In the aftermath of the U.S. presidential election, the Washington Post reporter David Finkel interviewed white evangelical voters in the small town of Sheffield, Ohio. The Leslie family had seen its annual income drop from $55,000 in 2001 to $35,000 in 2004. It did not affect their vote: Jobs will come and go, said Cary Leslie, but your character you have to hang on to that. It’s what you’re defined by. As far as they were concerned, that’s what defined the president. To know that he prays, Cary’s wife, Tara Leslie, and I really believe he does that’s a huge thing. Cary summed up his interpretation of the election in a simple sentence: It’s a victory for people like us.
When voters were asked to pick, from a list of seven possibilities, the one issue that mattered the most in deciding how to vote, 22 percent chose moral values, compared to 20 percent for the economy, 19 percent for terrorism, and 15 percent for Iraq. Many people found this result startling, in part because so much of the campaign’s public debate had revolved around the challenge of combating terrorism and the controversy over the war in Iraq. What is more, the conventional wisdom among political scientists and pollsters, strongly supported by recent history, has been that cultural/moral issues come to the fore only when there are no burning economic or foreign policy issues. The airwaves and op-ed pages soon were filled with pundits deploring (or less frequently, applauding) what seemed to be a widening cultural/religious divide between progressives and traditionalists. Many urged Democratic leaders to pay less attention to tofu-eating Vermonters and more to the salt-of-the-earth folks who attend Nascar races.
Then a counterreaction set in. Citing a potential for deep distortion, the polling director for ABC News revealed that he had argued in vain against including the term moral values as an option in his network’s poll. Pollsters noted that different ways of posing questions about the impact of moral values on voting decisions yielded dramatically different results. An exit poll by The Los Angeles Times suggested that the share of the electorate influenced by moral values was no higher in 2004 than in 1996, and only slightly higher than in 2000. The chief executive officer of America Coming Together, a 527 committee organizing get-out-the-vote efforts for Democrats, noted that in Ohio the share of the electorate regularly attending church actually declined from 45 percent in 2000 to 40 percent in 2004. On Dec. 5, The Washington Post published The Anatomy of a Myth on the front page of its highly Outlook section. And that, it seemed, was that.
Or maybe not. Probing beneath statistical aggregates to specific population groups offers reasons to believe that values actually were unusually influential in the 2004 presidential election.
Some scholars, like the political scientist Morris Fiorina and the sociologist Alan Wolfe, reject the thesis that the American people have become more polarized along moral and ideological lines during the past generation. In one sense they are right: the majority of the people remain clustered around the mainstream rather than the extremes. But while the shape of public opinion has remained relatively stable, its distribution between the major political parties has not. As recently as the 1960’s, the Democratic Party included substantial numbers of people who regarded themselves as conservatives. As recently as 1976, the Democratic presidential nominee enjoyed an edge among white born-again Christians.
Over the past four decades, the political parties have become internally more homogeneous, while the gap between them has widened. In 1976, for example, 29 percent of conservatives supported Jimmy Carter; in 2004, only 15 percent of conservatives supported John Kerry. Twenty-six percent of liberals supported Gerald Ford in 1976; only 13 percent supported George Bush in 2004. During this period, levels of religious observance became key determinants of ideological self-identification and partisan affiliation. Today, 54 percent of Americans who attend religious services once a week or more consider themselves conservatives, versus only 26 percent of those who never attend. Sixty percent of frequent attendees voted for George Bush over John Kerry, versus only 34 percent of nonattenders.
While we do not know precisely what moral values meant to the voters who selected this phrase as the principal determinant of their vote, we can draw some inferences from what we know about these voters. They tend to be white, male and married. Forty-two percent are white born-again Christians, twice the proportion of born-agains in the overall electorate. Fifty-seven percent regard themselves as conservative, and 59 percent as Republicans, versus 34 percent and 37 percent, respectively, for the electorate as a whole. Twenty-three percent said that they valued a candidate’s strong religious faith more than any other personal characteristic, compared to only 8 percent for the broader electorate. Eight in 10 of these voters supported President Bush over Senator Kerry.
Ever since Karl Rove’s famous lament about the four million missing evangelical votes in 2000, political commentators have been fixated on the Bush administration’s outreach to this group. The exit polls suggested that these efforts achieved at least moderate success. Overall, the president raised his share of the white Protestant vote from 63 percent in 2000 to 67 percent in 2004. The University of Akron’s John Green, one of the leading researchers on politics and religion, notes that the president’s share of the evangelical vote rose from 72 percent in 2000 to 78 percent this year, and that most of these gains came among evangelicals who reported attending church less than once a week.
