In the 1950s, the Fraternal Order of Eagles teamed up with movie director Cecil B. DeMille for a unique promotion of the epic movie The Ten Commandments. In a form of reverse product placement, the Eagles and DeMille donated monuments of the biblical Ten Commandments to communities all around the country. Rather than putting a product in the movie, the primary symbol of the movie was instead placed in prominent locations—in public parks, in front of courthouses, and in the case of Texas on the grounds of the state capitol. These monuments reflected the zeitgeist, as the 1950s brought public, even government-sanctioned, expression of religion to the fore in many ways. This was also the decade in which “In God We Trust” became the official national motto, and the Pledge of Allegiance was amended to include the words “under God.”
Those monuments stood for decades without causing a fuss. In recent years, however, they have led to court battles over whether their location on publicly owned land violates the constitutional prohibition on a government establishment of religion. In other words, fifty years ago these displays were so noncontroversial that they could safely be used as a marketing ploy for a big-budget Hollywood movie. Now they are the subject of litigation all the way to the Supreme Court.1
Something has changed.
In 1960, presidential candidate John F. Kennedy had to reassure Protestants that they could safely vote for a Catholic. (At the time 30 percent of Americans freely told pollsters that they would not vote for a Catholic as president.) At the same time, Kennedy won overwhelming support from his fellow Catholics, even though he explicitly disagreed with his church on a number of public issues. In 2004, America had another Catholic presidential candidate—also a Democratic senator from Massachusetts, also a highly decorated veteran, and also with the initials JFK. Like Kennedy, John (Forbes) Kerry also publicly disagreed with his church on at least one prominent issue—in this case, abortion. But unlike Kennedy, Kerry split the Catholic vote with his Republican opponent, and lost handily among Catholics who frequently attend church. Kennedy would likely have found it inexplicable that Kerry not only lost to a Protestant, but in George W. Bush, an evangelical Protestant at that. Writing about the religious tensions manifested in the 1960 campaign, political scientist Philip Converse described the election as a “flash of lightning which illuminated, but only momentarily, a darkened landscape.”2 Kerry’s candidacy was another flash of lightning, but the landscape it revealed had changed significantly. In 1960, religion’s role in politics was mostly a matter of something akin to tribal loyalty—Catholics and Protestants each supported their own. In order to win, Kennedy had to shatter the stained glass ceiling that had kept Catholics out of national elective office in a Protestant-majority nation. By the 2000s, how religious a person is had become more important as a political dividing line than which denomination he or she belonged to. Church-attending evangelicals and Catholics (and other religious groups too) have found common political cause. Voters who are not religious have also found common cause with one another, but on the opposite end of the political spectrum.
Again, something has changed.
This book is about what has changed in American religion over the past half century. Perhaps the most noticeable shift is how Americans have become polarized along religious lines. Americans are increasingly concentrated at opposite ends of the religious spectrum—the highly religious at one pole, and the avowedly secular at the other. The moderate religious middle is shrinking. Contrast today’s religious landscape with America in the decades following the Second World War, when moderate—or mainline—religion was booming. In the past, there were religious tensions, but they were largely between religions (Catholic vs. Protestant most notably), rather than between the religious and irreligious. Today, America remains, on average, a highly religious nation, but that average obscures a growing secular swath of the population.
The nation’s religious polarization has not been an inexorable process of smoothly unfolding change. Rather, it has resulted from three seismic societal shocks, the first of which was the sexually libertine 1960s. This tumultuous period then produced a prudish aftershock of growth in conservative religion, especially evangelicalism, and an even more pronounced cultural presence for American evangelicals, most noticeably in the political arena. As theological and political conservatism began to converge, religiously inflected issues emerged on the national political agenda, and “religion” became increasingly associated with the Republican Party. The first aftershock was followed by an opposite reaction, a second aftershock, which is still reverberating. A growing number of Americans, especially young people, have come to disavow religion. For many, their aversion to religion is rooted in unease with the association between religion and conservative politics. If religion equals Republican, then they have decided that religion is not for them.
