was unlikely that Obama lost any support among steadfast abortion opponents, as Obama had always dened himself as pro-choice. Finally, the additional national exposure on the stage of the Saddleback Church likely helped Obama to solidify further the perception of him as being a religious man as well as a unier in bringing together diverse coalitional elements through his willingness to participate in such a forum conducted by an evangelical pastor. Taken together with Obamas strong emphasis on his personal Christian faith in the earlier stages of the campaign and his campaign teams outreach programs, his willingness to participate in faith discussion forums worked to soften perceptions of the Democratic Party and its candidates as being the shock troops of secularism in American public life.
Religion and the Summer Interlude


Religion and the Running Mates

During the summer, with the primaries over and no major events until the conventions, increasing attention is paid to the possible choice of each candidates running mate. Because each presidential nominee is able to select his own running mate with very few constraints, the selection of the vice president offers an opportunity to understand the overall concerns of a campaign. Typically, a nominee chooses a running mate largely for one of several strategic reasons: to balance the ticket by compensating for something the presidential nominee lacks, to appeal to particular constituencies, or to enhance the probability of winning the Electoral College votes of a key state (typically the state in which the selected running mate resides). However, Obama and McCain each chose their running mate from states with very few electoral votes, with Obama selecting Joe Biden of Delaware and John McCain selecting Sarah Palin of Alaska (each state had only three Electoral College votes). Of course, balancing the ticket and appealing to particular constituencies need not be mutually exclusive endeavors. And there are a variety of ways to balance a ticket; a nominee may seek to do so ideologically, geographically, or socially in terms of gender, race, or even religion. Certainly, given his relative youth and inexperience serving in national government, Obama chose Joe Biden, in part, because of his length of public service in the Senate and his service on the Senates Foreign Relations Committee. In particular, Bidens experience in the area of foreign relations was viewed as a means to offset McCains perceived advantage over Obama in terms of foreign affairs and domestic security. In turn, McCain chose Sarah Palin, in part, because she was a woman and because of what appeared to be an opening among some Democratic women who were upset with Obama for preventing Hillary Clinton from becoming the rst female presidential nominee of a major political party. But religious considerations also played a major role in both Obamas and McCains selection of their running mates. The religious identities of the two vice presidential picks are important both as an indicator of the candidates wider electoral strategies and as an indicator of their broader public theology.7 Moreover, though in recent presidential elections it has been primarily the presidential candidates who have given voice to religious matters, in 2008, due perhaps to the predominance of other issues, on the one hand, and the personalities of the presidential candidates, on the other, the vice presidential candidates had a stronger religious voice in the general election campaign. Consequently, both the selection and the


The Disappearing God Gap?

campaign role of the two vice presidential candidates in 2008 illustrate the continuing signicance of religion in national electoral politics. Palin. McCain rst met Palin at a February 2008 meeting of governors. Some conservative thinkers and leaders were familiar with her, and through the spring of 2008 she frequently appeared on the conservative movements short lists for the vice presidential nominee.8 It is not entirely clear which other possible nominees were on McCains short list, but it is clear that a number of the candidates who were rejected, such as Tom Ridge of Pennsylvania or Joe Lieberman of Connecticut, were problematic precisely because they were on the record as supporting legalized abortion as well as having made or adopted other public stances that would not please the religious conservative base of the party. Clearly, McCain wanted to show a different side of the Republican Party, one outside of Washington, while at the same time identifying with the religious values and commitments of its base (Grunwald and Newton-Small 2008). Baptized a Catholic at her mothers parish in Idaho, Palin was a few months old when the family moved to Wasilla, Alaska, a town of some seven thousand people about forty miles north of Anchorage.9 A few years later, her mother began to attend the Wasilla Assemblies of God Church, taking her daughter with her. When Sarah was twelve years old, she was baptized by immersion in that same church (Brachear 2008). However, since 2002, she had worshipped regularly at Wasilla Bible Church, a nondenominational church that she joined, she said, because it was less extreme ( Johnson and Severson 2008), and because she was attracted by the childrens ministries (Miller and Coyne 2008).10 Nevertheless, she remained close to the Wasilla Assemblies of God Church, its former pastor, and its current leadership. And when she is in the state capital, she sometimes worships at the Juneau Christian Center, another Assemblies of God congregation. As someone who worshipped for much of her life at an Assemblies of God church (the same denomination in which Joshua DuBois, Obamas religious affairs chief, had served as a pastor), Palin could be described as the rst vice presidential nominee to come from the Pentecostal tradition (Sullivan 2008b).11 Still, it is unclear that the family ever had a particular or specic attraction to Pentecostalism per seor the Assemblies of God denomination or any denomination in particular. In this, perhaps, they seem to represent the libertarian or strongly individualist culture of the West and Alaska more particularly (Davey 2008). Religiously, she describes herself as a Bible-believing Christian rather than an evangelical or Pentecostal, although her religious commitments surely appealed to evangelicals or Pentecostals. And when asked, she describes her religious afliation as nondenominational, reecting a growing trend within
Religion and the Summer Interlude


religion in the United States of joining churches that lack traditional denominational ties (Allen 2008). Palins job was not to bring seasoned experience to governing after the election, but rather to help her partys nominee win the election by appealing to the base of the Republican coalition. In that sense, McCain got what he wanted. In August, John McCains base was much more excited by Palin than Obamas base was excited by Biden (Saad 2008; Jones 2008). McCains surprise choice gave him the rst edge in news coverage his campaign had ever enjoyed relative to Obamas campaign (Surprise Vice President 2008). In her address at the Republican National Convention, Palin needed to introduce herself to the assembled delegates and win their support even while introducing herself for the rst time to a national audience. As a result, she discussed her family background and her work as governor of Alaska. She described herself as a small-town person, someone who gets things done with a mavericks independence, like her running mate. She contrasted the character of the two presidential candidates, focusing on personalities and personal stories more than policies (Governor Palin 2008). By the end of September, however, the boost that Palin could give had reached its limit. The very fact that she was so strongly identied with the conservative Protestant base of the Republican Party meant that her benets never extended beyond the base. Palins public theologyfounded on Bible-believing, as she put itmeant that she could bridge gaps across various conservative Christian denominations in a way that a Mormon, for example, could not. But this sort of public theology could not do what mainline Protestantism had long done in the United States, or what much of the intellectual tradition or communitarian sensibility of Catholicism seeks to do, which is to nd language and images that can speak broadly within the public square. Controversy often follows the candidates vice presidential pick. Inevitably some are disappointed, and the opposition has good reason to look for ways of picking apart the oppositions choice in order to raise questions about their opponents judgment. In the case of the 2008 presidential election, controversy was not far behind Palins selection, as we will see in the next chapter. Biden. The process that resulted in Bidens selection as Obamas running mate took two months. Not only was the candidate interviewed extensively, so were his brother, sister, wife, and children (Zeleny 2008c). Because the process took a much longer time than Palins, it involved a sustained level of secrecy. Other nalists included Senator Evan Bayh of Indiana, Governor Tim Kaine of Virginia, and Governor Kathleen


The Disappearing God Gap?

Sebelius of Kansas. Of these the rst is Episcopalian and the other two are Catholic, itself interesting in terms of what it suggests about the strategy Obama and his team were contemplating (Zeleny 2008a and 2008b). Many friends of the Obama campaign had suggested Biden, pointing out that he had a compelling story tied to the Catholic culture and economic struggles with which he grew up and with which he continues to identify. They were also aware that Biden is one of the least afuent members of the Senate, which meant both an easier review process and some immunity from the kinds of attacks that wealthier senators have endured when running for the presidency (Clemons 2008). Biden attends mass weekly, and he is noteworthy for reecting an important trend among Catholic politiciansnamely, being a supporter of legal abortion.12 On the Sunday following the announcement of his candidacy, Biden was at mass in his parish in Delaware. A reporter in attendance noted that Bidens name was mentioned during the homily as having been selected as the vice presidential candidate, at which point Biden made the sign of the cross. Later in the mass, one of the prayers of the faithful was that Biden might nd increased wisdom and strength as he becomes more involved in the presidential campaign as candidate for vice president. Afterward, his pastor noted that Biden has some positions that dont go along with the Catholic Church. He added that while he understands that Biden is personally opposed to abortion, he doesnt want to impose his view. The reporter continued, noting that Bidens pastor added that some of the other parish priests and the bishop of the diocese have spoken with him about his pro-choice votes, but that they have never refused him Communion, as some other dioceses have done with politicians who support abortion rights (Memoli 2008). Bidens religious rhetoric and public theology is formed, like Palins, by his regional culture and origins. In contrast to Palins individualism, his religious language stakes out a more communitarian space. Where she is more focused on individual beliefs and personal morality, he is more focused on a communal dimension, emphasizing relationship and compassion, including a sense of limits. This perspective informs his statements in that he believes, as a Catholic, that human personhood begins at birth, but that he ought not impose that belief on others (Phillips 2008). His experience of faith is something that leads to a sense of peace: I get comfort from carrying my rosary, going to mass every Sunday. Its my time alone (Chaddock and Chaddock 2007). Biden credits much of his experience of life from having grown up in closely knit Catholic neighborhoods in Scranton (Biden 2007). He emphasizes culture and community: My idea of
Religion and the Summer Interlude


self, of family, of community, of the wider world comes straight from my religion. Its not so much the Bible, the beatitudes, the Ten Commandments, the sacraments, or the prayers I learned. Its the culture. Though the Obama campaign wanted someone who would be seen as a working class Catholic (Lizza 2008) it clearly also was looking for someone who would bring decades of experience of politics at a national level and foreign policy experience. In the fall campaign, Biden, who had 60 percent favorability ratings (compared to Palins 44 percent), was sent to stump for Obama in states where Hillary Clinton had done better than Obama and where Bidens image was strongnamely, in smaller cities in the Northeast as well as in Florida (Broder 2008e). In other words, his identity as part of an older generation of European ethnics from the northeast complemented Obamas younger, hybrid generational persona. Bidens Catholic identity as a mass-attending, family-oriented senator with a strong communal sense was part of this. His background and image as a Catholic was clearly seen as a way to increase the appeal of the Democratic ticket among white Catholics (Slater 2008a). While the overall appeal was not toward Catholicism per se but rather toward economic and other issues on a state-by-state basis, Bidens persona helped to frame and communicate these issues, offering at the same time an implicitand sometimes explicitpublic theology. On the stump, Biden often told the story of his father, who struggled with business failures yet lived sustained by a strong sense of community and community values (Broder 2008d). This in itself recalled the generation that came of age during the Great Depression and Second World War although Biden himself, born in 1945, is closer to the experience of the Baby Boomers. In any case, these images and values signaled experience and wisdom, and a strong sense of communityprecisely what the campaign wanted to communicate. In this, his experience as one of the longest serving members of the U.S. Senate, including involvement in U.S. foreign policy, seemed to combine with his larger personal narrative, suggesting a public theology that was engaged, communal in its vision, and sustained by faithful practice.

The Changing Issue Priorities of the American Electorate

Despite the interlude in the campaign between the primary and the general election seasons, it would be a mistake to believe that American political life was static during the summer months. Nowhere is this more evident than with regard to voter attitudes on issues. In particular, the


The Disappearing God Gap?

table 4.4

Rating of Current Economic Conditions by Religious Tradition (December 2007 and June 2008) Evan. Main. Prot. (%) Black Prot. (%) Roman Catholic (%) Jews (%) Relig. Unafl. (%) Total (%)

Rating of Current Economic Conditions Gallup, Dec. 2007 Excellent/good Fair Poor Time, June 2008 Excellent/good Fair Poor

Prot. (%)

29.9 48.0 22.2 25.0 39.4 35.6

32.0 42.1 25.9 22.5 34.7 42.8

15.3 33.3 51.4 10.9 26.1 63.0

29.1 46.1 24.8 21.4 36.7 42.2

25.0 60.0 15.0 15.8 36.8 47.4

27.7 41.1 31.1 13.4 39.0 47.6

28.9 44.5 26.9 21.0 36.5 42.5

perceptions of American voters were shifting as to what were the most important issues confronting the country and what the next president would primarily need to address. These changing perceptions tended to work to the advantage of the Obama campaign as the candidate of the nonincumbent party. First, over the months of the primary process, Americans were increasingly viewing the economy in more negative terms. As can be seen in the far right column of table 4.4, more Americans reported in December 2007 that they viewed current economic conditions as excellent/good than viewed it as poor. Only six months later, in June 2008, more than twice as many Americans viewed current economic conditions as poor than excellent or good. Thus, even before the economic meltdown of September 2008, a plurality of Americans had already assessed economic conditions in America as being very poor. These perceptions, however, varied somewhat by a respondents religious tradition. In December 2007, mainline Protestants were the most likely to report that current economic conditions were excellent or good, while Jews were the least likely to view economic conditions as poor. Black Protestants, on the other hand, were the most likely to indicate that they perceived current economic conditions as poor, with a majority doing so. By June 2008, however, most members of each religious tradition, with the exception of evangelicals, were reporting that current economic conditions were poor. Though a plurality of evangelical Protestants in June 2008 assessed economic conditions as fair, nearly an equivalent percentage viewed such conditions as poor, while only a quarter of evangelicals assessed the situation as excellent or good. Black Protestants continued
Religion and the Summer Interlude


table 4.5

Perspective on Current Economic Conditions by Religious Tradition ( June 2008) Evan. Main. Prot. (%) 5.2 29.7 65.1 Black Prot. (%) 0.0 29.6 80.4 Roman Catholic (%) 2.2 30.0 66.1 Jews (%) 5.6 22.2 72.2 Relig. Unafl. Total (%) 4.9 20.7 74.4 (%) 4.4 27.9 67.7

Perspective on Current Economic Conditions Getting better Staying the same Getting worse

Prot. (%) 6.3 28.0 65.7

Source: Time, June 1929, 2008.

to lead the way in assessing the economy as doing poorly. Thus, while there was some variation in perceptions of economic conditions by afliation with particular religious traditions, most Americans regardless of religious afliation perceived the economy as having gotten worse, rather than better, over the rst half of the election year. Moreover, as can be seen from table 4.5, the overwhelming majority of Americans did not see economic prospects getting better in the near future. More than two-thirds of all Americans assessed the U.S. economy as getting worse rather than better. Again, while religious tradition afliations colored such economic perceptions to a certain extent, it hardly created massive differences in such perceptions: the overwhelming majority of all Americans, regardless of religious tradition, perceived the American economy as getting worse by the advent of summer. Over the course of the summer months, the economy became by far the central issue of the campaign, while concerns about the war in Iraq waned in its wake. This can be seen in table 4.6. In early June, a plurality of Americans cited the economy as the most important issue, with the war in Iraq ranking second in importance. Other Americans cited health care, terrorism, or illegal immigration as the most important issue, but only about 10 percent of the American people did so in each case. By the end of July, an additional 6 percent of Americans cited the economy as the most important issue, with a corresponding six-point decline in the percentage of Americans citing the war in Iraq as the most important issue. By early September, the majority of Americans cited the economy as the major issue, while concern about the war in Iraq as the most important issue had dropped to a level comparable to concerns raised about health care and terrorism. Thus, by the start of the fall campaign, the economy was the major issue on the minds of most voterslong before the nancial crisis of late September, which only strengthened concerns about economic matters.


The Disappearing God Gap?

table 4.6

Most Important Issue over Time June 45 2008 (%) July 2729 2008 (%) 48 18 13 9 9 2 Sept. 57 2008 (%) 56 13 12 11 6 1 1

Economy War in Iraq Health care Terrorism Illegal immigration Other (volunteered) Unsure

42 24 12 11 8 1 1

Source: CNN/Opinion Research Corporation Poll.

All of this increased concern about the economy worked to the advantage of Obama. Not only are the Democrats typically perceived as the party that better promotes the interests of the economically disadvantaged, but McCains strengths rested more in the areas of national defense and securityissues that now ranked much lower in the minds of most voters as the fall election campaign was about to begin. By his own admission, McCain had noted that economic issues were not his strong suit. For example, in a Wall Street Journal interview, McCain acknowledged: Im going to be honest: I know a lot less about economics than I do about military and foreign policy issues. . . . I still need to be educated (Moore 2007). And later in a Republican candidates debate in January 2008, McCain reiterated the point in a level of honesty unusual in a politician: The issue of economics is not something Ive understood as well as I should.13

The summer interlude yields some apparently contradictory lessons about the importance of religion in the 2008 campaign. On the one hand, both major party hopefuls continued to try to posture themselves as candidates whom believers could trust, and both candidates in choosing vice presidential running mates considered religion as important in their selection. Obama consulted with unconventional partners, including evangelicals, and he brought on a vice presidential candidate with strong Catholic roots. McCain participated effectively in a national religious forum and made
Religion and the Summer Interlude


the surprise announcement of a little-known running mate with Pentecostal roots who might have strong appeal among most evangelicals. The polls remained tight, so both sides knew they had good reason to nd voting blocs wherever they could get them, including both familiar and untried religious territory. On the other hand, there were signs in the summer that the role of religion would be different than in earlier elections. Perhaps the best indicator was the diminished importance of issues that have motivated many religious voters in the past three decades. The culture wars over marriage, abortion, and similar matters did not rage with the same intensity as in the previous two elections. Health care, education, the war in Iraq, and, most importantly, the economy were emerging as the most salient issues for a wide swath of the electorate, pushing aside other concerns. Indeed, the voting publics concern with the economy was only reinforced by the dour state of nancial markets as the general election campaign heated up after Labor Day. As we discuss in greater detail later, by the time of the election nearly half of all registered voters reported that the economy was the most important factor in the determining their vote, compared to the six percent who cited issues such as abortion or same sex marriage as their chief concerns (Princeton Survey Research Associates 2008, 9). The gap was even wider in other surveys (e.g., FOX News Poll 2008), including the 2008 exit polls, which suggested the economy was the key issue for over 60 percent of the population. These patterns were signicantly different from 2004, when more voters cited moral issues than the economy as the linchpin to their vote.14 Given the importance of the economy in the election, it may seem that religion had little inuence on the election from the end of the summer to the rst week of November. After all, the differences in opinion about the economy among religious believers were quite weak; citizens from all faiths appeared to assess the economy similarly. Moreover, as we noted at the outset of the chapter, the end of the primary campaign marks a shift in focus from party loyalists to more moderateand numerousvoters. As a result, neither major party candidate could afford to focus as intensely on the niche issues that motivated important, yet comparatively small, portions of the party base. To do so would risk appearing out of the mainstream. Candidates were much more likely to begin posturing and framing issues to look attentive to the widest range of voters. And by all accounts, those voters were seriously worried about the economy. Yet the effects of the economy were complex, and in some ways can be overstated. Even when there is shared opinion about the state of the economy, the political response to those opinions may be wide-ranging. As


The Disappearing God Gap?

we shall see in the following chapters, even when religionists said much the same thing about nancial woes at the time, their varied responses in voting patterns suggests some persistence in the God gap. Whats more, to say that the economy overwhelms religious issues assumes that there is a hard distinction between the two. In the 2008 election, many religious and political elites complicated that distinction by linking religion and economic questions. For many voters, the message that there is religious signicance to economic matters carried weight throughout the fall campaign.

Religion and the Summer Interlude



Religion and the Fall Campaign

he nature of a presidential campaign, as well as public attention to it, shifts as the summer draws to a close. For most Americans, summer signies a lull in the presidential campaign, as attention to the race diminishes. The drama of the presidential nomination process has passed, and the time and attention of most voters are increasingly given to summer activities. Though some Americans continue to be relatively attentive to news reports related to the presidential campaign, most voters are largely withdrawn psychologically from it. Nevertheless, though they remain largely removed from the public eye, the campaign organizations are busy during the summer months preparing for the national convention and the fall campaign. Both the Obama and the McCain organizations worked with their respective national parties to insure smoothly functioning, and harmonious, national conventions. The campaigns negotiated mutually convenient times, places, and formats for presidential debates, rened their themes and issue stands, planned their campaign strategies, and developed schedules for the fall campaign. The announcement of the vice presidential running mates just prior to the convention signals the transition to the fall campaign.1 The news media, which continue to cover the campaign during the summer months, begin to increase their coverage at that point. Journalists begin initially to discuss the personal life and political experience of the newly named running mates in order to better acquaint themselves and the public with the vice presidential nominees. The party ticket is now complete, and only awaits confirmation at the national convention.

The National Conventions

Every four years, the Democratic and Republican parties hold presidential nominating conventions. These conventions historically have served two major functions: (1) to select the partys nominee for president, and (2) to draft a party platform that outlines the partys values and policy goals, lays out its vision for the future, and seeks to distinguish itself from the opposing party (Hershey 2009, 182185). But with the revolution in the presidential nomination process in the 1970s, national conventions today are now much more a coronation of the nominee already chosen earlier through the primary process.2

The Importance of the National Conventions

Nonetheless, national party conventions continue to be important political events. Party conventions have always been, and continue to be, important means of political advertisement.3 Because of television coverage, parties have increasingly tried to stage-manage their conventions. Schedules have been changed to accommodate the television audience; podiums and the visual backdrop have been altered to enhance the visual appeal of the televised image; and, the placement of delegates (e.g., women or minority delegates) and signs have been strategically positioned in front of the cameras to offer a particular image of the party (Polsby and Wildavsky 2004, 127129). Conventions are also carefully managed to transmit verbal messages and visual images that signal enthusiasm, appeal, and unity. The drafting of party platforms also continues to be an important, though potentially the most controversial and contentious, endeavor at the convention. Party platforms are important because they represent the partys principal attempt to dene itself; to state what it stands for and why its candidates should be elected (Wayne 2000, 165). Platforms articulate the partys goals and the policy positions that serve to characterize the party and differentiate it from the opposing party. Adopting the partys platform is a potentially contentious endeavor, because two often conicting goals are inherent in the platform-drafting process: winning the election and pleasing the partys core constituency (165). Obviously, any public disagreement over the substance of the platform can easily undermine the effort to project an image of a unied and enthusiastic party. As a result, party platforms are broad-based documents. They seek to appeal to independents and weak partisans of the other party on the basis of issues and values (166), while at the same time they target particular messages to

Religion and the Fall Campaign


specic groups whose support they seek either to retain or attract by inserting distinct planksidentiable sub-areas in the platformthat discuss their pet concerns (Domke and Coe 2008, 103). Today, conventions serve largely as the launching pad for the fall campaign; they are planned in order to sell the ticket and the party to a national audience (Polsby and Wildavsky 2004, 127). Most of the actual business conducted at party conventions happens prior to television coverage, with prime time coverage reserved for more entertaining activities meant to appeal to votersspotlighting rising political stars in the party, showing lms about the partys presidential nominee, and scheduling and coordinating speeches meant to convey desirable messages and themes. In so doing, the party seeks to attract voters to the party and its presidential candidate and thereby to receive a bounce in the level of public support for its candidate in the polls taken following the convention (127).

Religion at the National Conventions

Historically, religion has played only a relatively minor role at the national conventions, typically being limited to clergy or other religious leaders offering invocations and benedictions at the opening and closing of each days events. However, in 2008, religion became evident in new ways at the national party conventions. In particular, parties discovered new ways to bring religion into the national convention, and the media coverage of the two conventions suggested that religion played a much more prominent role in the Democratic Party convention. Religion and the Democratic Convention. Similar to previous conventions, the ofcial program of the 2008 Democratic National Convention in Denver (August 2528) began with an invocation and ended with a benediction delivered by a national religious leader or an individual active in his or her faith community. But beyond this common practice, religion played an unprecedented role at the Democratic convention. In fact, one observer claimed that in comparison to the 2004 Democratic convention, the 2008 gathering was a veritable religious revival meeting (Waldman 2008a). Much of the new religious tenor of the Democratic convention was the product of the planning of the Democratic National Committees Faith in Action Committee, created in response to the Republican victory in the 2004 presidential election. The Faith in Action Committee initiated several new faith events within their partys convention in an effort to dispel any existing perceptions that the Democratic Party was hostile or unfriendly toward religion.


The Disappearing God Gap?

At the 2008 Democratic convention, the committee sponsored, for example, its rst-ever multifaith gathering on Sunday afternoon just prior to the opening of the convention. The event brought together delegates, elected ofcials, local residents, musical guests, and religious leaders, with about three thousand in attendance at the Colorado Convention Center (Lopez 2008). Among those appearing were Democratic Party Chairman Howard Dean, House Majority Whip Jim Clyburn, and Colorado Governor Bill Ritter. The gathering was led by the Democratic conventions CEO, Leah D. Daughtry, a Pentecostal minister. In addition to keynote remarks, the program included readings from diverse religious texts, prayers, and gospel music. As desired, this event captured considerable media coverage. The interfaith gathering was only part of the larger effort to project the Democratic Party as friendly to religion. At least nine different faithrelated events were held. Democratic ofceholders talked about the role of faith in their lives at a lunch sponsored by the Faith and Politics Institute. In addition, there were the rst-ever Faith Caucus meetings as well as the sponsorship of several afternoon faith panels. In an obvious effort to address a key concern from the 2004 election, these panels were often framed as discussions about how Obama and the Democratic Party could integrate serious moral considerations into policymaking. On Tuesday afternoon of the convention, there were panel discussions related to Common Good on Common Ground, which focused on the moral issues of the day, and a panel on How an Obama Administration Will Engage People of Faith. On Thursday afternoon of the convention, the Faith Caucus convened a panel discussion on Moral Value Issues Abroad, which addressed how religious voters could work together to address pressing moral issues around the world, as well as a panel on Getting Out the Faith Vote which sought to address how to engage communities of faith in the 2008 election. In addition to these events, there were others held by the Network for Spiritual Progressives, the National Jewish Democratic Council, and the American Muslim Democratic Caucus. Religion and the Republican Convention. If Americans were asked, prior to 2008, which political party convention was most likely to engage in some kind of religious display, most would have chosen the Republicans. But in contrast to the media coverage given to the place of religion at the 2008 Democratic National Convention, hardly any coverage was given to religion at the 2008 Republican gathering in St. Paul, September 14. As usual, the convention opened and closed each day with an invocation and a benediction and the convention did include a performance by a Christian pop star, and a few speakers alluded to their faith (Stoddard 2008; Paulson
Religion and the Fall Campaign


2008). But beyond those scattered examples, there was virtually no media coverage related to religion at the Republican convention.4 Several different possibilities account for this lack of media coverage. First, part of what makes something newsworthy is its newness. The emphasis on religion at the Democratic convention was something fresh, and hence it was deemed newsworthy. Conversely, religious activities or expressions at the Republican convention might have simply been deemed an old story, and hence not worthy of coverage. Second, there were few, if any, religious events to coverand intentionally so. After all, McCain wanted to avoid close ties to the Religious Right, and yet he needed to energize his support among the evangelical Protestant core of the Republican Party. He apparently sought to accomplish both objectives through his choice of Sarah Palin as his running mate. Indeed, as she energized delegates with her speech at the convention, Palin may have served as the religious story of the GOP convention. Third, even if religion was less explicitly evident in the Republican than the Democratic convention, it may be that religion was implicitly evident in terms of the rhetoric employed. Scholars have long noted the fusion of religion and political themes within the public rhetoric of many American political leaders. In speeches at national conventions, politicians can combine political covenants, family values, and God and country into civic piety (Hart and Pauley 2005). In such cases, while religion may not be directly invoked with any great frequency, it may nevertheless be evident in the rhetorical undertones. Yet the subtlety of such discourse makes it more difcult to cover. Research has revealed that delegates to the national conventions not only differ from the public at large but differ across parties as well (Jackson and Crotty 1996, 7887). Delegates to the Republican convention are more likely to be religious than delegates to the Democratic convention (Layman 2001, 100110). Perhaps, then, any religious expressions at the Republican National Convention were more likely to be found in personal conversations among delegates than on stage.

Campaign Themes and Issues

After the conventions, the campaign intensies, as now the candidates primary responsibilities are to get the campaigns message out to the voters and to energize activists (Polsby and Wildavsky 2004, 138). Almost immediately, the focus of media coverage also changes, as journalists and


The Disappearing God Gap?

pundits begin to assess the strengths and weaknesses of the two tickets, identify which kinds of voters they believe are likely to hold the keys to electoral success for each campaign, sort out safe and battleground states, and point to important campaign events, particularly the presidential debates. There are some important differences between the general election campaign and the campaign for the partys nomination, besides the obvious. The calendar is different. The general election campaign typically begins on Labor Day and ends the rst Tuesday after the rst Monday in November, a total of approximately three months, while the nomination process may last a year or longer. Moreover, while the battle for the nomination is fought in a series of state-by-state campaigns spread out over several months (permitting the possibility of recovery if one falters in a particular state or on a particular day), in the general election all states cast their ballots on the same day. Finally, the rules are different. Typically in primary contests, one wants to win the most votes to be declared the winner, but in the general election campaign one must win a majority of Electoral College votesnot simply a plurality of total votes castin order to be elected president. As a result, there are really fty different elections occurring on Election Day.

Campaign Themes
Campaign themes and campaign issues work hand in hand. While issues can play an important role in shaping voting decisions, they typically do not do so in the manner posited by democratic theorists. Presidential candidates typically have detailed position papers outlining their viewpoints and particular positions on a whole range of issues. But while such position papers exist, they dont win elections. Most voters have neither the time nor the inclination to examine every issue carefully. As a result, campaigns typically seek to frame the criteria on which candidate choices are to be made. They establish themes that aim to inuence which issues should be considered and which particular qualities of the candidate should be taken into account when voters make their choice (Salmore and Salmore, 1989, 112). Obama. A key theme of the Obama campaign could be summed up as the interrelated concepts of hope and change. Obamas advertisements frequently included the tag line The Change We Need, which was linked to Obamas persistent emphasis on hope, as laid out most clearly in his bestselling book, The Audacity of Hope. The theme, which tapped into a common desire for a better future, also has important religious undertones, as illustrated in Obamas 2004 convention speech:5
Religion and the Fall Campaign


HopeHope in the face of difculty. Hope in the face of uncertainty. The audacity of hope!. . . . In the end, that is Gods greatest gift to us, the bedrock of this nation. A belief in things not seen. A belief that there are better days ahead.6 Of course, these are familiar themes for an out-party candidate in a presidential election. At least since Andrew Jackson in the early nineteenth century, candidates have crafted outsider campaigns that attempt to translate voter frustrations into a winning strategy focused on change. Obamas two immediate predecessorsBill Clinton and George W. Bushemployed this strategy. Clintons persistent focus on transforming the at economy in 1992 was coupled with his campaign image as the man from Hope, a play on his boyhood home of Hope, Arkansas. And a clear message of the Bush campaign in 2000 was that he represented a welcome change from the worst excesses of the Clinton administration, which put candidate Al Gore in the unenviable position of touting his experience as the incumbent vice president while also distancing himself from the administration. While Obama invoked these familiar themes of hope and change, his campaign was also positioned to use them in distinctive ways. First, his position on some key policy issues distinguished him even from his fellow top-tier Democratic contenders. Consider his early public opposition to the war in Iraq. To be sure, he was spared a public vote on the matter because he was serving in the Illinois legislature when authorization for the war was being debated in the U.S. Senate. Hillary Clinton and John Edwards both cast votes authorizing military action in Iraq. But Obamas absence from the U.S. Senate at the time arguably reinforced his outsider status, and ultimately Democratic primary voters preferred that status over Clintons (and ultimately McCains) stronger claims of experience. Obamas emphasis on change and hope was also aided by the fact that the incumbent presidenta president from the opposing partywas unusually unpopular at the end of his term. Obama was also able to use his considerable personal skills and charisma to buttress his claim as the most likely change agent. Liberal and conservative commentators alike, for example, noted that Obama had unusual rhetorical gifts; attendance at his campaign speeches far exceeded any other candidate from either political party, and his cadences and references suggested he t with a line of other prominent reformers in U.S. history, from Franklin D. Roosevelt to Martin Luther King. It is particularly relevant that Obama linked his campaign theme of change with a hopeful optimism, as epitomized in the slogan Yes, we can! Critics


The Disappearing God Gap?

charged that he was preaching change for changes sake and that his general optimism masked a lack of specic plans. Nevertheless, the combination of change, hope, and optimism was ready-made for the 2008 election cycle, especially as nancial markets soured toward the end of the general election campaign. It is difcult to overestimate another factor that reinforced Obamas dual status as outsider and change agent: race. As the rst African-American major party nominee for the presidency, Obama represented a unique and historic symbol of change. In fact, some leaders in the Democratic Party worried that the symbolism was too potent and argued that Obama could not win key states because of his race. The intraparty discord as a result of such claims may have backred on its chief proponents (which apparently included former president Bill Clinton), as it appeared to rally AfricanAmerican support for Obama in the primary elections. In any event, the debate about his race simply illustrates its importance as a consideration in the campaign, and perhaps as an advantage to Obama in shaping his image as both outsider and reformer. Here we see an interesting conuence of race and religion. Social scientists have a notorious time disentangling the effects of these two factors on political choices, but that is often because the two are inextricably linked. Such was the case with Barack Obama. By most accounts, Obama had thought a great deal about the role of religion in public life, including how his own Christianity ought to shape his policy preferences and political tactics. His own faith commitments should have been an advantage in the mass political culture of the United States, where a religious presidentor at least the perception of oneis almost a strategic prerequisite. Yet religion can cut both ways. The very public comments of Jeremiah Wright, his controversial pastor at Trinity United Church of Christ, raised many eyebrows in the electorate. Coupled with his work as a community organizer in a Chicago Catholic parish and his association (however brief and apparently inconsequential) with some radical activists, some voters voiced concerns over Obamas ultimate goals. Change is a powerful campaign theme, but it resonates with voters only if it has a reformist balance: not too reactionary, on the one hand, or too revolutionary, on the other. In the nal analysis, we argue that Obamas religion, among other factors, helped him convince the electorate that he embodied that balance. The irony is that many voters discovered that Obama was a religious manby itself a key advantageafter his pastors comments began to circulate. Obama managed to distance himself from Wright without losing his religious bona des; surveys after the episode suggest that more voters
Religion and the Fall Campaign


saw him as being religious than in late 2007 (though the public perceived him to be more religious than Hillary Clinton throughout the entire campaign). In a sense, Obama personied two common characteristics of public religion: on the one hand, its powerful ability to speak in a prophetic voice against perceived injustices; and, on the other, its stabilizing inuence on individuals and institutions. Indeed, one might argue that, far from generating discord, Obama managed to use religion in order to reinforce a second key theme: unity. This theme touched the desire among many that after the contentious partisanship of the Clinton and Bush presidencies, we could seek to emphasize what we share rather than focus on how we differ. This theme of unity was already evident in his address to the 2004 Democratic National Convention, which initially introduced Obama to millions of Americans across the country. Obama tied this theme to the famous biblical story of Cain and Abel: It is that fundamental belief: I am my brothers keeper. I am my sisters keeper that makes this country work. Its what allows us to pursue our individual dreams and yet still come together as one American family.7 Obama repeated this theme in his acceptance speech at the Democratic National Convention: Thats the promise of America, the idea that we are responsible for ourselves, but that we also rise or fall as one nation, the fundamental belief that I am my brothers keeper, I am my sisters keeper. At its heart, this theme of unity reects a civic gospel honed by an AfricanAmerican Christian spirit of forgiveness and reconciliationand a belief that in Christ all are one (Wellman 2008). The result is a radical vision of inclusiveness that replaces the conventional means of identifying oneself by whom one stands against politically. It is a new form of politics that seeks to overcome identity politics by contending that differences in gender, ethnicity, age, income, and sexual orientation no longer matter. McCain. At times it was difcult, even for McCain supporters, to discern a coherent message or set of issues in the Republicans varied and changing campaign. Even as late as August 2008, one commentator contended that the presidential campaign of John McCain has been run so poorly that Republicans should be allowed to sue it for negligence, with a major failure of the McCain campaign being that it had no discernible overarching theme or themes (Crowley 2008). At best, McCains overall campaign theme seemed to be I am not Bush. In general, McCains campaign lacked a central, positive message that voters and the media could latch on to, and following the election, some informed observers concluded that this failure was a major reason for McCains defeat (Berkowitz 2008, 1).


The Disappearing God Gap?

If there was any one major theme emphasized by the McCain 2008 presidential campaign it was its focus on candidate character and experience. Throughout both the primary and general election campaigns McCain attempted to portray himself as an honest politician who, despite his long service in Washington, remained untainted by the corruption that the public sometimes associates with Washington politics. Indeed, McCain frequently referred to his campaign as the straight talk express and emphasized his political independence by labeling himself a political maverick who was unafraid to disagree with his own party on key issues. In fact, McCains early television ads following the GOP convention, titled Original Mavericks, played up McCain and Palins maverick personas, and contended that they would change Washington (Rhee 2008). Other commentators, too, noted that the McCain campaign was pushing biography, not issues (Martin and VandeHei 2008). Furthermore, McCain attempted to turn his military record and long history of service in the U.S. Senate into a campaign asset, arguing that his political experience made him well suited to deal with todays complex and dangerous world. Asserted McCain, I will not be a president who needs to be tested. I have been tested (Morgenstern 2008, 1). By emphasizing his honesty and long record of public service, McCain was, at least by implication, contrasting himself with his opponent. Until the latter stages of the campaign, McCain generally refused to make a direct, frontal assault on Obamas record and character. In part, this refusal may have stemmed from McCains personal sense of integrity. But it may also have stemmed from a concern that a direct, personal challenge to Obama might be perceived as racist. In the end, McCain did run an advertising campaign that emphasized ObamaNot Ready to Lead. Nevertheless, some observers have suggested that McCains unwillingness to focus on Obamas lack of experience as well as his association with such controversial characters as Rev. Jeremiah Wright, Bill Ayers, and Tony Rezko may have cost him votes. On the other hand, even McCains allies complained that the campaign offered a myriad of confusing themes and seemingly could never settle on just how they wanted to present McCain to the American public (Shear and Eilperin 2008), as the campaign vacillated between pitching McCain as a committed conservative one day and an independent-minded reformer the next.