A plausible case can be made that the Catholic vote contributed at least as much to the president’s victory. A survey taken in July showed a dead heat (40-40) between Bush and Kerry among Catholics, with 18 percent undecided. In the end, however, the president raised his share of the Catholic vote from 47 percent in 2004 to 52 percent in 2004. In Ohio, where the Bush campaign sent large numbers of field workers to Catholic churches, the president received 55 percent of the Catholic vote, up from about 50 percent in 2000. John Green calculates that this shift netted Bush 172,000 votes, more than his eventual margin of victory in that state. In Florida, where the Catholic share of the total vote rose from 26 to 28 percent, Bush’s share of the Catholic vote rose from 54 percent in 2000 to 57 percent in 2004. In combination, these shifts represented an additional 400,000 votes, roughly the margin of victory in Florida.
Only 20 percent of Catholics supported denying Communion to pro-choice Catholic politicians; only 16 percent felt that these politicians had an obligation to vote the way the bishops recommend; only 7 percent thought that the views of the bishops would have a significant impact on the way they themselves voted. There is evidence, however, that another moral values issue, gay marriage, may have moved substantial numbers of Catholic voters toward the president. The mid-summer survey showed that fully 23 percent of Catholics strongly disapproved of making gay marriage legal and would definitely vote against a candidate they disagreed with on that issue, versus only 4 percent who strongly approved and were prepared to vote on the basis of that conviction. While President Bush voiced full-throated opposition to gay marriage, breaking with his own vice president to advocate a constitutional amendment banning the practice, John Kerry offered a more nuanced position, opposing gay marriage but leaving the matter to the states. To the extent that Catholics interpreted Kerry’s position as covert support for local gay marriage initiatives, this issue could well have moved some of them to support Mr. Bush.
Based on a survey taken the day after the election, the veteran Democratic pollster Stanley Greenberg concluded that Republican appeals based on moral values accounted for much of the movement away from the Democratic nominee among Hispanics as well. While the polls remain in dispute, it is clear that the president increased his share of the Hispanic vote from 35 percent in 2000 to between 40 and 44 percent this year. One plausible explanation: increasing numbers of Hispanics are evangelicals rather than Catholics, and 60 percent of Hispanic evangelicals voted for Mr. Bush. But among Hispanic Catholics, Bush raised his support from 31 percent in 2000 to 42 percent this year.
A similar dynamic appears to have been at work among African-Americans. The president received 16 percent of the black Protestant vote, up from 9 percent in 2000, and black Protestants who attend church more than once a week gave Mr. Bush 22 percent of their vote. Interviews with black ministers suggest that Republicans were making inroads among black religious leaders with appeals to biblical teachings on such issues as gay marriage and abortion.
Likewise, although students surveyed by the Center for Information and Research on Civic Learning and Engagement were notably more liberal and pro-Kerry than other age groups, fully 26 percent selected moral issues as the principal determinant of their vote. They did so in spite of the fact that the poll question specified abortion and gay marriage as instances of moral issues, which might well have narrowed the appeal of this option. Sixty percent of those who selected moral issues supported George Bush.
President Bush also succeeded in transforming the war in Iraq and the fight against terrorism into questions of basic values and American national identity. Earlier this year, the University of Virginia’s Center on Religion and Democracy conducted a survey probing public attitudes toward America’s role in the world. The center found that while most Americans see our country as a force for good in the world, a significant minority does not. These dissenters are not evenly distributed between the political parties: 30 percent of Democrats are highly critical of America’s global role, versus only 9 percent of Republicans. And religion reinforces partisanship. Americans who said that their religious beliefs were not at all important to them, most of whom are Democrats, were much more hostile toward America’s role in the world than were those for whom religious beliefs mattered. Of these critics, the survey’s author concluded that they differ sharply [from the majority] in certain universal judgments. To wit, their hope for the future is for a world of diminishing emphasis on country, for globalized standards of living and for flexible moral commitments grounded in secular truths. In their lack of a strong national identity and their secularism, they share more in common with many Europeans than with the majority of their fellow citizens.
In retrospect, this may have been the deepest moral values issue at stake in the 2004 election. While Republicans stood united in their belief in American exceptionalism, Democrats were badly divided. President Bush was able to rally his party by sounding the trumpet of American virtue on the global stage. By contrast, John Kerry struggled to bridge the gap between Tony Blair Democrats, who agreed with Bush’s principles but deplored his inept policies, and Michael Moore Democrats, who rejected root and branch the idea of a global fight against terrorism and for democracy.
As we learned in Vietnam, the moral dimension of foreign policy cannot prevail indefinitely against facts on the ground. Eventually, the efficacy of means trumps the nobility of ends. If the president brings the American occupation of Iraq to a successful conclusion, the Bush doctrine may well define a new values-based internationalism and reinforce the structural political advantage of the party that more fully embraces American exceptionalism. If he fails, the post-Vietnam understanding of limits to American power will likely rise from the ashes of Fallujah, once again redrawing the moral fault-lines of American politics.