Religious polarization has consequences beyond the religious realm, because being at one pole or the other correlates strongly with one’s worldview, especially attitudes relating to such intimate matters as sex and the family. Given that American politics often centers on sex and family issues, this religious polarization has been especially visible in partisan politics. A “coalition of the religious” tends to vote one way, while Americans who are not religious vote another.
The current state of religious polarization has led social commentators to use heated, even hyperbolic, language to describe the state of American society. The bestseller lists are full of books highly critical of religion, countered by pundits whose rhetoric decries a public square made “naked” by religion’s absence.3 In an overused metaphor, America is supposedly in the midst of a war over our culture.4
And yet, when one ignores these venomous exchanges, and looks instead at how Americans of different religious backgrounds interact, the United States hardly seems like a house divided against itself. America peacefully combines a high degree of religious devotion with tremendous religious diversity—including growing ranks of the nonreligious. Americans have a high degree of tolerance for those of (most) other religions, including those without any religion in their lives.
Religion’s role in America thus poses a puzzle. How can religious pluralism coexist with religious polarization?
The answer lies in the fact that, in America, religion is highly fluid. The conditions producing that fluidity are a signal feature of the nation’s constitutional infrastructure. The very first words of the Bill of Rights guarantee that Congress—later interpreted to mean any level of government—will favor no particular religion, while ensuring that Americans can freely exercise their religious beliefs. In the legal arena, debates over such matters as whether the Ten Commandments can be displayed on public property hinge on the interpretation of the Constitution’s words. More broadly, the absence of a state-run religious monopoly combined with a wide sphere of religious liberty has produced an ideal environment for a thriving religious ecosystem. Religions compete, adapt, and evolve as individual Americans freely move from one congregation to another, and even from one religion to another. In the United States, it seems perfectly natural to refer to one’s religion as a “preference” instead of as a fixed characteristic.
This state of flux has actually contributed to religious polarization. A fluid religious environment enables people seeking something different to leave one religion for another, to find religion for the first time, or to leave religion altogether. This churn means that people gradually, but continually, sort themselves into like-minded clusters—their commonality defined not only by religion, but also by the social and political beliefs that go along with their religion.
The malleable nature of American religion, however, means that these clusters are not bunkers. Instead, the same fluidity that contributes to religious polarization means that nearly all Americans are acquainted with people of a different religious background. Even if you personally have never gone through a religious change, you likely know someone who has. Furthermore, that someone is likely to be more than a passing acquaintance, but rather a co-worker, a close friend, a spouse, or a child. All of this religious churn produces a jumble of relationships among people of varying religious backgrounds, often within extended families and even households, which keeps religious polarization from pulling the nation apart.
The contrast between John F. Kennedy in 1960 and John Kerry in 2004 is thus doubly revealing. It not only highlights the new ways that religion divides American society but, more subtly, it also reminds us that old divisions are largely forgotten. In 1960, Kennedy faced overt hostility to his Catholicism, even in polite company. We find it no coincidence that this was also a time when there were many social barriers to relationships between Catholics and Protestants. John Kerry ran in a different world. By 2004 his Catholicism presented no problems for Protestants. We again find it hardly coincidental that in the years between Kennedy and Kerry, Americans of many different religious backgrounds increasingly came to connect with one another—as neighbors, friends, and spouses. That electoral flash of lightning in 2004 thus illuminated more than the changed political topography; it also exposed an altered social landscape. Interreligious personal connections have resulted in a social web interwoven with different religions and people with no religion at all—with implications far beyond presidential politics.