Campaign Issues
Of course, broad campaign themes alone are not sufcient. Candidates and parties must adopt substantive issue positions to provide some practical
Religion and the Fall Campaign


evidence for the particular themes they choose to emphasize. Given the lengthy duration of presidential campaigns, candidates frequently begin developing their policy statements and issue positions at least two years in advance of the election (Polsby and Wildavsky 2004, 147). Their goal is to frame the issues in ways that will be most attractive to their targeted voting blocs. One of the assumptions of issue framing is that there is rarely a single way to look at a policy issue: AIDS can be cast as a moral issue or a health issue; abortion can be viewed in terms of protecting the sanctity of life or promoting individual freedom; military combatants can be labeled as freedom ghters or guerillas. And since our values are heterogeneous, campaigns have an opportunity to shape the way voters think about an issue by setting the terms of the policy debate. Political scientists generally agree that issue framing is a potent political tactic (see, e.g., Schattschneider 1960; Gamson 1992). In political campaigns, choosing the right frames can attract voters and demobilize the opposition; choosing the wrong ones can alienate potential supporters and lead to weakened candidacies. Above all, this means carefully crafting a campaigns language to communicate the appropriate issue frames, often through analogies, metaphors, or other linguistic symbols (Kinder 1998). Moreover, a candidates position on an issue also communicates aspects of ones governing philosophy that reect a particular understanding of human nature and the role of community or the larger society within political life. This then is how the competing issue positions of Obama and McCain should be seenas competing stories about the world and the proper place of people in it. An examination of the two candidates (and their respective parties) language about several key issue areas reveals the competing framing narratives they offered as well as the challenges and opportunities the two candidates and parties faced.8 Economic and Social Equity Issues. During the 2008 campaign, Obama put forward a blizzard of specic policy proposals aimed at providing additional help for low-income persons. He advocated a $10 billion foreclosure prevention fund, the elimination of all income taxes for seniors with incomes of less than $50,000 a year, raising the minimum wage, an increase in the Earned Income Tax Credit program, new health care benets that would move the country toward a universal health care system, and the creation of 20 Promise Neighborhoods in high poverty areas.9 Obamas urry of ideas presented the persona of someone who cared and who was prepared to use government to address the needs of those who had previously gone largely unrecognized by government. Poverty was contextualized as both a collective and personal responsibility; economic need was


The Disappearing God Gap?

framed as demanding a social obligation rather than constructed as individual failure or tragedy. Obama often framed the economic issues of 2008 in moral terms precisely the kind of language that some liberal religious leaders had been urging since the beginning of the campaign (e.g., Wallis 2006). In a major speech on the economy in September 2007, for example, Obama echoed the words of Franklin Delano Roosevelt by calling for a re-appraisal of values and a rejuvenated faith in working collectively to solve economic problems. He pointedly rejected the hidden hand, self-interest, and survival of the ttest as appropriate economic values, and he insisted that interest group politics was skewing public policy toward the afuent, which was not only bad economics but offends our morals (Obama 2007). The apparent goal in this and other pronouncements was to shift the language of morality in the direction of issues that are typically on the Democratic agenda (e.g., progressive taxation, social welfare). That effort was in many ways a response to religious conservatives, who for several decades effectively deployed moralistic language on a different set of issues, including abortion, marriage, and other family values concerns. By contrast, McCain did not put forward nearly the same number of specic proposals on economic and social equity issues. He could not, because he had a different message and governing philosophy to communicate, one tied to the sanctity of the individual and the freedom of the individual from government claims. He advocated lowering the corporate tax rate with the aim of stimulating the economy, and he supported providing health care tax credits to help low income persons to purchase health insurance. McCains economic policy differed from Obamas in that McCain pledged not to raise taxes, even on high-income taxpayers, while Obama favored raising tax rates for the top 5 percent of taxpayers and expressed opposition to continuation of the Bush reduction in capital gains taxes scheduled to expire in the year 2010. In the early days of the presidential race, economic issues did not capture the publics attention. However, following the dramatic decline of the stock market and related crises of the nancial and housing industries in mid- to late September, economic policy dominated the presidential campaign. Both McCain and Obama agreed on the need for the federal government to provide a large scal stimulus to the economy to encourage consumer demand for goods and services. However, they differed on its size and the precise mechanism to be used to stimulate the economy. For example, Obama proved more sympathetic than McCain to providing federal loans or subsidies to the ailing American automobile industry. This reected the political allies of both candidates as much as their
Religion and the Fall Campaign


regional orientationthe industrial Midwest as contrasted to the resourcedriven, libertarian West. Nevertheless, McCains willingness to support big government solutions to economic problems may have cost him the votes of free-marketoriented members of his party. At a minimum, it undermined some of the earlier media ads and any later effort of the McCain campaign to frame Obama as a tax and spend liberal, since McCain also advocated that government spend hundreds of billions of dollars to address the nancial situation. Environmental and Energy Issues. In spite of considerable campaign rhetoric, there was signicant agreement between Obama and McCain on environmental and energy issues. Both emphasized the need to develop cleaner alternative energy sources. Both favored a cap and trade system for reducing carbon emissions that most believe are causing global climate change: such a system would cap the amount of allowable greenhouse gas emissions, and businesses would either have to reduce their emissions to meet those standards or purchase offsets that would compensate for their emissions by other activities. However, McCain made a campaign issue of calling for offshore drilling, while Obama supported only limited offshore drilling and only if it were linked to energy conservation efforts.10 McCain favored and Obama opposed removing the federal gasoline taxes for the summer of 2008, when gasoline prices shot up to over $4 a gallon. Both McCain and Obama recognized that environmental issues communicate more than simple policy stances. McCain tended to use language of the frontier, tied to a claim to the resources at hand. Such language was consistent with the public persona he presented and with his character as a politician from the West. Obama tended to use language tied to stewardship or creation care language, which links environmentalist goals with a divine mandate to care for the earth. Obamas campaign document on the environment clearly evokes a sense of moral obligation by making regular use of the language of responsibility and stewardship (Obama for America n.d.). The document also associates the care of the environment with commitment to family in that the campaign pledged, for example, to protect our children from environmental toxins and to reect a responsibility to our children to leave this Earth better than we found it. The moral language and its association with family once again appropriated a page from the Republican playbook, and while McCain explicitly acknowledged his belief in climate change and expressed a desire to address it, he seemed to have little response to the issue framing of the Obama campaign. Cultural Issues. On abortion, the Democrats clearly and consistently supported a womans right to choose. The Democrats platform declared


The Disappearing God Gap?

in unambiguous terms: The Democratic Party strongly and unequivocally supports Roe v. Wade and a womans right to choose a safe and legal abortion, regardless of ability to pay, and we oppose any and all efforts to weaken or undermine that right (Democratic National Platform 2008, 50). By contrast, the Republican platform declared: We support a human life amendment to the Constitution, and we endorse legislation to make clear that the Fourteenth Amendments protections apply to unborn children. We oppose using public revenues to promote or perform abortion and will not fund organizations which advocate it (Republican National Platform 2008, 52). The contrast between the two party platforms could not have been clearer. The two candidates, Barack Obama and John McCain, as well as their vice presidential running mates, fully supported their parties positions. These positionswhile rarely a subject of television adswere frequently articulated by all four candidates during the course of the campaign. In addition, while he was a member of the of Illinois state legislature, Obama opposed a bill that would have given legal rights and required medical care to any infant who survived an attempted abortion.11 A federal version, called the Born-Alive Infant Protection Act, passed the Senate in 2002 by unanimous consent and was signed into law by President Bush. Given Obamas opposition to the bill as a state legislator, his critics contended that he favored not only abortion but infanticide. In response to such accusations, Obama put forth a counterframe around the theme of both personal responsibility and compassion. Such a stance could protect his base of pro-choice supporters as well as attract pro-life voters. More importantly, it spoke to those who needed to hear that the candidate understood abortion to be a moral issue even if he supported legal abortion. Thus, the Obama campaign sought to change discussion about abortion in American society from whether it should be legal or illegal to lets work together to reduce the number of abortions (Zapor 2008). As Obama acknowledged in his acceptance speech at the Democratic National Convention, We may not agree on abortion, but surely we can agree on reducing the number of unwanted pregnancies in this country. Despite this acknowledgement, however, the campaign ran radio ads that championed Obamas right to an abortion position without discussing his abortion-reduction agenda (Waldman 2008b).12 While this meant that those who opposed abortion could frame him as pro-abortion, it also meant that those who favored abortion rights could hear his framed message as exible, accepting of pluralism, compassionate, and pro-woman. On same sex marriage the Democratic platform said, We support the full inclusion of all families, including same sex couples, in the life of our
Religion and the Fall Campaign


nation, and support equal responsibility, benets, and protections. We will enact a comprehensive bipartisan employment non-discrimination act. We oppose the Defense of Marriage Act and all attempts to use this issue to divide us (Democratic National Platform 2008, 52).13 The difference with the Republican platform was clear: Republicans have been at the forefront of protecting traditional marriage laws, both in the states and in Congress. A Republican Congress enacted the Defense of Marriage Act, afrming the right of states not to recognize same-sex marriages licensed in other states. Unbelievably, the Democratic Party has now pledged to repeal the Defense of Marriage Act (Republican National Platform 2008, 53). John McCains website proclaimed, John McCain believes the institution of marriage is a union between one man and one woman.14 Obamas website was silent on the question of same sex marriage. Elsewhere, both candidates stated their opposition to a constitutional amendment banning same sex marriages, and both favored civil unions with partners receiving legal benets normally reserved for married couples. However, McCain did take the position that a national same sex marriage ban might be in order if the courts overturned state laws banning same sex marriages. Both Obama and McCain supported giving federal funding to embryonic stem cell research. Aware that this position would put him at odds with religious conservatives, McCain indicated that a reason he supported such research to help spur medical advances was because these embryos would otherwise either be discarded or kept in a permanent frozen state (Foust 2008). There were similar policy differences between the two parties on other cultural issues, but the differences were more often expressed less clearly or with the use of code words. For example, the Republican platform stated, The public display of the Ten Commandments does not violate the U.S. Constitution and accurately reects the Judeo-Christian heritage of our country. We support the right of students to engage in studentinitiated, student-led prayer in public schools, athletic events, and graduation ceremonies, when done in conformity with constitutional standards (Republican Platform 2008, 5354). The Democratic platform was silent on these issues. Nevertheless, with a few exceptions (most notably the famous Saddleback Forum in August), McCain did not emphasize his generally conservative views on issues of personal morality. Instead, he tended to place somewhat greater emphasis on social justice issues, taking moderate or progressive positions on such matters as government efforts to halt climate change


The Disappearing God Gap?

(favoring such efforts), the need for national health insurance (proposing extending insurance coverage through the marketing of health insurance nationwide, the expansion of health savings accounts, and the elimination of the tax deductibility of employer-paid insurance), and immigration (supporting efforts to assimilate undocumented aliens). On family issues and issues around the proper public space for religion in a pluralist society, Barack Obama crafted his presentation to convey a sense of compassion tempered with a sense of responsibility, including stories of his own life or his role as a father. He also put forth a frame of inclusiveness, trying to communicate his identity as a person of faith while also communicating a sense of welcome for all. Balancing such frames as these can be quite difcult, and the campaign was no exception. McCain found it hard to get away from a sense of rigidity or lack of compassion, just as Obama found it hard to avoid accusations of permissiveness on abortion or even a lingering sense among many that he was a Muslim. Foreign Policy and Military Issues. There were sharp differences between both the candidates and the party platforms in 2008 on foreign policy and military issues. From the beginning of his candidacy for the presidency, Obama emphasized that only he had voted against the war in Iraq, and an important core of his supporters were those who adamantly opposed the war. With U.S. troops now in Iraq, Obama favored the establishment of a denite, eighteen-month period for the removal of American troops from that country and a recommitment of troops to Afghanistan (where, he contended, the real threat to American security could be found). However, his foreign policy positions did not emphasize the establishment of viable democracies in the Middle East as an announced, overt goal of his presidency, as Bush had done (although his critics noted that Bush had not consistently pursued this goal). On the other hand, McCain emphasized his distinguished military record, his support of the war in Iraq, and his advocacy of the military surge in that country. While McCain and the Republican Party platform called for a phased withdrawal of forces, they argued that there should be no xed timetable and troops should be withdrawn only as conditions in Iraq allowed. More generally, Obama stressed the importance of diplomacy and working multilaterally with allies, while McCain stressed the importance of military strength. Thus, the broader narratives of such foreign policy positions were clear and consistent: McCain expressed a kind of Western libertarianism and practical religion tied to individual morality in that just as individuals should best be left alone, so, too, the United States in its foreign policy should best go it alone; Obama, on the other hand, offered a competing frame of collective action and shared responsibility.
Religion and the Fall Campaign


During the early stages of the 2008 presidential campaign, foreign policy issues proved to be of major importance and clearly distinguished McCain from Obama, as the public opinion polls generally revealed that the public viewed McCain as having more experience than Obama in issues related to matters of foreign policy. However, with the nancial crisis unfolding in late September, economic issues came to dominate the campaign, and, as a result, issues related to foreign policy were largely moved to the back burner of public debate and discussion as attention became focused on economic matters. Faith-Based Initiatives. The Charitable Choice provision of the Welfare Reform Act of 1996, signed by President Clinton, specied that when the government decides to partner with community groups in addressing social welfare needs, no private agency should be excluded from consideration simply because of its religious character. In other words, it is the specic role of a social service agency, rather than its religious identity, that is to serve as the basis of assessment. All programs, whether religious or secular, are to be evaluated simply on the basis of how well they train the unemployed, care for children, or serve drug addicts. If a program is successful, it deserves funding consideration regardless of whether the agency is religious or secular. In addition, the provision seeks to safeguard the religious character of the faith-based organizations that choose to partner with the government in providing such services. The faith-based and community initiative program championed by President Bush went beyond the earlier Charitable Choice provision. First, it created an Ofce of Faith-Based and Community Initiatives directly in the White House and similar ofces in eleven executive branch departments of government. Second, it used these various ofces to give renewed emphasis to the earlier Charitable Choice provision by actively working to level the playing eld between secular and faith-based organizations in competing for government funds for their programs. Third, it proposed to apply the Charitable Choice provisions to the funding of nearly all health and social service programs in areas in which nonprot organizations provided services of some type currently funded by the federal government. Thus, Bushs proposal would have expanded opportunities for government funding for the social service activities of religious groups by allowing them to compete for funds without restricting their religious activities (Bartkowski and Regis 2001, 3; Burden 2007, 124). However, when the Bush administration introduced these steps in 2001, there was an intense political restorm in Congress around them, with Democrats largely standing in opposition to them.15


The Disappearing God Gap?

McCain also supported faith-based initiatives and indicated to The New York Times that such programs have been one of the more successful parts of the Bush Administration and I would continue it (The Times Interviews 2008). Because the faith-based initiative was so closely tied to the Bush administration, many Democrats expected Obama to disavow it. But in late July 2008, he stated that the faith-based ofce had never fullled its promise and announced that an Obama administration would x, expand and elevate the faith-based initiative [program] (Sullivan 2008c). Hence, Obama effectively took away a potentially distinctive position from McCain by endorsing the goals of the faith-based ofce while suggesting that the program had been poorly implemented. Conclusion. Obama understood that taking positions on particular issues framed his story, and he was not unaccustomed to parsing the story in a way that would make sense to people of faith. He spoke of the economy in terms of the collective good and looked at poverty not as individual failing but as a social responsibility. He contextualized the environment in terms of stewardship issues and concerns that would be recognizable to religious people who speak of the care of creation. He addressed family issues from the perspective of compassion and inclusion, supporting legal abortion but talking from time to time of personal responsibility and morality. Obama showed himself to be a master of recognizably religious rhetoric that felt inclusive and was less reminiscent of a language of personal morality and more of a religiously progressive public theology and worldview. McCain took a variety of positions that told a contrasting story. His frame, too, had a place within U.S. religious rhetoric and could easily be placed within both religious and regional categories. While McCain showed himself to be more expert with the language of personal morality, he had little opportunity to use it, given the major national questions of the day.

Campaign Organizations and Religious Groups

Political campaign organizations are composed of staff people who work on behalf of a candidate, formulating and working to implement the strategy needed to garner victory at the polls on Election Day. At the head of the organization is a campaign manager. Typically there is a deputy campaign manager directly below the campaign manager, and then below them are a number of departmental directors who are responsible for different aspects of the campaign. Typically, there are a variety of departments, including policy, fund-raising, legal, technology, scheduling and advance, outreach, and eld (or ground).16
Religion and the Fall Campaign


Presidential campaign organizational structures have varied over time. But typically as part of their outreach operations in recent years, presidential campaign organizations have had staffers who focus predominantly, if not solely, on promoting the candidate among designated religious groups.

The Obama Campaign Organization

One of the important differences between the Obama campaign in 2008 and the Kerry campaign in 2004 was the much greater effort the Obama campaign made to appeal to religious voters. When Mara Vanderslice served as director of religious outreach for John Kerrys presidential campaign in 2004, she had only one unpaid intern and no budget; she wasnt permitted to talk to the media and was basically ignored at every turn (Diamant 2008). By contrast, the Obama campaign sought to engage religious voters from the start. His campaign had an entire religious outreach department, with money to spend and the ear of the candidate (Zapor 2008); it had at least six full-time staffers working on outreach to religious communities (Diamant 2008). Joshua DuBois, as we noted in the previous chapter, headed up the religious outreach efforts of the Obama campaign as director of religious affairs. D. Paul Monteiro served as deputy director for religious affairs, while Mark Linton served as Catholic coordinator. All three were part of the Obama campaign structure throughout the primary campaigns. In June 2008, the Obama campaign hired Shaun Casey, associate professor of Christian ethics at Wesley Theological Seminary, as a senior advisor for religious affairs. He became Obamas evangelical coordinator. Terri LaVelle, staff director of the Democratic National Committees Faith in Action Initiative, focused on outreach to African Americans and women. In addition, two others worked on outreach among Jews, and another did outreach among Muslims. Aiding these full-time staffers were six to ten interns working full-time on religious outreach (Shaun Casey, personal correspondence).17 The Obama campaign engaged in three specic organized efforts to reach out to religious voters. One was called the Joshua Generation project, which was designed to reach out to young evangelical and Catholic voters that the campaign thought could be brought into the Obama camp. A website was launched with fanfare and a signicant amount of media coverage. But the campaign struggled to move the project from grand vision to on-the-ground reality. It was never fully operational, it was underfunded, its website never


The Disappearing God Gap?

had more than minimal content, and its impact was almost nonexistent (Shaun Casey, interview, March 5, 2009). Another project of the campaign was more successful. It consisted of Faith Forums, which Shaun Casey described as the most effective onthe-ground effort at religious outreach for the Obama campaign (Casey, interview). This approach was rst used successfully in the Iowa and South Carolina primaries and continued to be used in later primaries and in the fall campaign. It consisted of securing a public room in a local library or elsewhere and inviting as many persons as the campaign could identify as having a strong faith interest (church networks were helpful here). Typically anywhere from ve to twenty persons would attend. A video of Barack Obama speaking about his faith, his commitment to family, and his sense of values was shown. There would be a time for questions and comments, and then persons attending were asked to volunteer for the campaign. But regardless of whether or not those attending would volunteer, the word about Obama as a person of faith and deep moral values was spread. Hundreds of these Faith Forums were held. In the general election, the campaign often made use of surrogates at these forums, including Tony Hall, a former Democratic congressman from Ohio and an evangelical, and Douglas Kmiec, a Pepperdine University law professor, well-known Catholic, and former ofcial in Republican administrations. A third attempt at reaching religiously oriented voters had less success than the Faith Forums. These were Religious House Parties, in which volunteers invited friends and neighbors into their homes. The campaign developed a curriculum and a kit containing information on Obama and his religious commitment and values, and a short DVD of Obama was also included. Those hosting the house parties underwent training via a conference call. But the campaign assigned only one intern to this effort, and it never reached the numbers it would have needed to have a signicant impact (Casey, interview). It is easy to overstate the importance of the Obama religious outreach efforts to the campaign. They were not as extensive nor as high a priority as the news media at times made them out to be. The situation is somewhat like the old story of the dog that could count to tenthe surprising thing was not that he could only count to ten, but that he could count at all. As Shaun Casey said, We [the religious outreach people] were largely peripheral to the campaign, but in earlier Democratic campaigns we were not included in the campaign, not even peripherally. As limited and sometimes as fumbling as the religious outreach efforts were, Casey argues that they had measurable results and that in subsequent Democratic presidential campaigns more and better use will be made of religious outreach staff.
Religion and the Fall Campaign


The McCain Campaign Organization

While the Obama campaign made a greater effort at religious outreach than had previous Democratic presidential candidates, the McCain campaigns religious outreach efforts were fewer and less effective than those of previous Republican presidential campaigns. The problems were signaled already in early 2007. John McCain had hired two aides to spearhead religious outreach for the campaign, but they were dismissed in early April 2007, only three months after they had been hired. This dismissal was attributed to performance related considerations as well as to being part of a broader staff reshufing . . . that grew from weaker-thanexpected fundraising (Gilgoff 2007). Nevertheless, the two dismissed aides soon leveled some telling broadsides against the McCain campaign organizationwith a description that largely reected the experience of Mara Vanderslice within the Kerry campaign of 2004. Marlene Elwell, hired in December 2006 to be the national director of McCains Americans of Faith coalition, indicated after her dismissal that McCains top campaign strategists are intent on winning votes of religious voters without having to develop serious ties to faith communities (Gilgoff 2007). Elwell noted that the campaign staff refused to return scores of their phone calls or e-mails, that they, as staffers, were denied access to leaders of the McCain campaign, and that they were pressed to collect church directories, a controversial tactic, as the centerpiece of a strategy to win values voters (Gilgoff 2007). The other red staffer claimed that the campaign staff exhibited a contempt for Christians (Gilgoff 2007). As late as mid-May 2008, the McCain campaign still lacked any staffer who was devoted fulltime to faith outreach (Gilgoff 2008a).18 In part, this was the result of the campaign not making a clear-cut distinction between outreach to religious voters and outreach to social and scal conservatives more generally.19 Robert C. Heckman served as senior consultant for the McCain campaign and director of conservative outreach. Marlys Popma, who served as the national coordinator for evangelical and social conservative outreach, was hired to work full-time in May 2008 (Marlys Popma, telephone interview, March 6, 2009). Joshua Lynch served as the point person for work with Catholics, while Jenny Sutton worked with the Jewish community. Popma came with extensive campaign experience, having worked on the McCain primary campaign in Iowa (her home state) and several other primary states. She also had leadership positions in the Iowa Right to Life organization, and had worked on other Republican campaigns. She is a well-known gure in


The Disappearing God Gap?

Christian Right Republican circles. A number of full-time interns rounded out the stafng for outreach to religious communities. The McCain campaign made a series of targeted mailings to religiously motivated voters and held periodic conference calls with key religious leaders. Other mass media outreach efforts aimed at social conservatives were seen as having a special appeal among religious conservatives. Popma later reected that the McCain campaign was one of the more difcult campaigns on which she had ever worked. This was because evangelicals were dividing into what she called movement conservatives, that is, leaders who had long been active in socially conservative organizations and campaigns, and young evangelicals, that is, evangelicals who wished to broaden the political agenda to include such issues as global warming and poverty and felt older evangelicals tone on such issues as abortion and same sex marriage was overly divisive and off-putting (Popma, telephone interview). She reported, for example, that she nally concluded she needed to have separate conference calls with these two groupstheir concerns about the other and mutual distrust were that great. On whether or not the McCain campaign put less emphasis on reaching out to religious voters than the 2004 Bush campaign had, Popma reported that she remains convinced that John McCain himself was fully committed to reaching out to religious voters, but that his top campaign leaders felt that they could not expand upon what Bush had achieved among conservative religious voters and that at the end of day they would vote for McCain in any case. Thus, the campaign gave lower priority to reaching religious voters (Popma, telephone interview). In summary, it is hard to compare the Obama and McCain efforts at reaching religious voters, since the Obama campaign was moving away from a Democratic tradition of largely ignoring religious communities as targets for campaign appeals, and the McCain campaign was following earlier Republican campaigns that had had extensive outreach efforts targeted at religious communities. Consequently, relative to earlier Democratic presidential campaigns Obama appeared to do a great deal, and relative to earlier Republican presidential campaigns McCain appeared to do much less. What is clear is that both campaigns in 2008 made major efforts to reach out to many voters with deeply felt beliefs and attachments to faith communities.

Religion and the Battleground States

Earlier in this chapter we examined the relationship between religion and major campaign themes. It is important, however, to recognize that general
Religion and the Fall Campaign


campaign themes and policy positions must be applied within the framework of more specic strategic and tactical plans. Given the diversity of the American population and the complexity of its state and regional political systems, adopting a one size ts all campaign plan may not be particularly useful to presidential candidates. One important feature of the American political system that affects campaign strategy in important respects is the Electoral College. In every state except Maine and Nebraska, all of a states electoral votes go to the candidate who receives a plurality of popular votes in that state. This practice, known as the Unit Rule, profoundly affects campaign strategy in presidential elections. First, the Unit Rule enhances the strategic importance of politically competitive (i.e., closely contested) states since a small change in popular votes in those states can result in all of the states electoral votes being allocated to a given candidate. Second, the Unit Rule enhances the importance of states with large numbers of electoral votes since it raises the stakes for winning or losing those states. If we put those two features together it is clear that those states that are politically competitive and relatively populous are unusually important in presidential campaigns, often making the difference between winning or losing a given presidential election. It is little wonder, then, that these statesoften referred to as swing or battleground statestypically receive a disproportionate share of attention in presidential campaigns. Identifying battleground states can be challenging.20 Nevertheless, in the summer of 2008, there was widespread consensus that the presidential election would be largely won or lost in eight populous and politically competitive states: Colorado, Florida, Virginia, Missouri, North Carolina, Ohio, Pennsylvania, and Michigan.21 Throughout much of the campaign, these eight states received most of the media and campaign attention, and so we focus on them here, particularly in terms of the way in which religion differed across these eight states. First, as can be seen in table 5.1, the religious afliations of citizens in the various states differ widely. While no religious tradition enjoys majority support in any of the battleground states, evangelical Protestants constitute particularly large pluralities of citizens in three of the eight states: North Carolina, Missouri, and Virginia. It is not surprising, then, that both Obama and McCain made particularly large efforts to attract evangelical votes in these states by emphasizing the moral issues that have traditionally motivated evangelical voters. In other states, including Pennsylvania, Ohio, Michigan, and Florida, mainline Protestants and Catholics constitute sizeable percentages of voters, and in these states the two presidential candidates more frequently emphasized their positions


The Disappearing God Gap?

table 5.1 Religious Afliation in Battleground States: 2008 (Percentage)

Evan. Prot. Colorado Florida Michigan Missouri North Carolina Ohio Pennsylvania Virginia 23 25 26 37 41 26 18 31 Main. Prot. 19 15 19 18 21 22 25 20 Black Prot. 2 8 8 6 13 7 7 10 Rom. Cath. 19 26 23 18 9 21 29 14 1 2 1 Jews 2 3 1 1

Other Relig. 10 7 6 4 4 6 6 6

No Afl. 25 16 17 16 12 17 13 18

Source: Pew Forum 2008.


Less than 1 percent.

on economic issues and on a broader array of social issues (Silk and Walsh 2008). Indeed, the McCain camp fervently hoped to win the support of blue-collar Catholics who may have been disenchanted with Obamas liberalism. And while Jews never exceeded 2 percent of the total population in any of the battleground states, both campaigns prior to the fall election considered the Jewish vote to be potentially important, particularly in Pennsylvania, Florida, and Ohio, which were thought to be closely contested (Hostein 2006; Jackson 2008; Kampeas 2008). As noted previously, economic concerns dominated much of the 2008 presidential campaign, as voters were profoundly worried about rising unemployment, the availability of credit, and the September 2008 collapse of the stock market. Thus, in their respective appeals to battleground state voters, both Obama and McCain emphasized economic issues (Abramowitz and Barnes 2008). This does not mean, however, that the values voters, concerned about issues of personal morality, disappeared from the political landscape. Indeed, both Obama and McCain sought to appeal to evangelical voters by emphasizing issues of personal morality as well as issues of social justice in their campaigns. For his part, Obama made every effort to redraw the map of religious voters, attracting moderate evangelicals, particularly those in key southern states such as North Carolina, Virginia, and Florida. Part of Obamas strategy involved attempting to counter the Democrats image as a largely secular party, hostile to religion (Hagerty 2008). To do this, Obama sprinkled his speeches with civil religious or even biblical language, including referring to America in Reaganesque fashion as a shining city on a hill and arguing that his campaign had a higher purpose (Cooper and Healy 2008). So condent
Religion and the Fall Campaign


table 5.2 Religiosity in Battleground States: 2008 (Percentage)

Church Attendance Weekly or More Colorado Florida Michigan Missouri North Carolina Ohio Pennsylvania Virginia 29 37 38 43 49 36 39 41 Occasionally 34 32 34 30 30 35 36 34 Seldom/ Never 37 31 28 27 20 28 26 25 Very Important 44 57 54 59 69 55 54 59 Religious Salience Somewhat Important 29 25 30 27 20 30 29 25 Not at All Important 26 18 17 14 11 16 17 16

Source: Pew Forum 2008.

was Obama that, toward the end of the campaign, he brought such themes directly to voters living in McCain strongholds within such key states as Florida and North Carolina (Drogin and Barabak 2008). In addition, Obama made direct appeals to young, religious voters in swing or battleground states, using Facebook and other websites (Riley 2008). Table 5.2 presents two measures of religiosity in battleground statesreligious service attendance and perceived importance of religion. Not surprisingly, the two measures of religiosity tend to be positively related to each other. What the table reveals, however, is that there is considerable variation in the levels of religiosity found among those living in these different battleground states. Topping the list was North Carolina, where respondents scored highest on both measures of religiosity. At the other end of the spectrum, only slightly more than a quarter of Colorado respondents reported attending religious services weekly or more and less than half indicated that their religion is very important to them. To the extent that religious and moral appeals mattered in the 2008 campaign, it is likely that they would have a greater impact in battleground states scoring relatively high in terms of religiosity (e.g., North Carolina, Missouri, and Virginia) than in battleground states scoring lower (e.g., Colorado, Pennsylvania, and Ohio). The nal vote in the 2008 election, reported (by percentages) in table 5.3, suggests that religion may indeed have played an important role, though not necessarily a decisive role, in the battleground states. While McCain made a respectable showing in virtually all of the battleground states, he won only one of them, Missouri (and that by a very narrow margin). On the other hand, it is also true that McCain tended to do best


The Disappearing God Gap?

table 5.3 Popular Vote in Battleground States: 2008 (Percentage)

Obama Colorado Florida Michigan Missouri North Carolina Ohio Pennsylvania Virginia
Source: Pew Forum 2008.

McCain 45.9 48.4 41.1 49.5 49.6 47.1 44.3 47.3

Other 1.6 0.7 1.6 0.2 0.6 1.6 4.1 0.9

52.5 50.9 57.3 49.3 49.8 51.3 51.5 51.8

in those states (e.g., North Carolina, Virginia, Missouri, and Florida) with relatively large percentages of both conservative evangelicals and highly religious voters. None of this is meant to downplay the key role of economic factors, particularly in such economically stressed states as Pennsylvania and Ohio. But it also suggests that religion may also have played an important role in battleground states in 2008.

Changing Campaign Fortunes

Presidential campaigns are also shaped and colored by the resources available to the candidates as well as by national and international events that largely reside outside the control of the candidates. While resource needs can be anticipated by candidates and campaign organizations, the same cannot be said for important events that are unique to the particular campaign. While candidates cannot control these events, they can control the way in which they choose to respond to themthereby highlighting important, yet differing, qualities of the two candidates. Such differing levels of resources and differing responses to national and international events also can play important roles in shaping the outcome of the presidential election.

Campaign Funding
Obamas campaign had far more nancial resources at its disposal than McCains. This inequity stemmed from Obamas ability to raise large sums of money through the Internet by means of small nancial contributions,22 and from Obamas decision to decline public nancing
Religion and the Fall Campaign


and the spending limits tied to itmaking Obama the rst major party candidate to reject public nancing during the general election campaign. Initially Obama had agreed to limit his campaign spending as long as McCain did the same; McCain did accept public funding and the spending limit of $84.1 million in public funds. But in June, Obama opted out of using the public funding system. As a result, McCains campaign had limited funds and was constrained in terms of the total amount of money it could spend on the campaignwith Obama having raised nearly twice as much money as McCain (Center for Responsive Politics 2009). Obamas campaign strategy was to utilize his massive campaign funding and political organization to expand the geographic area where the campaign was fought. With his far greater resources, Obama enjoyed a substantial advantage in terms of paid media ads and full-time campaign staff, permitting him to be more active in a larger number of states. More specically, the Obama campaign was able to target some traditionally Republican states, believing that changing demographics combined with the challenging political environment for Republicans nationwide gave Democrats new opportunities in states such as Indiana, North Carolina, and Virginia (Zeleny 2008e). McCain, however, had to adopt a different strategy. He had to target traditional battleground states the campaign staff initially believed they could win to achieve a narrow Electoral College majority. As a result, while Obama was able to expand his inuence away from states that traditionally voted Democratic, McCain was ghting in the nal weeks of the campaign to gain support in formerly uncontested Republican states such as Indiana and North Carolina (Nagourney and Zeleny 2008b). Obama outspent McCain in key battleground states such as Ohio, Pennsylvania, and Florida, and also in traditionally Republican territory such as North Carolina and Indiana, where he purchased television ads and made campaign appearances (Goldenberg and MacAskill 2008). In fact, already by the end of May, the Obama campaign staff was more than double the size of the 2004 Bush reelection campaign staff and nearly three times the size of McCains staff (Mooney 2008). By July, Obama and the Democratic Party had nearly two hundred paid staff working in Florida, with more on the way, and the campaign anticipated that there would be two hundred paid staffers working in each of the battleground states of Michigan, Pennsylvania, and Ohio (Mooney 2008). By the fall campaign, Obama enjoyed nearly a 4:1 ratio over McCain in terms of eld ofces in most states; even in Iowa, which hardly constituted a battleground state in 2008, Obama had more than forty eld ofces in operation.23


The Disappearing God Gap?

The Economic Crisis

As the economic crisis burst onto the American scene in mid-to-late September, Obama beneted from the tendency of voters to blame the party in the White House for bad economic news (The Politics of Despair 2008). In early September, McCain actually led Obama in public opinion polling due, in part, to the bounce he received on the heels of the GOP convention and Palins presence on the ticket, but this advantage quickly dissipated on the heels of Wall Streets meltdown, and McCain continued to trail in public opinion polls throughout the remainder of the campaign. Not only did the electorate report that the economy was the primary issue, but Obama enjoyed a two-to-one margin over McCain among those voters who cited the economy as the major issue (Halloran 2008). Moreover, McCains actions in the wake of the crisis also brought doubt to his candidacy by those who thought it exhibited erratic and impulsive behaviorundesirable presidential traits in a time of crisis. A week after he had stated that the fundamentals of our [U.S.] economy are strong (The Politics of Despair 2008), McCain suspended his campaign, pulled his television advertising, returned to Washington, and participated in the congressional negotiations over a stalled $700 billion government bailout plan for Wall Street.

The Palin Controversies

The perception that McCain was acting erratically may have had its roots in his earlier decision to choose Sarah Palin as his running mate. From the very beginning, Palins selection led some to wonder about the degree to which she had been properly vetted, as her apparent lack of knowledge on some topics created doubts about her ability to assume responsibility should McCain become incapacitatedas well as doubts about McCains wisdom in choosing her. Her faltering interview with Katie Couric on the heels of the GOP national convention set in motion this interpretation, and additional reports (e.g., that she was unable to name the three nations that are part of the North American Free Trade Agreement) led to Palin becoming a butt of jokes on the late night comedy shows, including impersonations by Tina Fey on Saturday Night Live.24 There were sufcient questions about Palins background and qualications to lead some noted conservative opinion shapers to question her candidacy, or to use their questions about Palin to question McCains candidacy and nd reasons to vote with the opposition (e.g., Kathleen Parker and Christopher Buckley).
Religion and the Fall Campaign


In the end, McCains choice of Sarah Palin as his running mate may have helped energize support for his candidacy among his base of social conservatives, but she may have alienated many of the independent and swing voters he needed most.

Presidential campaigns try to set issue agendas to the advantage of their candidate. Much planning, organization, and human and nancial resources are collected in those efforts, and deployed in the frantic nal campaign stage that is bookended by the national conventions and Election Day. At the same time, campaigns know that unanticipated and uncontrollable events can enter into the campaign season at any time, disrupting plans and testing their adaptability and resourcefulness, particularly in the ten nal weeks. Candidates and campaigns must respond to those events, and further, to how the media interprets those events and the campaigns own reactions to them. In the 2008 election, the larger and better-funded Obama organization was far better situated at the start of the postconvention stage, and responded better to unforeseen events, particularly the new economic crisis as it unfolded in late September. Their own assets, aided by a generally sympathetic media, enabled them at the outset to better dene the terms of the election, and, in the last weeks of the campaign, to better respond to and shape perceptions as events unfolded. There are several specic examples. Relatively underfunded and understaffed, McCain needed direct campaign help from religious conservatives and their organizations. According to many observers, that need led to him seriously considering and eventually choosing Sarah Palin as his running mate. The surprising choice also accentuated his maverick reputation and, had he been able to link the impulsive Palin pick to his prior maverick decision to advocate an Iraq surge, which was turning out reasonably well, the campaign might have been successful in portraying McCain as the kind of leader t for the times. But true to fashion, McCain reacted in a similar manner to the dramatic events of the Wall Street meltdown and economic anxiety. Maverick started to become impulsive and risky, weakening McCain in the domain of economic policy, already his weakest policy area. And the style of his reaction to economic events began to cast his Palin choice in a different lightas perhaps another example of too great a risk and too poorly thought out. This affected all aspects of the campaign, including the religious angle. The media


The Disappearing God Gap?

presentation of Palins faith, combined with McCains obvious dependence on the older, more conservative, and traditionalist religious groups and leaders, made him look too dependent on a set of supporters who by some measures were declining in their political inuence. Obamas campaign, on the other hand, because of its greater nancial and human resources, could rely on superior nancing and stafng to resist for at least a few days many of the twists and turns of the campaign. His caution appeared measured, thoughtful, and mature, particularly in contrast to McCains approach in the economic are-up. In regard to religion, Obamas mainline ecumenism and self-built organization offered no controversy, and just enough credibility to gain slightly larger slices of religious voters. As is typical in presidential elections, both organization and events mattered. Obama was better organized and funded, and thus could afford to be conservative in campaign tactics and timing. McCain, underfunded and lightly staffed, needed to latch on to external movements and important new events to gain support and build momentum. By being dependent upon such externalities, for McCain to win every choice had to be perceived as courageous, and every external event had to be harnessed to his advantage. His was a high-risk strategy that did what it was likely to dofail.

Religion and the Fall Campaign



Religion and Election Day: Voter Mobilization in 2008

lection day brings to a close the presidential campaign. Months, if not years, of work on the part of the candidate and the candidates staff now hinge on the decisions of those who choose to go to the polls to cast their ballots. Religion affects the outcome of Election Day in two different ways: (1) in terms of who comes, and who does not come, to the polls, and (2) in terms of voting choices. With regard to the former, considerable efforts are made by candidate organizations, party organizations, and interest groups to excite, activate, and even physically bring voters to the polls. Varying rates of turnout among different segments of the electorate can alter election outcomes. With regard to the latter, only the internal decision-making process of the voter has the nal say for which candidate ones vote is cast. Seeking to get religious voters to the polls is hardly a new phenomenon in American elections.1 Given the prevalence, diversity, and size of such groups in American society, it is not at all surprising that campaign organizations view religious groups among the building blocs of their coalition of voters. Of course, presidential candidates are not alone in having a stake in getting congenial religious voters to the polls. So do political parties and interest groups, as they nd it useful to mobilize key religious communities to further their political purposes. And at times, so do religious communities, as they seek to achieve their policy goals by working for candidates, joining political parties and interest groups, and forming their own political organizations. Like other kinds of social groups, religious groups often seek to activate their own members politically.

As we noted in the introduction, political elites employ different means to engage, persuade, and activate voters. One such endeavor is to mobilize voters to go to the polls, an endeavor that may involve a variety of different strategies. But regardless of the strategy employed, the goal is to shape the composition of those who decide to come to the polls on Election Day. This chapter examines the relationship between religion and political mobilization in the 2008 campaign, particularly religiously based mobilization.2 Religion was a critical factor in the political mobilization of voters in the 2008 election, and in surprising ways. As we shall see in the next chapter, the overall relationship between various religious traditions and partisan voting patterns did not change in any dramatic way during this election cycle. Rather, what did change were patterns of mobilization and the emergence of new discourses and networks invested in religiously based mobilization that were broadly supportive of Democratic, rather than Republican, positions. As a result, what was different in 2008 was that the McCain campaign organization, the Republican Party, and conservative religious organizations did not mobilize their base of religious voters as successfully as did the Obama campaign organization, the Democratic Party, and more liberal religious organizations. Whether this is the start of a new trend, or simply a relatively unique occurrence, remains unclear.

Political Mobilization
In their efforts to successfully capture the presidency, political parties, interest groups, and even concerned citizens often hope to win elections, in part, by shaping the actual composition of the voting public. And because members of particular religious groups often exhibit relatively similar perspectives and belong to relatively cohesive local organizations (e.g., congregations) that lend themselves to mobilization, political campaigns frequently attempt to activate and mobilize different segments of religious groups. The logic of targeting particular social groups is based on a long-standing recognition that elections are fundamentally a group activity (Lazarsfeld, Berelson, and Gaudet 1944; Berelson, Lazarsfeld, and McPhee 1954; Stanley, Bianco, and Niemi 1986; Stanley and Niemi 1999). Given that the propensity for party coalitions to represent group conicts is a durable feature of American politics (Leege et al. 2002, 7), campaign strategists and candidates employ messages that relate to group politics, in
Religion and Election Day: Voter Mobilization in 2008


which aspects of group conict, perceptions, prejudices, and fears are put into play (Wielhouwer 2009). Central to most political mobilization efforts is personal contact with members of the electorate. Research has long shown that personal contacts are a highly effective means to elicit participation, and, as a result, political campaigns frequently use face-to-face as well as telephone contacts to persuade and mobilize voters (e.g., Gosnell 1926, 1927; Eldersveld and Dodge 1954; Eldersveld 1956; Huckfeldt and Sprague 1992; Wielhouwer and Lockerbie 1994; Gerber and Green 2000a, 2000b, 2001; McClurg 2004). Moreover, those contacted personally are signicantly more likely to become politically active, even after controlling for other factors related to political participation (Wielhouwer 2009, 413); they are more likely to vote, to campaign on behalf of a candidate, and to contribute nancially to campaigns (Rosenstone and Hansen 1993; Wielhouwer and Lockerbie 1994). However, the effectiveness of such contacts varies, in part, by the method employed, the locale in which it occurs, and the partisan nature of the election (see, e.g., Green and Gerber 2004, 2005). Overall, however, the people most likely to be contacted are those already predisposed to participate (namely, those who have high socioeconomic status, are registered to vote, and are socially connected to their communities) and to vote for the partys candidates (Wielhouwer 2009, 411). Parties in particular want to focus their efforts on individuals who are likely to support the party and its candidates, while seeking to avoid mobilizing individuals who support their opponents (Kramer 1970; Wielhouwer 2000). Thus, Democrats typically contact groups likely to vote Democratic, while Republicans typically focus on groups inclined to vote Republican (Rosenstone and Hansen 1993; Gershtenson 2003; Wielhouwer 2003; Panagopolous and Wielhouwer 2008). Research on the mobilization of religious voters has focused on various sources of contact. A number of studies have examined religious sources of mobilization effortsparticularly in terms of efforts aimed at religious conservativesincluding those by religious or moral-concerns interest groups and by clergy and friends at ones house of worship (Wilcox and Sigelman 2001; Guth et al. 1998, 2002, 2007). These studies have primarily analyzed religiously based means of contacts, including the use of voter guides and the urging of clergy to register and vote, revealing that efforts to mobilize religious voters over the past several elections have been relatively successful, especially when they involved direct contact with voters by religiously oriented interest groups or informal discussion about politics with others at their houses of worship.


