[W]e use the fourth quadrennial National Survey of Religion and Politics, conducted at the University of Akron in the spring and fall of 2004 and sponsored by the Pew Forum on Religion and Public Life. This survey of a national random sample of 4,000 respondents asked a range of religious questions seldom available in other surveys and, fortunately, also had a large battery of foreign policy questions. It is just this sort of evidence that has been lacking in the debate over the role of religion in foreign policy.
To go to the core of recent arguments, we examine the backing that America’s diverse religious communities provide for the Bush Doctrine, the president’s stress on military strength, preference for unilateral rather than multilateral action, willingness to engage in pre-emptive war (as in Iraq), and a tilt toward Israel in the Middle East. To measure this support, we use five items: an approval rating for Bush administration foreign policy, an assessment on whether the Iraq war was justified, whether pre-emptive war is ever justified, whether the United States should stress unilateral or multilateral action in international affairs, and, finally, whether America should favor the Israelis over the Palestinians. Although these questions tap different aspects of foreign policy, people respond to the package in consistent ways. In the jargon of social science, the questions scale nicely, forming a single dimension.
To simplify presentation in the accompanying table, we report the percentage of each religious group that falls in the top half of public support for the Bush Doctrine. Thus, a score above 50 percent is more favorable than average, a score below 50 percent is more opposed. […]

We find vast religiously correlated differences among citizens in support for the Bush Doctrine. As the first column shows, Latter-day Saints are most positive, with 82 percent falling in the top half of the scale. Aside from the Mormons, evangelicals as a group do, in fact, provide disproportionate backing for the president’s policies, as critics contend. Interestingly, Hispanic Protestants, largely evangelical in theology, also exceed the sample average. Mainline Protestants follow, barely scoring on the positive side, and white Catholics are split right down the middle. Virtually all other religious groups (including Jews) are much less favorable toward administration policy, with black Protestants, the agnostic/atheist coterie, and other non-Christians (Muslims, Hindus, Buddhists, etc.) concentrated toward the bottom of the scale.
In addition to these differences among traditions, we find striking divisions within the larger Christian traditions. In each, traditionalists are most in favor of the Bush Doctrine, centrists less so, and religious modernists dissent in large numbers. (The same divisions can be seen even within smaller traditions, such as Hispanic Protestants and Catholics, Jews, and black Protestants, but the sample numbers are too small to report with confidence.) Altogether then, both membership in a religious tradition and theological traditionalism within the Christian traditions have important consequences for foreign policy attitudes.
Of course, politicians don’t exhibit equal solicitude for the views of every citizen: they are much more attuned to voters and, especially, to political activists. The scores for voters (column 2) and activists (column 3) reveal some interesting findings. Voters overall are actually a bit more supportive of the Bush Doctrine than the citizenry at large (52 percent), but activists are less favorable (only 44 percent).
The same basic religious patterns hold among voters and activists that we saw among citizens generally, but with some important modifications. First, for both evangelical and Catholic traditionalists endorsement of the Bush Doctrine rises as political engagement increases. (Among mainline traditionalists it goes up among voters, but retreats among activists.) In contrast, for mainline and Catholic modernists the president’s backers decline in strength as engagement increases, a pattern that also appears among the smaller faiths, and especially in the secular and agnostic/atheist groups. Thus, religious divisions over foreign policy exhibited by citizens generally are even wider among voters and, especially, activists.
The cause of these patterns is a complicated issue that we cannot fully address here. We can, however, identify three important factors, all of which have some influence. First, there may be a doctrinal basis for these differences. Thus, evangelicals’ distinctive posture may reflect the influence of dispensational theology, biblical literalism, Christian exclusivism, or perhaps moral dogmatism—”black or white” thinking. Conversely, the absence of such beliefs—or the presence of liberal religious or secular perspectives—may explain opposition to the president’s policies.
Second, religious leaders may have directed their flocks toward or away from the Bush Doctrine. Here, too, evangelicals provide a good example, given the strong support many clergy voiced for the Iraq war, and their suspicions about international institutions such as the United Nations. On the other side, the criticism that many mainline and Catholic clergy, including the Pope, directed toward facets of the Bush Doctrine may have attenuated support in those communities, at least among those hearing the cues.
Finally, foreign policy attitudes may simply be an artifact of partisanship and ideology. For example, evangelicals are a core GOP constituency and naturally endorse policies adopted by their conservative president and party leadership. Other religious groups may react in much the same fashion, depending on their own location in the current party lineup. In this context, it is worth noting that support for the Bush Doctrine matches very closely the share of the vote each religious group gave the president in 2004.
We did a modest test of these possibilities by incorporating measures of religious doctrine, attention to religious cues, and partisanship into a statistical analysis. All else being equal, religious doctrine makes a substantial contribution to support for the Bush Doctrine: Biblical literalists, dispensationalists, believers in the existence of Satan, and those who see salvation exclusively in Jesus score higher on the scale. And moral dogmatism plays a role: citizens who argue that there is a single standard of right and wrong for all times and places are much more likely to support the president.
Religious cues also make an independent contribution, largely reflecting the policy stance of the “governing” authorities in each tradition. Among evangelicals, those whose ministers preach on the Iraq war and terrorism are more supportive of the Bush Doctrine, as is the case among Hispanic Protestants. Among virtually all other religious groups—including mainline Protestants and white Catholics—those hearing pastoral discourses on these topics are less supportive of the president, sometimes substantially so. This effect is often even greater among political activists than among voters.
Finally, partisanship and ideology also shape assessments of the Bush Doctrine, even aside from the impact of religious doctrine and leadership cues: Republicans and conservatives score high on the scale, Democrats and liberals, much lower. In this regard, remember that President Bush actively courted evangelicals and other traditionalists before and during the 2004 campaign, and foreign policy was part of the pitch. Senator Kerry and the Democrats countered with appeals to other religious groups, apparently with some success.
In sum, American religious groups—and not just evangelicals—do indeed hold distinctive views on the Bush Doctrine. Evangelicals and traditionalists of all sorts are the strongest adherents, while the non-religious, religious modernists, and minority faiths are the most negative. These divisions become sharper as political engagement increases, and theology, religious leaders, and political identifications all play a role in deepening the chasm.
What may be most interesting is that the results within groups are as graphic as between groups, showing just how much liberal Protestants and Catholics resemble seculars rather than their putative co-religionists.