Over the last fifty years, American religion has thus experienced two countervailing transformations. The first is the emergence of a new religious fault line in American society. Left on its own, such a fault line could split open and tear the nation apart. The second change, however, is precisely why the fault line has not become a gaping chasm. Polarization has not been accompanied by religious segregation—either literally or even metaphorically. To the contrary, rather than cocooning into isolated religious communities, Americans have become increasingly likely to work with, live alongside, and marry people of other religions—or people with no religion at all. In doing so, they have come to accept people with a religious background different from theirs. It is difficult to demonize the religion, or lack of religion, of people you know and, especially, those you love. Indeed, interreligious relationships are so common that most Americans probably pay them little mind, and consider them unremarkable. But their very commonness makes them remarkable indeed.
Polarization and pluralism are the principal themes in the recent history of American religion, but they hardly exhaust all that has changed, is changing, and will change in the nation’s religious environment. The sheer vitality of religion in America means that it is ever evolving, although that evolution takes place against a backdrop of some constants too. We begin by asking how and why American religion became polarized, and close by asking—and answering—how polarization and pluralism can coexist. But to get from polarization to peaceful pluralism, we consider a number of other questions along the way:
To what extent do Americans engage in religious mixing and matching?
Which religions win, and which lose, in the religious marketplace? Historically, who have been the winners, and who the losers?
What keeps people in their congregations, and why do they switch from one to another? How have religious entrepreneurs responded to the second aftershock, which is pushing people—young people especially—away from religion?
How has religion engaged three major trends in American society: the revolution in women’s rights, rising income inequality, and growing ethnic and racial diversity?
What happened to cause religious devotion to be so strongly associated with partisan politics, and what will the future likely hold for the connections between religion and politics?
How does politics happen, or not, inside a congregation? How can religiosity be so closely associated with partisan politics when overt politicking from the pulpit is rare?
Who is right: those who make the case for the positive contribution of religion to civil society, or those who make the case against?
Any discussion of religion in America must begin with the incontrovertible fact that Americans are a highly religious people. One can quibble over just how religion, and religiosity, should be gauged, but, by any standard, the United States (as a whole) is a religious nation. In general, Americans have high rates of religious belonging, behaving, and believing—what social scientists call the three Bs of religiosity.5 Eighty-three percent of Americans report belonging to a religion; 40 percent report attending religious services nearly every week or more;6 59 percent pray at least weekly; a third report reading scripture with this same frequency. Many Americans also have firm religious beliefs. Eighty percent are absolutely sure that there is a God. Sixty percent are absolutely sure that there is a heaven, although fewer (52 percent) have this level of certainty about life after death. Slightly fewer, 49 percent, are certain that there is a hell.
Yet it is also important to note that not every American is so religious, or religious at all. After all, 15 percent never attend religious services, 17 percent do not identify with a religion, 20 percent are not certain about the existence of God, 40 percent are not sure there is a heaven, and 48 percent are not certain there is life after death.
When we put these basic facts together, a picture of religion in America comes into focus. Americans overwhelmingly, albeit not universally, identify with a religion. Identity, however, does not necessarily translate into religious activity because not all who identify with a religion frequently attend religious services, or engage in other religious behavior. The vast majority of Americans also believe in God, but Americans are less sure about life beyond the grave. Ever an optimistic people, Americans are more likely to envision heaven than hell. In fact, more Americans are certain about heaven than are certain about life after death. When we probe further, we find that Americans believe in a God who is loving and not very judgmental. Sixty-two percent say they “very often” feel God’s love in their life, while only 39 percent say that they feel God’s judgment this frequently. Americans’ God is more avuncular than angry, and it turns out (as we shall see in Chapter 13) that this sort of everyday theology has real implications for the ways in which Americans get along with one another. This is merely one example among many that we shall discuss in which Americans’ religiosity and community connections are closely tied together.
By any objective standard, this profile shows the reasonably high religiosity of the United States. That profile appears even stronger when the United States is compared to the rest of the planet, especially other industrialized, democratic nations. The United States ranks far ahead of virtually all other developed nations in terms of all three Bs of religiosity. To take just one example, Figure 1.1 displays how the United States compares to the rest of the world in a measure of religious behavior, namely the weekly attendance of religious services. Indeed, in this global ranking of religious observance America edges out even the Iran of the ayatollahs.