The Disappearing God Gap?

Other research has focused more on the role of political parties in mobilizing religious voters (e.g., Layman 2001; Leege et al. 2002: Monson and Oliphant 2007). At times, this research has examined more indirect means of voter mobilization than voter contact. For example, some research indicates that the mobilization of religious conservatives into the Republican coalition has been largely the result of changes in the composition of party elites and represented a mass-level response to the changing conict (e.g., between religious conservatives and secularists or between economic and social conservatives) that has occurred among party elites and activists (e.g., Layman 2001), while some claim that the growing proportion of religious conservatives in the Republican coalition can be attributed to the manipulation of symbols of cultural threat by the Republican Party to motivate evangelicals to political participation (Leege et al. 2002). Still, there is some research on the mobilization of religious voters through reported contacts by political parties (e.g., Monson and Oliphant 2007; Wielhouwer 2009). Since 1956, the American National Election Studies have asked questions related to party contacts, with the 2000 and 2004 elections revealing the highest levels of personal contacts by political parties, with nearly 45 percent receiving some contact from one or both major political parties in the 2000 and 2004 elections (Wielhouwer 2009, 411). Not surprisingly, given the desire to mobilize only those voters most likely to support a partys candidates, those associated with certain religious traditions are more likely to report contacts with one political party than the other (see, for example, table 14.2 in Wielhouwer [2009]). Despite this long-standing recognition of the importance of personal contacting, the presidential election campaign in 2004 nevertheless witnessed a return to more traditional methods of voter mobilization specically that of personal contacting to register and turn out voters (Conway 2005). This renewed emphasis was tied, in part, to a reduced emphasis on persuading swing voters in favor of maximizing turnout of the partys base (Fiorina 1999). With this shift in emphasis came signicant changes in voter mobilization methods. For example, the Republican National Committee inaugurated a plan to employ more personalized methods of contacting, one component of which was the careful targeting of potential supporters based on political proling. In this political proling, the electorate was divided into dozens of microtargeting segments, which dictated the kinds of messages the voter was to receive (Sosnick, Dowd, and Fournier 2006; Monson and Oliphant 2007). Christian radio stations, talk radio programs, direct mailings of printed materials, videos, and DVDs enabled candidates to target specic messages to more narrow,
Religion and Election Day: Voter Mobilization in 2008


and conned, audiences rather than softening or distilling such appeals so as to be acceptable to broader audiences. Such narrowcasting of campaign messages is designed to target those individuals most receptive to the particular message transmitted and constituted a more cost-effective means of political persuasion (Conway 2005, 60). The successful application of these political mobilization techniques in the 2004 election enabled the Republican Party both to expand its base and to capture the votes of swing voters in the electionthereby contributing to the reelection of George Bush (Conway 2005, 60). In 2008, the Obama campaign was committed to mobilizing new voters, particularly the young. However, paying people to register voters is generally too cost prohibitive, as it is generally believed that such dollars can be expended more fruitfully on advertising, mailings, or telephone calls to registered voters.3 But if a campaign can mount a voter registration drive of sufcient scale in which volunteers do the bulk of the work, then such a drive may be highly benecial to the campaign. Perhaps due to the candidates background as a community organizer, the Obama campaign not only drew volunteers by the thousands throughout the presidential primaries but also trained such volunteers to become de facto organizers through its Vote for Change registration effort (Hayes 2008). The question then is whether the development and renement of these particular efforts were associated with any changes in the mobilization of religious voters in the 2008 election. Were different segments of religious voters contacted in 2008 when compared to 2004? And were there any signicant changes in the sources of such contacts?

Religious Constituencies and Political Contacts in the 2008 Election

When examining efforts to mobilize religious voters politically, there are three different, but parallel, mobilization efforts that can be made to get religious voters to the polls on Election Day (Green 2007, ch. 7). First, the campaign organizations of the two major candidates, along with their associated national party committees, operate major voter registration and get-out-the-vote efforts that target religious communities. Second, there are a host of religiously based organizations, some linked to the Christian Right and others to the Religious Left, that work to get voters to the polls on behalf of particular candidates or issues. And nally, there are efforts on the part of clergy and other members of local congregations to mobilize their fellow congregants.


The Disappearing God Gap?

Sources of Voter Contacts

Though it is not unusual for presidential campaigns to work to mobilize religious voters, there was much discussion about the concerted effort of the Bush campaign to do so in the 2004 presidential election, as Karl Rove, Bushs campaign chief strategist, sought to increase turnout among evangelical Protestants. However, Roves effort was not limited to evangelicals; it extended to other religious traditions as wellincluding Catholics, Jews, black and Latino Protestantsand to weekly worship attenders more generally (Green 2007, 142). These efforts included the targeting of religious voters through direct mail, telephone calls, e-mails, and personal contacts. And though the Democratic Party in 2004 experienced various problems related to its religious outreach programs, including unsuccessful efforts to appoint a religious liaison at the DNC and a short-lived voter guide project, it nevertheless aggressively sought to mobilize key religious constituencies, particularly black Protestants, Latino Catholics, Jews, and other religious minorities (Green 2007, 143). In 2008, however, the McCain campaign largely sought to mobilize religious groups most likely to support the Republican nominee through mobilizing social conservatives, not by targeting certain specic religious groups as such. The Obama campaign, on the other hand, sought to engage and mobilize religious voters in part through Faith and Values tours during the fall campaign, and it, too, sought to mobilize those religious voters most likely to support its candidate. Both campaigns, however, generally viewed mainline Protestants and Catholics as being crucial to the success of their campaigns, as members of both religious traditions have largely split their ballot between the Democratic and Republican nominees across the past several elections (Smidt et al. 2005; Green et al. 2007). In order to assess these efforts at voter mobilization, the postelection component of the Henry Institute National Survey on Religion and Public Life asked respondents whether or not they had been contacted during the campaign by a variety of possible sources of contacts. The series included contacts by political candidates and by political parties (collectively these contacts will be described as partisan contacts). However, party and candidate organizations are not alone in making efforts to contact voters; a host of other organizations not specically tied to candidates or parties were also engaged in campaign activities and voter mobilization efforts in the 2008 elections (e.g., organized labor, environmental groups, business groups, education groups, medical groups). Many of these organizations sought to collect and dispense relatively large sums of money in

Religion and Election Day: Voter Mobilization in 2008


efforts to identify, register, and stimulate voting turnout among carefully targeted segments of the electorate. So survey respondents were also asked whether they had been contacted during the campaign by a number of different kinds of interest groups; this series included contacts by a conservative issue group, a liberal issue group, a moral or religious group,4 a labor union, a business group, or an environmental group. And nally, houses of worship can serve as sites in which those who attend regularly may be mobilized to become engaged politically. It is not unusual for local houses of worship to sponsor voter registration drives or for pastors to encourage their members to participate in the political process. Worship attendance facilitates political contacting, as those who worship regularly are more prone to engage in informal political discussion with friends, more likely to hear political cues from clergy, or more likely to pick up voting guides that are left on church tables. Consequently, our survey also sought to ascertain to what extent local congregations may have served as sites in which their members were contacted in some fashion to participate in the presidential election. These various potential sources of voter contact are examined in table 6.1 according to the religious tradition with which the respondent was afliated. In addition, comparable data from the 2004 presidential campaign are also presented in order to discern whether there were any important differences that occurred in political contacting among such religious groups over the past two presidential elections. Partisan Sources. Members of different religious traditions clearly received different levels of candidate and party contacts during the 2008 presidential elections. Contacts made by political parties were more common than contacts made by specic candidate organizations, and overall more than three-fths of the American public reported that had received at least one contact by either a political party or a political candidate during the course of the fall election. Overall, Jews were the most likely to report some contact by either a political candidate or a political party, while black Protestants were the least likely to report any contacts from such sources. Mainline Protestants were slightly more likely to report partisan contacts than evangelical Protestants, the religiously unafliated, or Catholics. Table 6.1 also provides comparable data for the 2004 presidential election. Given the renewed recognition of the importance of personal contacts, it is probably not too surprising that respondents reported higher levels of partisan contact in 2008 than in 2004. Rather, what is surprising is the level of growth in such contacts: just a little a little more than onehalf of Americans in the 2004 presidential election had reported some


The Disappearing God Gap?

table 6.1 Source of Voter Contact by Religious Tradition: 2004 and 2008
Evan. Prot. (%) 2008 Partisan contacts Candidate Party At least one Interest group contacts Conservative Liberal Religious

Main. Prot. (%)

Black Prot. (%)

Rom. Cath. (%) Jews (%)

Unafliated (%) All (%)

45 47 62 22 9 18 40 26 11 28 42 80

45 57 66 17 14 16 40 13 4 21 30 79

26 48 56 10 6 10 25 30 19 38 46 71

35 50 57 15 12 17 37 17 7 15 37 79

50 67 73 29 36 20 67 20 14 21 29 86

40 44 59 10 10 9 27 13 9 14 26 70

40 50 61 17 11 15 36 19 9 21 34 75

At least one b Congregational contacts Clergy Urge Voter Guides Discuss with friends At least one Overall at least one b 2004 Partisan contacts Candidate Party At least one Interest group contacts Conservative Liberal Religious a At least one b Congregational contacts Clergy urge Voter guides Discuss with friends At least one Overall at least one b

36 41 52 6 5 22 32 43 10 35 60 78

37 45 54 3 8 12 31 27 5 23 39 73

18 29 39 4 6 19 37 49 21 29 64 75

31 44 52 3 6 15 35 36 13 19 48 77

46 63 71 4 27 25 50 39 2 39 51 90

34 38 49 3 10 8 33 6 3 7 11 59

34 41 51 4 7 15 33 32 9 23 44 73

Sources: 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

Includes both moral and religious groups; bIncludes unions, environmental, and business groups.

Religion and Election Day: Voter Mobilization in 2008


such contact, while more than three-fths did in the 2008 presidential election. However, patterns of contacts among members of different religious traditions remained virtually the same across the two elections. Just as was true with regard to the 2008 presidential election campaign, Jews were the most likely to report some contact by either a political candidate or a political party in the 2004 contest, while black Protestants were the least likely to report any contacts from such sources. Mainline Protestants, as would be true in 2008, slightly led evangelical Protestants and Catholics in partisan contacts, with the religiously unafliated trailing slightly behind. Organizational Sources. In the 2004 election Bush received substantial support from groups linked to the Christian Right. Organizations such as Focus on the Family, the Family Research Council, the Traditional Values Coalition, and Concerned Women for America sought to energize their members, in part, by distributing voter guides. Conservative Catholic organizations, such as Catholic League, Priests for Life, and Catholic Answers, were also activeas well as other new organizations that emerged to oppose same sex marriage, an issue that burst on to the political agenda in 2004. And though the Christian Coalition at the national level was already in disarray by 2004, some of its state and local organizations were still relatively active. Compared to the Religious Right, the Religious Left was not nearly as organized in the 2004 campaign (Kellstedt, Smidt, Green, and Guth 2007). Certainly efforts had been made in the 1990s to form national organizations of liberal religious activists, often primarily as an effort to counter the electoral efforts of the Christian Right. Probably the best known of these organizations were the Interfaith Alliance and Call to Renewal. Liberal Catholic organizations were also present, including Pax Christi, Catholics for Political Responsibility, Catholics for the Common Good, and Catholics for Free Choice. Still, despite the presence of such groups, organizations on the Religious Left have not exerted any major inuence on American electoral politics over the past several decades. However, following the 2004 election, considerable efforts were made both by the Democratic Party as well as by various leaders and authors outside the party to decouple two particular ideas about the relationship between religion and politics: rst that the Republican Party was the only party that advocated policies that reected religious values and, second, that the Democratic Party was a party of secularists. We have discussed some of the efforts made by the Democratic Party in previous chapters. However, other efforts were made by groups that worked to bring the


The Disappearing God Gap?

voice of faith to progressive public policy issues, on the one hand, and on the other by partisan groups on the Democratic side that sought to show people of faith that there was space for them within the Democratic Party. These endeavors sought either to weaken the hold of the Republican Party among religious members of the electorate or to mobilize certain kinds of religious members of the electorate to show up at the polls on Election Day and cast their ballot for candidates of the Democratic Party. In fact, a substantial number of books were released in early 2008 that sought to accomplish these particular objectives. These books, of course, had been in the works for some time, prepared by their writers, agents, and publishers in full anticipation of the 2008 campaign year. These books included Catholic E. J. Dionnes Souled Out (January 2008), evangelical Protestant Tony Campolos Red Letter Christians: A Citizens Guide to Faith and Politics (February 2008), and mainline Protestant Amy Sullivans The Party Faithful (February 2008). These books reect an important debate that was occurring among many Americans regarding the religious legitimacy of certain political options. Much of this debate was led by Catholic intellectuals or was addressed toward Catholics. For example, in June 2008, Michael Sean Winterss book Left at the Altar: How the Democrats Lost the Catholics and How the Catholics Can Save the Democrats was released, followed by Robert P. Joness Progressive and Religious (August 2008). But within conservative Christian circles, Douglas Kmiecs book Can a Catholic Support Him? Asking the Big Questions about Barack Obama (September 2008) was perhaps the most notable. Kmiec was a former Reagan Democrat who had served in the Reagan administration and was widely known for his passionate opposition to abortion, and his support of Obama launched considerable media discussion that even led on one occasion to him being denied communion.5 However, this debate was not limited to Catholics. Part of this new and growing conversation was occurring among evangelicals, one that seemed for many commentators to have the marks of a generational shift. A new tone was being set by Campolos book as well as by best-sellers such as Brian McLarens Everything Must Change: Jesus, Global Crises, and a Revolution of Hope (October 2007), David Gushees The Future of Faith in American Politics: The Public Witness of the Evangelical Center (January 2008), and Ron Siders The Scandal of Evangelical Politics: Why Are Christians Missing the Chance to Really Change the World? (February 2008). Evangelical writers produced many texts besides these, and some were notable in their attacks on the Religious Right per se (e.g., Boyd 2007; Balmer 2007).6
Religion and Election Day: Voter Mobilization in 2008


These books served to contribute to the life and vibrancy of various organizations that had been built since the 2004 election, as they lent legitimacy not only to the new organizing, but in some cases to the organizers themselves.7 Given the linkages involved, the most appropriate way to classify these various groups is not always clear, but they tended to be either organizations composed largely of people from a particular faith tradition who sought to advance the connection of their faith tradition to the policies of the Democratic Party or a group of Democratic partisans belonging to a variety of different faith traditions who sought to make room for religious people within the party. Examples of such newly formed groups included Faith in Public Life, the Matthew 25 Network, the Church Ladies Project, Faithful Democrats, and Catholic groups such as Catholic Democrats, Catholics United, and Catholics in Alliance for the Common Good.8 These organizations generally had a relatively young staff and leadership as well as sophisticated, web-savvy organizers. Whether their roots were within the evangelical, mainline, or Catholic traditions, or even an emerging space at the margins of those traditions, they viewed themselves as building a network of religious progressives, and the sum total of these developments provided a very different sort of context for religious framing and mobilization during the campaign. Given these developments, one might anticipate that there would be far greater numbers of political contacts made by religious organizations in the 2008 campaign than the 2004 campaignor at least greater numbers of such contacts made by religious organizations that might be viewed as being more on the left side of the political spectrum. Consequently, we examine in table 6.1 responses to possible sources of interest group contacts in terms of the religious afliations of the respondents. As is evident from the table, a little more than one in three potential voters reported some contact from an interest group during the 2008 presidential campaign, slightly higher than the one-third that did so in 2004. Jews, Catholics, and evangelical Protestants were the most likely to report at least one such contact during the 2008 campaign, with two-thirds of all Jews reporting at least one contact by an interest group, while only two-fths of all evangelical Protestants and mainline Protestants did so, with Catholics trailing closely behind. On the other hand, black Protestants fell substantially below the national level in terms of reporting any interest group contact. With regard to the nature of these interest group contacts, members of most religious traditions reported higher levels of contact by conservative than by liberal organizations in the 2008 presidential election campaign, with Jews and the religiously unafliated being the exception to


The Disappearing God Gap?

this pattern. Likewise, most such constituencies reported higher levels of contacts by religious interest groups than by liberal interest groups, though once again this was not the case for Jews and the religiously unafliated. Overall, voter contacts by interest groups were up only slightly between the 2004 and 2008 presidential electionstrailing substantially the increased levels of contacts posted by candidates and political parties. Perhaps the most striking difference between the 2004 and 2008 elections was the apparent increase in reports related to contacts from both conservative and liberal issue groups. It is unclear, however, to what extent this apparent change is a methodological artifact of a slightly different wording employed in the two surveys. In 2004, respondents were asked whether any liberal issue group, like Americans Coming Together had contacted them, whereas in 2008 respondents where simply asked whether any liberal issue group had contacted them. Similarly, in 2004, respondents were asked whether any conservative issue group, like the Club for Growth had contacted them, whereas in 2008 they were simply asked whether any conservative issue group had contacted them. But even assuming that some of this increase may be a function of question wording, it is likely, given the magnitude of the increase in reported contacts, that some of the change was real. Nevertheless, regardless of whether there were increased levels of contacts made by conservative and liberal issue groups, there was no substantial overall increase in reported contacts by religious groups between the 2004 and 2008 elections. Whereas religious interest group contacts more than doubled between 2000 and 2004 (Guth et al. 2007), such an increase was not evident between the 2004 and the 2008 elections. Thus, despite the apparent increase in electoral activity on the part of the Religious Left in the 2008 election, the data reveal a leveling off of contacts by religious interest groups across the past two elections, suggesting either a decline in the level of contacts on the part of religious organizations linked to the Christian Right or the lack of on the ground contact efforts on the part of the newly formed progressive religious interest groups.9 Congregational Sources. Finally, table 6.1 presents reported forms of contacts made within congregational settings. Overall, only a minority of the general public received any single form of religious communication from their house of worship. For example, only about one in ve reported that they had discussed the election with a friend in church, about the same proportion indicated that they had heard their clergy or other religious leader urge them to vote, and fewer than one in ten found a voter guide in church.
Religion and Election Day: Voter Mobilization in 2008


Political discussion in church was the most commonly reported congregational contact, but those afliated with different religious traditions varied considerably in terms of such reported activity. Given Obamas candidacy, it is not surprising that black Protestants led the way with more than a third reporting having discussed the election at church with their friends. Evangelical Protestants also reported a relatively high level of election discussion at church, while Catholics were the least prone to discuss politics with friends at their house of worship. While ministers, priests, rabbis, and other religious leaders may often admonish their parishioners to register and vote, the extent to which they do so varies considerably by religious tradition. Black Protestants are the most likely to report such admonitions, with evangelical Protestants trailing somewhat behind. More surprising, however, is the relatively low level of such reports among mainline Protestants. Historically, mainline Protestantism has been more affected by, and associated with, the social gospel (a theological perspective that stresses structural reform, not just spiritual conversion and personal sanctication, as a means for social change and helping the disadvantaged). At its core, the social gospel implies a need for pastors to become politically involved in order to assist the less fortunate (Olson 2000, 49), and such clergy have sought historically to encourage their parishioners to be politically engaged (Guth et al. 1997; Smidt 2004). Consequently, such low levels of urging on the part of mainline Protestant clergy are surprising. Press accounts have frequently highlighted the use of voter guides to communicate with religious voters. Such guides report candidates past votes or current stances on issues of special concern to those publishing the guides. These voting guides have been found in elections over the last three or four decades, having been deployed by pro-life groups at least since the 1970s. One of the contributions of these guides to political mobilization is the legitimacy they provide for the voters candidate choices. Though voter guides are most frequently linked to conservative religious movements such as the Christian Coalition of America and the Family Research Council, which titled its voter guide the Christian Voter Guide,10 they are not limited to conservative causes. Among Catholics, voter guides have been a contentious issuelargely for ecclesiological reasons.11 Nonetheless, the 2008 election marked the eighth presidential election cycle in which the United States Conference of Catholic Bishops released a document called Faithful Citizenship (Carr 2008),12 a document that is not a voting guide but a statement explaining the Catholic position on election year issues. There are some voter guides geared toward Catholics, ones that tend to take conservative stances based largely on the abortion issue.13


The Disappearing God Gap?

Less than one in ten reported nding a voting guide at church. Once again, black Protestants were the most likely to report the presence of such guides in their houses of worship, while mainline Protestants were the least likely to do so. And while press accounts have frequently focused on voter guides within evangelical Protestant houses of worship, evangelicals trailed considerably behind black Protestants in this regard. Finally, even with the presence of the U.S. Conference of Catholic Bishops document and other published voter guides for Catholics, relatively few Catholics reported voter guides in their houses of worship. Finally, it should be noted that there was a noticeable decline in some church-related activities between the 2004 and 2008 elections.14 Specically, there was a substantial drop in clergy urging their parishioners to register and vote, as nearly one-third of respondents reported hearing such encouragements in 2004 while less than one-fth did so in 2008.15 Particularly noteworthy is the fact that this decline was not limited to any particular religious traditions, as evangelical Protestants, mainline Protestants, black Protestants, Catholics, and even Jews all reported fewer encouragements by their religious leaders to register and vote in 2008 than they did in 2004. Whether this was a function of the growing wariness of continued religious involvement in politics on the part of clergy, a function of an election that appeared to be less closely contested than the previous two elections, or a function of some other factors is unclear. But whatever its sources, the net decline in such urgings contributed to a decline in the overall percentage who reported at least one type of contact related to the presidential election within their congregational setting, as the overall percentage dropped by ten points in 2008. What can we conclude, then, about religious activity in 2008? Clergy encouragement of voting and political discussion in church continued at a fairly high level, though at a somewhat lower level than in 2004. However, such activity varied by tradition, with black Protestants and evangelical Protestants leading the way. Voter guides were as evident as they were in 2004, but less so than in 2000, especially among evangelicals. Overall, then, religious contacting at the congregational level continued to be relatively extensive, even if less so than previously, as levels of contact were virtually equivalent to those of interest groups. However, if the levels of contacting associated with moral and religious organizations were shifted from the analysis of interest group contacts and placed within a broader religious category that included both the activities of moral and religious interest groups and that of houses of worship, then the overall activity of religious groups in making voter contacts would far exceed the reach reported by the remaining interest groups.
Religion and Election Day: Voter Mobilization in 2008


Partisan Bias in Voter Contacts

While almost all religious groups reported increased levels of contacts by political parties in the 2008 presidential election, it still is unclear whether there were any important changes in terms of which political parties were reported to have made these contacts with these various constituencies. Hence table 6.2 examines all such contacts in terms of which political party or its proponents were reported to have made them. More than one-quarter of the general public in the 2008 election reported contacts by Democratic candidates and a similar proportion contacts by Republican candidates, while a little more than one-third reported contacts by each of the two major parties. Overall, therefore, neither party exhibited a major advantage in terms of the extent to which such contacts were made. And though reports of contacts were higher in 2008 than in 2004, there continued to be rough parity between Democratic and Republican contacts in 2008. Differences begin to emerge, however, when one examines the particular religious groups who reported such contacts. Not surprisingly, Democratic candidates and the Democratic Party were more likely to contact members of some religious groups than other religious groups, and this was also true for Republican candidates and the Republican Party. For example, Republicans were more likely than Democrats to contact evangelical Protestants, while the pattern was reversed for Jews and the religiously unafliated. And Democrats and Republicans contacted mainline Protestants and Catholics at about relatively equivalent levels in 2008, just as they had done in 2004. However, when comparing partisan contacts in 2004 and 2008, one would anticipate that those particular religious groups viewed as most crucial to the McCain campaign would likely report increased contacts from Republican Party sources, while those specic religious groups most crucial to the Obama campaign would likely report increased contacts from Democratic Party sources. And the data presented in table 6.2 reect these expectations, as traditionalists within the three major Christian religious traditions reported substantial increases in Republican Party contacts between 2004 and 2008, with traditionalists among mainline Protestants and non-Hispanic Catholics reporting an increase of over 10 percent in such GOP contacts across the two elections. Moreover, while traditionalists among evangelical Protestants reported a modest increase in Democratic Party contacts between 2004 and 2008, the increased level of Republican Party contacts among traditionalist mainliners and Catholics was coupled with virtually no increased levels of contacts from the


The Disappearing God Gap?

table 6.2 Partisan Direction of Campaign Contacts by Religious Tradition

2004 Candidate Religious Tradition General public Evangelical Prot. Traditionalist Centrist Modernist Mainline Protestants Traditionalist Centrist Modernist Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Cath. Traditionalist Centrist Modernist Jews Other religions Unafliated 23 23 23 25 17 25 19 26 27 18 15 9 25 26 26 21 44 25 25 Dem (%) 24 31 35 26 23 29 28 28 33 14 8 11 25 30 24 23 16 25 22 29 23 25 21 17 33 32 29 41 21 24 22 32 29 35 27 48 25 29 Rep Party Dem (%) 28 35 41 25 29 34 34 35 34 17 15 16 34 41 35 28 29 25 23 28 29 30 26 32 36 36 35 38 36 21 22 31 35 29 30 50 31 25 Rep Candidate Dem (%) 27 34 39 26 36 35 39 34 35 24 12 24 30 47 25 25 29 37 25 34 32 32 33 29 39 31 45 36 38 24 24 37 31 33 49 64 48 36 Rep 2008 Party Dem (%) 34 40 46 34 23 38 46 40 28 44 33 8 41 54 38 36 36 51 26 Rep

Sources: 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

Democratic Party. Conversely, a substantially larger percentage of centrist and modernist evangelical Protestants, centrist mainline Protestants, and modernist Catholics reported Democratic Party contacts in 2008 than reported them in 2004. Here again the pattern reects the political landscape of the 2008 election, as these are among those segments of the three major religious traditions in which Obama would likely have the greatest opportunity to capture some religious votes and win at the margins within these larger religious traditions. Hispanic Protestants were important to Bushs reelection in 2004, and the Republican Party and the McCain campaign hoped to retain their support. But the illegal immigration issue had the potential to move sizable segments of Hispanic Protestants into the Democratic column in 2008, as many Republicans had taken what most Hispanics saw as an
Religion and Election Day: Voter Mobilization in 2008


anti-immigration stance. Consequently, it is not too surprising that Hispanic Protestants in 2008 reported substantially greater contacts from both parties than what they reported in 2004. More surprisingly, however, black Protestants did not report any greater number of Democratic Party contacts in 2008 than in 2004, though they reported substantially more GOP contacts. Perhaps this lack of increased Democratic Party contact was simply a function of the realization that African Americans were likely to go to the polls anyway to support the rst African American to capture the nomination of a major political party. Hispanic Catholics reported a slightly greater level of contacts from the Democratic Party in 2008 than in 2004, but a markedly lower level of GOP contacts. Meanwhile, Jews and the religiously unafliated received increased levels of contacts from both major parties in 2008, but with Democratic contacts increasing at much greater levels than Republican. However, the data presented in table 6.2 do not reect the partisan bias of all possible contacts madethat is, when one takes into account contacts not just by candidates and political parties but also by interest groups or friends at ones house of worship on behalf of a particular presidential candidate as well. Hence, table 6.3 examines all reported contacts on behalf of Obama and McCain. First, it is clear from table 6.3 that many voters reported contacts from both sides of the political divide; in fact, more than twice as many voters reported contacts from both parties as from only one. Thus, overall, neither candidate held any major advantage in terms of the extent to which such contacts were made with voters overall, as there was parity between reported Democratic and Republican contacts in the 2008, as well as the 2004, presidential election. Second, the net Republican advantage in contacts (in which Democratic-only contacts are subtracted from Republican-only contacts) shifted in some important ways from 2004 to 2008. Republicans held a much smaller net advantage among evangelical Protestants in 2008 than they did in 2004, as a larger number of evangelicals reported contacts from both political parties in 2008 than reported them in 2004. This may reect the greater outreach to religious voters noted earlier by the Obama campaign and Democratic-aligned advocacy groups. As mentioned earlier, mainline Protestants and Catholics were viewed by both the Obama and the McCain campaigns as being crucial to their success on Election Day. Overall, Republicans enjoyed a slightly larger net advantage over Democrats among mainline Protestants and non-Hispanic Catholics in the 2008 election than they enjoyed in the


The Disappearing God Gap?

table 6.3 Overall Contact Bias by Religious Traditions: 2004 and 2008
2004 Dem Only (%) General public Evangelical Protestants Traditionalist Centrist Modernist Mainline Protestants Traditionalist Centrist Modernist Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Cath. Traditionalist Centrist Modernist Jews Other religions Unafliated 15 8 7 10 2 14 14 13 15 4 28 7 18 13 16 24 31 17 19 Both (%) 30 33 35 30 27 35 34 32 41 31 20 24 33 34 39 21 39 27 24 Rep Only (%) 15 23 28 13 17 15 19 16 11 13 5 15 17 26 16 14 14 14 11 Net Rep (%) 0 15 21 3 15 1 5 3 4 9 23 8 1 13 0 10 17 3 8 Dem Only (%) 15 9 7 12 3 12 5 18 11 20 21 24 16 2 13 34 29 18 18 Both (%) 35 41 44 36 40 42 41 44 40 38 25 14 39 48 36 38 50 32 32 2008 Rep Only (%) 15 18 22 15 7 16 15 17 14 12 18 1 18 31 16 13 7 10 10 Net Rep (%) 0 9 15 3 4 4 10 1 3 8 3 23 2 29 3 21 22 8 8

Sources: 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

previous election. However, both campaigns wanted to mobilize those segments of both traditions most likely to cast ballots for their particular candidate. Consequently, Republican contacts would likely be concentrated among the traditionalist and centrist elements of both traditions, while Democratic contacts would likely be focused on the centrist and modernist segments of each tradition. Thus, this slightly larger overall GOP advantage in 2008 masks some important differences within both traditions, as traditionalist mainline Protestants and Catholics had much higher Republican bias in their contacts in 2008 than they did in 2004, while modernist Catholics had a much greater Democratic bias in their 2008 contacts. Finally, Democrats enjoyed a much greater net advantage in their contacts among religious minorities in 2008 than in 2004. Whereas
Religion and Election Day: Voter Mobilization in 2008


Republicans had held an overall net advantage of nearly ten points among Hispanic Protestants in 2004, Democrats enjoyed a similar overall net advantage among them in 2008. Hispanic Catholics shifted even more dramaticallyfrom a net GOP advantage of almost ten points in 2004 to a Democratic advantage of more than twenty points in 2008. Similar gains in Democratic contact advantages could be found among Jews as well as among those of other religions. Only among black Protestants did this increased Democratic advantage not prevail in 2008. Overall, therefore, the 2008 presidential campaign witnessed a continuation in the growth in voter contacts. Such contacts were not random in nature, as candidates and parties targeted those groups most likely to support them. Overall, this net growth in voter contacts did not work to the particular advantage of any one party over the other in terms of the electorate as a whole. But different religious groups, particularly those more likely to support a particular party and its candidates, witnessed substantial increases in voter contacts on behalf of that specic party and candidate.

Voter Turnout
Of course, political contacts are primarily a means to an end: to prompt voters to turn out on Election Day. Nevertheless, a little less than onethird who reported being contacted during the campaign (approximately 30 percent in both 2004 and 2008) never made it to the polls on Election Day. Still, those who were contacted were nearly twice as likely to vote as those who were not contacted (approximately 70 percent compared to 40 percent, respectively)a pattern evident in both 2004 and 2008 (data not shown). Table 6.4 presents the percentage of those respondents afliated with different religious traditions who votedbroken down as well by whether they reported having been contacted during the campaign. The table presents the patterns for both the 2004 and 2008 presidential elections in order to assess whether the patterns evident in the past election were similar to, or different from, those in the prior election. Overall, a higher percentage of Americans receive some form of contact than choose to vote, as not all who report some kind of contact necessarily vote on Election Day, and the proportion of Americans who reported some kind of contact (whether from parties, candidates, interest groups, or their houses of worship) was identical in the 2008 and the 2004 election (74 percent) (data not shown). On the other hand, the proportion


The Disappearing God Gap?

table 6.4 Voter Turnout by Political Contact Controlling for Religious Tradition:
2004 and 2008 2004 No Religious Tradition General Public Evangelical Protestants Traditionalist Centrist Modernist Mainline Protestants Traditionalist Centrist Modernist Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalist Centrist Modernist Jews Other religions Unafliated Voted (%) 61 64 68 57 60 69 80 70 62 49 50 43 67 82 69 54 87 58 52 Contact (%) 74 72 71 72 85 79 81 79 76 63 54 53 72 81 73 62 89 66 64 Contact (%) 38 34 41 30 32 42 73 43 29 19 40 11 51 80 55 32 60 40 36 Voted (%) 62 64 71 55 47 71 71 73 69 58 59 54 73 80 69 75 88 60 54 Contact (%) 65 68 75 62 71 73 84 71 82 68 68 64 74 81 70 81 93 68 66 2008 No Contact (%) 44 35 50 32 26 59 65 68 41 22 35 39 68 67 65 33 86 45 34

Sources: 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

of the American public who voted in the 2008 election was up slightly, by one point, over the proportion in the 2004 election.16 Nevertheless, those who reported some kind of voter contact were also the most likely to have voted; this is true across all categories of religious tradition afliation. Still, even among those who reported having been contacted during the course of the campaign, those afliated with different religious traditions exhibited different turnout rates on Election Day. Jews were the most likely to report having votedand this pattern holds for both the 2004 and the 2008 presidential elections. Hispanic Catholics were the least likely, along with the religiously unafliated, to have voted in the 2008 election. With regard to the three largest Christian traditions, non-Hispanic Catholics were the most likely to have voted in the 2008 presidential
Religion and Election Day: Voter Mobilization in 2008


election, six percentage points over their level in 2004. Mainline Protestants also came to the polls at a higher rate in 2008 than in 2004but only marginally so. On the other hand, the percentage of evangelical Protestants who voted in 2004 was not only lower than what was evident among the other two major Christian traditions, but their level of turnout did not increase between 2004 and 2008. However, the greatest level of increase in levels of voter turnout was evident among the ethnoreligious traditionsnamely, black Protestants, Hispanic Protestants, and Hispanic Catholics. Given the candidacy of Barack Obama, the rst minority nominee of a major political party, it is perhaps not too surprising that members of these ethnoreligious groups ocked to the polls on Election Day. Those afliated with each of these groups exhibited an increase in voter turnout of nine percent or more between the 2004 and 2008 elections. There was, however, considerable variation within the major Christian traditionsboth in terms of reported contacts and in terms of turnout on Election Day. Traditionalists within each of the three major traditions were the most likely to have voted; this is true for both the 2004 and the 2008 presidential elections. Moreover, overall, centrists within each major Christian tradition were more likely to have voted than modernists within the same tradition; this, too, was true in both of the past two elections.17 On the other hand, the major changes in turnout between the 2004 and 2008 presidential elections transpired among the modernist segments of each of the three major religious traditionswith modernist mainline Protestants and Catholics turning out at much higher rates in 2008 than they did in 2004. Modernist evangelical Protestants, on the other hand, declined in their level of turnout in 2008 compared to 2004. But since the relative size of modernist mainline Protestants and modernist Catholics is far greater than the size of modernist evangelical Protestants (see table 4.2), the presence of modernist religionists within the three major Christian traditions was far greater in the 2008 presidential election than in the 2004 election.

The Religious Composition of the Electorate

From another vantage point, one can see that the efforts at voter contacts are, in the end, simply intended to shape the composition of the electorate in particular ways so as to alter the likelihood of victory for particular parties and candidates. Analyzing the percentage of particular religious groups who voted provides important information about religion and the 2008 presidential election, but it does not reveal other important


The Disappearing God Gap?

information that is needed to understand the outcome of voting on Election Daynamely, (1) the relative proportion that such religious groups comprise in terms of all who voted in the presidential election, (2) how that composition may be different in 2008 from that in 2004, and (3) how the changing religious composition of those casting ballots on Election Day compares and contrasts with changes in other sociodemographic factors across the two elections. There were some important indications prior to the 2008 fall campaign that high levels of voter turnout would likely prevail in the 2008 election, and, as a result, the likely composition of those casting ballots in the two elections might exhibit some important differences. For example, the 2008 caucus and primary elections witnessed record turnouts, as more than twenty state contests set turnout records. Overall, nearly 55 million Americans voted in the 2008 nominating elections, easily surpassing the 31 million who voted in 2000, the last time when both major parties had contested races (Patterson 2009). And there were some indications that younger voters might turn out in higher numbers in 2008 than they had in previous elections. For example, the number of ballots cast by those under thirty years of age in the Iowa caucuses was three times greater than it had been in 2004 (Patterson 2009). However, though many young adults were clearly enthusiastic about the Obama candidacy, younger voters have frequently been a source of disappointment for candidates counting on their votes on Election Day, as young adults frequently fail to cast their ballots when November rolls around. Table 6.5 examines the changing social composition of voters in the 2008 and 2004 elections based on the results of exit poll data gathered on Election Day. Since the categories employed and reported are fairly broad in nature, with each category capturing a substantial portion of the electorate, it is unlikely that major changes in the composition of the electorate will be evident from one election to the next when examining these particular categories. Nonetheless, several important changes are evident in terms of those who constituted the voting public in 2008 compared to the voting public in 2004. First, the exit poll data do suggest that younger voters formed a slightly larger portion of voters in the 2008 election than they did in 2004as the percentage of those voting who were twenty-nine years of age and younger increased by one point from 2004 to 2008. Similarly, the electorate comprised a larger number of minorities in 2008 than in 2004. The percentage of white voters dropped to below 75 percent for the rst time in American history, while the proportion of African Americans, Latinos, and other ethnic minorities jumped three percentage points from 23 percent in 2004
Religion and Election Day: Voter Mobilization in 2008


table 6.5 Composition of the Voting Electorate: 2004 and 2008

2004 (%) Age 1829 3044 4564 65 Gender Male Female Race White Black Hispanic Other Education Did not complete high school High school graduate Some college College graduate Postgraduate study Religion Protestant/other Christian Catholic Jewish Something else None Church attendance More than weekly Weekly Monthly A few times a year Never 16 26 14 28 15 12 27 15 28 16 54 27 3 7 10 54 27 2 6 12 4 22 32 26 16 4 20 31 28 17 77 11 8 4 74 13 9 5 46 54 47 53 17 29 38 16 18 29 37 16 2008 (%)

Sources: 2004:; 2008:

to 26 percent in 2008. Each of these changes in composition would seemingly work to the advantage of Obama, as younger voters and racial minorities were far more likely to express support for him than McCain. In addition, there were several other changes in voter composition that worked to help Obama. First, the percentage of voters who reported


The Disappearing God Gap?

none in terms of their religious afliation increased by two points. But perhaps more signicantly, the percentage of those voting who reported that they attended church greater than weekly dropped by four percentage points between the two elections. Given these patterns in the exit poll data, it is likely that the religious composition of the electorate in terms of our more rened categories of religious afliation also changed across the two elections. In conducting this analysis, we are shifting away from an examination of exit poll data back to our national surveys as a means to ascertain whether the religious composition of the electorate did change in signicant ways. These data are examined in table 6.6. Based on this analysis, it is clear that the composition of the electorate in terms of religious tradition afliation did shift in some important ways between 2004 and 2008. Several patterns are particularly noteworthy. First, the relative proportion of evangelical Protestants, whose members are the most likely to cast their ballots for Republican candidates, dropped considerably. Given that evangelical Protestants turned out to vote at the same rate across the two elections (see table 6.4), this drop can only be attributed to an increased level of turnout from other segments of the electorate. This relative decline among evangelical Protestants also stands in contrast to the pattern among mainline Protestants and Roman Catholics, the other two large Christian traditions; both groups constituted approximately the same proportion of the electorate in 2008 as they had in 2004. Second, in terms of religious afliations, the increase in turnout occurred primarily among those afliated with the ethnoreligious traditionsnamely, black Protestants, Hispanic Protestants, and Hispanic Catholicsand also with the other, largely non-Western, religions. Clearly, the candidacy of Barack Obama spurred minorities within the American electorate to ock to the polls on Election Dayand those afliated with these groups were most likely to cast their ballots for Barack Obama and the Democratic Party. Overall, therefore, these data conrm what the exit poll data revealed in table 6.6. Third, the proportion of religious traditionalists among those casting ballots in the 2008 election was also lower than the proportion found in 2004, while the proportion of religious modernists voting rose over the last two elections. When one adds the percentages for traditionalist evangelical Protestants, traditionalist mainline Protestants, and traditionalist non-Hispanic Catholics, the resultant percentage is 26.5 percent in 2004, but 24.9 percent for 2008. Conversely, their modernist counterparts in each tradition made up 11.7 percent of the electorate in 2004, but
Religion and Election Day: Voter Mobilization in 2008


table 6.6 The Religious Composition of the Voting Electorate: 2004 and 2008
2004 (%) Evangelical Protestants Traditionalist Centrist Modernist Mainline Protestants Traditionalist Centrist Modernist Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalist Centrist Modernist Jews Other religions Unafliated 28.3 18.0 8.4 1.9 19.6 4.6 9.6 5.4 1.6 5.5 2.7 19.9 3.9 11.6 4.4 2.6 5.5 14.2 2008 (%) 25.3 16.0 7.8 1.5 19.5 4.6 9.4 5.5 2.2 7.9 3.9 19.8 4.3 10.4 5.1 1.5 6.0 13.8

Sources: 2004: National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

12.1 percent in 2008. Again, this shrinking of the traditionalist segment, coupled with the growth of the modernist segment, would seemingly have worked to the electoral advantage of Obama over McCain.