The United States also has an equally high degree of belonging and believing. For instance, 38 percent of Americans report being an active member of a church or religious organization, compared to only 16 percent of Australians, 9 percent of Italians, and 4 percent of the French. Likewise, while nearly half, 47 percent, of Americans affirm that religion is “very important” in their lives, only 17 percent of the Swiss, 12 percent of the Dutch, and 9 percent of Swedes say the same.7
COMPARED TO OTHER INDUSTRIALIZED NATIONS, THE U.S. HAS A HIGH RATE OF WEEKLY ATTENDANCE AT RELIGIOUS SERVICES
SOURCE: WORLD VALUES SURVEY, 2005–2007.
Americans’ high religiosity is thrown into especially sharp relief with a comparison to our close cultural cousins, the British. While 54 percent of the British say they never pray, only 18 percent of Americans say the same. A third of Americans believe that scripture is the actual word of God, compared to only 9 percent of the British.8
One measure of Americans’ religiosity illustrates particularly nicely how religion permeates the lives of many but is absent from those of others. Almost half (more precisely, 44 percent) of the American population reports saying grace or a blessing before meals at least daily, while almost precisely the same percentage (46 percent) says grace occasionally, or never.9 (In this context, we use the Christian term “grace” as a shorthand for all prayers said before meals, in whatever religious tradition.) We are hard-pressed to think of many other behaviors that are so common among one half of the population and rare among the other half—maybe carrying a purse.
We single out grace saying because it turns out to be an excellent indicator of overall religiosity that, in turn, predicts many other attitudes and behaviors. For example, grace saying reappears in Chapter 11, where we show that the frequency of saying blessings before meals has a strong connection to one’s partisan politics.
The Faith Matters Surveys
We can tell you about grace saying and many other aspects of religiosity in the United States because we have conducted extensive surveys of Americans in which we asked a wide-ranging set of questions about their religious lives, as well as their civic involvement, social relationships, political beliefs, economic situation, and demographic profile. The first survey was administered to a randomly selected, nationally representative sample of 3,108 Americans in the summer of 2006, and then followed up with a second, separate survey with as many of the same people as we could find (1,909, to be precise) in the spring and summer of 2007. Known as the Faith Matters surveys, together they constitute one of the most detailed examinations ever undertaken of Americans’ religious and civic lives. As will become apparent, interviewing people more than once has turned out to be extremely valuable in understanding religious change. Because of the dynamism in American religion, even the short period of time that elapsed between the first and second interviews provides insights into small but significant shifts in various aspects of Americans’ religious lives.10 Throughout this book, we make repeated reference to the data collected in the two Faith Matters surveys. Usually we report on results from the 2006 survey, since it has a larger and more representative sample, and also included a more extensive set of questions. Whenever it is appropriate, however, we report results from the 2007 survey. In either case, we always identify the survey to which we are referring.
Throughout this book, we rely on the Faith Matters surveys heavily but not exclusively. The coin of our realm is convergent validation—that is, testing interpretations against as many sources of data as possible—so whenever possible, our key claims have been confirmed with the General Social Survey, the National Election Studies, the Pew Religion and Public Life surveys, and other comparable, publicly available sources of data. We have listened carefully to what the data have to say; the data are loudest when they speak in harmony.
Anyone who wishes to describe and then analyze the state of religion in America has to grapple with the fact that there are an enormous number of religious faiths, and myriad denominations and other subgroupings within those faiths. The aforementioned religious fluidity has meant schisms, mergers, the founding of new faiths, and the arrival of faiths from other nations. Take, for example, Lutheranism, an important group within Protestantism. Lutheranism is not monolithic. There are the Evangelical Lutheran Church in America (ELCA), the Missouri Synod Lutherans, the Wisconsin Synod Lutherans, and still other Lutheran denominations. While these denominations share common Lutheran DNA, as they all have Martin Luther as a progenitor, they nonetheless exhibit significant differences in worship, practice, and theology. And that is just the many varieties of Lutheranism. Multiply the same phenomenon across Presbyterians, Methodists, Baptists, Pentecostals, Jews, and so on and you begin to see the complexity of trying to discuss each denomination separately. Further complicating matters, there is also a growing trend toward Christian churches with no denominational label at all.
The situation, though, is not hopeless, as American religion can be usefully analyzed using a taxonomic system that, to paraphrase Albert Einstein, is as simple as possible, but no simpler. The huge array of denominations can be grouped into a more manageable number of religious traditions. To use a biological metaphor, religious traditions are like a genus, while the individual denominations are like species.
Evangelical, Mainline, and Black Protestants
Protestantism presents the greatest challenge to any system of religious morphology, as no other category better illustrates the mutability of religion in America. To return to biological taxonomy: If a religious tradition is a genus, then Protestantism is analogous to a religious family. Within that family there are three significant genera: evangelical, mainline, and Black Protestant.
Evangelical Protestants comprise one of the most significant religious traditions in America—particularly for understanding change in American religion. Historian Mark Noll notes that evangelicalism dates as far back as the early eighteenth century, when a movement began within Protestantism to find a “true religion of the heart.”11 Evangelicalism was the dominant strain within American Protestantism through most of the nineteenth century. Then, in the late 1800s and early 1900s, Protestants split over a debate between fundamentalists and modernists, a split that still echoes today. For our purposes, an evangelical Protestant is someone who, knowingly or not, has taken the fundamentalists’ side in that debate. During this period, writes sociologist Christian Smith, Protestant churches increasingly adopted “liberal theology, biblical higher criticism, and an increased skepticism about supernaturalism.”12 The result was a parting of the ways between these questioning modernists and the fundamentalists, who held fast to a more traditional, and thus conservative, interpretation of scripture.
While the fundamentalist–modernist debate raged within religious denominations, it also spilled over into American society more generally. One important rallying cry for the fundamentalists was a rejection of evolution as an explanation for the origin of man. The issue came to a head in 1925 with the famous Scopes Monkey Trial, over the question of whether evolution could be taught in the public schools of Tennessee. Few today remember that the fundamentalists won the battle in the courtroom, as the state’s anti-evolution statute was upheld. Better remembered is that they lost the war of public opinion beyond the courtroom, as their beliefs were subjected to national ridicule. In the wake of this derision fundamentalists largely retreated from engagement with wider American society.
Fundamentalists began to reemerge from their self-imposed exile with the founding of the neo-evangelical movement in the wake of the Second World War. With Billy Graham as their most public face, the neo-evangelicals were moderates within the fundamentalist wing of Protestantism who sought to soften the hard edge of fundamentalism and reengage with American society. They maintained orthodox Protestant beliefs, but shed the anti-intellectualism and insularity that had come to characterize fundamentalists in the wake of their post-Scopes withdrawal. This new style of conservative Protestantism has become the norm, such that in public parlance the “neo” came to be dropped from their name. The term “evangelicalism” now encompasses all theologically conservative Protestants (except Black Protestantism, as explained below), whether they be Billy Graham–like neo-evangelicals, members of “seeker-sensitive” megachurches, traditional fundamentalists, or Pentecostals.13
Because they are an amorphous group defined by admittedly blurry boundaries, one can debate just who counts as an evangelical. The label is not necessarily one that people willingly adopt for themselves, even if their belonging, believing, and behaving all align with the standard scholarly usage of the term. This was brought home to us when we interviewed members of the Saddleback megachurch (see Chapter 2). We asked a number of people at this high-profile church, widely identified as quintessentially evangelical, how they described their religious affiliation. Overwhelmingly, they said “Christian,” not “evangelical.” Similarly, many people reasonably identify themselves as belonging to a specific denomination, like the Missouri Synod Lutherans, rather than a nebulous movement like evangelicalism.