Religious pluralism characterizes the United States, and religious factors have long been critical in voter mobilization. Religion is crucial for voter segmentation and appeals, communication channels, and one-on-one persuasionas well as for political identity and self-understanding. Religion also provides a language for framing of issues and political action in general. Consequently, knowing the multilayered relationship between religion and politics is critical for understanding political mobilization and outcomes, and the 2008 election was no exception to this requirement.


The Disappearing God Gap?

Ultimately, however, it is the electorate who elects, and what matters is the shape of the electorate. In many ways, voter turnout in 2008 largely followed well-established patterns, with those religious groups typically exhibiting higher or lower levels of turnout continuing to do so in the past election. Mainline Protestants and non-Hispanic Catholics voted at higher rates than did evangelical Protestants, who, in turn, voted at a higher rate than did the religiously unafliated. In these relationships, voter mobilization followed well-established patterns in the 2008 election. Nevertheless, the election revealed some important new developments, as a higher proportion of minorities cast ballots in 2008. African Americans, Latinos, and other persons of color are growing rapidly within the population, becoming an ever-increasing portion of the potential electorate on Election Day. But their growing numbers mean little if they fail to come to the polls. But in 2008, with a minority candidate at the top of the Democratic ticket, they turned out in much higher numbers than in previous electionscontributing to a voting base more highly disposed to cast ballots for the Obama candidacy. Whether this increased level of voter turnout among people of color can be sustained is not totally clear. On the one hand, in the long run, the changing demographics of the American electorate currently works to the electoral advantage of the Democratic Party, as minorities are likely to continue to grow as a proportion of the potential American electorate. On the other hand, in the short run, sustaining such relatively high levels of voter turnout among people of color in future presidential elections may prove to be a challenge, now that the initial enthusiasm for securing the election of the rst minority candidate has been attained. At the same time, the religious base of the Republican Party did not match the increased level of turnout scored by its Democratic rivals in the 2008 election. Evangelical Protestants typically exhibit a lower turnout rate than mainline Protestants and Catholics, due in part to a theological tradition that resists expecting too much from government or from political engagement. But while evangelical Protestants voted at the same rate in the 2008 election as they did in the 2004 election, they constituted a somewhat lower proportion of the electorate given the increased turnout among religious minorities. Had evangelical Protestants matched the turnout rates of mainline Protestants or Catholics, the election certainly would have been much closer and might have even contributed to a possible McCain victory (though the latter would have depended on the particular states in which such increased turnout occurred). In addition, religious traditionalists composed a somewhat smaller proportion of the electorate in 2008 than in 2004, while religious modernists
Religion and Election Day: Voter Mobilization in 2008


composed a higher proportion of the electorate in 2008, as did the religiously unafliated. Perhaps this increased level of turnout among religious modernists reected the increased efforts at religious mobilization by the Obama campaign compared to earlier Democratic presidential campaigns, or perhaps it simply reected a somewhat higher level of voter turnout among younger adults, as younger adults are more likely to hold religious views that are relatively modernist when compared to their elders. But regardless of the basis, the increased levels of turnout among religious modernists also served to enhance the likelihood of an Obama victory in the 2008 presidential election. Nevertheless, while the composition of the electorate helps to shape the outcome of the election contest, it does not do so as directly as how such religious groups ultimately choose to cast their ballots in the election. Turnout shapes election outcomes, but in the end it is the actual votes cast that determine the winner of the election contest. Consequently, the mobilization of religious groups and their level of turnout is only half the story; the other half is how those who came to the polls on Election Day decided to cast their ballots. It is to this other part of the story to which we now turn our attention.


The Disappearing God Gap?


Religion and Election Day: Voting Patterns

s discussed in the introduction, religious afliation has historically been the prime means by which religion has inuenced voting. This relationship, particularly during the nineteenth century, has been documented by many historians (Swierenga 2007, 2009). And because race and ethnicity were so closely tied to religious afliation in the United States, the major political parties could largely be viewed as coalitions of various ethnoreligious groups (Kleppner 1979). Moreover, even after the New Deal realignment of the 1930s, a realignment spurred by economic issues, religious afliation remained a key contributing factor in vote choice (e.g., Lazarsfeld, Berelson, and Gaudet 1944; Berelson, Lazarsfeld, and McPhee 1954).1 However, by the end of the twentieth century, the linkage between religion and politics had become more complex. Already by the late 1980s, new political differences had emerged related to the expression of traditional religious beliefs and practices (J. Green 2007, 210). No longer were partisan differences most clearly evident between religious traditions. Now they were also increasingly evident within themwith traditionalists within the various religious traditions falling more on one side of the political spectrum and modernists more on the other. For example, those who attended worship services once or week or more were more Republican, while the less observant were more Democratic (Kellstedt et al. 1996, ch. 14; Kohut et al. 2000). As a result, the two major parties coalitions were changing; the Republican Party was attracting support from the traditionalists within the various religious communities, and the Democratic Party was attracting increased support from those in the same

communities who were less traditionally religious, along with increased support from the religiously unafliated (Layman 2001). Thus, though the importance of religion in shaping voting decisions in the 2004 presidential election may have surprised many commentators and analysts, its underlying patterns had already been well documented. Nonetheless, the God gap generated considerable public attention and discussion, and there was a fair amount of debate and discussion over its meaning and implications. In fact, many interpreted the results as reecting a massive cultural divide. Yet just four years later, in the postmortems of the 2008 presidential election, relatively little attention was given to the role of religion. Public commentary focused on the historic election of an African American and the economic crisis. For many, the election conrmed that economic issues are at the heart of political life and that religious values are secondary, if not epiphenomenonal, in nature politically. Did this relative lack of attention and discussion suggest, therefore, that the God gap in American voting had diminished signicantlyor even dissipated? And what role, if any, did religious voters play in the election of Barack Obama? Furthermore, even if religion played only a modest role in shaping voting decisions in the 2008 presidential election, was this decline in relative importance largely a function of the uniqueness of the 2008 election or a portent that religion will likely be even less of a factor in shaping the outcome of future elections?

Religion and the 2008 Presidential Election

Voting is a very blunt instrument for expressing policy preferences. Particularly within a two-party system, members of different groups often decide to support the same candidatebut for very different reasons. For example, values voters (i.e., those for whom cultural issues are primary) might vote for McCain because they perceive him to be more socially conservative than Obama, but others might come to support McCain because they view him to be stronger and more experienced in matters of foreign affairs. Consequently, it may be helpful initially to examine how different issues priorities shaped voting decisions in the 2008 presidential election.

Issue Priorities and Voting Choice

On Election Night in 2004, media commentators were surprised to learn from exit polls that a plurality of voters had chosen moral values from a


The Disappearing God Gap?

list of seven issues as being the most inuential in their voting decision (moral values were chosen by just over one-fth of the exiting voters, slightly ahead of economy and jobs and terrorism). This prompted a snap judgment by many analysts that President Bush had won reelection simply on the basis of support by deeply religious voters who favored Bushs conservative stance on issues such as abortion and same sex marriage. Others quickly pointed out that moral values hardly dominated the issue priorities of voters and that, in any close election, numerous groups may be seen as providing the required margin of victory in the election. While providing voters with an extensive listing of different issue concerns can be very informative in ascertaining the specic kinds of issues that animate them, our analysis will focus on three broadly based agendas that reect different kinds of issue prioritiesnamely, economic issues, foreign policy issues, and cultural issues. Though these are analytical categories rather than specic issues, when these specic issues are placed in their appropriate categories they capture roughly the same distribution patterns (see, e.g., Green 2007, 70). In the 2004 presidential election, roughly one-quarter of the electorate (26 percent) cited various cultural issues as having the highest priority in their voting decision, but most voters said their highest priority involved either economic (36 percent) or foreign policy (38 percent) issues (data not shown). More important, those who expressed different issue priorities cast their ballots in fundamentally different ways (see table 7.1). Social issue voters preferred Bush over Kerry by large margins, while Kerry captured the economic issue voters by similarly large margins; foreign policy voters were much more divided in their preferences. What impact, then, did issue priorities have on voting decisions in the 2008 election? Much of the postelection analysis in 2008 pointed to the role of economic issues in Obamas victory. In order to ascertain which issues animated voters in the 2008 election, the Henry Institute National Survey of Religion and Public Life asked respondents a series of questions about what kinds of issues were important to their votes, seeking to tap the relative weight of economic, foreign policy, and cultural issues. Clearly economic issues dominated, as more than one-half of those voting (58 percent) cited them as the most important kind of issue, up from the one-third (36 percent) who cited economic issues as being the most important issue in 2004 (data not shown). Still, despite this overwhelming focus on economic issues in the 2008 election, a quarter of the electorate in 2008 (25 percent) cited cultural issues as being the most important kind of issue in the electiona level that virtually matched the proportion who cited such issues in 2004 (data not shown).
Religion and Election Day: Voting Patterns


table 7.1 Presidential Vote by Most Important Issue: 2004 and 2008
Most Important Issue Economic Issues Year and Vote 2004 Bush Kerry Total 2008 McCain Obama Total 36 64 100 63 37 100 56 44 100 32 68 100 55 45 100 70 30 100 (%) Foreign Policy Issues (%) Cultural Issues (%)

Sources: 2004: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

As in 2004, those who held different issue priorities supported different candidates for president in 2008. Nearly two-thirds of those who cited economic issues as the highest priority in the election cast their ballots for Obama. Roughly the same proportion who cited foreign policy issues as having the highest priority voted for McCain. However, the relationship between issue priorities and vote choice is not readily obvious, since voters taking different sides on a particular issue might still cite that issue as particularly important to them. Perhaps this explains why those who cited cultural issues as having the highest priority voted overwhelmingly for Bush in 2004, but only marginally so for McCain in 2008. This did not represent any diminished support for McCain among social conservatives; rather, it reected a growth in attention to cultural issues among liberals.2 Although substantial numbers of voters arrive at their presidential voting decisions months prior to the fall campaign, campaigns do matter. Though nearly half of those surveyed in the Henry Institute National Survey of Religion and Public Life (46 percent) reported that they had made their voting decisions prior to the national conventions, and though an additional 17 percent reported that they had made their decisions after the conventions but before Labor Day, as many as one-third (37 percent) indicated that they made their voting decisions sometime during the course of the fall election campaign (data not shown). In fact, as many as 7 percent of all Obama voters reported having decided to vote for him on Election Day itself!


The Disappearing God Gap?

Clearly, events that occur during the course of the campaign can shape voting decisions. In 2008, events with respect to the economy were particularly important. The bankruptcy of the Lehman Brothers investment house, a Wall Street giant, was announced on September 15. The global nancial crisis that ensued was the largest in seventy years and the rst that occurred in the midst of a presidential campaign (Schier and Box-Steffensmeier 2009, 62). Not surprisingly, therefore, those who made their voting decisions after Labor Day were more likely to cite the economy as the most important issue of the campaign than those who made their voting decisions earlier (65 percent to 55 percent, data not shown). Those afliated with different religious traditions expressed somewhat different issue priorities in the 2008 presidential election. Overall, a majority of members of almost every religious tradition cited economic issues as the most important issue of the election; only among the religiously unafliated and the composite group other faiths did a plurality, rather than a majority, cite economic matters as the most important issues of the campaign (see table 7.2). Overall, cultural issues tended to outweigh foreign policy issues for members of most religious traditions; only among Jews, black Protestants, and mainline Protestants did foreign policy issues rank second in importancethough trailing far behind economic issues. Indeed, one-third of Hispanic Protestants and evangelical Protestants picked social issues as did nearly one-third of the religiously unafliated. Cultural issues dropped in relative importance among evangelical and Hispanic Protestants across the two elections, while these issues grew in relative importance among the religiously unafliated. Clearly, then, religion affects politics, in part, through the way in which afliation with a religious community shapes the expression of different issue priorities and agendas. Though religious afliations shape issue agendas, so, too, do religious practices and beliefs. Traditionalists within each major religious tradition were the most likely to cite cultural issues as being the most important factor in their vote choice. Catholic traditionalists even ranked cultural issues higher in priority than economic issues, and traditionalist evangelical Protestants were equally likely to cite cultural and economic issues as the most important. Still, the fact that the same broad issue agendas can be promoted for different policy ends is reected in the fact that both religious traditionalists and religious modernists were more likely than religious centrists to cite cultural issues as being the most important though modernists tended to express quite different policy positions on such social issues than religious traditionalists (data not shown).
Religion and Election Day: Voting Patterns


table 7.2 Issue Priorities by Religious Tradition: 2004 and 2008

2004 Religious Tradition Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalist Centrist Modernist Jews Other faiths Unafliated Econ. (%) 25 18 40 25 38 38 33 44 48 65 47 37 16 41 43 33 21 34 FP (%) 34 30 40 41 42 31 46 43 14 16 30 43 28 46 47 57 46 48 Soc. (%) 42 52 20 34 21 32 21 12 39 19 23 20 55 13 11 10 33 18 Econ. (%) 52 42 68 67 63 41 73 62 60 88 80 56 36 63 61 73 41 46 2008 FP (%) 17 16 19 10 19 26 15 22 7 7 5 18 19 21 14 27 24 24 Soc. (%) 32 42 13 24 18 33 13 16 33 4 16 25 44 17 26 0 35 30

Sources: 2004: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

Religion and Vote Choice

The most crucial way in which religion may shape electoral behavior is in terms of the relationship between religion and vote choice. Given the waxing and waning of the size of different religious groups within the free market system of American religious life, different religious groups have played prominent roles in different party systems. Moreover, from one political era to the next, the members of a particular religious group may nd themselves more fully or less fully aligned with a major political party. Since the advent of survey research in the mid-1930s, researchers have noted differences in voting preferences along lines of religious afliation, though patterns have shifted over time (Kellstedt et al. 2007).3 But while the specic nature of the relationship may change, religious afliation has always been an important factor in American politics that has shaped and given color to each political order.


The Disappearing God Gap?

In the 1930s, evangelical Protestants were largely Democratic;4 today they are largely Republican. Mainline Protestants were until a few decades ago the core of the Republican Party; but their movement toward the Democratic Party has now left them largely a swing constituency. Catholics, on the other hand, were strong Democrats in the 1930s and 1940s, but they, too, have now also become roughly equally divided in terms of partisanship. Black Protestants have also changed; largely Democratic since the time of Franklin Roosevelt and the New Deal, they have become even more strongly Democratic in their support during the 1960s and the decades following. And while the religiously unafliated have always tended to be Democratic, they have been moving in an increasingly Democratic direction since the early 1990s. And over this same period, the linkages between religion and politics have, at least for the largest religious traditions, changed from largely conicts between religious communities to also include conicts within them (Kellstedt et al. 2007, 292). Table 7.3 reports the two-party presidential vote in 2008 across the religious landscape, and it provides the comparable two-party vote for each religious category in 2004. These data reveal several important patterns. First, there continue to be important differences in voting choice based on patterns of afliation with the major religious traditions. Compare, for example, the McCain vote among all evangelical Protestants, mainline Protestants, black Protestants, and the religiously unafliated. Even using these broad categories of religious tradition helps explain differences in presidential voting, revealing the continuing political importance of religious belonging. Second, there are systematic political differences within all the major religious traditions based on the division between religious traditionalists and modernists, regardless of their particular religious tradition. Traditionalists voted more heavily for McCain, while modernists across all traditions were the least supportive of McCain. This pattern held true for the candidates in 2004 and in 2008, although in the most recent election centrist mainline Protestants were slightly more likely to have voted for Obama than modernist mainline Protestants. In short, the basic patterns in religious voting found in the 2004 election continued to hold true in the 2008 election. While a few religious groups exhibited double-digit changes in their pattern of partisan voting (namely, black Protestants, modernist evangelical Protestants, and centrist mainline Protestants), what is particularly striking is the relative similarity in levels of party voting found across the two elections. In other words, there was far more continuity than change in the patternsand
Religion and Election Day: Voting Patterns


table 7.3 Presidential Vote in 2008 by Religious Tradition

Reported Vote 2004 Religious Tradition Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalist Centrist Modernist Jews Other faiths Unafliated Bush (%) 77 85 66 59 51 65 51 39 63 17 31 53 79 51 33 27 45 28 Kerry (%) 23 15 34 41 49 35 49 61 37 83 69 47 21 49 67 73 55 72 McCain (%) 76 85 67 42 46 66 39 43 57 7 31 53 71 54 37 29 41 24 2008 Obama (%) 24 15 33 58 54 34 61 57 43 93 69 47 29 46 63 71 59 76

Sources: 2004: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey on Religion and Public Life.

magnitudeof religious voting from Bushs victory in 2004 to Obamas in 2008.5 Thus, despite Obamas convincing victory, the results of the election reveal little evidence of any fundamental shift in the structure of faith-based voting. In addition, one is also struck by the relative lack of swing constituenciesthat is, religious groups in which candidates won by small margins. There was much speculation during the campaign about whether Obama would attract a substantial increase in support among evangelical Protestants, partly because his campaign along with his progressive allies had targeted them and partly because, as we saw in chapter 3, McCain struggled for their support. As was true with mainline Protestants and Catholics, Obama did best with the modernist segment of those afliated with the evangelical Protestant tradition. However, the size of the modernist component of that tradition is much smaller than its traditionalist


The Disappearing God Gap?

component (1.7 percent versus 14.8 percent; see table 2.2), and, as a result, evangelical Protestants in the end supported McCain at virtually the same level they supported Bush in 2004. The closest division in the two-party vote occurred among mainline Protestants and non-Hispanic Catholics. More than half of mainliners cast their ballots for Obama, while less than half of non-Hispanic Catholics did so. However, even when evangelical Protestants, mainline Protestants, and Roman Catholics are examined in terms of religious traditionalism, only centrist Catholics exhibit a partisan division of less than ten percentage points. All other religious groups examined in table 7.3 reveal one candidate or the other capturing the votes of that religious community by more than ten points. Clearly, the foundation of Obamas victory was the strong support he received from racial, ethnic, and religious minorities. Not surprisingly, Obama did very well among black Protestantsobtaining more than 90 percent of their votes. But he also did very well among Hispanic Catholics, Jews, and the religiously unafliated, securing about 70 percent or more of their vote totals. Obama also did well in the composite category of other religious faiths, with nearly three-fths of their votes. McCain, on the other hand, did well among evangelical Protestants as a whole, but particularly so among traditionalists. Centrist evangelicals and Catholics, as well as Hispanic Protestants, were also generally supportive of McCain. All other religious groups voted for Obama.6 In the preelection component of the Henry Institute National Survey on Religion and Public Life, respondents were asked whether they had voted in the 2004 election, and if so, for whom they had voted. When one compares the 2004 reported votes of the respondents with their reported votes in 2008, one nds that only 12 percent of the respondents were swing voters, those who voted for the candidate of one political party in the previous election and for the candidate of the opposing party in the subsequent election (data not shown). In other words, nearly nine of ten voters cast a presidential vote in 2008 that reected their partisan choice in 2004. Again, these patterns substantiate the adage that elections are won at the margins, as the subterranean changes in voting patterns do not shift markedly from one election to the next. Overall, 7.3 percent of those voting in both elections shifted from Bush to Obama, while 4.3 percent reported that they had moved from Kerry to McCain (data not shown). If traditionalists tend to vote Republican, and modernists are more inclined to support Democrats, then one might anticipate that it would be the centrists in each religious tradition who are
Religion and Election Day: Voting Patterns


most up for grabs. In fact, they were the most likely to swing, as centrists in each of the three major religious traditions were the most likely to have swung from Bush to Obama. Moreover, among all religious groups, those who swung from Bush to Obama exceeded those who swung from Kerry to McCain; in fact, no religious group exhibited a level of swing from Kerry to McCain that exceeded that from Bush to Obama among their coreligionists (data not shown). Among those who voted for a candidate of a different party in 2008 than in 2004, economic issues were overwhelmingly cited as the most important issue of the campaign. And among all voters who ranked economic issues as most important, more than 10 percent swung from Bush to Obama (while less than 3 percent swung from Kerry to McCain). On the other hand, less than 5 percent who cited foreign policy issues or social issues as the most important swung their votesand only among those who cited foreign policy issues did the swing from Kerry to McCain (3.6 percent) exceed the swing from Bush to Obama (1.8 percent) (data not shown). Clearly, the issue priorities of voters were crucial in affecting whether or not voters partisan choice shifted from 2004 to 2008.

The Religious Composition of the Presidential Vote Coalition

Another way in which to assess the relationship between religion and the contemporary American party system is to examine the voting coalitions supporting each of the presidential candidates. When groups are analyzed on this basis, the contribution that each makes to a political partys electoral coalition is determined by three factors: the size of the group in the electorate, its turnout rate, and the proportion of the vote that it gives to that party (Axelrod 1972; Leege et al. 2002, 231236). Considered in this way, evangelical Protestants have clearly grown in importance over the past several decades in terms of their proportionate contribution to the total votes received by Republican presidential candidates, while their proportionate contribution to the Democratic presidential coalition of votes has declined somewhat (Smidt 2008, 2223). Between 1972 and 1980, evangelical Protestants contributed a little more than onefth (22 percent) of all the votes cast for GOP presidential candidates, but that proportion increased to a little more than one-quarter during the elections of the 1980s and then rose to nearly one-third in the 19922000 period. Thus, evangelical Protestants have grown steadily in terms of their political importance to the GOP, though, given their overall size in the population, they still contribute an important number of votes to the Democratic candidate for president.


The Disappearing God Gap?

Mainline Protestants historically were the core of the GOP coalition of voters, but their contribution to the GOP vote total has steadily declined over the past several decades. During the 1970s, mainline Protestants contributed more than one-third of all the votes secured by the GOP nominee for president, whereas by the turn of the millennium that number had diminished to less than one-fth (Smidt 2008, 23). Two different factors contributed to this drop: declining GOP voting coupled with the declining size of mainline Protestantism itself. In contrast, the contribution of Catholics to the Democratic and Republican presidential coalitions has hardly changed at all since the early 1970s (Smidt 2008, 24). Catholics cast basically one out of every ve votes for Democratic candidates for president between 1972 and 1980, and they continued to do so during the 1980s and 1990s. Over that same span of time, they also continued to contribute a relatively equivalent, and equally steady, proportion to the Republican candidate. During the 1970s, white Roman Catholics provided a little more than one out of every ve votes cast for Republican nominees, and their contributions hovered around the same level during the 1980s and the 1990s. The religiously unafliated, on the other hand, have grown somewhat in importance as a component of the Democratic voting bloc, while clearly declining in importance within the Republican voting bloc (Smidt 2008, 24). During the 1970s and 1980s, the religiously unafliated were a relatively important component of the GOP coalition of voters, providing about 15 to 17 percent of all GOP votes for president, but their contribution to the GOP coalition of voters declined in the 1990s, and in 2004 they provided less than one in ten of all votes cast for Bush, while over the same period of time, their overall contribution of votes provided for Democratic presidential candidates has increased slightly. Table 7.4 examines the relative contribution of each religious tradition to the total votes cast for McCain and Obama in 2008 and compares such results with the proportion each gave to the Bush and Kerry vote totals in 2004. The table details the religious components of the party coalitions in 2008, with each column adding to 100 percent in terms of the religious sources of each presidential candidates vote. Again, a pattern of continuity between the 2004 and 2008 presidential campaigns is evident. The Democratic Coalition. As the rst viable African-American presidential candidate, Obama had special appeal to religious minoritiesparticularly African Americans. With his candidacy, black Protestants turned out in larger numbers in 2008jumping by two points as a percentage of the electorate compared to 2004. These black Protestants voted overwhelmingly for Obama, and, as a consequence, black Protestants accounted for
Religion and Election Day: Voting Patterns


table 7.4 The Religious Composition of the Presidential Vote Coalition

2004 % Religious Constituency Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalists Centrists Modernists Jews Other faiths Unafliated Totala Bush Voters 39.8 27.8 10.0 2.0 18.4 5.5 9.0 3.9 2.6 2.6 1.9 19.8 5.8 11.3 2.7 1.4 5.3 8.1 99.9 % Kerry Voters 12.1 5.0 5.5 1.5 18.9 3.1 9.2 6.5 1.6 13.2 4.4 18.5 1.6 11.3 5.7 4.1 5.6 21.7 100.1 % All Voters 26.0 16.6 7.7 1.7 18.6 4.3 9.1 5.2 2.1 7.7 3.2 19.2 3.7 11.3 4.2 2.7 5.5 14.9 99.9 % McCain Voters 39.8 26.7 11.6 1.5 17.9 5.8 7.6 4.5 2.2 1.5 5.4 18.8 5.4 9.7 3.7 0.7 5.2 8.6 100.1 2008 % Obama Voters 10.4 3.9 4.8 1.7 17.6 2.5 10.1 5.0 1.4 16.6 9.4 14.0 1.9 6.8 5.3 1.6 6.4 22.7 100.1 % All Voters 23.7 14.2 7.9 1.6 17.7 4.0 9.0 4.7 1.8 9.7 7.5 16.2 3.5 8.1 4.6 1.2 5.8 16.3 99.9

Sources: 2004: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey of Religion and Public Life.

Percentages greater or less than 100 due to rounding.

nearly 17 percent of all Obamas ballots. Together, Hispanic Catholics, Jews, and those of other religious faiths constituted 14.5 of the electorate in 2008, providing another 17 percent of Obamas votes, but only 11 percent of McCains. Though the Democratic preference of these particular religious minorities was not something new in 2008, it was larger and more cohesive than in previous elections. Taken together, these religious minorities provided more than one-third of all ballots cast for Obama. Another important component of the Obama coalition was the votes cast by the religiously unafliated. In fact, the religiously unafliated have been the primary source of presidential votes for the Democrats, providing the single largest source of vote for Kerry in 2004 and for Obama in 2008.


The Disappearing God Gap?

Other important sources of votes in the Democratic coalition were centrist mainline Protestants, with signicant contributions also made by modernist mainline Protestants and by centrist and modernist Catholics, while centrist evangelical Protestants fell just below the 5 percent level. Overall, however, the percentage of the total votes that the Democratic presidential candidates received from those afliated with the three largest Christian traditions fell from 49.5 percent in 2004 to 42.0 percent in 2008reecting, in part, the increased level of turnout among religious minorities who strongly supported Barack Obama. Nevertheless, the 42.0 percent that the three largest Christian traditions contributed to the Obama coalition was larger than the contribution of either the unafliated (22.7 percent) or religious minorities (35.4 percent). The Republican Coalition. While the Democratic coalition of voters was religiously and ethnically diverse, the religious base of the Republican coalition was largely anchored in the three largest, predominantly white, Christian traditions of evangelical Protestantism, mainline Protestantism, and Roman Catholicism.7 More than three out of every four votes received by McCain (76.5 percent) were cast by those afliated with one of these three traditions, approximately the same percentage of the Bush coalition in 2004 (78.0 percent). In contrast, as we noted earlier, those afliated with these same three traditions accounted for only 42.0 percent of the votes within the Obama coalition. And whereas Hispanic Catholics, black Protestants, Jews, persons from other religious faiths, and the religiously unafliated provided Obama with 56.7 percent of the votes he received, those afliated with these religious groups accounted for only 21.4 percent of McCains votes. Evangelical Protestants were the core of the Republican coalition in 2004, and they were again in 2008. Evangelical Protestants accounted for two out of every ve votes Bush received in 2004, and they provided for the exact same proportion of all McCains vote in 2008. In fact, traditionalist evangelical Protestants alone provided more than one out of every four votes McCain received in 2008. Mainline Protestants accounted for more than one-sixth of all of McCains votenearly the same percentage that mainline Protestants represented within the Obama coalition, and virtually the same percentage they accounted for in terms of the Bush vote in 2004. Non-Hispanic Catholics provided for nearly one-fth of all McCain votes in 2008, down slightly from their level of contribution within the Bush coalition in 2004. One the other hand, religious traditionalists within each of the three major Christian traditions, rather than evangelical Protestants, could also
Religion and Election Day: Voting Patterns


be viewed as the core of the Republican coalition, as they accounted for nearly two out of every ve votes McCain received (37.9 percent)nearly identical to the level of support provided by evangelical Protestants as a whole. Centrists within the three major traditions provided another onequarter (28.9 percent) of all McCains votes. Thus, overall, traditionalist and centrist evangelicals, mainliners, and non-Hispanic Roman Catholics accounted for two-thirds (66.8 percent) of all votes cast for the Republican presidential candidate in 2008. Conclusion. The 2008 presidential election reveals little evidence of any fundamental shift in religious voting. Overall, the religious structure of the Democratic and Republican coalition of voters remained virtually the same from the 2004 election to the 2008 election. Rather than the 2008 presidential election representing some fundamental shift in allegiances, the story was more one of variation within the basic structure of that vote (cf. Green 2009). In other words, Obama won by improving marginally the proportion of votes he captured within many, though not all, religious groups. Of course, the net result of these marginal shifts was sufcient to enable him to capture the White House in a rather convincing fashion. This was the case in spite of concerted efforts by the Obama campaign to appeal to religious voters and difculties of the McCain campaign in appealing to them.

The Relative Importance of Religion in Voting Choices

To this point, we have simply examined the relationship between religion and voting behavior in isolation from other factors that also serve to shape the voters decisions on Election Day. But just how important a factor was religion in shaping voting decisions compared to other variables such as age, gender, or income?

Social versus Religious Factors in the Recent Election

While religion was clearly associated with voting decisions in the 2008 election, was it the most important factor shaping electoral decisions? Table 7.5 addresses this question, in part, by examining the percentage of the two-party vote cast for Barack Obama and John McCain in 2008 by various sociodemographic variables. The table simply presents the percentage of voters casting ballots for McCain within categories of seven different variables. Each of the variables examined (see table 7.5) is signicantly correlated (at the .001 level) with vote choice.8


The Disappearing God Gap?

table 7.5 Selected Sociodemographic Variables and Presidential Vote: 2008

% McCain Race White Black Hispanic Other Age Under 30 344 4564 65 Marital status/gender Single female Single male Married female Married male Income $40,000 or less $40,001$80,000 $80,001$125,000 Over $125,000 Education Less than high school graduate High school graduate Some college College graduate Some postcollege education Religious tradition Evangelical Protestants Mainline Protestants Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Jews Other faiths Unafliated Religious traditionalism Traditionalists Centrists Modernists 65 59 32 (354) (476) (350) .27*** 76 46 57 7 32 53 29 41 24 (280) (209) (21) (115) (89) (190) (24) (69) (192) .44*** 27 47 51 45 43 (113) (336) (381) (184) (167) .13*** 42 45 58 39 (406) (487) (174) (114) .11*** 31 33 51 59 (271) (231) (330) (346) .24*** 33 49 44 55 (205) (346) (411) (208) .15*** 55 2 33 40 (829) (129) (151) (67) .34*** (N ) Eta***

Source: 2008 Henry Institute National Survey on Religion and Public Life. *** Statistically signicant at .001 level.

Religion and Election Day: Voting Patterns


If demographics are destiny in terms of which political party prevails in our two-party system, then the patterns revealed in table 7.5 raise two important questions about the future strength of the Republican Party. First, whites were signicantly more likely than African Americans and Hispanics to vote for McCain. While whites constitute the vast majority of the electorate, as we noted earlier, they fell below 75 percent of the electorate for the rst time and their share of the electorate is likely to decline further in future elections. This alone represents a considerable challenge for the GOP, given its persistent and growing difculties in attracting minority support. Second, while the highly anticipated surge in younger voters did not fully materialize, voters under the age of thirty did vote overwhelmingly for Obama. Over the past several decades, young Americans had actually been more likely to vote Republican than their elders, while Democrats generally fared better among whites who had reached voting age before World War II (Abramson, Aldrich, and Rohde 2006, 113). However, in 2004, younger voters between ages eighteen to twenty-nine were the most supportive of Kerryalthough only 55 percent of young voters cast their ballots for him (Abramson et al. 2006, 113114). But in 2008, Obama captured two-thirds (67 percent) of the young voteproviding Obama a margin over McCain nearly four times greater than the nine-point margin that Kerry enjoyed among such young voters in 2004 (Rothenberg 2008). Thus, the presidential elections of 2004 and, particularly, 2008 suggest a new trendone in which the youngest voters are the most likely to support the Democratic candidate. Thus, the situation is now reversed from that of the Reagan era; it is the oldest voters who are now the most likely to support the Republican candidate. Should these patterns continue to prevail, they would have important long-term consequences for American electoral politics. Not only would it reveal the emergence of a cohort of young voters much more highly disposed to the Democratic than the Republican Party, but this more youthful Democratic cohort would increasingly grow in political importance and inuence as the older, more Republican, voters increasingly pass from the political scene through natural processes of mortality. Gender differences in American voting behavior emerged in the 1980 presidential election, and differences grew through the 2000 election (Abramson et al. 2006, 112). Initially, during the Reagan era, some feminists hoped that women would play a major role in defeating the Republicans, but the presence of a gender gap in voting does not necessarily help the Democratic Party. For example, in the 2004 presidential election, women largely split their vote between Kerry and Bush (51 percent of women voted


The Disappearing God Gap?

for Kerry), while men voted primarily for Bush (44 percent for Kerry). In the 2008 presidential election, however, there was more of a marital gap than a gender gap.9 Among those not married, Obama captured an overwhelming majority of support (two-thirds or more of votes cast), while McCain enjoyed majority support among married votersboth male and female. This support for McCain among married voters may reect in part the greater appeal of the Republican McCain to persons holding more traditional cultural values, as they are likely to be found disproportionately within their ranks. Historically, American politics has not exhibited the same level of class-based conict as is found in many other industrialized Western nations. And while social class has remained an important inuence in shaping voting choice outside the United States, the inuence of social class on the vote has declined over the past several decades (Nieuwbeerta and DeGraaf 1999). However, some have suggested class-based voting is reemerging within advanced industrial societies (Evans 1999). For example, differences in educational attainment could well serve as a basis of political cleavage in which the information-rich and technologically sophisticated members of the electorate vote differently from those who are information-poor and unskilled. Moreover, Stonecash (2000, 124) argues that class divisions have grown, not declined, since the 1950s, at least when class divisions in voting are measured by income groups. However, in 2008, the data presented in table 7.5 reveal that family income and education were both curvilinearly related to vote choice, with those at the lowest and those at the highest levels of family income being most likely to cast their ballots for Obama. Thus, the partisan division was less a pattern of the rich versus the poor than a coalition of the relatively rich and the relatively poor versus the middle. Finally, table 7.5 examines the pattern of voting for McCain within categories of the two religious variables we have been analyzing throughout the past two chaptersreligious tradition afliation reecting the politics of religious belonging and religious traditionalism reecting the new politics of religious behaving and believing. In terms of the strength of the relationships between voting and each of the seven variables examined in table 7.5, the two religious variables were among those most strongly related to vote choice. In fact, of all the variables analyzed, religious tradition, even without incorporating the traditionalist-modernist divide, is most strongly correlated with vote choice. Religious traditionalism ranked third in terms of the magnitude of its associated eta value, just behind race and just ahead of the combined gender and marital status variable. Obviously, in terms of the simple
Religion and Election Day: Voting Patterns


relationships examined, religion is more strongly related to the vote than other factors frequently examined in seeking to explain differences in voting patterns within the American electorate. This helps to explain why religious considerations played a signicant role in both campaigns.

Relative Importance of Religion across Elections

Of course, the analysis presented in table 7.5 has its limits. It does not take into account the simultaneous inuence of each factor in shaping voting decisions. Nor does it provide a broader time perspective than the past two elections. Two questions remain. First, does religion remain an important factor explaining variation in voting choices once all other factors have been taken into account? Second, have religious factors, relative to other factors, grown or lessened in importance in shaping presidential election outcomes over the past four elections? These questions are addressed in table 7.6, which presents the result of a Multiple Classication Analysis.10 This analysis assesses the relative inuence of seven variables in shaping the presidential vote cast by voters in the 1996, 2000, 2004, and 2008 presidential elections, the same variables analyzed in the previous table.11 Because religious tradition has historically been the prime way in which religion has related to electoral politics, we rst conducted the analysis (model 1) using only the religious tradition variable (and omitting the religious traditionalism variable) to ascertain its relative inuence in relationship to the other social variables; then we redid the analysis using both the religious tradition and religious traditionalism variables (model 2). Doing so allowed us to ascertain to what extent the politics of religious beliefs and behavior may be supplanting the politics of religious belonging as an inuence on voting decisions. The statistics presented in the table are the resultant beta coefcient associated with each sociodemographic variable in each of the last four presidential elections. The beta coefcient reects the relative importance of that particular variable in accounting for variation in vote choice, once the effects of the other sociodemographic variables have been taken into account (that is, held constant). The larger the beta coefcient, the greater the inuence of the variable. Several important conclusions can be drawn from table 7.6. First, the rank order of the relative importance of the various variables did not change markedly across the four elections examined. For example, the beta coefcients for both age and income have been relatively small in magnitude and stable across time, with age and income exhibiting the lowest beta coefcients in the 2008 election. Accordingly, the data indicate that once


The Disappearing God Gap?

table 7.6 The Relative Effects of Selected Sociodemographic Variables on Presidential

Vote Choice over Time: 19962008 Model 1 1996 2000 2004 2008 Beta Beta Race Age Marital status and gender Income Education Religious tradition Religious traditionalism R2 .17 .07 .14 .06 .04 .23 x .15 .13 .08 .19 .08 .06 .30 x .20 Beta Beta .18 .10 .09 .03 .08 .31 x .19 .29 .07 .12 .07 .16 .29 x .25 Model 2 1996 2000 2004 2008 Beta Beta .19 .07 .16 .07 .03 .18 .20 .18 .15 .09 .20 .09 .07 .24 .18 .22 Beta Beta .23 .10 .09 .05 .07 .18 .27 .23 .31 .06 .13 .07 .15 .21 .18 .28

Sources: 1996: Second National Survey of Religion and Politics, Bliss Institute, University of Akron; 2000: Third National Survey of Religion and Politics, Bliss Institute, University of Akron; 2004: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron; 2008: Henry Institute National Survey of Religion and Public Life.

the other variables have been taken into account, the inuence of age and income differences within the electorate have been relatively unimportant factors in shaping peoples voting decisions across the past four elections. In contrast, the relative importance of marital status and gender has changed considerably over the same period. The combined effects of gender and marital status ranked third in relative importance in the 1996 election and second in relative importance in the 2000 election, but their importance has diminished in the past two electionsranking only fourth out of the six variables analyzed. Thus, the gender or marital gap in presidential voting peaked in the 2000 election, and its relative importance has diminished subsequentlyalthough it continues to have some inuence. Race and education have grown in relative importance over the past four elections. Education ranked last in relative importance for the 1996 and 2000 elections, and was only marginally more important than income in the 2004 election, ranking second from the bottom. However, education jumped in importance between 2004 and 2008. It was the third most important of these seven variables by the 2008 election (when religious tradition is used as the sole religion variable). Race, however, has shifted in relative importance as an explanatory variable over the same period of time, ranking second in relative importance in the 1996 election, third in the 2000 election, and then second again in the 2004 election. However,
Religion and Election Day: Voting Patterns


with Barack Obama at the head of the Democratic ticket, race jumped in importance in the 2008 election, tying religious tradition as the most important explanatory variable of the six analyzed. Finally, when religious tradition is entered as the lone religious variable, it consistently ranks as the most important variable in shaping electoral decisions for the three elections between 1996 and 2004with the magnitude of the beta coefcient far exceeding its closest rival in accounting for variation. In the 2008 election, it was tied with race as the most important variable. Clearly of all the variables examined across the past four presidential elections, religious tradition ranks as the most important for shaping electoral decisions. However, when religious tradition and religious traditionalism are both entered in the analysis together (model 2), the effect of the religious tradition variable diminishes in importance, but it hardly disappears. With both religious variables in the analysis, the values of the beta coefcients for the sociodemographic variables change only slightly from what was evident when religious tradition was the sole religious variable. Only the coefcients of the race variable change substantially across the two analysesand then only in the 2004 and 2008 elections. What is more important for this second analysis presented in table 7.6 is the relative importance of the religious tradition measure versus the religious traditionalism measure across the four elections. Several patterns are particularly noteworthy. First, even with the presence of two religious variables in the analysis, both variables remain relatively important, if not the most important, explanatory variables in explaining vote choice across the four elections. This is critical since it is religious traditionalism that measures the God gap within the electorate, while religious tradition captures differences in religious afliation or belonging. Clearly both religious tradition and religious traditionalism have been at work in shaping electoral choice in the last four electionsand at roughly equivalent levels of effect. Second, already in the 1996 election, religious traditionalism slightly outweighed religious tradition in importance. Thus, the God gap has been present, and has been relatively important, for some time. Though many observers noted the relationship between frequency of attendance at worship services (a component of our traditionalism variable) and vote choice in 2004, such a pattern appeared to be already operative in the 1996 election and may well date to sometime before that election. In other words, religious traditionalism outweighed religious tradition (and all other sociodemographic variables) in shaping vote choice already several elections prior to the much noted 2004 election.


The Disappearing God Gap?

Third, the relative importance of the two religious variables uctuates from one election to the next, with each variable continuing to be relatively important in accounting for differences in voting choices across the four elections. In each of the elections examined, the relative importance of the two variables shifts back and forth. Thus, in the 2000 election, religious tradition outweighed religious traditionalism contrary to the 1996 pattern, but in 2004 the pattern reversed itself again, only to be reversed once again in 2008. This shifting of the relative importance of religious tradition and religious traditionalism across the past several elections suggests that both religious factors are, relatively speaking, rmly rooted within the American electorate todayshifting in relative importance simply with the particularities of the specic election itself. In the short run, either religious factor might be somewhat more important in one election than the next, but it is highly unlikely that either factor will disappear as a major factor in shaping election outcomes. And nally, with the high level of support given to Obama in 2008 by members of the various ethnoreligious minority groups, there is a decline in the relative importance of the God gap in explaining the voting behavior of Americans in the 2008 election. Moreover, given that African Americans and Hispanics generally gave overwhelming support to Obama while whites were more evenly divided, there was a corresponding increase in the relative importance of race in the 2008 election as well. This is captured by the magnitude of the beta coefcient for the race variable, which, with both religious variables included in the analysis (model 2), outranked both religious tradition and religious traditionalism as explanatory variables. As already noted, the data examined in table 7.6 suggest that religious traditionalism has been a relatively constant, and relatively important, contributing factor in voting decisions across the past four elections. However, it has been a major contention of this volume that the politics of religious belief and behavior is something relatively new and that it has been superimposed on the long-established relationship of the old politics of religious belongingeven though the data presented in table 7.6 do not reveal it. Consequently, it is necessary to examine some earlier presidential elections in order to place the last four presidential elections into a broader contextone that will more clearly reveal whether or not religious traditionalism is a new factor in presidential voting. Fortunately, the Anti-Semitism Study of 1964,12 a national survey of Americans conducted during the presidential campaign of 1964, contains both fairly detailed denominational afliation measures (which permit the creation of a relatively accurate religious tradition measure) as well as a
Religion and Election Day: Voting Patterns


number of measures that tap religious belief and religious behavior identical to the ones used to create the religious traditionalism measure employed in table 7.6.13 The Anti-Semitism Study also asked respondents how they had voted in the 1960 presidential election and what their voting preferences were in the 1964 presidential election, with Election Day only several weeks away. The resulting relationship between religious tradition and the vote, as well as the relationship between religious traditionalism and the vote within each religious tradition, is presented in table 7.7. The corresponding pattern found in 2008 is also included in the table for purposes of comparison; the percentages presented for 2008 are identical to those found in table 7.3. Several noteworthy patterns can be discerned from the data. First, in contrast to contemporary patterns, the greater proclivities of mainline Protestants to vote Republican and the greater proclivities of Catholics to vote Democratic in the 1960s are clearly evident. While evangelical Protestants and mainline Protestants cast their ballots in a relatively similar fashion in the Nixon-Kennedy contest of 1960, mainline Protestants were more inclined to cast their ballot for Goldwater in 1964 than evangelical Protestants. Less than one in ve Catholics voted for Nixon and less than one in ve were inclined to vote for Goldwater. Likewise, the greater Republican tendencies among the religiously unafliated are also apparent, with one-half to one-third reporting that they voted Republican or intended to vote Republican across the two elections. Second, what is also particularly noteworthy is the lack of any real differences between and among the traditionalists, centrists, and modernists within each religious tradition. Not only are the percentage differences between the traditionalist segment of each religious tradition and its modernist segment very modest (with the largest difference being 8 percent among mainline Protestants in 1960), but traditionalist Protestants are very different from traditionalist Catholics. Clearly, it is religious tradition afliation, not religious traditionalism, that was shaping voting decisions in the 1960s. These same data are once again analyzed in table 7.8, but this time controlling for the same sociodemographic variables examined in table 7.6. The table presents the resultant beta coefcients for each of the seven variables included in the analysiswith the data for 2008 once again presented in the table for purposes of easy comparison (these are the same beta coefcients found in table 7.6). Once again, several important differences between the 1960s and the present are revealed in the table. For example, with population changes over the past ve decades, racial and ethnic differences are now much


The Disappearing God Gap?

table 7.7 Vote Choice by Religion Tradition and Traditionalism

(% Voting Republican) 1960 Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Black Protestants Catholics Traditionalists Centrists Modernists Jews Unafliated 59 62 55 60 57 60 57 52 12 16 16 15 18 6 49 1964a 42 44 41 43 47 50 45 44 2 19 17 18 22 3 33 2008 76 85 67 42 46 66 39 43 7 53 71 54 37 29 24

Sources: 1960 and 1964: The Anti-Semitism Study; 2008 Henry Institute National Survey of Religion and Public Life.

Indicates voter preference at the time of the survey.

more important in explaining differences in voting patterns than when the American electorate was far more homogenous in racial composition. Similarly, the growth in the importance of gender as a factor shaping voting over the past ve decades is suggested by the greater magnitude of the beta scores for the combined marital status and gender variable in 2008 compared to the 1960s. However, by contrast, age is a far less important explanatory variable today than during the 1960s. And nally, the relative importance of income and education in shaping voting decisions remain much the same today as they did in the early 1960s. But with regard to religious tradition and religious traditionalism, three important patterns emerge. First, the Protestant-Catholic divide of the 1960 presidential election (which involved the candidacy of Catholic John Kennedy) made religious tradition by far the most important variable that year. Nevertheless, even in the absence of a Catholic on the ticket of either party in 1964, religious tradition continued to far outweigh the other variables in its relative importance. And nally, as seen in the beta scores associated with the religious traditionalism measure, religious traditionalism was not a factor in explaining differences in presidential
Religion and Election Day: Voting Patterns


table 7.8 The Relative Effects of Selected Sociodemographic Variables on Presidential Vote Choice over Time: 19602008
1960 Beta Race Age Marital status and gender Income Education Religious tradition Religious traditionalism R2 .09 .11 .05 .07 .14 .37 .03 .22 1964a Beta .08 .13 .05 .09 .12 .27 .01 .16 2008 Beta .31 .06 .13 .07 .15 .21 .18 .28

Sources: 1960 and 1964: The Anti-Semitism Study; 2008 Henry Institute National Survey of Religion and Public Life.

Presidential preference rather than vote cast.

vote choice less than half a century ago. Not only did religious traditionalism rank last in terms of the magnitude of its beta scores, but the value of such scores was virtually zero. Clearly, then the political importance of religious traditionalism is something that has only emerged in American politics sometime over the course of the past several decades.

The Continuing Signicance of Religion in Future Elections

The analysis presented to this point suggests not only that elements of a new religious order based on the new politics of religious believing and behaving are becoming increasingly evident within American electoral politics but that this new order is largely superimposed on top of the old order of the politics of religious belonging. Clearly, these different means by which religion can order political alignments are hardly mutually exclusive, as religious traditions and traditionalism can both matter, and importantly so. As a result, facets of both the ethnocultural and the theological restructuring models are currently evident within American politics. In this section, we assess whether the new religious order of the politics of religious believing and behaving is likely to replace the older order of the politics of religious belonging, whether the converse is likely to occur, or whether both orders will likely continue to prevail over the next several elections. As a result, we shift our analysis initially from the voting


The Disappearing God Gap?

behavior of the electorate in the 2008 election to the reported partisan identications of Americans following the election. Analysts of American voting behavior have long differentiated between ones vote choice and ones partisan identication. Partisan selfidentication is conceptualized as a psychological variable that can be differentiated from the behavior of casting a vote; one could identify as a Republican and still decide on Election Day to vote for Obama (or identify as a Democrat and vote for McCain). There is a strong relationship between partisan identications and vote choice, of course, but the two are analytically distinct. Moreover, casting a ballot for a candidate of the opposite party does not typically change ones partisan identication; most, if not all, Republicans who cast their vote for Obama likely continued to view themselves as Republicans following the election and will continue to vote primarily for Republicans in future elections. Thus, partisan identications are less prone to change from election to election than vote choice. As a result, in our effort to assess the possible continuing inuence of religion in American electoral politics, it is better to analyze the partisan identications of respondents than their voting decisions in the 2008 election. Table 7.9 examines the partisan identications of Americans following the 2008 presidential election, broken down in terms of both religious tradition afliation and religious traditionalism within the three largest Christian traditions.14 However, since presidential voting was analyzed in terms of a two-party vote, but partisan identications are analyzed in terms of three options (Democrat, independent, Republican), the percentage of identiers with a particular political party will always be less than the percentage who voted for the presidential candidate of that party. Consequently, it is more helpful to examine the overall pattern of distribution for partisan identications than the absolute percentage of identiers with each particular party within each religious group. Overall, a higher percentage of Americans identied with the Democratic Party (43 percent) than with the Republican Party (33 percent) in the weeks following Election Day in 2008 (data not shown). And within every religious tradition analyzed except for evangelical and mainline Protestants, the percentage of those who identify as Democrats exceeds the percentage who identify as Republicans. Still, overall, the data suggest that the differences in presidential voting found earlier in the chapter with regard to different religious groups are largely evident in the reported partisan identications of such groups. First, not only did three out of four evangelical Protestants cast their ballots for McCain in 2008, but such evangelicals overwhelmingly identify as
Religion and Election Day: Voting Patterns


table 7.9 Partisan Identication of Religious Groups Following the 2008 Election
Republican Religious Constituency Evangelical Protestants Traditionalists Centrists Modernists Mainline Protestants Traditionalists Centrists Modernists Hispanic Protestants Black Protestants Hispanic Catholics Non-Hispanic Catholics Traditionalists Centrists Modernists Jews Other faiths Unafliated (%) 56 64 44 48 40 61 33 32 36 7 6 36 52 36 21 27 27 16 Independent (%) 20 14 26 26 20 14 19 27 16 7 33 20 20 18 25 13 26 41 Democrat (%) 25 21 30 26 40 25 48 41 48 86 62 44 28 46 54 60 47 42 Total (%) 101a 99a 100 100 100 100 100 100 100 100 101 100 100 100 100 100 100 99a

Source: Henry Institute National Survey on Religion and Public Life.


Percentages greater than or less than 100 due to rounding.

Republican rather than Democratic. A majority of evangelical Protestants continue to identify as Republicans, with the percentage of those who identify with the Republican Party more than doubling the percentage who identify with the Democratic Party. Thus, while a Democratic presidential candidate may, in the near future, be able to improve on the 24 percent of votes obtained by Obama in 2008, any such improvement will be limited in scope and will likely have to be drawn from the group of evangelicals who identify as independents. Mainline Protestants, as a whole, are evenly divided in their partisan identications (with two-fths of mainliners reporting identication with each major party). Thus, it is the independents among mainline Protestants who will likely tip the partisan voting of such religionists in the next presidential elections. Not only were mainline Protestants an important swing constituency in the 2008 election, but for the next several presidential elections at least they will likely continue to be largely up for grabs, with the presidential candidate of either major party potentially able to win a majority of mainline Protestants votes.


The Disappearing God Gap?

Non-Hispanic Roman Catholics voted more heavily for McCain than Obama (53 percent versus 47 percent) in the 2008 election. However, despite their Republican presidential voting, such Catholics continue to identify somewhat more closely with the Democratic Party than the Republican Party. The magnitude of the Democratic advantage in partisan identications among non-Hispanic Catholics is relatively small compared to most other remaining religious traditionswhile their overall size makes them a relatively large bloc of voters. Thus, despite their somewhat Democratic proclivities, non-Hispanic Roman Catholics are also likely to be viewed by presidential candidates as an important swing religious constituency over the next several elections. The religiously unafliated continue to identify more heavily with the Democratic than the Republican Party, and they are likely to continue to vote much more Democratic than Republican in the near future. Nevertheless, there may be some possibilities for Republican inroads among the unafliated, as nearly an equivalent proportion of the unafliated claim to be independents as claim to be Democrats. And most other religious groups (particularly Hispanic Catholics and black Protestants) are overwhelmingly Democratic in their partisan identications suggesting a continuation of high Democratic voting among these groups in future presidential elections. Only among Hispanic Protestants do even a third of these religious groups report that they identify with the Republican Party. Finally, the partisan identications of religious traditionalists are much more Republican than Democratic in nature. In fact, a majority of all traditionalists, regardless of religious tradition, identify as Republicans outnumbering Democrats by more than two to one. The exception to this doubling pattern is found among traditionalist Catholics, where Republican partisan identications come close to, but do not exceed, a two-toone ratio to Democratic identications. Among modernists, Democratic partisan identications tend to be more prevalent than Republican ones, but not by nearly such a ratio. In fact, among modernist evangelical Protestants, Republican identications continue to be much more prevalent than Democratic identications. Among modernist mainline Protestants, a slightly larger percentage identify as Democrats than as Republicans. Only among modernist Catholics does one nd a Democratic partisan identication ratio that matches the Republican partisan identication advantages found among traditionalists, regardless of religious tradition. Centrists tend to fall between traditionalists and modernists in terms of their identication with the Republican Partybut much closer to
Religion and Election Day: Voting Patterns


the modernists than to the traditionalists within their respective traditions. Actually, among centrist evangelical and mainline Protestants, the percentage of Democratic partisan identication exceeds that found among the modernist segments of their respective traditions. Only among non-Hispanic Catholics do the partisan identications of centrists fall neatly between those of the traditionalists and modernists within their ranks. Thus, overall, the data patterns found in table 7.9 suggest that the partisan differences in voting patterns analyzed earlier are also rooted in the partisan identications along the lines of both religious afliation and religious traditionalism. These patterns do not suggest any disappearance of the God gap in American politicsat least in the near future. Rather, the data suggest that both the old politics of religious belonging and the new politics of religious believing and behaving will likely continue to be present in American presidential elections for some time. Although this analysis of partisan identications may be helpful in assessing the likely continuation of the God gap in American politics, it does not provide any perspective with regard to generational differences that might be found within the American electorate. The passing away of one generation and the emergence of a new generation could alter the way religion relates to politics. Hence, our nal analysis examines age differences in voting for Obama within categories of the four largest religious traditions and within the traditionalist, centrist, and modernist categories of religious traditionalism. These data are presented in table 7.10. Generally speaking, there are few, if any, generational differences in presidential voting within the four largest religious traditions.15 For example, only one in ve evangelical Protestants aged eighteen to twentynine reported than they had voted for Obama, about the same percentage reported by other evangelical Protestants regardless of their age.16 Likewise, the youngest segment of mainline Protestants were only slightly more likely to vote for Obama than the oldest segment of mainliners, and were actually less likely to do so than either of the two middle-aged groups of mainline Protestants. Once again, like evangelical Protestants, there was not any real clear age differences in voting for Obama. More distinct age differences, however, are found among non-Hispanic Catholics and among the religiously unafliated. Younger Catholics were actually much more likely to have voted for McCain than older Catholics, as only one-quarter of non-Hispanic Catholics aged eighteen to twentynine reported that they had voted for Obama.17 This patterns stands in stark contrast to the much more evenly divided partisan voting found within each of the other three Catholic age groupings.


The Disappearing God Gap?

table 7.10 Voting for Obama by Large Religious Traditions and Religious
Traditionalism Controlling for Age Age 1829 (%) Religious traditions Evangelical Protestants Mainline Protestants Catholics Unafliated Religious traditionalism Traditionalists Centrists Modernists 38 71 72 30 52 68 41 59 71 31 53 58 22 52 26 98 20 55 45 73 28 58 53 65 23 49 47 52 3044 (%) 4564 (%)


Source: Henry Institute National Survey on Religion and Public Life.

Only among the religiously unafliated does one nd a clear progression in diminished Republican voting as one moves from the youngest to the oldest age categories. While only a little more than one-half of the unafliated sixty-ve years and older reported that they had voted for Obama, nearly two-thirds of those between forty-ve and sixty-four years of age did so, and nearly three quarters of those between thirty and fortyfour. But among the unafliated who were under thirty, nearly all reported that they had cast their ballots for Obama. Thus, given the voting patterns evident among the youngest voters in each of the four largest religious traditions, it would appear that, if Obama received unusually strong support among the young, it was primarily so among the religiously unafliated young (and likely the young among the ethnoreligious minorities who, as a whole, strongly supported the Obama candidacy). Finally, table 7.10 also examines age differences in voting for Obama among those exhibiting different levels of religious traditionalism. Once again, generational differences are not particularly evident. Among religious traditionalists, the percentage of those voting for Obama varied from 30 to 40 percent, regardless of the age of those exhibiting such traditionalist religious beliefs and behavior. And while nearly two-fths of the youngest age group of traditionalists voted for Obama, a slightly higher proportion of those aged forty-ve to sixty-four did so. Thus, young traditionalists were hardly distinctive in their support for Obama. The same conclusion must be drawn with regard to religious modernists. Roughly 60 to 70 percent of religious modernists voted for Obama,
Religion and Election Day: Voting Patterns


regardless of age. Those religious modernists who were sixty-ve or older were the least likely to have voted for Obama (though well over half did so), but those modernists under thirty were hardly more likely to do so than those who were thirty to forty-four years of age or those forty-ve to sixty-four. Centrists, on the whole, were much more evenly divided in their voting for McCain and Obama, though a majority of centrists in each of the three oldest age categories voted for Obama. Only among religious centrists does one nd a sizable difference between the voting patterns of those under thirty years of age and those who were older than thirty. More than twothirds of those centrists under the age of thirty voted for Obama, while only a little more than one half to three-fths of older centrists did so. Thus, given the voting patterns found in table 7.10, one does not nd any consistent generational difference in voting patterns that cuts across either religious traditions or religious traditionalism. Overall, religion tends to trump age in terms of how respondents voted in 2008. In other words, young evangelical Protestants voted much more like other evangelical Protestants than they did other younger voters as a whole; the same can be said for young mainline Protestants and even for young Roman Catholics. Likewise, those religious traditionalists under the age of thirty voted more like other religious traditionalists than they did other younger voters as a whole. In conclusion, therefore, the different patterns of partisan identications evident across different religious traditions, coupled with the lack of partisan differences found across different age groups within those traditions, suggest that the partisan differences associated with the old politics of religious belonging are likely to continue for numerous elections to come. Similarly, the different patterns of partisan identications across different levels of religious traditionalism, coupled with the general lack of partisan differences across different age groups within those different levels, suggest that the partisan differences associated with the new politics of religious belief and behavior are also likely to continue for the next several presidential elections. While the God gap may have diminished somewhat in the 2008 presidential election, it appears to be rmly rooted within the American electorate in terms of patterns of partisan identication. Moreover, the differences in partisan attachments and voting behavior associated with religious traditions and religious traditionalism remain largely untouched by generational differences within the electorate. Consequently, the patterns of religious voting that prevailed in the rst three presidential elections of the twenty-rst century are likely to continue to be evident over the course of the next several presidential elections as well.


The Disappearing God Gap?


The God Gap Revisited

residential elections culminate on a Tuesday in early November, but they are not dened merely by that moment in time. Elections are better described as a process, not a single grand event. Indeed, American presidential elections take candidates and voters down a remarkably long and winding path. To be sure, the weeks just prior to the November election can be decisive for voters, but their choices have already been reduced and rened in an election cycle that can last two or more years. Long before campaigns begin, hopeful contenders begin to plot strategy. They build or strengthen networks within political parties, establish exploratory committees, test the fund-raising waters, and begin crafting messages to target key voting blocs. At this early stage, well before most ordinary voters take serious notice of the candidates, party elites and activists hold signicant sway in paring down the eld. The remaining candidates are then subjected to a grueling competition through the primary elections and caucuses, the state-by-state contests that give voice to the broader electorate and ultimately yield party nominees. In this book we have focused on religion as part of that complex and dynamic process. In so doing, we have assumed at least two things about the role of religion in the 2008 election. First, we have assumed that campaigns and elections are a volatile process, requiring a comprehensive analysis over time. To understand the process fully, scholars must avoid focusing entirely on any specic point in the campaigneven Election Day itselfbecause to do so risks inating the importance of a narrow range of factors that seem particularly salient at that moment. Presidential elections are governed by myriad formal rules, cultural expectations, and, in our mass media age, the ever-present need to raise vast sums of money. The conuence of rules, expectations, and candidate decisions at certain

points in the election process can have profound implications for the future direction of the campaign.1 Second, we have assumed that the role of religion reects the complexity and dynamism of the election process itself. In some instances, religion appears to take a back seat to other factors (e.g., the economy, the war in Iraq, Barack Obamas race). At other times, it is front and center (e.g., the Jeremiah Wright controversy, Mitt Romneys Mormon faith, Sarah Palins Pentecostalism). Yet when looking at the campaign as a process, we can see that different moments are often connected in subtle, yet important, ways. Obamas membership in a prominent AfricanAmerican church, especially during the Jeremiah Wright controversy, may have reinforced the anxieties of some voters about the issue of race, but it may have also backred for those who sought to make his religious associations a wider campaign issue. Palins nomination as a vice presidential candidate, which was widely perceived as a gesture toward evangelical conservatives and provided an initial bump in the polls, may have had the unintended effect of alienating other voters later in the process. As we have shown, the nature and scope of presidential elections provide countless openings for religion to affect a campaign, and clearly religious believers and groups attempt to do so. Their efforts are a familiar part of the history and political culture in the United States. Yet every election poses its own set of challenges and opportunities to candidates. In this chapter we take a step back to consider the patterns that emerge in the interplay of religion and politics in the 2008 election. On the one hand, we conclude that many of the patterns in cultural attitudes and norms, faith-based mobilization, voter turnout, vote choice, and partisanship follow the trend lines of presidential elections over the past two decades. As a result, we nd continuing evidence of a God gap. On the other hand, beneath the surface of these patterns, we discover fresh efforts at voter mobilization and evidence of shifting voting blocs that might have implications for the future role of religion in presidential elections.

Continuity in the Electoral Role of Religion

We began this book with a discussion of similarities between 2008 and previous elections, focusing especially on continuities that have little to do with religion. Those continuities suggest a metaphor of geologic time: the occasional dramatic disruption notwithstanding, most underlying change from election to election is incremental and even imperceptible.


The Disappearing God Gap?

As in the previous several presidential elections, essentially the same distribution of issue preferences, partisan identications, and voting blocs along religious lines emerged in 2008. As we have shown, religious afliation and level of traditionalism remained powerful predictors of attitudes on policy, turnout, and nal vote choice. In particular, partisan identications and partisan choices continue to be strongly associated with both religious traditions and religious traditionalismand these associations seem to be largely unaffected by the voters age or other characteristics, signifying a trend with staying power. In addition, the proportions of faith-based blocs in the McCain and Obama electorates were comparable to Republican and Democratic coalitions of past elections. Religion also helped shape the decisions of party elites and the rank and le early in the primaries and at other points in the process. In every state, candidates sought to piece together winning coalitions by appeals to key voting blocs. In some states, religion-based blocs were relatively inuential, and in tight races their efforts on behalf of candidates paid off. Mike Huckabees strength in some early states is illustrative. His campaign resonated with many religious conservatives, including a small, but tightly networked and well-organized, group of homeschooling families, giving him a boost in Iowa and key southern states and giving his competitors pause. By contrast, GOP candidates such as Rudy Guiliani, who favored abortion rights and civil unions for same sex couples, simply avoided competing in primaries and caucuses in which social conservatives formed an indispensable voting bloc. From the beginning of presidential history, citizens have been suspicious of candidates perceived to hold weak or unorthodox religious values, and 2008 was no exception. That hesitancy came through particularly in the primary elections. It helps explain concerns about Romneys membership in the Latter Day Saints church, the early successes in GOP primaries of the amiably evangelical Huckabee, the rumors about Obamas connections to Islam, and even the failed campaign of the uneasily religious Giuliani. In fact, the average voters insistence that presidential candidates must be religiousand religious in a mainstream wayis something akin to what political scientists call a standing decision, a nonnegotiable starting point for many voters in considering candidates for ofce.2 This stability across elections challenges the expectations of many that the God gap would either narrow or play no role in the 2008 contest. Some argued, for example, that the Democrats efforts to widen the range of religiously salient issues would attract new religious voters and perhaps close the gap altogether by realigning faith-based voting blocs (Dionne 2008a; Sullivan 2008a; Wallis 2008). Yet our ndings suggest that the
The God Gap Revisited


God gap persisted in 2008, as voting behavior and partisan identications continued to be strongly associated with a citizens level of religiosity. Moreover, there were few fundamental shifts in religion-based partisan alignments, and the effects of any new inux of religious (even the religiously unafliated) votersincluding younger oneswas marginal. Others suggested that a variety of nonreligious factorsObamas race, the low popularity of the incumbent president, the sour economywould overwhelm religion as a determinant of voting behavior in the general election. Religion, and more specically the God gap, would be simply irrelevant in the 2008 campaign. There is no doubt that the economy, the low popularity of George W. Bush, and Obamas considerable appeal mattered profoundly to voters. Nevertheless, even these factors did not signicantly alter the role of religious belief, behavior, and afliation in the 2008 election compared to previous elections. Evangelical Protestants were the cornerstone of the Republican vote, while the religiously unafliated and religious minorities anchored the Democratic coalition. Religious traditionalists generally voted for McCain and continued to identify with the GOP, modernists tended to do the opposite, while centrists remained swing voters who generally broke for the Democrats in this election. These are now familiar patterns in presidential contests.

Change in the Electoral Role of Religion

While continuity was apparent throughout the process, one need only scratch the surface to see indicators of change. Perhaps the changes are invisible to a casual observer, but they are still measurable, and their aggregate effect over time can be signicant. Consider, for example, the campaigns respective efforts at faith-based voter mobilization. It is a truism that turnout wins elections. Hence a variety of groupsinterest groups, parties, and candidates themselves devote enormous time and resources to mobilizing potential voters. They do not seek to mobilize everyone, only those most likely to vote for them. This means that campaigns must make decisions to focus on and put resources toward some groups rather than others. In the past, the conventional wisdom has been that Republican candidates were more likely than Democrats to nd fertile ground among people of faith. And there is considerable evidence in support of this perspective. In 2004, for example, the Bush campaign identied the need to mobilize evangelicals as paramount, and ultimately gained a ten


The Disappearing God Gap?

percentage point increase in evangelical voters over the previous election (Keeter 2007). The campaign itself used intensive microtargeting techniques to identify and communicate with those evangelical voters (Monson and Oliphant 2007). But in 2008, McCain adopted a different strategy. While seeking the votes of social conservatives, he did not rely as heavily as his predecessors on religious groups for turning out that bloc of voters. In fact, with a few exceptions, he usually appeared downright uncomfortable with faith-based appeals. But this relative lack of faith-based mobilization was not simply a reection of the personality traits of the GOPs standard-bearer. Importantly, other trends in religious mobilization appear to have nally caught up to the Republican Party. For over a decade, the workhorses of religious organizationsthe Christian Coalition, the Family Research Council, and other Christian Right groupshave been in decline. These groups made important contributions to Republican successes throughout the 1990s, most notably the dramatic victories in the House of Representatives and Senate in 1994 (Rozell and Wilcox 1995) and in various state legislatures (Cleary and Hertzke 2006). By 2008, however, these stalwarts of the Republican coalition had signicantly reduced their budgets and staffs and, in the case of the Christian Coalition, dismantled its chapter system across the countrya system that served as the backbone of mobilization efforts. Coupled with McCains rocky relationship with Christian Right leaders, these organizations difculties became McCains problems, too. This is not to say that diminished Christian Right organizations led inevitably to lower levels of faith-based mobilization in 2008. After all, the Republican Party and the McCain campaign itself tried to ll the void. But both the GOP and the McCain campaign faced their own resource problems, especially in comparison to the unprecedented size of Obamas war chest. Without the accustomed aid of robust Christian Right groups, the Republican Party was unable to use religion for a full-scale, precinctby-precinct, voter mobilization effort. In contrast to previous Democratic nominees and to the McCain campaign, the Obama campaign used significant resources to court religious voters both in the primaries and the general election. We did not, however, nd evidence of the same level of religious microtargeting by Obamas (or even McCains) campaign compared to Bushs in 2004. Perhaps this was largely a function of the fact that the religious groups Obama would most likely target are not as clearly identiable as Bushs traditionalist base; the Religious Left, as Olson (2007, 54) argues, is anything but a unied political movement.3 Still, Obamas campaign developed an active infrastructure for religious
The God Gap Revisited


mobilization, and that may have provided benets in an election that was decided at the margins. One area of strength in the Obama campaign and weakness in the McCain campaign pertained not to religious mobilization but rather to religious demobilization. Political scientists have long known that casting doubt among likely voters for ones opponent is often more effective for a candidate or party than generating support for ones own campaign (e.g., Burnham 1970). As Leege et al. (2002) argue, the GOP in recent decades realized that it lacked a coalition to rival the size of the Democratic Party, so Republican partisans used potent cultural symbols and values frequently reinforced with the language of religionto diminish the inuence of the Democratic coalition.4 As we suggest throughout this book, there is power in the campaign technique of framing, that is, dening the terms in which voters think about issues and candidates. In 2008, however, it seems that efforts at demobilization through framing and other techniques had less potency for the GOP than for the Democratic Party. Not even Obamas ties to Jeremiah Wright, his controversial pastor at the time, negatively affected his level of support in the primary process, despite the potentially explosive combination of race and religion that Wright represented. In fact, immediately after Wrights most inammatory comments were aired publicly, Obamas percentage of likely voters actually increased in some key primary states, and the association of Obama and Wright never seemed to gain traction as an election issue during the rest of the campaign. On the other hand, Obamas efforts to appear more open and welcoming to religious voters undermined any GOP efforts to portray his possible presidency as one that would be determined to reduce the role of religion in American politics and American public life. In addition to voter mobilization and demobilization, there were some subtle yet important differences in vote choice and voting coalitions. First, Obama and the Democratic Party enjoyed double-digit gains in voting among modernist evangelicals, centrist mainline Protestants, and black Protestants. While the relationships remained largely consistent with those evident in the past several elections, it was the magnitude of the change for each group that was unusually high. Moreover, turnout was notably higher among some key groups, thereby amplifying their inuence compared to previous elections. Again, black Protestants (along with the religiously unafliated, who formed over one-fth of Obamas winning coalition) are a key example of increased turnout from previous elections. A closer look at Obamas winning coalition hints at the intriguing emergence of ethnoreligious groupings as key blocs in presidential elections. African-American Protestants and, perhaps most signicantly, Latino


The Disappearing God Gap?

Catholics both turned out at higher levels and voted Democratic at higher levels than in previous elections. Of course, disentangling the effect of race from religion is very difcult, and it is entirely possible that both Latinos and African Americans were responding to the symbolism of Obamas racial identity (or some other set of factors). Nevertheless, we do see signicant variation within ethnic categories when divided by religion, most clearly represented by the contrast in vote choice and partisanship between Latino Catholics and Latino Protestants. Moreover, it is far from clear that the swelling of voter turnout among the various ethnoreligious groups in the 2008 campaign, largely in favor of Barack Obama, can be sustained now that a minority candidate has been elected presidentand it was these particular groups that largely provided Obama with his margin of victory. With Obamas election, the political playing eld has been permanently changed. Just as Catholics ocked to the polls to help elect John Kennedy in 1960, but did not turn out to vote at quite the same level in subsequent elections (nor were as prone to vote for some presidential candidate who was Catholic simply because of that candidates Catholicism), so, too, it is likely that subsequent candidates of the Democratic Party, and even Obama himself in his reelection campaign, will nd it more difcult to attain this swelling of voter turnout among religious minorities in future presidential contests.

Religion and Future Elections

Throughout this book, we have examined political differences between groups of voters based on their religion. Just how wide that political cleavage actually is depends on how religion is dened, whether in terms of religious traditions or religious traditionalism. But what is not clear is whether the political divisions related to the relatively new religious traditionalism (the politics of religious belief and behavior) will necessarily replace the old ethnoreligious divisions (the politics of religious belonging). What can these two perspectives teach us about the 2008 election and beyond? As we have noted in previous chapters, these two perspectives are not necessarily mutually exclusive. Recent elections, as well as the 2008 election, provide evidence for the political relevance of both ethnoreligious and theological factors (Green 2007; Green et al. 2007). How then will religion undergird electoral politics in the future? Will the relatively new cleavages of religious traditionalism continue to erode the old cleavages based on religious traditions, or will the divisions related
The God Gap Revisited


to religious traditionalism begin to dissipate and the old cleavage based on religious traditions reemerge? Or will they both become irrelevant? In addressing this matter, it is helpful to differentiate between the next several elections and elections a generation from now. In the long term, elections are likely to be shaped by macro forces such as demographic changes, domestic and international events, and the changes that prevail in American religious life, discussion of which would take us beyond the connes of this chapter. In the short run, however, the current situation is likely to continue namely, one in which both religious traditions and religious traditionalism shape electoral cleavages (Green 2007, 170).5 There are several important reasons for this. First, both parties have distinct religious constituencies within their respective coalitions. These constituencies are linked to each party on the basis of their issue priorities and their partisan identicationswith such ties having developed slowly over time. It is highly unlikely that these issue priorities or partisan identications will change dramatically in the short run. Second, candidates and political parties will work to exploit these differences, as they will seek to contact and mobilize those voters who are most likely to support their candidacies. Moreover, the development of microtargeting of communication messages enables candidates and parties to advance their particular messages among those religious voters most congenial to such messages. Still, there are short-term considerations that will likely affect the long-term relationship between religion and electoral politics. First, it remains to be seen whether, and to what extent, the Democratic Party will continue to seek to recapture and reconnect with religious voters, particularly those who may be otherwise inclined to support their candidates. Certainly, Obama sought to do so in 2008, and to the extent that he was successful he helped to narrow the God gap, even if only modestly. However, as we show in chapters 6 and 7, it is far from clear that it was the new faith-friendly Democratic voters who were responsible for Obamas election. Did religious voters respond to the God-talk and the religious outreach of the Democratic Party, or did they respond largely like many others voters who simply longed for some kind of change? Certainly, some religious voters (e.g., ethnoreligious minorities and religious modernists) voted for Obamaand in larger numbers than in previous elections. However, it is unclear whether they did so because of who Obama was and what he represented or because of his willingness to use God-talk and his efforts at mobilizing religious voters. It is also unclear now whether the 2008 election effort on the part of Obama and the Democratic Party will prove to be something unique to


The Disappearing God Gap?

the last election or something that will continue in future elections. Despite the partys success in 2008, there are antireligion voices within the Democratic Party who respond negatively to what some have labeled the religious industrial complex and who see themselves battling for the soul of the Democratic Party (Posner 2008). The highly diverse coalition of faith communities within the Democratic Party, including a signicant component of highly educated, relatively secular activists within the party, makes efforts at party unity challenging. Barack Obama, with his facility for framing morally complex issues in an ecumenically spiritual way (Posner 2008), was successful in doing sobut that may be something that other Democratic presidential candidates are far less able to accomplish. To what extent then was turnout in the 2008 election shaped by the unique persona of the winner? Were Barack Obamas unique personal story, religious sensitivity, and identity as a person of color so special as to have shaped and dened the 2008 election in ways that simply would not have happened without him? And even if not, are such efforts likely to occur with future Democratic candidates and as successfully? Moreover, even the emergence of some new issue agenda or the dressing of current political issues in more theological garb may not be sufcient to reduce the God gap in the near future. Green (2007, 168) has argued that issues are central to the political meaning of [the religion gaps] because they are a principal means by which religious groups are connected to voter coalitions. Some might therefore argue that, once issues that generally motivate religious voters (e.g., abortion) fall off the agenda, we might expect a diminution of the religion gap in future elections. But in many ways, the 2008 presidential election was a good test of this assertion. After all, many of the timely issues that confronted voters in the 2008 election were not the typical lightning rods for distinctively faith-based mobilization. The souring economy was the chief example. But even with the overwhelming importance of economic issues in 2008, the religion gaps persisted, and so did signicant variations within religious groups based on policy attitudes. Does that mean that the issue basis of political cleavages based on religion will continue in future elections? Perhaps. But to suggest that issues will matter is not necessarily to imply that the religion gaps will look as they currently do. Much depends on the framing of and mobilization around the issues. Few issues are inherently religious in nature. An issue only becomes religiously salient if it is framed in specic faith-based terms for an audience that is motivated to act. Often this framing is accomplished by labeling some aspect of the policy issue as a fundamental moral
The God Gap Revisited


concern. Many of the familiar issues that have animated the so-called values voters in the pastfor example, abortion and same sex marriagewere prominent for those voters in 2008. The 2008 election also witnessed the early and tentative steps by some religionists to redene additional issuesincluding the environment, the death penalty, and even the economy itselfin moral terms. While those efforts gained only marginal traction in 2008, they may portend a restructuring of religion gaps in the future. To see the possibilities, one need only look at other points in our history when issues as varied as alcohol consumption, communism, or the gold standard became matters of fundamental political importance to religious voting blocs. In some instances, the issue exploded onto the religion agenda; in others, it emerged slowly over time. But in any case, the point is simply that myriad issues have been redened by political elites or social movement activists in ways that mobilize religious voters. As a result, we can imagine that newly framed issues would not necessarily diminish political cleavages based on religion, but simply shift around the issue-based blocs that ll those gaps. This kind of issue framing has often driven the theological gap in electoral politicsthat is, the difference in voting patterns based on voters level of devotion and adherence to traditionalist beliefs and values. But we also see hints in 2008 of emerging ethnoreligious groupings that may have an impact on future elections. Consider Latino voters. Pollsters and pundits alike discussed in great detail the role of Latinos in the wake of the election. But what we nd is that the more specic category of Latino Catholic was an even more relevant political story. With their share of the total electorate at less than 4 percent (see table 6.6), their role may seem unimpressive. Yet their strong turnout in Obamas favor, as well as their geographic concentration in some key states, gave Latino Catholics an amplied voice in the election. It is an open question whether Latinos or any other demographic group will combine with distinctive religious identities to form new voting blocs in the future. Perhaps the emerging spirituality in American politics, which places greater emphasis on individual religious choice and expression than institutional afliation and authority (Hammond 1992; Fuller 2001), will open opportunities for new gaps in the religio-political landscape. If so, we would expect any partisan realignments to evolve gradually, rather than sweep through the electorate and shift in a single election.6 After all, for such shifts to be sustained, they usually need to be institutionalized. Because the emerging spirituality is ambivalent about religious institutions, it would likely need to build its own networks for mobilization.