The solution to this definitional ambiguity is to identify evangelicals by their congregation’s denominational affiliation (or, as the case may be, the absence of such an affiliation).14 Therefore, when we refer to “evangelicals,” we mean people who report identifying with one of a large number of denominations that generally endorse the tenets of evangelicalism.15 For our purposes, evangelicals also include people who attend a nondenominational church, since, in recent years, a large number of nondenominational churches are evangelically inclined (e.g., the typical megachurch is both evangelical and nondenominational).16
While evangelicals are the heirs to the fundamentalists, mainline Protestants are descendants of the modernists. Loosely speaking, mainline Protestant denominations are more liberal theologically than their evangelical counterparts. Importantly, they are more likely to emphasize the Social Gospel—that is, the belief that a Christian’s priority should be the reform of social institutions—than personal piety. As described by political scientists Kenneth Wald and Allison Calhoun-Brown:
Stressing Jesus’s role as a prophet of social justice, the mainline tradition sanctifies altruism and regards selfishness as the cardinal sin. In this tradition, which extends membership to all and understands religious duty in terms of sharing abundance, the Bible is treated as a book with deep truths that have to be discerned amidst myth and archaic stories.17
The term “mainline” connotes that these are the denominations that have historically been the closest thing to establishment churches in America: the Episcopalians (the American branch of the Anglican Communion), for example, and the Congregationalists (successors to the Puritans).18 By the 1950s, when our story opens, mainline Protestant denominations—Methodist, Lutheran, Presbyterian, Episcopalian, Congregationalist, and a few others—represented the dominant religious tradition in America, but as we shall see, that dominance would change dramatically in the ensuing half century.
While the split between evangelical and mainline Protestants centers on theology, Black Protestantism—the third tradition within the Protestant family—is instead defined by race. Black Protestantism is a legacy of racial segregation. As detailed in Chapter 9, the Black Church (the term used to refer to all historically African American congregations and denominations) has a long and distinctive history in the United States. Black Protestants generally blend an evangelical focus on personal piety with a strong dose of Social Gospel. Just as importantly, the Black Church is an inherently racialized institution—race is integral to Black Protestants’ theology, iconography, and worship. The result is a unique religious tradition.19
Catholics, Jews, and Mormons
Of course, Protestantism does not exhaust the many varieties of religion in the United States. Catholics are also a major share of the religious population, but they are more easily recognized through self-identification, since their denomination, religious tradition, and self-identity are all one and the same. Catholics thus use the same label to describe themselves as do academics and other observers. You might say that Catholics know who they are.
Jews and Mormons20 can also be easily recognized through self-identification. While they are each a much smaller share of the population than Protestants or Catholics, both are highly distinctive traditions that, because of their size, are often neglected in analyses of the American religious environment.21
The sheer variety of American religious traditions means that, even after classifying most Americans into these religious traditions, there are still a small number of people spread across a wide array of different religions. These include Sikhs, Hindus, Buddhists, Muslims, and many more. Many of these other faiths are a growing presence on the American religious scene, having grown from roughly one percent in the 1970s to between 2 and 3 percent today. But being, at most, 3 percent of the population still means they collectively comprise a small proportion of the national population, with each individual group being smaller still. Since the Faith Matters survey was administered to a randomly selected representative sample of the United States, it contains the correct proportion of each group. But the absolute number of these other faiths is too small to permit reliable analysis.22 We are thus limited in what we can report about these disparate faiths.
No Religion/The “Nones”
The final category consists of people who report no religious affiliation, those who have come to be called the “nones.”23 That is, when asked to identify with a religion, they indicate that they are “nothing in particular.” These nones are not necessarily hard-core secularists, as we shall discuss in Chapter 4. This category, though, does include that small fraction of the American population who describe themselves as either atheists or agnostics, although these labels turn out to have little common usage.24 While atheism has recently gained prominence, particularly on the bestseller lists, self-identified atheists and agnostics comprise a vanishingly small proportion of the U.S. population. For instance, in the 2006 Faith Matters survey precisely five people out of 3,108 chose either label.