The Disappearing God Gap?

Historically, the religious basis of political cleavages in American politics has been based on patterns of religious afliation. What difference does it make whether the divide is based on religious belonging or religious traditionalism? The consequences of the particular religious divide in American politics move far beyond the relative partisan advantages or disadvantages associated with it. While a political divide based on patterns of religious belonging carries with it certain potential problems, there are some potentially unfortunate consequences that ow from a political divide based on a religious traditionalismas it can all too easily lead to the perception that there is a religious-secular conict in American politics. Under such circumstances, political debate easily becomes more heated and relationships with those afliated with the opposite party less civil. Such polarization then gets translated into interparty conict, both as election-oriented organizations (Layman 2001) and as parties in government (Guth and Kellstedt 2001; Guth 2007),7 making it more difcult to deliberate during campaigns or represent interests broadly through governing. Moreover, this increased party polarization may well be linked to a decline in civility of American political debate (Layman, Carsey, and Horowitz 2006, 101), as well as to a declining interest in politics, lower levels of political trust, and ultimately to lower levels of engagement and participation among those who occupy the center of the political spectrum (Dionne 1991; King 1997; Shea 2003; Fiorina, Abrams, and Pope 2005). One response often given to these perceived dangers is to encourage the privatization of religion and moral convictions, making them irrelevant to public debate. But this answer only serves to foster a thin consensus in a complex society in which deeper values and traditions, once part of public debate, have been increasingly pushed into the realm of personal life (Walzer 1994). The basic assumption is that public order is best maintained in civil society by encouraging people to resist speaking explicitly about beliefs that require a religious commitment others might not share (Audi 1997). Yet by denying the appropriateness of expressing ones fundamental beliefs, this privatization of religious and moral convictions serves, in turn, to reduce the motivation for banding together, for organizing around deeply held values, and attempting to bring these commitments to bear on public issues (Wuthnow 1999, 22). From the beginning of American history, the practices of religion and democracy have been intertwined in an uneasy yet mutually benecial alliance. The promise of the institutional disestablishment of religion was twofold. On the one hand, it ensured that democracy could ourish
The God Gap Revisited


without the dominance of a single religious power. On the other hand, it made it possible for a wide plurality of religious traditions and perspectives to ourish by forbidding the state to support any specic religion. Still, disestablishment did not separate religion from politics, nor should it. All public policies embody particular normative goals and values; in this sense, all governments legislate morality. Whether or not they are derived from a particular religious faith, ones values affect the way one views the ends and purposes of government, the policy goals it should pursue, and the kinds of procedures it should employ. It is therefore inevitable that religious values, like all other values, help to shape political life. But if religion in publicand electorallife poses the danger of incivility and hard-to-bridge chasms and if banning religion to the private sphere is unacceptable, what is the answer? Our basic answer is to challenge the belief that religion necessarily leads to greater incivility and deeper and harder-to-bridge political debates. This often asserted conclusion can be challenged on two bases. First, there is the simple fact that some of the most bitter, and hard-to-reconcile, political divisions in American political history were not marked by religious versus secular or religion versus religion divides. Think, for example, of the bitter division over American military involvement in Vietnam during the 1960s and early 1970s which resulted in massive demonstrations, thousands of arrests, and students being shot down at Kent State University. Religion played little role in these struggles. Second, religion can be a source for reconciliation and a nding of common ground as well as source of division and bitterness. Lincolns second inaugural address used religious themes to urge reconciliation of a tragically divided nation; the civil rights struggles of the 1960s used religious values and motivations to draw the nation together in legislating equal rights for all Americans. Today, President Obama uses religious themes on occasion in an attempt to nd common ground on some divisive issues, and he uses theological discourse to gesture toward our common hopes and aspirations (Thiemann 2009). We are not saying religion in the public, electoral realm poses no dangers, only that seeking to ban religion from the public realm carries with it its own problems. The positive or negative role that religion playslike so many other forces in our societydepends on the forms and ways in which religion enters the public realm. The 2008 election was a watershed in many ways, not least of which was the victory of the rst African American as president of the United States. The same election, however, tapped into familiar themes about the


The Disappearing God Gap?

role of religion in American political culture. That religion can breed electoral conict is beyond question, but it is also important to remember its role as a source of unity and civic responsibility. In many ways, religion generates citizen participation in its highest forms, including the exercise of the franchise. Whatever its future shape, that role is unlikely to diminish.

The God Gap Revisited



1. To a certain extent, this is an overstatement. First, religious belief and religious afliation are related. In addition, early analyses of voting behavior tended to focus on social characteristics and group memberships, and such assessments may be a function, too, of the type of questions employed in survey research. Nevertheless, the work of ethnoreligious historians who focused on the linkages between religion and political behavior in the nineteenth century, which we discuss later, primarily emphasized the religious afliations of voters. 2. Of course, the relative proportion of social demographic groups evident among those who choose to turn out to vote can jump more dramatically from one election to the next. For example, it is not surprising, given the historic candidacy of Barack Obama as the nominee of a major political party, that the proportion of African Americans who voted in the 2008 presidential election was much higher than in previous elections. 3. Historically, mainline Protestantism simply reected the major denominations linked to the culturally dominant Protestant faith tradition within American society. However, following the fundamentalist-modernist split within many Protestant denominations in the 1920s, Protestantism in the United States became divided as many evangelical Protestants left their historic denominations and formed what they deemed to be more theologically orthodox denominations. In so doing, evangelical Protestantism became, at least in the short term, located largely outside the mainstream of American society, while mainline Protestantism remained at the center of American culture. Today, mainline Protestantism is largely linked to membership in the National Council of Churches. For a discussion of the differences between evangelical and mainline Protestantism, see Smidt (2007). 4. Ascriptive afliations represent afliation on the basis of birth rather than choice. In other words, to be Irish was to be Catholic or to be Norwegian was to be Lutheran.

5. Religious movements seek change. Some movements seek to return religious institutions to their previous practices or traditional elements of the religious faith, while other movements favor the adoption of new forms of belief or practice. For example, some Catholics are involved in a movement that seeks a return to the Latin Mass, while other Catholics are involved in movements that seek to allow priests to marry. 6. It is not uncommon for potential presidential candidates to provide a commencement address at a college or university associated with a religious group. For example, former Democratic governor of New York Mario Cuomo provided a commencement address at the University of Notre Dame, while Senator John McCain delivered a commencement address at Liberty University, as a means of trying to mend fences with certain religious constituencies. 7. Such quotes from the Bible and hymnals are frequently used in ways quite different from their original meaning. For example, this quote is taken from a hymn that says there is power, power, wonder-working power in the blood of the Lamb (emphasis added, with the blood of the Lamb referring to the blood of Jesus). 8. The next two paragraphs draw heavily from Wielhouwer (2009). 9. This survey was commissioned by the Paul B. Henry Institute for the Study of Christianity and Politics at Calvin College and was conducted by Opinion Access Corp. of Long Island, New York. The survey was funded, in part, through a grant received from the Lynde and Harry Bradley Foundation.

Chapter 1
1. The literature on political culture is quite extensive. The classic work is Almond and Verbas The Civic Culture (1963). A prominent contemporary analysis of political culture is Robert Putnams Bowling Alone (2000), which examines the role of civil society (including religion) in generating social capital, i.e., norms, values, and networks that help citizens work together on matters of mutual importance. 2. In this volume, the terms Religious Right and Christian Right will be used interchangeably. While the former term is more encompassing than the latter, the overall composition of the Religious Right is drawn predominantly from Christian sources. Consequently, since the two terms largely capture the same phenomenon, we use the two terms interchangeably for stylistic purposes. 3. There has been little change in key beliefs and behaviors over the past few decades. National surveys over the last half century document the relative stability of various facets of American religious life, including consistent belief in God (over 90 percent; Gallup and Lindsay 1999, 23; Braiker 2007) and religious service attendance (Smidt et al. 2008, tables 3.2 and 3.3). 4. On the wall metaphor, see Dreisbach (2002). 5. Data analysis by the authors. Source: Fourth National Survey of Religion and Politics, Bliss Institute, University of Akron, November 2004 (N = 4,000); Henry Institute National Survey of Religion and Public Life, AprilMay, November 2008 (N = 3,002).

Notes to Pages 1229


6. Henry Institute National Survey of Religion and Public Life, AprilMay 2008. 7. Ibid. 8. Moreover, clergy do not run afoul of the tax code if they wish to speak generally about key issues or political philosophies, provided they do not endorse specic candidates in the process. 9. Eisenhowers statement has been referenced in many places as evidence of various claims: that he was not personally religious, that he was eclectic in his own religiosity, that he sought to exploit religion in general, and so on. In context, however, it does appear that Eisenhower did have a particular type of religion in mind, that is, religious faith that is open to American political institutions and culture. See Henry (1981). 10. However, this perception of greater religiosity was based on examining responses only among those who felt they were able to rate the religiosity of the particular candidate. A majority of respondents (53 percent) reported that they knew nothing about Romneys religiosity. If these responses were added to the mix (as in table 3.1), Romney would not be viewed as the most religious. 11. It may be more than mere coincidence that both President Clintons approval ratings and President Bushs approval ratings rose substantially in the aftermath of the Oklahoma City bombings and the 9/11 attack. This would be consistent with the priestly function of civil religion. Some political observers have even suggested that the presidency has characteristics of an imperial institution. See, for example, Schlesinger (1973). 12. It is important to note, however, that this does not necessarily suggest the converse, i.e., that voters perceive that the Democrats are unfriendly to religion. A plurality of voters (42 percent) say that Democrats are friendly toward religion, and only 12 percent say they are unfriendly. The claim here about the GOP is relative: Republicans are perceived as being more friendly than Democrats toward religion.

Chapter 2
1. Historians have sometimes used the word ethnocultural instead to describe these groups. Here, the terms will be used interchangeably, though typically the word ethnoreligious will be employed to emphasize the religious nature of such groups. 2. While the Eastern Orthodox tradition is a separate Christian tradition, typically national surveys capture relatively few respondents who report afliations with some branch of Eastern Orthodoxy (e.g., Greek Orthodox, Russian Orthodox, Romanian Orthodox, Syrian Orthodox, etc.). As a result, those who report such afliations are placed in the other category. 3. While the label Hispanic/Latino is more correct, it is an awkward construction to employ. As a result, rather than employing both words at the same time, we use Hispanic and Latino interchangeably throughout the volume; the single term is intended to capture all those who fall within both designations. 4. The discussion in this section of the chapter draws in part from the commentary found in Guth (2009).


Notes to Pages 2945

5. The data for the table are drawn from the Fourth National Survey of Religion and Politics, conducted for the Bliss Institute for Applied Politics at the University of Akron. This survey not only contains a larger battery of religious measures than any other poll but also has a large postelection sample (N = 2,730) that allows us to pinpoint the nature of religious voting. (Henceforth references to the Fourth National Survey or the 2004 Survey of Religion and Politics indicate this survey, including in tables throughout the book.) 6. This designation is used, at times, to indicate that black Catholics are included within the larger Catholic category. 7. More specically, traditionalists can be characterized by adherence to historic beliefs of their faith, high levels of religious observance, and identication with sectarian religious movements; modernists subscribe to more heterodox religious beliefs, are less religiously observant, are more likely to identify with more liberal or ecumenical religious movements, and are more eager to include modern beliefs and practices within their religious worldview; centrists then fall between traditionalists and modernists in these matters. Seven identical survey questions found in both the 2004 National Survey of Religion and Politics and in the 2008 Henry Institute National Survey of Religion and Public Life were used to construct the traditionalism measure in each of the two surveys: views of Gods existence, views on biblical authority, views of salvation, level of church attendance, frequency of private prayer, level of participation in religious small groups, and level of religious giving. A factor analysis of these seven items in both surveys revealed a single underlying dimension of traditionalism, as a principal components factor analysis produced a single factor with an eigenvalue greater than 1.0. The resultant factor scores were then divided into thirds, with the third exhibiting the highest factor scores labeled traditionalists, the middle third centrists, and the lowest third modernists. As a result, modernist mainline Protestants are those mainliners by denominational afliation whose scores on the traditionalism factor fell within the bottom one-third of such scores. Previous analyses have shown that these categories are quite robust, with different means of calculation producing very similar results. For examples of alternative approaches, see Guth et al. (2006) and Green et al. (2007). 8. Political socialization can be dened as the process by which citizens learn about political leaders, governmental institutions, and political processes, and acquire their political beliefs and practices (Pearson-Merkowitz and Gimpel 2009). 9. The Fifth National Survey of Religion and Politics, conducted in the summer of 2008 did not nd such a marked movement away from the GOP among mainline Protestants, instead revealing a more even division between identication as Democrats and Republicans (private correspondence with John Green). It is not clear whether such differences are a function of (1) different house effects associated with sampling strategy of the different survey organizations employed, or (2) the employment of similar, but not necessarily identical, protocols in assigning respondents to the mainline Protestant category.

Notes to Pages 4653


10. Religion may also shape the ways in which different issue positions are interrelated and held together. However, we have chosen not to address this particular matter in this chapter. 11. Because our 2008 survey was limited in the number of political issue questions asked, we have supplemented issue questions asked in 2008 with certain issue questions posed in 2004. When identical questions were asked in both years, both distributions are provided. 12. There is some variation in what the types of issues we consider in this section are called. Sometimes they are referred to as social morality issues or simply moral issues. We prefer the term cultural issues, since it best captures the cultural, value-ladenness of the issues considered here. Social morality seems too broad a term, since issues surrounding opposing poverty and environmental pollution, for example, can also be seen as having both social and moral dimensions. The term moral issues is even broader and could include issues of war and peace, the use of torture, HIV/AIDS assistance, and trade policies, since all such issues haveat least in our minds and those of many othersa clear moral dimension. 13. This widespread support for these practices continued even though prayers at public school graduation ceremonies were held to be unconstitutional by the Supreme Court in 1992 (Lee v. Wiseman, 505 U.S. 577 [1992]), and in several closely decided cases the Supreme Court has at times approved religious displays in public places while sometimes holding them unconstitutional (e.g., Van Ordern v. Perry, 545 U.S. 677 [2005] and McCreary County v. ACLU of Kentucky, 545 U.S. 844 [2005]). 14. Mara Vanderslice had earlier served as director of religious outreach for John Kerrys presidential campaign. Eric Sapp had been an aide to Senator Edward Kennedy on the Health, Education, Labor and Pensions Committee and then later worked for Representative David Price (D-NC). 15. While the term white is somewhat problematic, Euro-American or even Caucasian is too cumbersome when used regularly in the text.

Chapter 3
1. Candidate Howard Dean only took such a step during the 2004 primaries; John Kerry, the eventual nominee, only reluctantly did so during the general election campaign. 2. For example, in his campaign for the presidency in 2000, McCain criticized Rev. Jerry Falwell and Bob Jones University. 3. The survey did not ask respondents about their perceptions of the religiosity of Huckabee nor their evaluations of him. Consequently, perceptions related to the religiosity of Huckabee cannot be examined here. 4. The other two response options were mostly unfavorable and very unfavorable. 5. Campbell Brown in her words of introduction to the event. 6. This survey is described in fuller detail at the conclusion of the introduction.


Notes to Pages 5699

Chapter 4
1. Technically speaking, one simply presumes, given the primary election outcomes, that McCain and Obama will be the nominees for president of their respective parties. The certainty is established only after the parties national convention delegates have voted ofcially to make each candidate their nominee. 2. Black liberation theology has some similarities to, but also some differences from, the more widely known liberation theology. The latter became prominent in the Latin American Church in the 1960s. As is the case with black liberation theology (sometimes called simply black theology), it emphasizes the need to offer relief to the poor, oppressed, and disenfranchised of this world. However, black theology emphasizes the interpretive framework of the Exodus experience for African Americans, given their history of slavery in the United States, while liberation theology has a very strong anticapitalist, Marxist underpinning, which is not the case with black liberation theology. Due to these Marxist leanings, liberation theology was condemned by Pope John Paul II and has subsequently waned in inuence. In addition to the Cone interview with Terry Gross cited later in this paragraph, see Cone (1970). 3. A plurality (44 percent) responded that the Democratic Party was neutral, while more than one-fth (21 percent) indicated that they thought the party was unfriendly toward religion. The corresponding gures for the Republican Party were 33 percent who perceived the party to be neutral toward religion and only 8 percent who thought the GOP was unfriendly toward religion (data not shown). 4. Actually, at an earlier 2006 Saddleback Global Summit on AIDs and the Church, Hillary Clinton had appeared in person, while John McCain, Barack Obama, Mitt Romney, Mike Huckabee, and John Edwards had participated via video due to scheduling conicts. At the summit, the various candidates were asked to share their plans for addressing HIV/AIDS if they were elected president (Warren 2008). 5. Certainly, in terms of abortion and same sex marriage, Rev. Warren can be labeled a social or cultural conservative. For example, prior to the 2004 election, he authored a letter to pastors emphasizing the need to combat gay marriage and abortion rights, and Warren was an active supportive of Proposition 8 on the California ballot in 2008, which amended the state constitution to ban same sex marriage. On the other hand, he has also been vocal and active in efforts to address problems like AIDS, global poverty, and the environment. In fact, Warren had been selected by Time as one of 15 World Leaders Who Mattered Most in 2004 (Steptoe 2005), and Newsweek designated him in 2006 as one of 15 People Who Make America Great, an award given to people who devote themselves to helping others (Adler et al., 2006). 6. The study examined religion-focused election coverage in forty-eight different news outlets between June 1 and Oct. 15, 2008. The full report can be found at 7. In many ways, one can argue that the choice of ones running mate is itself a reection of the nominees public theology. Public theology in the sense used

Notes to Pages 107124


here overlaps with the concept of social theology (Guth et al. 1997; Hoover 1998; Smidt 2004), a term that recalls the Social Gospel movement at the turn of the twentieth century (Rauschenbusch 1907). As generally used today within social science literature, the term refers to how religious people understand social action or civic engagement in light of their theological understanding (Putnam 2000; Smidt 2003). 8. One of many such lists on blogs appeared on Justin Quinns U.S. Conservative Politics Blog dated June 30, 2008: Draft Sarah Palin! 9. See (accessed December 18, 2008). 10. Sermons at the Wasilla Bible Church tend to avoid politics, though a minister at the Wasilla Assemblies of God Church famously questioned whether those who voted for John Kerry in 2004 would be saved. See alaska/wasilla.html (accessed December 18, 2008). 11. She could also be described as the rst ex-Catholic nominee. It is interesting to note that conservative Catholic blogs discussed why she left Catholicism, relating it to the unsettled nature of Catholicism just after the Second Vatican Council ended in December 1965, before her second birthday (see http://www and http://, for example). Others cited certain connections Palin has as anti-Catholic ( 12. It is exactly this issueabortionthat had led to Biden being named by many conservative Catholicsnotably many bishops as wellas being unacceptable, precisely as a Catholic who is seen as not following the demands his faith would place on him as a politician. 13. 14. This is not to say that concerns about morality policy were nonexistent in the 2008 elections. Indeed, three statesArizona, Florida, and Californiapassed ballot initiatives that banned same sex marriage, and several other states addressed issues from abortion to gambling to the right to die. The claim here is a relative one: these kinds of cultural issues did not affect presidential vote choice to the same degree as previous elections.

Chapter 5
1. The announcement of the nominees choice for running mate has not always been made prior to the partys convention. However, during the past several decades, this has been the predominant pattern. 2. In the 1970s, a revolution in the party presidential nomination process was initiated through the widespread adoption of state presidential primaries as a means to select delegates to the national convention. And whereas the proceedings of each partys convention received gavel-to-gavel coverage (every day of the convention from its opening in the morning through its closing at night) by all three major television networks during the 1950s and 1960s, today there is far less continuous, and less network, coverage of the conventions.


Notes to Pages 125135

3. For example, the Democratic National Convention Committee (2008) reported that it expected some 15,000 members of the media to attend the 2008 Democratic National Convention. 4. A fairly exhaustive web-based search revealed no media coverage of specic religious events at the Republican National Convention. 5. This theme might also have been implicitly linked to the theme of unity, as perhaps the best we can do initially is simply hope (and then work) for unity. 6. This statement is also a New Testament allusion. See Hebrews 11:1: Now faith is the substance of things hoped for, the evidence of things not seen. 7. This biblical story can be found in Genesis 4. The quote is from verse 9. 8. In drafting this portion of the chapter, we primarily used the 2008 Democratic and Republican national party platforms and the McCain and Obama campaign websites in analyzing the policy positions taken in the 2008 campaign. 9. However, despite advancing these specic policy proposals that looked toward government to address various economic problems, Obama tried to appeal to more conservative segments of the American people by acknowledging that government cannot solve all our problems and that more than money is needed to fulll Americas promise; it will require a renewed sense of [personal] responsibility, as the essence of Americas promise entails individual and mutual responsibility (Obama 2008b). 10. In fact, the Republican convention and some of his rallies were marked by chants of Drill, baby, drill! 11. This characterization of Obamas position is subject to some dispute. For a background to the particular bill and differing interpretations of what it sought to accomplish, see Zorn (2004). 12. The Obama campaign also failed to combat effectively the charge that he opposed giving protection to infants who survive abortions (Waldman 2008b). 13. However, in his acceptance speech at the Democratic convention, Obama once again sought to nd common ground by stating, We know there are differences on same-sex marriage, but surely we can agree that our gay and lesbian brothers and sisters deserve to visit the person they love in a hospital and to live lives free of discrimination. 14. At the time, this information related to his presidential campaign was on www The website now features materials related to his reelection to the Senate. 15. By executive order, Bush was able to establish the White House Ofce on Faith-Based and Community Initiatives as well as to establish ofces within various departments of government that worked, in part, to ensure that religious groups were not penalized in their activities or in their applications for funding of social service programs simply because they were religious in nature. 16. The eld department is responsible for necessary on-the-ground organizing that allows staff and volunteers to contact voters personally through canvassing, phone calls, and hosting local events.

Notes to Pages 135151


17. Moreover, in those eighteen states designated by the Obama campaign as battleground states, the campaign had a full-time, paid constituency director and some of these states also had a person responsible for outreach to religious groups. 18. This statement is technically accurate, as religious outreach was tied to outreach to social conservatives (as reected in the title of Marlys Popma). However, Marlys Popma and Joshua Lynch were already serving on the McCain campaign by mid-May. Nevertheless, McCains campaign staff who focused on outreach to different faith communities were far fewer in number, and more narrowly focused, than similar staff of the Obama campaign. 19. It may have also been a function of having less funding available for use in the campaign. See the discussion later in the chapter related to the different levels of funding available for use in the Obama and McCain campaigns. 20. This is due to several factors: polling results may vary, popular preferences may shift over the course of a campaign, and observers may disagree over how populous a state must be before it warrants the label battleground state. 21. Toward the end of the campaign, however, some observers concluded that Michigan was solidly in the Obama camp (and, indeed, McCain pulled his resources from that state) and that two smaller states, Indiana and Nevada, were approaching battleground state status. 22. Obama was clearly more successful than McCain in fund-raising. During the 2008 portion of this presidential election, Obama was able to raise a total of $745 million for his campaign compared to McCains $368 million. Moreover, 88 percent ($656 million) of Obamas funding came through individual contributions, while only 54 percent ($199 million) of McCains did so (Center for Responsive Politics 2009). 23. 24. Moreover, for weeks, the McCain-Palin campaign had to deal with the fallout from the disclosure that the Republican National Committee was billed for $150,000 in wardrobe purchases for the Palin familya discovery that was widely ridiculed and undercut Palins hockey mom appeal.

Chapter 6
1. Scholars sometimes differentiate between internal and external mobilization. Internal mobilization relates more to psychological dispositions to participate (e.g., interest in the campaign), while external mobilization focuses more on the efforts outside the individual to get that person to the polls. In this chapter, we focus predominately on external mobilization. 2. To address these matters, this chapter draws upon data from the Henry Institute National Survey on Religion and Public Life, which in its postelection survey asked respondents a series of questions related to campaign contacts they received during the election campaign. This national survey is an ideal source for analyzing religious mobilization as individual voters experienced it. It is a random sample telephone survey with 3,002 respondents conducted from April 8 to May 10, 2008, with 1,515 of the original respondents resurveyed in the three weeks following the election.


Notes to Pages 152165

3. Conway (2005, 52) notes that door-to door canvassing costs between $7 and $17 per vote, with only one additional vote added for every fourteen contacts made. 4. Respondents were also asked whether or not they had been contacted by specically Christian groups. For our discussion here, we have combined responses to moral and religious groups and specically Christian groups and have simply labeled them religious groups. 5. See Hudson 2008. For the denial of communion incident see Dionne 2008b. As Dionne points out, like many Catholic Democrats, Kmiec was profoundly attracted to Ronald Reagan. For him, ve words in Reagans 1980 acceptance speech summarized the essence of a Catholic view of politics: family, work, neighborhood, peace and freedom. In an interview with Dionne, Kmiec argued that, thirty-ve years after Roe, opponents of abortion need to contemplate whether a legal prohibition of abortion is the only way to promote a culture of life. . . . To think you have done a generous thing for your neighbor or that you have built up a culture of life just because you voted for a candidate who says in his brochure that he wants to overturn Roe v. Wade is far too thin an understanding of the Catholic faith. Kmiec, a critic of the Bush administrations Iraq policy, added that Catholics should heed the broad social teaching of the church, including its views on war. 6. While books on the left-liberal side of the spectrum were seemingly much more numerous, they were not the only new materials on religion and politics. Deal Hudson, director of, contributed his Onward, Christian Soldiers: The Growing Political Power of Catholics and Evangelicals in the United States (March 2008). Another Catholic voice, known for his strong words against politicians supportive of abortion, is Catholic Archbishop Charles Chaput of Denver who wrote Render Unto Caesar: Serving the Nation by Living Our Catholic Beliefs in Public Life (August 2008). 7. Perhaps the clearest illustration of the long-established conventional wisdom that authorship of a successful book can establish credentials of movement leaders is Jim Wallis, whose book The Great Awakening: Reviving Faith and Politics in a PostReligious Right America (January 2008) launched the season of book releases on these topics in the election year that would culminate in the election of Obama as a religious progressive. Walliss Gods Politics (2005) had helped start the momentum that led to the Call to Renewal Convention where Barack Obama rst addressed religious progressives in 2006. 8. Faith in Public Life was founded in 2006. As with these other organizations, it was formed in the aftermath of the 2004 election, amid concern that progressive voters of faith were not being well represented. See Matthew 25 was led by Mara Vanderslice, director of religious outreach for the Kerry-Edwards campaign in 2004 (see Benen 2008). As with most of these organizations, it continued to exist after the election, taking on a new role in public policy debates. See The National Congress of Black Women and the National Council of Womens Organizations established the Church Ladies Project as an effort to engage African-American women in

Notes to Pages 168174


religious organizations to take electoral action. See http://womensorganizations .org/index.php?option=com_content&task=view&id. For Faithful Democrats, see &id=982&Itemid=79. The organization was founded in 2006. For Catholic Democrats, whose mission is to work within the Democratic Party to bring about a more just ordering of society, see Catholics United was founded in August 2004 as one of the rst such organizations. See Catholics in Alliance for the Common Good ( was founded in 2005 by Alexia Kelley, who had previously worked for the Democratic National Committee as director of religious outreach from October 2004 to early 2005 ( and, before that, for the Catholic Campaign for Human Development. 9. Indeed, most of the new progressive religious groups are not likely to be mass membership organizations, but rather ones composed of leaders and a small number of better-educated supporters disaffected by the Christian Right. 10. The Christian Coalitions voter guide may be found at See also Many other voter guides were produced by conservative Christians, such as that found at index.html. On the other hand, FRC and the Christian Coalition have successfully built themselves into a voter guide brand. Even while the Christian Coalition is a shadow of its former self, it continues to produce and disseminate its voter guide, so central is the guide to its role. See Cooperman and Edsall (2006). 11. This is because Catholic theology considers the nature of the church to be tied to particular churches, understood as a group of congregations linked in communion to a bishop as successor to the apostles. In this light, the role of the local diocesan or eparchial bishop means that Catholic parishes are not the same as independently operated congregations, particularly in the case of the evangelical congregations that usually distribute Christian Coalition or FRC voting guides. There are 177 territorial dioceses of the Western, or Latin Rite, Catholic Church in the United States whose bishops belong to the United States Conference of Catholic Bishops. There is also a separate (Latin Rite) diocese for the military, and another 17 eparchies of Eastern Catholic Churches (including the Byzantine, Maronite, Melkite, Ruthenian, and other churches). See comm/statisti.shtml. 12. The most recent edition of this statement was issued in November 2007 and may be accessed at 13. These include The Voters Guide for Serious Catholics issued by Catholic Answers, a Catholic apologetics organization based in San Diego. The full text is available at Other materials include A Brief Catechism for Catholic Voters, issued by conservative Catholic broadcaster EWTN in 2002, catechism.htm, and How to Vote Catholic at guide.html.


Notes to Pages 175176

14. This nding is also conrmed in a report from the Pew Research Center for the People and the Press (2008b). 15. In addition, there was also a decline in voter registration activities in houses of worship (from 7 percent in 2004 to 4 percent in 2008). These data, however, are not included in the discussion or analysis of this chapter. 16. The data for both surveys are weighted to reect national turnout rates as survey respondents typically overreport their level of turnout on Election Day. 17. The exceptions to this pattern are modernist evangelicals who surpassed centrist evangelicals in voter turnout in 2004 and modernist Roman Catholics who surpassed centrist Roman Catholics in voter turnout in 2008.

Chapter 7
1. Another issue in the 1932 election, repeal of Prohibition, was arguably tied to the Catholic vote, with especially strong support within urban European ethnic working classes. The passage of Prohibition had always been very largely a Protestant issue, and its defeat could be seen in ethnoreligious terms. 2. For example, the proportion of those expressing pro-life positions among those citing social issues in 2004 was greater than the proportion in 2008. 3. Much early survey research, however, used rather broad categories of religious afliation (e.g., Protestant, Catholic, Jew), but even these broad categories did reveal differences in voting patterns across religious categories. 4. In large part, this was due to the sizable proportion of evangelicals who resided in the South and were highly Democratic in their partisan identications. 5. Perhaps the most important shift took place among mainline Protestants, as they moved from being marginally Republican in their voting in 2004 to more fully Democratic in their voting patterns in 2008. 6. Reported results of the 2008 National Survey of Religion and Politics generally reveal very similar results to the ndings presented here, but they differ with regard to three patterns: the Henry Institute survey reveals modernist evangelicals and centrist mainliners to have voted Democratic by sizable margins, while the National Survey of Religion and Politics reveals that only 45 percent of modernist evangelicals and 51 percent of centrist mainliners cast their ballots for Obama. On the other hand, the Henry Institute survey reveals that nearly half of centrist Roman Catholics voted for Obama, while the National Survey of Religion and Politics reveals that only 34 percent did so. See Green (2009). 7. These traditions are predominantly white in the sense that our analysis has removed Hispanics Protestants and Hispanic Catholics from their ranks, with black Protestants constituting another separate religious tradition in our analysis thereby, in terms of this classication system, leaving the resulting members of these traditions predominantly white. 8. Technically, the values for the measure of association reported in table 7.4 are eta statistics derived from one-way analysis of variance. The dependent variable was the mean score for the ethnoreligious groups derived from a dichotomous variable in which a vote for Obama was scored a value of 0 and a vote for McCain was

Notes to Pages 177204


scored a value of 1. In the case of dichotomous variables scored in this fashion, the mean score is also the percentage of respondents who exhibit that particular characteristic, and hence the data are presented as percentages rather than as mean scores. 9. Some of the difference between single women and married women in their level of support for Democratic candidates is attributable to the larger portion of black women, who vote overwhelmingly Democratic, within the single women category. 10. Multiple Classication Analysis (MCA) enables one to use categorical variables in multivariate analysis and avoids the necessity of using dummy variables in the analysis. In this way, one can obtain a summary beta value for the nominal-level variable as a single variable, as opposed to beta value for particular categories of such variables estimated in comparison to the suppressed category of the nominal-level variable. 11. Of course, political scientists have long recognized political party identication as the single most important variable shaping voting decisions. Should party identication be entered into this multivariate analysis, it would overwhelm the relative inuence of all the sociodemographic variables involved. However, when these same social variables are used to account for variation in political party identications, the religious tradition variable far outweighs all the other sociodemographic variables in explaining differences in partisan identications. 12. The Anti-Semitism Study of 1964 was a national survey (N = 1,975) conducted in October 1964 by Charles Glock, Gertrude Selznick, Rodney Stark, and Stephen Steinberg of the University of California at Berkeley. The primary focus of the study was the examination of various religious and personal beliefs, attitudes, and behavior that might contribute to anti-Semitic orientations. 13. The variables used to create the traditionalism measures included the only religious behavior question asked in the survey, church attendance, along with four religious belief items: certainty of belief in God, salvation through Jesus Christ alone, certainty of existence of the devil, and certainty of life after death. These ve items were factor analyzed, and they loaded on one factor. As we had done with the previous traditionalism measures analyzed, we once again divided the resulting factor scores into three equal parts, with traditionalists being classied in terms of the third with the highest factor scores and modernists the third with the lowest such scores. 14. In terms of the analysis presented here, those who claimed to be independents but reported that they leaned toward a political party were classied as identiers with that party. 15. Of course, one can categorize age differences in various ways. The specic age categories that are employed in table 7.10 were chosen to reect the same age categories that were employed in exit poll reports for the 2008 election. 16. There were some media reports immediately following the election that Obama had doubled his support among young white evangelicals (those ages 18 to 29) compared to Mr. Kerry. The report did not provide any data as to what percentage of young evangelicals were found to have voted for Obama, and


Notes to Pages 207218

though not stated explicitly, it is likely that these gures are based on exit poll data that are still not publicly available. It should be noted, however, that the operation measure employed to identify evangelicals in exit polls is not based on denominational afliation, which might well account for the discrepancy between the survey data analyzed here and these media reports. See Goodstein (2008b). 17. It is not totally clear why younger Catholics were more likely to have voted for McCain than older Catholics. Part of the explanation may be the relative number of non-Hispanic Catholic respondents who fell into the eighteen to twenty-nine age category (N = 19), as smaller numbers contribute to increased instability in the resulting percentages. Moreover, this higher level of reported voting for McCain among young non-Hispanic Catholics cannot be attributed to patterns of religious traditionalism, as none of these younger Catholics were traditionalists in terms of their religious beliefs and behavior.

Chapter 8
1. Social scientists sometimes use the metaphor of a path to describe how actors respond to institutional processes such as elections (e.g., Pierson 2004). At any given point, a candidates or voters decision in response to the rules of the road or new information or events can have profound effects farther down the path. Some scholars even speak of elections and other political and economic processes as path dependent; institutional rules and the choices of actors dene the path that profoundly shapes future options. While the concept of path dependence is controversial (Thelen 2004), the metaphor of a path nevertheless has the virtue of reminding us that process matters in electoral campaigns. 2. The concept of a standing decision has generally been applied to the role of party identication in determining vote choice among loyal partisans (e.g., Key 1966). 3. For an examination of some of the ambiguities related to the notion of the Religious Left as well as some of its internal diversity, see Kellstedt et al. (2008). 4. It was not that Republican partisans chose to create cultural issues to use. Rather, these cultural issues arose largely because citizens and groups sought changes in existing policies (e.g., the right to an abortion, removal from public schools of the collective reciting of an opening prayer to open the school day, same sex marriage, or even getting some form of creationism into the curriculum of public schools). But once these cultural issues arise, party leaders can then choose to either employ them for partisan advantage or ignore them and focus on other matters that may provide bigger dividends politically. 5. The discussion of this paragraph draws heavily on the discussion found in Green (2007, 170173). 6. Political scientists call this kind of realignment secular, in contrast to critical. The usage of secular here should not be confused with its other connotation of nonreligious. 7. The discussion in this paragraph draws heavily on that in Layman, Carsey, and Horowitz (2006, 101).

Notes to Pages 218231



Abramowitz, Michael, and Robert Barnes. 2008. Candidates Swap Money Punches. Washington Post. October 21. http:/ headline/nation/ 6071382.html. Accessed October 22, 2008. Abramson, Paul R., John H. Aldrich, and David W. Rohde. 2006. Change and Continuity in the 2004 Elections. Washington DC: CQ Press. Adler, Jerry, et al. 2006. The Giving Back Awards: 15 People Who Make America Great. Newsweek. July 3: 5586. Allen, Jr., John. 2008. McCains VP Choice a Womanand a Post-denominationalist. National Catholic Reporter. August 30. Accessed December 18, 2008. Almond, Gabriel, and Sidney Verba. 1963. The Civic Culture: Political Attitudes and Democracy in Five Nations. Princeton, NJ: Princeton University Press. Ammerman, Nancy. 1997. Congregation and Community. New Brunswick, NJ: Rutgers University Press. Audi, Robert. 1997. Liberal Democracy and the Place of Religion in Politics. Pp. 166 in Audi and Nicholas Wolterstorff, eds., Religion in the Public Square. Lanham, MD: Rowman & Littleeld. Axelrod, Robert. 1972. Where the Votes Come From: An Analysis of Electoral Coalitions, 19521968. American Political Science Review 66 (March): 1120. Bacon, Perry, Jr. 2007. Clinton, Edwards and Obama Discuss Their Faith at Forum; All Say Religion Informs Politics, Personal Lives. Washington Post. June 5: A3. . 2008. Huckabee Steps Back into the Pulpit at Evangelical Church in N.H. Washington Post. January 2: A7. Bacon, Perry, Jr., and Juliet Eilperin. 2008. Huckabee: Evangelical Christians Now Have a Chance to Lead GOP. Washington Post. January 13: A17. Bacon, Perry, Jr., and Shailagh Murray. 2008. Opponents Paint Obama as an Elitist; Clinton, McCain Try to Score Off Bitter Remark. Washington Post. April 12: A4. Bacon, Perry, Jr., and Michael D. Shear. 2008. Romney Releases His Delegates, Backs McCain. Washington Post. February 15: A06.

Bailey, Holly. 2007. Perfect Stranger: On Paper, Mike Huckabee Could Be the GOP Front Runner; In Reality, He Barely Registers. Newsweek. September 3: 36. Balmer, Randall Herbert. 2007. God in the White House: A History: How Faith Shaped the Presidency from John F. Kennedy to George W. Bush. New York: HarperOne. Balz, Dan. 2007. In Speech on Faith, Romney Vows to Serve No One Cause. Washington Post. December 7: A1. . 2008a. Little New Hampshire Could Hold Signicance for Both Parties. Washington Post. January 8: A1. . 2008b. White Male Vote Especially Critical; Questions Remain over Obamas Ability to Appeal to Demographic. Washington Post. March 17: A5. . 2008c. Decisive Win Cant Forestall a Daunting Task. Washington Post. April 23: A1. . 2008d. Fla., Mich. Delegates Each Get Half a Vote; Compromise Prompts Anger from Clinton Campaign. Washington Post. June 1: A1. Balz, Dan, and Jon Cohen. 2006. Clinton and Giuliani Have the Early Edge for 08, Poll Shows. Washington Post. December 14: A3. . 2008. McCain Moves into Lead; Obama Gains on Clinton; Giuliani Falls to 4th in National Poll. Washington Post. January 14: A1. Balz, Dan, and Juliet Eilperin. 2008. GOP Race Is Close in S.C.; Huckabee and McCain Lead; Winner Will Receive Critical Boost. Washington Post. January 19: A1. Balz, Dan, and Anne E. Kornblut. 2008a. Edwards, Giuliani Withdraw from Race; Ex-NY Mayor Endorses McCain. Washington Post. January 31: A1. . 2008b. McCain Dominates Big States; Clinton and Obama Trade Victories; Democrats: NY Senator Withstands Push by Surging Rival in Key Battlegrounds. Washington Post. February 6: A1. . 2008c. Obama Claims Nomination. Washington Post. June 4: A1. Balz, Dan, Anne E. Kornblut, and Shailagh Murray. 2008. Obama Is Big Winner in S.C.; Clinton 2nd after Bitter Campaign. Washington Post. January 27: A1. Balz, Dan, and Shailagh Murray. 2008a. Democrats Set to Vote in S. Carolina; Obama Leads, but Polls Show a Racial Divide. Washington Post. January 26: A1. . 2008b. Results Refocus Democratic Campaign; Clinton, Obama Clash over Superdelegates. Washington Post. March 6: A1. Balz, Dan, and Michael D. Shear. 2007. GOP Field United On War, Divided on Social Issues. Washington Post. May 4: A1. Banerjee, Neela, and Michael Luo. 2008. McCain Cuts Tie to Pastors Whose Talks Drew Fire. New York Times. May 23: A17. Banks, Adelle M. 2008. McCains Faith Journey Largely Unspoken. Religion News Service. August 20. Barker, David C., Adam B. Lawrence, and Margit Tavits. 2006. Values, Partisanship, and Candidate Centered Politics in U.S. Presidential Nominations. Electoral Studies 25: 599610. Bartkowski, John, and Helen Regis. 2001. Charitable Choices: Religion, Race, and Poverty in the Post-welfare Era. New York: New York University Press.