Figure 1.2 displays the percentage of the American population in each religious tradition, sorted from largest to smallest. The largest group are evangelicals, with roughly 30 percent of the U.S. population fitting that classification. While evangelicals grew in the 1970s and 1980s, their proportion of the population has been slowly declining since about 1990.
The single largest denomination is the Roman Catholic Church. Catholics comprise about a quarter of the U.S. population, a proportion that has remained steady for decades. However, as we discuss in Chapter 9, Catholics’ steady share of the population obscures a dramatic change within American Catholicism. Over the last few decades, large numbers of “Anglo”—that is, non-Latino—Catholics have been dropping out of or disengaging with the Catholic Church, without being replaced by other Anglo converts. During the same period, however, the number of Latino Catholics has grown tremendously. Given current trends, this demographic transformation means that the American Catholic Church is on its way to becoming a majority-Latino institution, as we discuss in Chapter 4.
RELIGIOUS TRADITIONS IN THE UNITED STATES
SOURCE: FAITH MATTERS SURVEY, 2006.
The third largest “religious” group in the United States is actually defined by the absence of a religious affiliation—the “nones.” There are more nones (17 percent) than mainline Protestants (14 percent), a striking fact given that the mainline wing of Protestantism once represented the heart and soul of American religion and society. Significantly, the ranks of the nones have been growing, while the mainline Protestants’ share of the population has been shrinking.
Note also that Jews, one of the oldest religious traditions, rank right alongside Mormons, one of the newest (both are roughly 2 percent of the population).
Divvying up the population into religious traditions is only one way to make sense of the American religious landscape. Think of religious traditions as being like the “flavor” of one’s religion. And just as flavors come in varying levels of intensity, so can religion. Religious intensity can also be referred to as “religiosity,” and measured with a series of questions that tap into different ways of being religious, including both behaving and believing. In the next few pages we describe the way we measure religiosity in detail. Our discussion gets into statistical matters that some readers would probably prefer to skip over and this can be done without a loss of continuity. We provide this level of detail for the practical reason that religiosity is a recurring matter in the pages to follow. However, it also speaks to a fundamental question that goes right to the heart of any study of religion: What does it mean to be religious?
The specific questions we use to measure religiosity include the following:
• How frequently do you attend religious services?
• How frequently do you pray outside of religious services?
• How important is religion in your daily life?
• How important is your religion to your sense of who you are?
• Are you a strong believer in your religion?
• How strong is your belief in God?
Taken together, these questions run the gamut of ways that a person might be religious. They include the public activity of attending religious services, the (typically) more private activity of praying outside of religious services, the salience of religion in one’s life, and how strongly someone believes in God.
We have combined these questions into a single measure called the “religiosity index,” because when analyzing an overall concept they are more illuminating in combination than individually. Any single measure of religiosity, no matter how good, will inevitably misclassify a few people. Church attendance is, for example, an excellent measure of religious commitment in most cases, but for the elderly or infirm it can be misleading. Strength of belief in God is usually a good indicator, to take another example, but there are some exceptions. As St. Augustine said, “Doubt is but another element of faith,” so for some deeply religious people, the absence of doubt is not the best measure of religious commitment. Combining multiple indicators, assuming that each of them is fairly accurate, produces an even more reliable overall measure, for the same reason that diversification improves the performance of a stock portfolio. In loose terms, the religiosity index is a weighted average of responses to these questions. Those that contribute more to the common thread holding them all together—in this case, religiosity—receive more weight. (For our statistically savvy readers: We have created a factor score of these six items.25)
Since the nature of what it means to be religious is inevitably fraught with ambiguity and controversy, it is important to keep a few things in mind about this method of measuring religiosity. First, some readers may wonder whether, say, frequency of attendance at worship services (a public activity) really taps into the same concept as frequency of prayer (often done in private), and whether these two types of behaving really align with measures of belief like the existence of God. It is true that these index items are logically distinct, but in practice they are tightly bound together. Nearly all people (99 percent) who say that religion is very important in their lives are also “absolutely certain” that they believe in God; most people (79 percent) who attend religious services also pray at least once per day. And so on.