Belkin, Douglas. 2008. Democrats Highlight Their Religion in Voter Outreach. Wall Street Journal. September 18. .htmlmod=rss_Politics_ And_Policy. Accessed January 16, 2009. Bellah, Robert Neelly. 1967. Civil Religion in America. Daedalus 96 (1): 121. Bellantoni, Christina. 2008. Clinton Idol Tells Her to End Race; McGovern Moves to Obama Camp. Washington Times. May 8: A1. Benen, Steve. 2008. Obama, Evangelicals, and the Matthew 25 Network. June 11. The Carpetbagger Report, Accessed March 5, 2009. Berelson, Bernard R., Paul F. Lazarsfeld, and William N. McPhee. 1954. Voting: A Study of Opinion Formation in a Presidential Campaign. Chicago: University of Chicago Press. Berkowitz, Jeff. 2008. Two Major Reasons Why McCain Lost the Presidency. Public Affairs. October 29. Accessed February 2, 2009. Biden, Joseph. 2007. Promises to Keep: On Life and Politics. New York: Random House. Blunt, Sheryl Henderson. 2007. The Giuliani Choice: Conservative Leaders Doubt His Support among Evangelicals Will Last. Christianity Today. June: 20. Bolce, Louis, and Gerald DeMaio. 2002. Our Secularist Democratic Party. The Public Interest 149 (Fall): 320. Boller, Paul F. 1984. Presidential Campaigns. New York: Oxford University Press. Bolton, Alexander. 2006. Clinton Hires Faith Guru. The Hill. December 13. http:// html. Accessed November 29, 2008. Boorstein, Michelle. 2008. Altar Egos; Clergymen, Once Courted by Candidates, Have Become Liabilities, Raising Questions about the Role of Religion in Politics. Washington Post. June 7: B9. Boyd, Gregory. 2007. The Myth of a Christian Nation: How the Quest for Political Power Is Destroying the Church. Grand Rapids, MI: Zondervan. Brachear, Manya. 2008. How Religion Guides Palin. Chicago Tribune. September 6.,0,3371419. story. Accessed December 17, 2008. Bradberry, Leigh. 2009. The Religious Dimension of Voting in the 2008 Presidential Primaries. Paper presented at the Fifth Biennial Paul B. Henry Institute Symposium on Religion and Politics, Calvin College, Grand Rapids, Michigan, April 2325. Braiker, Brian. 2007. Gods Numbers. Newsweek. March 30. http// id/17879137/ site/newsweek. Accessed October 1, 2007. Brewer, Mark D., Rogan Kersh, and R. Eric Petersen. 2003. Assessing Conventional Wisdom about Religion and Politics: A Preliminary View from the Pews. Journal for the Scientic Study of Religion 42 (Mar): 125136. Broder, John M. 2008a. Romney in Crucial Victory over McCain in Michigan. New York Times. January 16: A1. . 2008b. Clinton Wins in Florida, But without Any Delegates to Sweeten the Victory. New York Times. January 30: A16.



. 2008c. For First Time, Superdelegates Favor Obama. New York Times. May 10: A1. . 2008d. Fathers Tough Life an Inspiration for Biden. New York Times. October 23. Accessed December 18, 2008. . 2008e. Hitting the Backroads and Having Less to Say. New York Times. October 30. Accessed December 17, 2008. Brodsky, Alyn. 2000. Grover Cleveland: A Study in Character. 1st ed. New York: St. Martins Press. Brown, Steven P. 2002. Trumping Religion: The New Christian Right, the Free Speech Clause, and the Courts. Tuscaloosa, AL: University of Alabama Press. Burden, Barry. 2007. Personal Roots of Representation. Princeton, NJ: Princeton University Press. Burnham, Walter D. 1970. Critical Elections and the Mainsprings of American Politics. New York: W. W. Norton. Buzz for a Baptist. 2007. Christian Century 124 (26): 5. Campbell, Angus, Philip Converse, Warren Miller, and Donald Stokes. 1960. The American Voter. New York: John Wiley. Campbell, David E., ed. 2007. A Matter of Faith: Religion in the 2004 Presidential Election. Washington, DC: Brookings Institution Press. Canellos, Peter. 2006. Romney Is Rising Fast as a Serious Contender. Boston Globe. December 12: A2. Carnes, Tony. 2004. Swing Evangelicals: Democrats Seek to Show That They Also Have Faith-based Values. Christianity Today. February 1. ct/2004/february/11.15.html. Carr, John. 2008. Faithful Citizenship: History, Context, Directions, Dangers. Journal of Religion and Society, Supplement Series 4. .pdf. Accessed March 6, 2009. Center for Responsive Politics. 2009. Candidate Comparison: Source of Funds 2008 Cycle. Accessed March 10, 2009. Chaddock, Gail, and Russell Chaddock. 2007. Joseph Biden: A Frank and Abiding Faith. Christian Science Monitor. August 27. Accessed December 18, 2008. Chan, Kenneth. 2008. Church-hosted Forum Reveals Hearts, Minds of White House Hopefuls. The Christian Post. August 17. Accessed October 24, 2008. Cleary, Edward L., and Allen D. Hertzke, eds. 2006. Representing God at the Statehouse: Religion and Politics in the American States. Lanham, MD: Rowman & Littleeld. Clemons, Steve. 2008. Joe Biden Wont Forget the Number of Houses He Owns. Hufngton Post. August 22. Accessed December 20, 2008. Cone, James H. 1970. A Black Theology of Liberation. Philadelphia: J.B. Lippincott.



Conway, M. Margaret. 2005. Political Participation in the 2004 Presidential Election. Pp. 4973 in William Crotty, ed., A Dening Moment: The Presidential Election of 2004. Armonk, NY: M.E. Sharpe. Cook, David. 2007. Huckabee Sees New Life in Presidential Bid After Iowa Straw Poll. Christian Science Monitor. August 17: 25. Cooper, Michael. 2008a. Iowa Results Lead Dodd and Biden to Quit Race. New York Times. January 4: A14. . 2008b. A Strong Endorsement for McCain from a Former President. New York Times. February 19: A21. Cooper, Michael, and Jack Healy. 2008. Candidates Battle for Vote-rich Pennsylvania. New York Times. October 28. 29campaign.html. Accessed October 28, 2008. Cooper, Michael, and Michael Luo. 2008. Exchanges in Latest Debate Highlight a New Dynamic in Republican Field. New York Times. January 11: 14. Cooper, Michael, and Megan Thee. 2008. Resurgent McCain Is Florida Victor; Giuliani Far Back. New York Times. January 30: A1. Cooper, Michael, and Sarah Wheaton. 2008. Obama Cutting into Clintons Edge among Superdelegates. New York Times. February 29: A18. Cooperman, Alan, and Thomas B. Edsall. 2006. Christian Coalition Shrinks as Debt Grows. New York Times. April 10. article/2006/04/09/AR2006040901063.html. Accessed January 25, 2009. Crowley, Monica. 2008. The Top Three McCain Campaign Screw-ups (And How to Fix Them). Human Events. August 4. Accessed February 7, 2009. Cummings, Jeanne. 2008. Obama Asks S.C. Black Voters to Have Faith. January 14. Curl, Joseph. 2008a. Romney Facing Uphill Battle; Delegate Tally a Challenge. Washington Times. February 7: A08. . 2008b. McCain Winning Back Republicans; Likability Factor, Democrats Fight Boost His Appeal. Washington Times. April 18: A04. Current Polls in Florida. 2008. Washington Post. January 29: A8. Davey, Monica. 2008. Little Noticed College Student to Star Politician. New York Times. October 24. Accessed December 18, 2008. Democratic National Convention Committee. 2008. Media at the 2008 Convention. http://www/ Accessed February 2, 2009. Democratic National Platform. 2008. Renewing Americas Promise. http://www Accessed April 3, 2009. Democratic Rivals at Ease with Faith Talk. 2007. Christian Century. June 26. 124 (13): 12. den Dulk, Kevin R. 2006. In Legal Culture, but Not of It: The Role of Cause Lawyers in Evangelical Legal Mobilization. Pp. 197219 in Austin Sarat and Stuart Scheingold, eds., Cause Lawyering and Social Movements. Palo Alto: Stanford University Press.



Diamant, Jeff. 2008. Democrats Step Up Outreach to Faith-based Voters. October 28. Accessed November 21, 2008. Diamond, Sara. 1998. Not by Politics Alone: The Enduring Inuence of the Christian Right. New York: The Guilford Press. Dinan, Stephen. 2007a. Arkansas Huckabee Joins Race. Washington Times. January 29: A4. . 2007b. Giuliani Wary of Repeal of Roe; Abortion One of Few Differences Seen at GOP Presidential Debate. Washington Times. May 4: A1. Dionne, E. J. 1991. Why Americans Hate Politics. New York: Simon & Schuster. . 2008a. Souled Out: Reclaiming Faith and Politics after the Religious Right. Princeton, NJ: Princeton University Press. . 2008b. For an Obamacon, Communion Denied. Washington Post. June 2. http:// html. Accessed October 22, 2008. Dionne, E. J., and Ming Hsu Chen, eds. 2001. Sacred Places, Civic Purposes: Should Government Help Faith-based Charity? Washington, DC: Brookings Institution Press. Domke, David, and Kevin Coe. 2008. The God Strategy: How Religion Became a Political Weapon in America. New York: Oxford University Press. Domke, David, D.V. Shaw, and D. Wackman. 1998. Moral Referendums: Values, News Media, and the Process of Candidate Choice. Political Communication 15: 301321. Dreisbach, Daniel L. 2002. Thomas Jefferson and the Wall of Separation between Church and State. New York: New York University Press. Drogin, Bob, and Mark Z. Barabak. 2008. McCain, Obama Zero in on Battleground States. The Morning Call. October 22. campaign.663783oct21,0,7384911.story. Accessed October 22, 2008. Durkheim, Emile. 1915. The Elementary Forms of Religious Life. London: George Allen & Urwin. Eichel, Larry. 2007. Religious Rights Giuliani Dilemma. Philadelphia Inquirer. October 21: A1. . 2008. McCain, in S.C., Looks for a Better Outcome; He Was Walloped in 2000 but Things Are Different Now. Philadelphia Inquirer. January 18: A12. Eilperin, Juliet, and Michael D. Shear. 2008. Schwarzenegger Endorses McCain; Senator Trying to Appeal to Liberals without Alienating Conservatives. Washington Post. February 1: A10. Eldersveld, Samuel J. 1956. Experimental Propaganda Techniques and Voting Behavior. American Political Science Review 50: 154165. Eldersveld, Samuel J., and Richard W. Dodge. 1954. Personal Contact or Mail Propaganda? An Experiment in Voting Turnout and Attitude Changes. Pp. 532542 in Daniel Katz, Dorwin Cartwright, Samuel Eldersveld, and Alfred M. Lee, eds., Public Opinion and Propaganda: A Book of Readings. New York: Dryden. Evans, Geoffrey. 1999. Class Voting: From Premature Obituary to Reasoned Appraisal. Pp. 121 in Geoffrey Evans, ed., The End of Class Politics? New York: Oxford University Press.



Feldmann, Linda. 2006. Two Years till Election Day, and the Race Is On. Christian Science Monitor. November 15: 1. . 2007. Can the Religious Left Sway the 08 Race? Christian Science Monitor. June 6: 1. . 2008. McCain Still Fighting for Conservative GOP Support. Christian Science Monitor. February 15: 2. Fields, Suzanne. 2007. Nobody Here but Believers: Democratic Seekers Hit the Sawdust Trail. Washington Times. March 12: A19. Fiorina, Morris. 1999. Whatever Happened to the Median Voter? Paper presented at MIT Conference on Parties and Congress. Cambridge, MA. http://www.stanford .edu/~morina/Fiorina%20Web%20Files/MedianVoterPaper.pdf. Accessed May 9, 2009. Fiorina, Morris, Samuel Abrams, and Jeremy Pope. 2005. Culture War? The Myth of a Polarized America. New York: Pearson Longman. First GOP Debate Airs Differences on Abortion, Evolution. 2007. Christian Century. May 29. 124 (11): 11. Formisano, Ronald P. 1983. The Transformation of Political Culture: Massachusetts Parties, 1790s1840s. New York: Oxford University Press. Foust, Michael. 2008. McCain and Obama: Where They Stand on Key Social Issues. BP News. June 4. Accessed January 26, 2009. Fowler, Robert Booth, Allen D. Hertzke, Laura R. Olson, and Kevin R. den Dulk. 2004. Religion and Politics in America: Faith, Culture, and Strategic Choices. 3rd ed. Boulder, CO: Westview Press. FOX News Poll. 2008. FOX News Poll: Obama Grabs Nine Point Lead over McCain. October 22.,2933,443343,00.html. Accessed March 21, 2009. Frank, Thomas. 2004. Whats the Matter with Kansas? New York: Metropolitan Books. Fuller, Robert. 2001. Spiritual but Not Religious: Understanding Unchurched America. New York: Oxford. Gallup, George, Jr., and D. Michael Lindsay. 1999. Surveying the Religious Landscape: Trends in U.S. Beliefs. Harrisburg, PA: Morehouse. Gamson, William. 1992. Talking Politics. New York: Cambridge University Press. George, Jason. 2007. Faith in God Unites All Presidential Candidates. Chicago Tribune. December 30. Gerber, Alan S., and Donald P. Green. 2000a. The Effect of a Nonpartisan Get-out-theVote Drive: An Experimental Study of Leaetting. Journal of Politics 62: 846857. . 2000b. The Effects of Canvassing, Telephone Calls, and Direct Mail on Voter Turnout: A Field Experiment. American Political Science Review 94: 653663. . 2001. Do Phone Calls Increase Voter Turnout? A Field Experiment. Public Opinion Quarterly 65: 7585. Gershtenson, Joseph. 2003. Mobilization Strategies of the Democrats and Republicans, 19562000. Political Research Quarterly 56 (3): 293308. Gibbs, Nancy, and Michael Duffy. 2007. Leveling the Praying Field. Time. July 23: 2834.



Gilgoff, Dan. 2007. Fired McCain Campaign Aides Sound Off. U.S. News and World Report. May 31. Accessed February 16, 2009. . 2008a. McCains Catholic Organizing: Any News? May 19. http:// Accessed February 16, 2009. . 2008b. Why the Christian Right Fears Obama. USA Today. June 16: 12A. Goldenberg, Suzanne, and Ewen MacAskill. 2008. John McCain Refocuses Campaign on Key Battleground States. The Guardian. October 8. 2008/oct/08/uselections2008.johnmccain3. Accessed March 10, 2009. Goldman, Julianna, and Kim Chipman. 2008. Obamas Focus on Faith Offers Promise, Pitfalls for Democrats. June 3. news?pid=20601070&ref=home&sid=aq5fx9T6K_qU. Goodstein, Laurie. 2008a. Spotlight Recasts Church Leaders and Their Support. New York Times. May 24: 12 . 2008b. Obama Made Gains among Younger Evangelicals Voters, Data Show. New York Times. November 7. .html?ref=politics. Accessed November 18, 2008. Gorski, Eric. 2007. Obama Reaches Out to Religious Voters. October 16. Accessed March 6, 2009. Gosnell, Harold F. 1926. An Experiment in the Stimulation of Voting. American Political Science Review 20: 869874. . 1927. Getting Out the Vote: An Experiment in the Stimulation of Voting. Chicago: University of Chicago Press. Governor Palin at the Republican National Convention. 2008. Acceptance speech aired on National Public Radio. September 3. .php?storyId=94258995. Accessed December 18, 2008. Greeley, Andrew M. 2003. Religion in Europe at the End of the Second Millennium. New Brunswick, NJ: Transaction. Green, Donald P., and Alan S. Gerber. 2004. Get Out the Vote! How to Increase Voter Turnout. Washington, DC: Brookings Institution. , eds. 2005. Recent Advances in the Science of Voter Mobilization. The Annals of the American Academy of Political and Social Sciences 601: 69. Green, John C. 2007. The Faith Factor: How Religion Inuences American Elections. Westport, CT: Praeger. . 2009. What Happened to the Values Voter: Believers and the 2008 Election. First Things 191: 4248. Green, John C., and James L. Guth. 1993. From Lambs to Sheep: Denominational Change and Political Behavior. Pp. 100117 in Leege and Kellstedt. Green, John C., Lyman A. Kellstedt, Corwin E. Smidt, and James L. Guth. 2007. How the Faithful Voted: Religious Communities and the Presidential Vote. Pp. 1536 in Campbell. Green, John C., and Mark Silk. 2003. The New Religion Gap. Religion in the News 6 (Fall): 3.



Gross, Terry. 2008. Reverend James Cone on Black Liberation Theology. Fresh Air. National Public Radio. March 31. 89236116. Grunwald, Michael, and Jay Newton-Small. 2008. Why McCain Picked Palin. Time. August 29.,8599,1837510,00.html. Accessed December 17, 2008. Guth, James L. 1983. Southern Baptist Clergy: Vanguard of the Christian Right. Pp. 117130 in Robert C. Liebman and Robert Wuthnow, eds. The New Christian Right: Mobilization and Legitimation. New York: Aldine Publishing Company. . 2007. Religion and Roll Calls: Religious Inuences on the U.S. House of Representatives. Presented at the annual meeting of the American Political Science Association, Chicago, August 30September 2. . 2009. Religion in the 2008 Election. Pp. 117136 in Janet Box-Steffensmeier and Steven E. Schier, eds., The American Elections of 2008. Lanham, MD: Rowman & Littleeld. Guth, James L., John C. Green, Corwin E. Smidt, Lyman A. Kellstedt, and Margaret M. Poloma. 1997. The Bully Pulpit: The Politics of Protestant Clergy. Lawrence, KS: University Press of Kansas. Guth, James L., and Lyman A. Kellstedt. 2001. Religion and Congress. Pp. 213233 in Corwin E. Smidt, ed., In God We Trust: Religion and American Political Life. Grand Rapids, MI: Baker Academic. Guth, James L., Lyman A. Kellstedt, John C. Green, and Corwin E. Smidt. 1998. Thunder on the Right? Religious Interest Group Mobilization in the 1996 Election. Pp. 169192 in Allan J. Cigler and Burdett A. Loomis, eds., Interest Group Politics, 5th ed. Washington, DC: CQ Press. . 2002. A Distant Thunder? Religious Mobilization in the 2000 Elections. Pp. 161184 in Allan J. Cigler and Burdett A. Loomis, eds., Interest Group Politics, 6th ed. Washington, DC: CQ Press. . 2007. Getting the Spirit: Religious and Partisan Mobilization in the 2004 Elections. Pp 157181 in Allan J. Cigler and Burdett A. Loomis, eds., Interest Group Politics, 7th ed. Washington, DC: CQ Press. Guth, James L., Lyman A. Kellstedt, Corwin E. Smidt, and John C. Green. 2006. Religious Inuences in the 2004 Presidential Election. Presidential Studies Quarterly 36 (June): 223242. Hagerty, Barbara Bradley. 2008. Obama Redraws Map of Religious Voters. All Things Considered. National Public Radio. October 24. content/news/2008/10/Obama_redraws_map_of_religious_voters. Accessed October 27, 2008. Halloran, Liz. 2008. McCain Suspends Campaign, Shocks Republicans. U.S. News and World Report. September 24. 2008/09/24/mccain-suspends-campaign-shocks-republicans.html. Accessed February 7, 2009. Hallow, Ralph Z. 2007a. Romney Declares Candidacy Ofcial; Running for GOP Nod as Outsider. Washington Times. February 14: A4.



. 2007b. Evangelical Leaders Warm to Romney: Minister Huckabee Shunned. Washington Times. October 11: A4. . 2008a. McCain to Seek Support from Top Conservatives; Will Address a Winter Meeting Today. Washington Times. March 7: A4. . 2008b. McCain Courts Skeptics in Conservative Alliance. Washington Times. March 8: A1. Hammond, Phillip. 1992. Religion and Personal Autonomy: The Third Disestablishment. Columbia: University of South Carolina Press. Hansen, Collin, and Sarah Pulliam. 2008. Not Easy to Command: Primaries Are Revealing Split between Evangelicals and Their Spokesmen. Election 2008 entry. Christianity Today. March. 52 (3): 19. Harper, Tim. 2008. Huckabee and Obama Triumph in Iowa Race: Illinois Democratic Senator Takes a First Step to Becoming the First Black U.S. President; Ordained Baptist Minister Shakes up Republican Race. Toronto Star. January 4: A1. Harris, Fredrick C. 1999. Something Within: Religion in African-American Political Activism. New York: Oxford University Press. Hart, Roderick P., and John L. Pauley II. 2005. The Political Pulpit Revisited. West Lafayette, IN: Purdue University Press. Harwood, John. 2008. In Superdelegate Count, Tough Math for Clinton. New York Times. April 7: A18. Hayes, Christopher. 2008. Obamas Voter Registration Drive. The Nation. August 13. http://www.the Accessed March 10, 2009. Healy, Patrick. 2008a. Clintons Campaign Sees Value in Keeping Former President in Attack Mode. New York Times. January 25: A22. . 2008b. Obama and Clinton Brace for a Long-distance Run. New York Times. February 7: A1. Healy, Patrick, and Jeff Zeleny. 2008a. At Debate, Two Rivals Go After Deant Clinton. New York Times. January 6: A1. . 2008b. As Support Fades, Clinton Focuses on Next Contest. New York Times. May 8: A1. Helman, Scott. 2006. Romney Consults Evangelical Leaders. Boston Globe. November 2. . 2008. Obama Rolls in Miss. Election; Democratic Rivals Sparring on Whose Wins Mean More. Boston Globe. March 12: A1. Helman, Scott, and Sasha Issenberg. 2008. McCain Courts Conservatives after Romney Quits Candidacy; GOP Puts Focus on US Security. Boston Globe. February 8: A1. Helman, Scott, and Susan Milligan. 2008. McCain Trounces Romney; Clinton Edges Obama in N.H.A Comeback on Each Side. Boston Globe. January 9: A1. Helman, Scott, and Joseph P. Williams. 2008. Obama Wins Big in N.C. Vote; Clinton Clings to Indiana Lead in Last of the Major Contests. Boston Globe. May 7: A1. Henry, Patrick. 1981. And I Dont Care What It Is: The Tradition-history of a Civil Religion Proof-text. Journal of the American Academy of Religion 49: 149. Hershey, Majorie Randon. 2009. Party Politics in America. 13th ed. New York: Pearson Longman.



Hoover, Stewart. 1998. Religion in the News: Faith and Journalism in American Public Discourse. Thousand Oaks, CA: Sage Publications. Hostein, Lisa. 2008. In Final Days, Campaigns Step up Bid for Pennsylvania Jewish Vote. Jewish Telegraphic Agency. October 25. content/news/2008/10/in_final_days_campaigns_step_up. Accessed October 29, 2008. Hout, Michael, and Claude Fischer. 2002. Why Americans Have No Religious Preference. American Sociological Review 67 (April): 165190. Huckfeldt, Robert, and John Sprague. 1992. Political Parties and Electoral Mobilization: Political Structure, Social Structure, and the Party Canvass. American Political Science Review 86: 7086. Hudson, Deal. 2008. Douglas Kmiec and the Lure of Obama. CatholiCity: The Catholic Church Simplied. .html. Accessed October 22, 2008. Hunter, James Davison. 1991. Culture Wars: The Struggle to Dene America. New York: Basic Books. . 1994. Before the Shooting Begins: Searching for Democracy in Americas Culture War. New York: Free Press. Inglehart, Ronald. 1990. Culture Shift in Advanced Industrial Society. Princeton, NJ: Princeton University Press. Issenberg, Sasha. 2008a. In Michigan Primary, Anything Can Happen; Balloting Changes Add Uncertainty for Voter Analysts. Boston Globe. January 15: A8. . 2008b. McCain Likes the Math; Huckabee Could Use MiracleDespite Tension, Arizona Senators Camp Cheer Rival. Boston Globe. February 12: A10. Its Clinton and McCain, Poll Says; Most Like Rudy, Dont Know Barack and Name Hillary as Fave Dinner Date. 2007. New York Daily News. January 26: 6. Ivers, Gregg. 1995. To Build a Wall: American Jews and the Separation of Church and State. Charlottesville, VA: University of Virginia Press. Jackson, David. 2008. Jewish Voters Could Tip Battleground States. USA Today. August 4. .htm. Accessed October 22, 2008. Jackson, John, and William Crotty. 1996. The Politics of Presidential Selection. New York: HarperCollins College. Jelen, Ted G. 2009. Religion and American Public Opinion: Social Issues. Pp. 217242 in Corwin E. Smidt, Lyman A. Kellstedt, and James L. Guth, eds., The Oxford Handbook of Religion and American Politics. New York: Oxford University Press. Jelen, Ted G., and Clyde Wilcox. 1995. Public Attitudes toward Church and State. Armonk, NY: M.E. Sharpe. Jensen, Richard. 1971. The Winning of the Midwest: Social and Political Conict, 18881896. Chicago: University of Chicago Press. Johnson, Kirk and Kim Severson. 2008. In Palins Life and Politics, Goal to Follow Gods Will. New York Times, September 17. politics/06church.html?_r=1&adxnnl=1&oref=slogin&partner=rssnyt&emc=rss&page wanted=all. Accessed December 21, 2008.



Jones, Jeffrey. 2008. Intense Political Week Brings Decline in Swing Voters. Gallup Poll. September 4. Accessed December 18, 2008. Judis, John B., and Ruy Teixeira. 2002. The Emerging Democratic Majority. New York: Scribner. Kampeas, Ron. 2008. Obama Labors to Gain Jewish Trust of His Friends and His Policies. Jewish Telegraphic Agency. October 26. content/news/2009=8/10obama_ labors_to_gain_Jewish. Accessed October 27, 2008. Keeter, Scott. 2006. Election 06: Big Changes in Some Key Groups. Pew Research Center for the People and the Press. Accessed October 22, 2008. . 2007. Evangelicals and Moral Values. Pp. 8092 in Campbell. Keeter, Scott, and Gregory Smith. 2007. How the Public Perceives Romney, Mormons. Washington DC: Pew Forum on Religion and Public Life. Kellstedt, Lyman A. 2008. Seculars and the American Presidency. Pp. 8199 in Gaston Espinoza, ed., Religion, Race, and the American Presidency. Lanham, MD: Rowman & Littleeld. Kellstedt, Lyman A., and John C. Green. 1993. Knowing Gods Many People: Denominational Preference and Political Behavior. Pp. 5371 in Leege and Kellstedt. Kellstedt, Lyman A., John C. Green, James L. Guth, and Corwin E. Smidt. 1996. Grasping the Essentials: The Social Embodiment of Religion and Political Behavior. Pp. 174 192 in John C. Green, James L. Guth, Corwin E. Smidt, and Lyman A. Kellstedt, eds., Religion and the Culture Wars. Lanham, MD: Rowman & Littleeld. . 1997. Is There a Culture War? Religion and the 1996 Election. Paper presented at the annual meeting of the American Political Science Association. Washington, DC, September. Kellstedt, Lyman A., John C. Green, Corwin E. Smidt, and James L. Guth. 2007. Faith Transformed: Religion and American Politics from FDR to G. W. Bush. Pp. 270295 in Mark Noll and Luke Harlow, eds., Religion and American Politics, 2nd ed. New York: Oxford University Press. . 2008. A Gentle Stream or a River Glorious? The Religious Left in the 2004 Election. Pp. 232256 in Campbell. Kennedy, Helen. 2007. How Edwards Sees Faith. New York Daily News. March 4: 28. Key, V. O. 1966. The Responsible Electorate: Rationality in Presidential Voting 19361960. New York: Vintage Books. Kiely, Kathy. 2008. Pennsylvania Primary Puts a Spotlight on Divide among Democrats; A More Pressing Question for the Party Than Who Will Win: How Will It Heal for This Falls Campaign? USA Today. April 21: 1A. Kinder, Donald R. 1998. Communication and Opinion. Annual Review of Political Science 1: 167197. Kindy, Kimberly. 2008. In Rebuking Minister, McCain May Have Alienated Evangelicals. Washington Post. May 29: A8. King, David C. 1997. The Polarization of American Parties and Mistrust of Government. Pp. 155178 in Joseph S. Nye, Jr., Philip D. Zelikow, and David C. King, eds., Why People Dont Trust Government. Cambridge, MA: Harvard University Press.



Kleppner, Paul. 1970. The Cross of Culture: A Social Analysis of Midwestern Politics 18501900. New York: Free Press. . 1979. The Third Electoral System, 18531892: Parties, Voters, and Political Cultures. Chapel Hill: University of North Carolina Press. Kohut, Andrew, John C. Green, Scott Keeter, and Robert Toth. 2000. The Diminishing Divide: Religions Changing Role in American Politics. Washington, DC: Brookings Institution Press. Kornblut, Anne E., and Dan Balz. 2008. Obama Pressed in Pa. Debate; Gaffes Are a Focus as He Spars with Clinton. Washington Post. April 17: A1. Kornblut, Anne E., and Matthew Mosk. 2007. Obamas Campaign Takes In $25 Million; He Nearly Matches Clinton, With Twice as Many Donors. Washington Post. April 5: A01. Kornblut, Anne E., and Shailagh Murray. 2008. Clinton Takes Pennsylvania; Victory Ensures That Tight Nominating Contest Will Keep Going. Washington Post. April 23: A1. Kornblut, Anne E., and Michael D. Shear. 2008. Two Races, One Big Day: Candidates Crisscross Country in Late Push before Pivotal Voting. Washington Post. February 5: A1. Kramer, Gerald. 1970. The Effects of Precinct-level Canvassing on Voting Behavior. Public Opinion Quarterly 34: 560572. Kranish, Michael. 2007. Evangelical Christians Remain Wary of Top GOP Candidates; Crucial Voting Bloc Appears Divided. Boston Globe. October 20: A1. Kurtz, Howard. 2008. Soft Press Sharpens Its Focus on Obama. Washington Post. March 3: C1. Layman, Geoffrey C. 2001. The Great Divide: Religious and Cultural Conict in American Party Politics. New York: Columbia University Press. Layman, Geoffrey, Thomas Carsey, and Juliana Menasce Horowitz. 2006. Party Polarization in American Politics: Characteristics, Causes, and Consequences. American Review of Politics 9: 83110. Layman, Geoffrey, and John C. Green. 2005. Wars and Rumors of Wars: The Contexts of Cultural Conict in American Political Behavior. British Journal of Political Science 36 (1): 6189. Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet. 1944. The Peoples Choice: How the Voter Makes Up His Mind in a Presidential Campaign. New York: Duell, Sloan and Pearce. Leege, David C., and Lyman A. Kellstedt. 1993. Rediscovering the Religious Factor in American Politics. Armonk, NY: M.E. Sharpe. Leege, David C., Kenneth D. Wald, Brian S. Krueger, and Paul D. Mueller. 2002. The Politics of Cultural Differences: Social Change and Voter Mobilization Strategies in the Post New Deal Period. Princeton, NJ: Princeton University Press. Lengell, Sean. 2008. For Obama, a Super Saturday; Wins in Nebraska, Washington State and Louisiana Cut Hillarys Lead in Delegates. Washington Times. February 20: A1. Lipset, Seymour Martin, and Earl Raab. 1981. The Election and the Evangelicals. Commentary 71 (March): 2532.



Lizza, Ryan. 2008. Bidens Brief. The New Yorker. October 10. http://www.newyorker .com/reporting/2008/10/20/081020fa_fact_lizza?currentPage=all. Accessed December 17, 2008. Lopez, Aaron J. 2008. Dems Message: Faith is What We Live. 2008 Denver Democratic National Convention. Rocky Mountain News. August 25. http://www.rockymountainnews .com/news /2008/aug/25/dems-message-faith-is-what-we-live/. Accessed February 7, 2009. Luo, Michael. 2007a. Vying for Values Voters. New York Times. October 16: 18. . 2007b. Religious Right Divides Its Vote. New York Times. October 21: 1. . 2007c. From Back of the G.O.P. Pack, Huckabee Is Stirring. New York Times. November 9: 20. . 2007d. Crucial Test for Romney in Speech on His Religion. New York Times. December 6: A36. . 2008. McCain Extends His Outreach, but Evangelicals Are Still Wary. New York Times. June 9. Accessed November 8, 2008. Luo, Michael, and Michael Cooper. 2008. Focus Shifts to S. Carolina for Romney and Rivals. New York Times. January 17: 26. Luo, Michael, and Jeff Zeleny. 2008. Obama and Clinton Flush with Cash from February. New York Times. February 29: A19. Manza, Jeff, and Clem Brooks. 1999. Social Cleavages and Political Change: Voter Alignments and U.S. Party Coalitions. New York: Oxford University Press. Martin, Jonathan, and Jim VandeHei. 2008. McCain, Palin Push Biography, Not Issues. September 9. .html. Accessed February 7, 2009. Martin, William. 1996. With God on Our Side: The Rise of the Religious Right in America. New York: Broadway Books. Massa, Mark Stephen. 2003. Anti-Catholicism in America: The Last Acceptable Prejudice. New York: Crossroad. McCain, John. 1999. Faith of My Fathers. New York: Random House. McClurg, Scott D. 2004. Indirect Mobilization: The Social Consequences of Party Contacts in an Election Campaign. American Politics Research 32: 406443. McCullough, David. 2001. John Adams. New York: Simon & Schuster. McIntire, Mike, and Christopher Drew. 2008. As Developer Heads to Trial, Questions Linger Over a Deal With Obama. New York Times. March 2: A25. Memmott, Mark. 2008. Obama Side on Defensive over Canadian Memo. USA Today. March 4: 8A. Memoli, Michael. 2008. Abortion Politics: Biden Never Refused Communion. MSNBC First Read. August 24. .aspx. Accessed December 20, 2008. Menendez, Albert J. 1977. Religion at the Polls. Philadelphia: Westminster Press. Merida, Kevin. 2008. Empty, Open Arms: John McCain Wants Conservatives by His Side; Fine, They Say, Just Move This Way. Washington Post. February 28: C1. Miller, Lisa, and Amanda Coyne. 2008. A Visit to Palins Church. Newsweek. September 2. Accessed December 18, 2008.



Milligan, Susan. 2008. Clinton Notches Win; Romney Rolls to Big Victory in Nevada Caucuses; Candidates Capture Second Straight State. Boston Globe. January 20: A1. Mr. Obamas Prole in Courage. 2008. New York Times. Editorial. March 19: A18. Monsma, Stephen V. 1996. When Sacred and Secular Meet: Religions, Nonprot Organizations, and Public Money. Lanham, MD: Rowman & Littleeld. Monsma, Stephen V., and J. Christopher Soper, eds. 1998. Equal Treatment of Religion in a Pluralistic Society. Grand Rapids, MI: Eerdmans. . 2008. The Challenge of Pluralism: Church and State in Five Democracies. Lanham, MD: Rowman & Littleeld. Monson, J. Quin, and J. Baxter Oliphant. 2007. Microtargeting and the Instrumental Mobilization of Religious Conservatives. Pp. 95119 in Campbell. Mooney, Brian C. 2008. Obamas Paid Staff Dwarng McCains. Boston Globe. July 20. dwarng_mccains. Accessed March 10, 2009. Moore, Edmund Arthur. 1956. A Catholic Runs for President: The Campaign of 1928. New York: Ronald Press. Moore, Stephen. 2007. Reform. Reform. Reform. John McCain Explains His Eclectic and TroublingEconomic Philosophy. Wall Street Journal. November 26. http:// Morgenstern, Claire. 2008. John McCain Tackles Plan for All the Major Campaign Issues. Tartan; Carnegie-Mellon University. October 27. http://www.thetartan .org/2008/10/27/news/mccain. Accessed January 20, 2009. Murray, Shailagh, and Anne E. Kornblut. 2008. Women, Latinos Propel Clinton to First Place. Washington Post. January 20: A1. Murray, Shailagh, and Matthew Mosk. 2008. Clinton Lent Her Campaign $5 Million; Delegate Race with Obama Is Nearly Even. Washington Post. February 7: A1. Murray, Shailagh, and Peter Slevin. 2008. As Minister Repeats Comments, Obama Tries to Quiet Fray. Washington Post. April 29: A1. Nagourney, Adam. 2006. Strategy, Not Drama, in Saying I Intend to Run in 08. New York Times. December 27: 23. . 2008a. Obama Takes Iowa in a Big Turnout; Huckabee Victor. New York Times. January 4: A1. . 2008b. Republican Field Rides Full Force into Florida Test. New York Times. January 21: A1. . 2008c. Thompson Ends Campaign for GOP Nomination. New York Times. January 23: 19. . 2008d. Texas and Ohio to Clinton; McCain Is In as Choice for the G.O.P. New York Times. March 5: A1. Nagourney, Adam, and Laurie Goodstein. 2007. Mormon Candidate Braces for Religion as Issue. New York Times. February 8. Nagourney, Adam, and Patrick Healy. 2007. Democratic Hopefuls Split over the Best Path to Peace. New York Times. February 3: A1. Nagourney, Adam, and Mark Leibovich. 2008. Clinton Ends Bid with Clear Call to Elect Obama. New York Times. June 8: A1.



Nagourney, Adam, and Marc Santora. 2007. Republican Candidates Hold First Debate, Differing on Dening Partys Future. New York Times. May 4: 20. Nagourney, Adam, and Megan Thee. 2007. Poll Finds G.O.P. Field Isnt Touching Voters. New York Times. December 11: 1. Nagourney, Adam, and Jeff Zeleny. 2007. Romney Wins Iowa Straw Poll by a Sizeable Margin. New York Times. August 12: 18. . 2008a. For Democrats, Increased Fears of a Long Fight. New York Times. March 16: A1. . 2008b. End of Campaign Centers on Turf Carried by Bush. New York Times. October 28: A1. National Association of Evangelicals. 2004. For the Health of the Nation: An Evangelical Call for Civic Responsibility. Washington, DC: National Association of Evangelicals. Nieuwbeerta, Paul, and Nan Dirk DeGraaf. 1999. Traditional Class Voting in 20 Postwar Societies. Pp. 2357 in Geoffrey Evans, ed., The End of Class Politics? New York: Oxford University Press. Nivoli, Pietro S., and David W. Brady, eds. 2006. Red and Blue Nation, Vol. 1. Washington, DC: Brookings Institution Press. Obama, Barack. [1995] 2004. Dreams from My Father. Reprint, New York: Crown. . 2006. Call to Renewal Keynote Address. June 28. http://obama . 2007. Our Common Stake in Americas Prosperity. Organizing for America. September 17. obam_24.php. Accessed May 11, 2009. . 2008a. Transcript of Obamas Speech. cnn.politics. March 18. www.cnn .com/2008/POLITICS/03/18/obama.transcript/index.html. . 2008b. Acceptance Speech of Barack Obama to the 2008 Democratic National Convention. Denver, CO: August 28. Obama for America. N.d. Promoting a Healthy Environment. Obama Biden. http://www Accessed May 11, 2009. Olson, Laura. 2000. Filled with Spirit and Power: Protestant Clergy in Politics. Albany: State University of New York Press. . 2007. Whither the Religious Left: Religiopolitical Progressive in Twenty-rst Century America. Pp. 5379 in J. Matthew Wilson, ed., From Pews to Polling Places: Faith and Politics in the American Religious Mosaic. Washington, DC: Georgetown University Press. Olson, Laura R., and John C. Green. 2006. The Religion Gap. PS: Political Science and Politics 39: 455459. Page, Susan. 2007a. Giuliani: Can Hero of 9/11 Win Over His Own Party? Views on Social Issues Could Cost Him Republican Partys Nomination. USA Today. February 1: 1A. . 2007b. Giuliani, Clinton Maintain Leads in Double Digits. USA Today. November 7: 7A. Panagopolous, Costas, and Peter W. Wielhouwer. 2008. The Ground War 20002004: Strategic Targeting in Grassroots Campaigns. Presidential Studies Quarterly 38 (2): 347362.



Parton, James. 1874. Life of Thomas Jefferson: Third President of the United States. Boston: J. R. Osgood and Company. Patterson, Thomas 2009. Voter Participation: Records Galore This Time, but What about Next Time? Pp. 4463 in Steven S. Smith and Melanie J. Springer, eds., Reforming the Presidential Process. Washington, DC: Brookings Institution Press. Paulson, Michael. 2008. Faith-talk at the Republican Convention. Boston Globe. September 3. re.html. Accessed October 14, 2008. Pearson-Merkowitz, Shanna, and James G. Gimpel. 2009. Religion and Political Socialization. Pp. 164190 in Corwin E. Smidt., Lyman A. Kellstedt, and James L. Guth, eds., The Oxford Handbook of Religion and American Politics. New York: Oxford University Press. Peel, Roy V., and Thomas C. Donnelly. 1931. The 1928 Campaign: An Analysis. New York: Richard R. Smith. Pew Forum on Religion and Public Life. 2003. Religion and Politics: Contention and Consensus. Washington, DC: Pew Forum/Pew Research Center. . 2007. Public Expresses Mixed Views of Islam, Mormonism. September 25. Accessed March 14, 2008. . 2008a. 2008 National Election Pool Exit Poll as Reported by MSNBC. http:// (object name 2008 Iowa Caucus Exit Poll Results). Accessed December 3, 2008. . 2008b. 2008 National Election Pool Exit Poll as Reported by MSNBC. http:// (object name 2008 New Hampshire Exit Poll Results). Accessed December 3, 2008. . 2008c. Religious Composition of Iowa. .php?StateID=4 (object name 2008 Religious Composition of Iowa). Accessed December 3, 2008. . 2008d. More Americans Question Religions Role in Politics. August 21. Washington, DC: Pew Research Center/Pew Forum. DocID=334. Accessed October 25, 2008. . 2008e. How the News Media Covered Religion in the General Election. November 20. Accessed December 3, 2008. Pew Research Center for the People and the Press. 2002. Americans Struggle with Religions Role at Home and Abroad. Washington, DC: Pew Research Center/Pew Forum. . 2007. Clinton and Giuliani Seen as Not Highly Religious; Romneys Religion Raises Concerns. Sept 5. Washington, DC: Pew Research Center/Pew Forum. . 2008a. Obama Weathers the Wright Storm, Clinton Faces Credibility Problem. March 27. Accessed October 22, 2008. . 2008b. High Marks for the Campaign, a High Bar for Obama. November 13. Accessed April 14, 2009. Phillips, Kate. 2008. As a Matter of Faith, Biden Says Life Begins at Conception. New York Times. September 8. html?_r=1&ref=politics. Accessed December 17, 2008. Pierson, Paul. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton, NJ: Princeton University Press.



The Politics of Despair. 2008. Economist. September 18. world/unitedstates/displaystory/cfm?story_id12262221. Accessed March 10, 2009. Polsby, Nelson, and Aaron Wildavsky. 2004. Presidential Elections: Strategies and Structures of American Politics. 11th ed. Lanham, MD: Rowman & Littleeld. Posner, Sarah. 2008. Battling for the Soul of the Democratic Party. Religion Dispatches. December 8. Accessed March 22, 2009. Press, Bill. 2005. How the Republicans Stole Christmas: The Republican Partys Declared Monopoly on Religion and What Democrats Can Do to Take It Back. New York: Doubleday. Princeton Survey Research Associates. 2008. Newsweek Poll: Palins Impact and Obamas Vulnerabilities, October 24. methodology.pdf. Accessed March 21, 2008. Putnam, Robert D. 2000. Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Shuster. Quinn, Justin. 2008. Draft Sarah Palin! U.S. Conservative Politics Blog. June 30. http:// Accessed December 18, 2008. Rauschenbusch, Walter. 1907. Christianity and the Social Crisis. New York: Hodder and Stoughton. Republican National Platform. 2008. 2008 Republican Platform. .com/2008Platform/. Accessed April 3, 2009. Rhee, Foon. 2008. McCain Plays on Maverick Theme. Boston Globe. September 8. http:// Accessed February 6, 2009. Riley, Jennifer. 2008. Young Religious Voters Notable Player in Swing States. August 14. young-religious-voters-notable-player-in-swing-states-14/index.html. Accessed October 22, 2008. Rosen, Gary. 2007. Narrowing the Religion Gap? New York Times Magazine. February 18. Section 6: 11. Rosenstone, Steven J., and John M. Hansen. 1993. Mobilization, Participation, and Democracy in America. New York: MacMillan. Rosin, Hanna. 2007. Closing the God Gap. The Atlantic. January/February. http://www Accessed February 4, 2009. Rothenberg, Stuart. 2008. Is 2008 a Realigning Election? Numbers Offer Some Clues. November 11. 11/is_2008_a_realigning_election.html. Rother, Larry, and Michael Luo. 2008. Groups Respond to Obamas Call for National Discussion about Race. New York Times. March 20: A21. Rozell, Mark, and Gleaves Whitney, eds. 2007. Religion and the Bush Presidency. Boulder, CO: Palgrave MacMillan. Rozell, Mark, and Clyde Wilcox, eds. 1995. God at the Grassroots: The Christian Right in the 1994 Elections. Lanham, MD: Rowman & Littleeld.



Rutenberg, Jim. 2008. McCain and Obama Agree to Attend Megachurch Forum. New York Times. July 21. html. Saad, Lydia. 2008. Biden Choice Does No Harm but May Not Help Much. Gallup Poll. August 23. .aspx. Accessed December 18, 2008. Sabar, Ariel. 2008. In Sealing Nomination, Obama Makes History. Christian Science Monitor. June 5: 1. Saddleback Presidential Candidates Forum. 2008. August 16. http://transcripts TRANSCRIPTS/0808/16/se.02.html. Salmore, Barbara G., and Stephen A. Salmore. 1989. Candidates, Parties, and Campaigns: Electoral Politics in America. 2nd ed. Washington, DC: CQ Press. Santora, Marc, and Adam Nagourney. 2008. On to Michigan, G.O.P. Rivals Turn to the Economy. New York Times. January 10: A1. Saul, Michael. 2008. Hil Primed for Super Tues. Tally: Heavyweight States Leaning Her Way. New York Daily News. January 27: 16. Scammon, Richard M., and Ben J. Wattenberg. 1970. The Real Majority. New York: Coward- McCann. Schattschneider, E. E. 1960. The Semisovereign People. New York: Holt, Rinehart, and Winston. Schier, Steven E., and Janet M. Box-Steffensmeier. 2009. The General Election Campaign. Pp. 5578 in Box-Steffensmeier and Schier, eds., The American Elections of 2008. Lanham, MD: Rowman & Littleeld. Schlesinger, Arthur M. 1973. The Imperial Presidency. Boston: Houghton Mifin. Scott, Janet. 2008. The Story of Obama, Written by Obama. New York Times. May 18: 1A. Seelye, Katharine Q., and Jeff Zeleny. 2008. Obama, Now on the Defensive, Calls Bitter Words Ill-chosen. New York Times. April 13: A1. Shea, Daniel. 2003. Schattschneiders Dismay: Strong Parties and Alienated Voters. Pp. 287299 in John C. Green and Rick Farmer, eds., The State of the Parties. Lanham, MD: Rowman & Littleeld, 4th ed. Shear, Michael, and Juliet Eilperin. 2008. McCain Driving Debate, But Some Fear Swerving. Washington Post. June 22. content/article/2008/06/21/ AR2008062101570.html. Accessed February 7, 2009. Shear, Michael D., and Alec MacGillis. 2007a. Romneys Cash Beckons Iowans to Straw Poll. Washington Post. August 10: A01. . 2007b. Romney Hopes to Ease Qualms on His Faith. Washington Post. December 4: A4. Shear, Michael D., and Peter Slevin. 2008. Huckabee Leaves Race. Washington Post. March 5: A1. Silk, Mark. 2007. The Democrats Get Religion. Religion in the News. 10 (12): 89, 19. Silk, Mark, and Andrew Walsh. 2008. A Past without a Future? Parsing the U.S. Catholic Vote. America Magazine. November 3. 2009/10/a_past_without_ a_future_parsing. Accessed November 3, 2008.



Sinderbrand, Rebecca. 2008. Clinton, Obama Put Politics Aside to Discuss Faith. CNN April 14. Accessed March 6, 2009. Slater, Wayne. 2008a. McCains Quiet Campaign: The Catholic Vote. Dallas Morning News. March 31. . 2008b. Democrats at Convention Tout Religious InclusionUp to a Point. Dallas News. August 24. national/stories/082508dnpolslatercol.c118238.html. Accessed December 18, 2008. Slevin, Peter. 2007. Michigan Primary Move Splits Democrats; Candidates Stay Away, but Others Say States Voice Should Match its Size. Washington Post. October 9: A6. Slevin, Peter, and Perry Bacon Jr. 2007. Home-school Ties Aided Huckabees Iowa Rise; Early Backers Rallied Conservative Network. Washington Post. December 17: A1. Smidt, Corwin E. 1988. The Mobilization of Evangelical Voters in 1980: An Initial Test of Several Hypotheses. Southwestern Political Review 16 (Fall): 333. . 2001. Religion and American Public Opinion. Pp. 96117 in Corwin E. Smidt, ed., In God We Trust? Religion and American Public Life. Grand Rapids, MI: Baker Academic. , ed. 2003. Religion as Social Capital: Producing the Common Good. Waco, TX: Baylor University Press. , ed. 2004. Pulpit and Politics: Clergy in American Politics at the Advent of the Millennium. Waco, TX: Baylor University Press. . 2007. Evangelical and Mainline Protestants at the Turn of the Millennium: Taking Stock and Looking Forward. Pp. 2951 in Matthew Wilson, ed., From Pews to Polling Places: Faith and Politics in the American Religious Mosaic. Washington, DC: Georgetown University Press. . 2008. Evangelicals and the American Presidency. Pp. 125 in Gaston Espinosa, ed., Religion, Race, and the American Presidency. Lanham, MD: Rowman & Littleeld. Smidt, Corwin E., James L. Guth, Lyman Kellstedt, and John C. Green. 2005. Religion in the 2004 American Presidential Election. Pp. 438465 in Thomas Pludowski, ed., American Politics, Media, and Elections: Contemporary International Perspectives on the U.S. Presidency, Foreign Policy, and Political Communication. Turin and Warsaw: Adam Marszalek and Collegium Civitas Press. Smidt, Corwin E., Kevin R. den Dulk, James M. Penning, Stephen V. Monsma, and Douglas L. Koopman. 2008. Pews, Prayers, and Participation: Religion and Civic Responsibility in America. Washington, DC: Georgetown University Press. Smith, Gregory, and Peyton M. Craighill. 2006. Do the Democrats Have a God Problem? Washington, DC: Pew Forum on Religion and Public Life. Sorauf, Frank. 1976. The Wall of Separation: Constitutional Politics of Church and State. Princeton, NJ: Princeton University Press. Sosnick, Douglas B., Matthew J. Dowd, and Ron Fournier. 2006. Applebees America: How Successful Political, Business, and Religious Leaders Connect with the New American Community. New York: Simon & Schuster. Stanley, Harold W., Wiliam T. Bianco, and Richard G. Niemi. 1986. Partisanship and Group Support over Time: A Multivariate Analysis. American Political Science Review 80 (3): 969976.



Stanley, Harold W., and Richard G. Niemi. 1999. Party Coalitions in Transition: Partisanship and Group Support, 195296. Pp. 162180 in Herbert F. Weisberg and Janet Box-Steffensmeier, eds., Reelection 1996. New York: Chatham House. Steensland, Brian, Jerry Park, Mark Regnerus, Lynn Robinson, W. Bradford Wilcox, and Robert Woodberry. 2000. The Measure of American Religion. Social Forces 79 (1): 291318. Steptoe, Sonja. 2005. Rick Warren: A Pastor with a Purpose. Time. April 18: 108. Stoddard, Ed. 2008. Faith on Full Display at Republic Convention. Reuters. September 2. Accessed October 12, 2008. Stonecash, Jeffrey. 2000. Class and Party in American Politics. Boulder, CO: Westview Press. Sullivan, Amy. 2008a. The Party Faithful: How and Why Democrats Are Closing the God Gap. New York: Scribner. . 2008b. Obamas Play for the Faithful. Time. June 12. time/politics/article/0,8599,1814206.00.html. Accessed January 7, 2009. . 2008c. Why Obama Seized the Faith-based Mantle. Opinion. USA Today. July 28. Accessed January 30, 2009. Summers, Mark W. 2000. Rum, Romanism, and Rebellion: The Making of a President, 1884. Chapel Hill: University of North Carolina Press. Surprise Vice President Selection Finally Puts McCain in the Coverage Lead. 2008. Dow Jones Insight. September 4. Accessed December 18, 2008. Sweet, Lynn. 2008. Obama Invites Christian Leaders in for Private Talks. Chicago SunTimes. June 11. leader.html. Swierenga, Robert. 1990. Ethno-religious Political Behavior in the Mid-nineteenth Century: Voting, Values, Cultures. Pp. 146171 in Mark Noll, ed., Religion and American Politics: From the Colonial Period to the 1980s. New York: Oxford University Press. . 2007. Ethnocultural Political Behavior in the Mid-nineteenth Century: Voting, Values, Cultures. Pp. 145168 in Mark A. Noll and Luke E. Harlow, eds., Religion and American Politics: From the Colonial Period to the Present, 2nd ed. New York: Oxford University Press. . 2009. Religion and American Voting Behavior, 1830s to 1930s. Pp. 6994 in Corwin E. Smidt, Lyman A. Kellstedt, and James L. Guth, eds., The Oxford Handbook of Religion and American Politics. New York: Oxford University Press. Thee, Megan. 2007. Superdelegates for Clinton (Undecideds No. 1). New York Times. November 8: A28. Thelen, Kathleen Ann. 2004. How Institutions Evolve: The Political Economy of Skills in Germany, Britain, the United States, and Japan. Cambridge: Cambridge University Press.



Thiemann, Ronald F. 2009. Obama as Public Theologian: Old Wine in New Wineskins. The Immanent Frame: A Blog of the Social Science Research Council. blogs/immanent_frame/2009/05/28/obama-as-public-theologian. Accessed May 29, 2009. The Times Interviews John McCain. 2008. New York Times. July 13. http://www.nytimes .com/ 2008/07/13/us/politics/13text-mccain.html?_r=2#. Accessed February 13, 2009. Tocqueville, Alexis de. 2003. Democracy in America and Two Essays on America. Trans. G. E. Bevan and I. Kramnick. London: Penguin. Trinity United Church of Christ. 2008. About Us. Accessed November 4, 2008. Verba, Sidney, Kay L. Schlozman, and Henry E. Brady. 1995. Voice and Equality: Civic Volunteerism in American Politics. Cambridge, MA: Harvard University Press. Wald, Kenneth D., and Allison Calhoun-Brown. 2007. Religion and Politics in the United States. 5th ed. Lanham, MD: Rowman & Littleeld. Waldman, Steven. 2008a. Faith Leaders Speak Their Minds at Denver Events. Capital Journal. Wall Street Journal. August 28. faith-leaders-speak-their-minds-at-denver-events/. Accessed October 23, 2008. . 2008b. Obama Is Losing Support among Moderate Evangelicals. Capital Journal. Wall Street Journal. September 5. obama-is-losing-support-among-moderate-evangelicals/. Accessed October 23, 2008. . 2008c. Obamas Fascinating Interview with Cathleen Falsani. http:blog Accessed November 20, 2008. Wall, Tara. 2008. McCain Resurrected? Counting on Conservative Courting to Fill the Gap. Washington Times. March 11: A17. Wallis, Jim. 2006. Gods Politics: Why the Right Gets It Wrong and the Left Doesnt Get It. New York: Harper. . 2008. The Great Awakening: Reviving Faith and Politics in a PostReligious Right World. New York: Harper. Walzer, Michael. 1994. Thick and Thin: Moral Argument at Home and Abroad. Notre Dame, IN: University of Notre Dame Press. Warner, R. Stephen. 1993. Work in Progress toward a New Paradigm for the Sociological Study of Religion in the United States. American Journal of Sociology 98: 10441093. Warren, Rick. 2008. Saddlebacks Civil Forum with McCain and Obama. Washington Post. August 13. saddlebacks_civil_forum. Accessed January 2, 2009. Wayne, Stephen J. 2000. The Road to the White House 2000: The Politics of Presidential Elections. Boston: Bedford/St. Martins. Weber, Max. [1930] 1992. The Protestant Ethic and the Spirit of Capitalism. Trans. Talcott Parsons. Reprint, New York: Routledge. Weisman, Jonathan. 2008. Romney Took McCains Words for a Spin. Washington Post. January 16: A7.



Welch, William M. 2008. Obamas Ties to Minister May Be A Big Problem, Some Say; Senator Has Rejected Racial Comments. USA Today. March 17: 4A. Wellman, James. 2008. The Civic Gospel of Hope. Seattle Times. February 7. http:// Accessed February 3, 2009. Wielhouwer, Peter. 2000. Releasing the Fetters: Parties and the Mobilization of the African-American Electorate. Journal of Politics 62: 206222. . 2003. In Search of Lincolns Perfect List: Targeting in Grassroots Campaigns. American Politics Research 31 (Nov.): 632669. . 2009. Religion and American Political Participation. Pp. 394426 in Corwin E. Smidt, Lyman A. Kellstedt, and James L. Guth, eds., The Oxford Handbook of Religion and American Politics. New York: Oxford University Press. Wielhouwer, Peter W., and Brad Lockerbie. 1994. Party Contacting and Political Participation, 19521990. American Journal of Political Science 38: 211229. Wilcox, Clyde, and Lee Sigelman. 2001. Political Mobilization in the Pews: Religious Contacting and Electoral Turnout. Social Science Quarterly 82 (September): 524535. Wilgoren, Jodi, and Bill Keller. 2004. Kerry and Religion: Pressure Builds for Public Discussion. New York Times. October 4. politics/campaign/07memo.html. Accessed May 26, 2009. Williams, Rhys, ed. 1997. Cultural Wars in American Politics: Critical Reviews of a Popular Myth. New York: Aldine DeGruyter. Williams, Roger. 1963. Mr. Cottons Letter Lately Printed, Examined and Answered. Pp. 313396 in The Complete Writings of Roger Williams, vol. 1. New York: Russell and Russell. Wilson, J. Matthew. 2009. Religion and American Public Opinion: Economic Issues. Pp. 191216 in Corwin E. Smidt, Lyman A. Kellstedt, and James L. Guth, eds., The Oxford Handbook of Religion and American Politics. New York: Oxford University Press. Witte, John Jr. 2000. Religion and the American Constitutional Experiment: Essential Rights and Liberties. Boulder, CO: Westview Press. Wuthnow, Robert. 1988. The Restructuring of American Religion. Princeton, NJ: Princeton University Press. . 1998. Loose Connections: Joining Together in Americas Fragmented Communities. Cambridge, MA: Harvard University Press. . 1999. Democratic Liberalism and the Challenge of Diversity in Late Twentiethcentury America. Pp. 1935 in Neil Smelser and Jeffrey Alexander, eds., Diversity and Its Discontents: Cultural Conict and Common Ground in Contemporary American Society. Princeton, NJ: Princeton University Press. Zapor, Patricia. 2008. Yes He Did! Obama Victory Changes Religion/Politics Debate. Catholic Register. November 5. Accessed February 3, 2009. Zeleny, Jeff. 2008a. Obama Urges U.S. to Grapple with Race Issue. New York Times. March 19: A1. . 2008b. Obama Remarks Called Out of Touch. New York Times. April 12.



. 2008c. Obamas Search to Find a Running Mate Sends Researchers on the Road. New York Times. July 30, 2008. politics/30veep.html. Accessed December 18, 2008. . 2008d. Behind the Scenes of the Vice-presidential Selection Process. New York Times. August 26. .html?r=1. Accessed December 18, 2008. . 2008e. Obama Battles Block by Block to Get Voters to Polls. New York Times. October 11. Accessed March 16, 2009. Zeleny, Jeff, and Patrick Healy. 2008. One on One, Democrats Set Aim at G.O.P. New York Times. February 1: A1. Zeleny, Jeff, and Carl Hulse. 2008. Kennedy Chooses Obama, Spurning Plea by Clintons. New York Times. January 28: A1. Zeleny, Jeff, and Adam Nagourney. 2008. An Angry Obama Renounces Ties to His Ex-pastor. New York Times. April 30: A1. Zorn, Eric. 2004. Keyes Saddles Wrong Horse for Senate Race. Chicago Tribune. August 17. .html. Accessed June 14, 2009.




The letter t following a page number denotes a table. abortion. See also under issue positions as general election issue, 6, 48, 6062, 64t, 67, 132, 229, 240n14, 241n12 public funding for, 147 right to choose, 76, 77, 98, 121, 123, 125, 127, 147, 223, 240n12, 243n6, 247n4 sanctity of life, 25, 38, 70, 84, 106, 123, 147, 155, 173, 176, 192, 230, 239n5, 243n5, 245n2 Adams, John, 20, 21 Almond, Gabriel, 235n1 atheist, 18, 21, 29, 3435, 36t, 37, 44, 108. See also secular; unafliated with any religion Baptist, 21, 24, 28 battleground state, 139, 155160, 242n17, 242nn2021 Bellah, Robert, 31, 38 black Protestants, 30, 4445, 245n7 mobilization of, 119120, 170188 passim and partisanship, 10, 15, 4650, 5254, 169, 216t, 217 and policy issues, 5663, 6667, 129, 130t, 195200 passim and primary voting, 99, 100t as targeted voter, 152, 157t and vote choice, 30t, 197213 passim, 226 Blaine, James G., 2223 Bush, George H. W., 83, 94 Bush, George W. campaigning by, 14, 88, 89, 100, 108, 155, 160, 169, 172, 193, 194, 198207 passim, 224225 presidency of, 45, 13, 79, 120, 140, 142, 149150, 168, 179, 236n11, 241n15, 243n5 reelection of, 6, 4547, 101 Call to Renewal, 172, 243n7 campaign funding, 159160, 162163, 242n19, 242n22 Campolo, Tony, 173 Carsey, Thomas, 247n7 Carter, Jimmy, 24, 83 Casey, Shawn, 152, 153 Catholic(s), 4, 7, 9, 26, 235nn45, 237n6, 244n11 candidate who is, 2124, 32, 34, 35t, 38, 76, 107, 125, 127128, 131, 213, 224, 240nn1112 mobilization of, 119120, 126, 170189 passim, 244n14

and partisanship, 10, 4450, 5255, 6869, 169, 216220 passim, 243n5, 244n8, 245n3 and policy issues, 5667, 129t, 130t, 195, 196t, 200 and primary voting, 75, 83, 85, 86, 99 as targeted voter, 152, 154, 156, 157 centrist religious belief, 47t, 50t, 237n7 and partisanship, 5355, 216220 passim and policy issues, 6465, 195, 196t, 200 and vote choice, 197213 passim, 245n6 and voter turnout, 179184 passim, 188t, 245n7 Chaput, Charles, 243n6 Christian Coalition, 6, 76, 172, 225, 244nn1011 Christian Right, 3, 6, 78, 80, 168, 172, 175, 225, 235n2, 244n9. See also Moral Majority; Religious Right church attendance, 26, 40, 158, 235n3, 246n13 and partisan identication, 33t, 68, 191 and 2008 election vote, 186t, 187 and vote choice, 3, 32, 36, 45, 69, 86, 87, 104, 169, 210, 237n7 Church of Jesus Christ of Latter Day Saints. See Mormon civil religion, 38, 40, 106, 142, 157, 236n11 civil rights, 41, 60, 85, 232 civic role of religion, 2531 Cizik, Richard, 70 clergy activism in political process, 166, 168, 170, 171t, 175177 endorsement of candidate, 29, 31, 236n8 Cleveland, Grover, 2223 climate change, 98, 146, 148, 155 Clinton, Bill, 13, 122, 140, 142, 150, 236n11 Common Good Strategies, 69 Cone, James H., 239n2 conservative religious belief, 11, 121, 145. See also orthodox religious belief; traditionalist religious belief as courted by Republicans, 25, 126, 155, 167

involvement in 2008 campaigns, 31, 78, 79, 84, 88, 106, 162 conventions. See Democratic National Convention; party conventions; Republican National Convention Conway, M. Margaret, 243n3 Council for National Policy, 9899 creation care, 146, 151 cultural issues, 5962, 64t, 6970, 132, 144149 as most important issue in 2008, 6567 and vote choice, 192195, 196t, 200, 238n12, 247n4 culture war, 11, 12, 39, 60, 132 Daughtry, Leah, 137 debates, 121, 131, 134, 139. See also faith forum; religious forum Democratic National Committee, 73, 103, 169, 244n8 Democratic National Convention, 107, 136137, 142, 147, 241n3, 241n13 demographic, 95, 160, 185, 189. See also voters base of support, 43, 93, 96 and vote choice, 204220, 223, 230231, 234n2, 246n11 Dionne, E. J., 173, 243n5 DNC. See Democratic National Committee Dobson, James, 93 Donnelly, Thomas C., 21 Dreisbach, Daniel L., 235n4 DuBois, Joshua, 70, 120, 125, 152 Durkheim, Emile, 9 economy. See also under issue positions concerns about, 56, 58, 67, 79, 128, 157, 159, 161, 162, 230, 241n9 as election issue, 3, 5, 38, 49, 68, 128133, 150, 192, 222, 224, 229 as most important issue in 2008, 192195, 196t, 200 education as election issue, 132 homeschool, 77, 83, 84, 108, 223 parochial school, 23, 2829, 37, 83, 108



education (continued) public school, 23, 28 as demographic variable, 205t, 207, 209, 213, 214t Eighteenth Amendment. See Prohibition Eisenhower, Dwight, 34, 236n9 Electoral College, 96, 124, 139, 156, 160 environment, 238n12, 239n5. See also under issue positions as election issue, 69, 70, 146, 230 as most important issue in 2008, 66t religious tradition and view of, 5759 passim, 64t Episcopalian, 43, 112, 113 ethnoreligious/ethnocultural perspective of religion and politics, 9, 10, 214, 227, 234n1 evangelical Protestant(s), 24, 25, 4445, 75, 123, 131, 234n3, 243n6 candidate who is, 3436, 3839, 125, 127, 132, 223 mobilization of, 14, 15, 119120, 170189 passim, 224, 244n11, 245n17 and partisanship, 8, 4655 passim, 6870, 138, 167, 169, 215220 passim, 224 and policy issues, 5667 passim, 129, 130t, 195, 196t, 200 as targeted voter, 152159 passim, 222 and vote choice, 30t, 77, 197213 passim, 225, 226, 245n6, 246n16 exit poll, 86, 132, 185, 187, 192, 246n15, 247n16 faith-based initiatives, 28, 6061, 90, 106, 241n15. See also under issue positions Faith Caucus, 137 faith forum, 70, 8485, 91, 98, 108, 113, 153. See also debates; religious forum Faith in Action Committee, 136 Faith in Action Initiative, 68, 152 Falwell, Jerry, 3, 6, 2425, 238n2 Family Research Council, 76, 79, 172, 176, 225, 244nn1011 nancial crisis, 16, 67, 195 Focus on the Family, 76, 93, 172

foreign policy, 6267 passim, 79, 98, 99, 106, 124, 128, 131, 192. See also under issue positions as most important issue in 2008, 192195, 196t, 200 framing (to inuence voters), 13, 144151, 174, 188, 226, 229, 230 FRC. See Family Research Council free trade, 57t, 5859, 64, 95, 161, 238n12 gay marriage, 67, 155, 240n14. See also under issue positions as conservative issue, 70, 239n5 as election issue, 6, 106, 122, 132, 172, 193, 244n13 religious tradition and view of, 6061, 64t gay rights, 48, 6062, 76. See also gay marriage; homosexual Gerson, Michael, 13 God and country religious perspective, 114, 138 grassroots support, 79, 83, 86, 95, 105, 106 Green, John C., 229, 237nn78, 247n5 Gushee, David, 173 Guth, James L., 236n4, 237n7 Hagee, John, 31, 100, 101 Hamilton, Alexander, 21 health care, 16, 130, 131t, 132, 144, 145 insurance, 56, 149 hegemony (cultural/religious), 20, 21, 23, 32, 3435, 37, 223 Henry Institute National Survey on Religion and Public Life, 16, 169, 193, 194, 199, 237n7, 238n11, 242n2, 245n6 Hispanic Catholics, 4547, 50, 52t, 245n7 mobilization of, 182184 and partisanship, 216t, 217, 226 and policy issues, 196t as proportion of 2008 electorate, 187, 188t, 226 and vote choice, 169, 198205 passim, 230 voted in 2008 election, 184 Hispanic Protestants, 4547, 50, 245n7 mobilization of, 181184



and partisanship, 216t, 217, 226 and policy issues, 195, 196t, 200 as proportion of 2008 electorate, 187, 188t and vote choice, 169, 198205 passim voted in 2008 election, 184 HIV/AIDS, 70, 109, 144, 238n12, 239n45 home foreclosure. See housing market homosexual, 25, 61. See also gay marriage; gay rights Hoover, Herbert, 2123 Horowitz, Juliana Menasce, 247n7 housing market, 5, 67, 144, 145 Hudson, Deal, 243nn56 Hunter, James Davison, 1112 ideology (of political party), 7, 135, 147148 immigration, 130, 131t, 149, 179180 Internet, 152, 158, 159, 174, 241n8 Iowa caucuses, 65, 81t, 8385, 153 Iraq War, 3, 68, 140, 149, 222, 243n5 desire to end, 5, 16, 73 as most important issue in 2008, 130132 passim religious tradition and view of, 6264 Islam. See Muslim issue positions, 143144, 162. See also specic issue abortion, 146147, 149, 151 economic, 144146, 151, 157 energy, 146 environment, 146, 151 faith-based initiatives, 150151 foreign policy, 149150 gay marriage, 147148 social equity, 144146, 151, 157 Jefferson, Thomas, 20, 21, 28, 32 Jew(s), 9, 28, 4445 candidate who is, 3436 mobilization of, 119, 170188 passim and partisanship, 8, 10, 4650 passim, 5254 passim, 69, 169, 216t and policy issues, 5663 passim, 6667, 129, 130t, 195, 196t, 200

and primary voting, 100 as targeted voter, 152, 154, 157 and vote choice, 30t, 198213 passim, 245n3 Jones, Robert P., 173 Joshua Generation, 111, 152 Kellstedt, Lyman A., 247n3 Kennedy, D. James, 3, 6 Kennedy, Edward (Ted), 93, 238n14 Kennedy, John F., 22, 32, 36, 38, 212, 213, 227 Kerry, John, 45, 16, 45, 49, 70, 101, 107, 152, 154 campaigning by, 193, 199201, 206207, 238n14, 238n1, 240n10, 243n8, 246n16 King, Martin Luther, 41, 110, 140 Kmiec, Douglas, 120, 153, 173, 243n5 labor unions, 8, 73, 90, 169170 Latter-day Saints, 10 Layman, Geoffrey, 247n7 Leege, David C., 226 liberal religious belief, 11, 31, 145. See also modernist religious belief; progressive religious belief liberation theology, 97, 110, 239n2 libertarian, 125, 146, 149 Lincoln, Abraham, 32, 38, 110, 232 Lucado, Max, 120 Lutheran, 21, 234n4 mainline Protestant(s), 2425, 4445, 234n3, 237n7 candidate who is, 108 mobilization of, 120, 126, 170189 passim and partisanship, 10, 43, 4650 passim, 5155 passim, 69, 169, 215220 passim, 237n8 and policy issues, 5666 passim, 129, 130t, 195, 196t, 200 and primary voting, 75, 91, 99 as targeted voter, 156, 157t and vote choice, 30t, 197213 passim, 226, 245nn56



maverick (political), 112, 126, 143, 162 McLaren, Brian, 173 Methodist, 21, 43 Middle East policy, 3839, 6263, 79, 101, 120, 149 mobilization of voters through congregation, 170, 171t, 175177 by Democratic party, 85, 120, 137, 178182 by interest group, 170174 by personal partisan contact, 166169, 170171 of religious voters, 14, 2425, 43, 164181, 188190, 222, 224230 by Republican party, 76, 178182 modernist religious belief, 4, 11, 12, 50t, 237n7, 246n13. See also liberal religious belief; progressive religious belief and partisanship, 47t, 48, 5355, 216220 passim, 224228 passim and policy issues, 64, 195, 196t, 200 and vote choice, 197213 passim, 245n6 and voter turnout, 179190 passim, 245n7 moral character (of political candidate), 20, 2224. See also religiosity of candidate Moral Majority, 2425. See also Christian Right; Religious Right Mormon(s), 3436 passim, 50, 70, 126 candidate who is, 15, 34, 36, 7678 passim, 80, 84, 88, 90, 222, 223 and primary voting, 8889 Multiple Classication Analysis, 208220, 245n8, 246nn1011 Muslim(s), 3437 passim, 44, 50, 101, 108 rumors about Obama being, 91, 115, 116, 121, 149, 223 as targeted voter, 152 NAFTA, 95, 161. See also free trade National Association of Evangelicals, 70 New Deal, 10, 53, 59, 191, 197 New Hampshire primary, 65, 8587 Nixon, Richard, 212 Olson, Laura, 225 orthodox religious belief, 11, 48, 234n3.

See also conservative religious belief; traditional religious belief Parsley, Rod, 100, 101 party conventions, 31, 135136, 162, 194, 240nn12. See also Democratic National Convention; Republican National Convention delegates to, 135, 137, 138, 239n1 (see also superdelegate) party platform, 2324, 135, 136, 147, 148, 149, 241n8 Peel, Roy V., 21 Pentecostal, 70, 125, 132, 137, 222 Pietist, 10, 104 policy goals (based on religion), 19, 5667 Popma, Marlys, 154155, 242n18 populist, 80, 85 posturing (to inuence voters), 13, 20, 2425, 111, 120, 131132 prayer in public schools, 60, 61t, 148, 238n13, 247n4 Presbyterian, 21, 43 primaries, 107, 185, 221225 passim, 240nn12. See also Iowa caucuses; New Hampshire primary District of Columbia, 94 Florida, 8790 passim, 92, 98, 103 Indiana, 103 Kansas, 94 Louisiana, 94 Maryland, 94 McCain and, 143 Michigan, 8790 passim, 98, 103 Montana, 103 Nevada, 8790 passim North Carolina, 103 Obama and, 168, 141 Ohio, 9496, 98 Pennsylvania, 98 Potomac Primary, 94 Puerto Rico, 103 results at conclusion of, 114, 124, 239n1 Rhode Island, 9496 South Carolina, 8791 passim, 153 South Dakota, 103



Super Tuesday, 90, 9294, 95 Texas, 9496 Vermont, 9495 Virginia, 94 Washington, 94 private prayer, 2627, 115, 237n7 progressive religious belief, 11, 48. See also liberal religious belief; modernist religious belief Prohibition, 2324, 39, 230, 245n1 public nancing (of campaigns), 159160. See also campaign funding Putnam, Robert, 235n1 Quinn, Justin, 240n8 race, 109. See also racism as campaign issue, 90, 91, 93, 97 of candidate, 6, 141, 180, 184, 189, 192, 201, 222, 224, 227, 229, 232 as variable effecting vote choice, 204214 passim racism, 105, 100 Reagan, Ronald, 2425, 68, 120, 173, 206 Religion Clause. See separation of church and state religiosity of candidate, 1824, 3241, 70, 8083, 117t, 153 of Democrats, 116, 141142 of Republicans, 223, 236nn910, 238n3 religious belief/behavior, 4, 8, 29, 41. See also centrist religious belief; ethnoreligious/ethnocultural perspective of religion and politics; modernist religious belief; traditionalist religious belief and issue views, 195, 196t and partisanship, 10, 5355, 191, 214220 passim, 227228, 231 and vote choice, 197203 passim, 205t, 207213 passim, 247n17 religious belonging (as predictor of political view), 9, 43, 44, 227, 231, 207222 passim religious forum, 15, 74, 121123, 131. See also debates; faith forum

religious interest group, 14, 166, 243n8, 244n9 Religious Left, 168, 172, 175, 225, 247n3 Religious Right, 2425, 76, 138, 173, 235n2. See also Christian Right; Moral Majority religious tradition (afliation), 10, 4350, 5171, 99, 234n1. See also specic religious faith and partisanship, 174, 215220, 223224, 227233, 246n11 and view of economic conditions, 128130 and vote choice, 196204, 207214 Republican National Committee, 103, 126, 167, 242n24 Republican National Convention, 107, 137138, 143, 161, 241n4 Right to Life, 123, 154 RNC. See Republic National Committee Robertson, Pat, 6, 83 Roman Catholic. See Catholic Roosevelt, Franklin D., 140, 145, 197 Saddleback Church, 70, 121123 Sapp, Eric, 69, 238n14 Schlesinger, Arthur M., 236n11 secular, 4, 49, 67, 85, 172, 229232 passim. See also atheist; unafliated with any religion separation of church and state, 22, 2731, 36, 40, 98, 111, 123, 231232 Sider, Ron, 173 signaling (to inuence voters), 13 Smidt, Corwin E., 234n3 Smith, Alfred E., 2124 passim, 32 social gospel, 71, 105, 176, 240n7 social justice (equity), 56, 58, 69, 110, 142, 148, 157, 239n7. See also under issue positions social services, 2829, 5658, 64t social welfare, 57, 66, 67, 122, 145 Southern Evangelicals, 10, 44, 68, 100, 245n4 stem cell research, 56, 77, 106, 148 Strang, Cameron, 120 straw poll, 73, 77, 79



Strider, Burns, 70 Sullivan, Amy, 173 superdelegates, 75, 90, 95, 98, 103 Super Tuesday. See primaries swing vote, 68, 101102, 167, 168, 197200 passim, 216, 217, 224 taxes, 56, 57t, 58, 76, 88, 89, 106, 122, 144, 145 theological restructuring perspective of religion and politics, 9, 214, 227 Tocqueville, Alexis de, 25, 30 traditionalist religious belief, 4, 11, 12, 50t, 237n7, 246n13. See also conservative religious belief; orthodox religious belief and partisanship, 47t, 4849, 5355, 215233 passim and policy issues, 6465, 195, 196t, 200 and vote choice, 191, 197204 passim and voter turnout, 178, 181184 passim, 187189 passim Trinity United Church of Christ, 37, 96, 97, 108, 110, 141 unafliated with any religion, 4445. See also atheist; secular candidate who is, 3435, 36t mobilization of, 170189 passim and partisanship, 4650 passim, 5254 passim, 216219 passim and policy issues, 57t, 6063 passim, 66t, 67, 129t, 130t, 195, 196t, 200 and primary voting, 86, 100 as targeted voter, 157t and vote choice, 30t, 197213 passim, 192, 226

unemployment, 16, 67, 88, 157 Values Voters, 39, 45, 79, 119, 154, 157, 192, 230 Vanderslice, Mara, 69, 152, 154, 238n14, 243n8 Verba, Sidney, 235n1 voter guides, 166, 169177 passim, 244nn1011, 244n13 voters. See also demographic African American, 67, 91, 185, 186t, 189, 205t, 206, 211, 234n2 educational level of, 7, 93, 94, 186t, 205t, 207 Hispanic, 94, 185, 186t, 189, 205t, 206, 211, 230, 236n3 women, 78, 94, 96, 186t, 205t, 246n9 working class/blue collar, 94, 95, 96, 205t young, 49, 85, 93, 96, 99, 158, 168, 185186, 190, 205t, 206, 224, 246n16 Wallis, Jim, 71, 243n7 War on Terror, 62, 63t, 130, 131t Warren, Rick, 70, 121, 239n5 Weber, Max, 9 wedge issue, 55, 56, 59, 62, 63, 67 Wielhouwer, Peter W., 235n8 Winters, Michael Sean, 173 worship services. See church attendance Wuthnow, Robert, 1112 Wright, Jeremiah, 31, 37, 96, 97, 102, 103, 105, 109, 110, 121, 141, 143, 222, 226 Zorn, Eric, 241n11