Second, keep in mind that this index does not rest on the inclusion or exclusion of any one particular item. For example, excluding how frequently someone prays does not change the substantive results contained herein one whit (likewise for any other individual item in the index). The fact that religiosity does not hinge on a single measure underscores that together these different questions are tapping into a common underlying concept.
Third, readers may wonder whether these particular questions favor one religious tradition over another. This is a common concern when social scientists study religion, as religiosity is sometimes measured with questions that are normative within Protestantism, specifically for evangelicals. Some such indices include items that ask whether the Bible is inerrant, or whether the respondent has ever been “born again.” Such questions are as distinctively Protestant as keeping kosher is distinctively Jewish. Our religiosity index avoids the problem of parochialism by including only items that could apply to all religious traditions. Still, we acknowledge the concern that perhaps this particular religiosity index is inadvertently biased toward evangelical Protestantism, or some other religious tradition.
There is, however, a conundrum in trying to determine whether this, or any other, method of measuring religiosity favors one religious tradition. How would we tell whether it is biased toward one tradition? Because people in that tradition score more highly on it. But do members of that tradition show up as highly religious because of the idiosyncrasies of the index, or because—no matter the measure—they truly are more religious? We will see that evangelicals, Black Protestants, and Mormons all rank high on the religiosity index. Is that because the index is somehow rigged to pick up the particular ways that members of these three traditions live their religions? In this case, the validity of our religiosity index as a general purpose gauge is bolstered by the fact that the same index, with questions worded identically, shows Muslims in Britain to have an extremely high level of religiosity.26 An index that is allegedly biased toward evangelical Protestants, Black Protestants, and Mormons in the United States could hardly also be biased toward Muslims in Britain.
A less formal but perhaps just as convincing test of this, or any other, way of empirically measuring religiosity is to ask whether it matches our intuitive sense of what it means to describe someone as religious. If you know someone who attends religious services frequently, prays often, has a strong belief in God, holds religion to be important, believes that religion defines her identity, and says that she strongly believes in her religion, would you not describe her as highly religious? And, likewise, would you describe someone who does not do or believe these same things as not being religious? That, perhaps, is the most convincing test of all.
While not the only possible way of measuring a concept as multifaceted as religiosity, the index is an empirically tractable, conceptually coherent, and intuitively compelling method of doing so. One important question that will recur in the following pages is this: Which matters more, the flavor of a person’s religion or the intensity? Does a highly devout Catholic have more in common with, say, a lapsed Catholic or a devout Jew? The answer will vary, of course, but for some matters we will see that intensity actually matters more than flavor. In that sense religiosity itself (as distinct from membership in a particular denomination or sect) turns out to be increasingly important in contemporary America.
Comparing the Most and Least Religious Americans
Consider the religiosity index to be a measurement tool, like a thermometer. Using it enables comparisons between Americans of varying levels of religiosity. Putting the most religious Americans (those in the top 20 percent) next to the least religious (bottom 20 percent) reveals that, on some matters, they differ dramatically. On others, there are few differences at all.
The most and least religious Americans differ, for example, on how “spiritual” they consider themselves to be. While 4 percent of the least religious describe themselves as very spiritual, 80 percent of the most religious do. Among rank-and-file Americans spirituality and religiosity go hand in hand. Americans’ attitudes on evolution are also sharply divided by religiosity. Less than 2 percent of the most religious Americans believe that “human beings have developed over millions of years from less advanced forms of life but God had no part in this process,” compared to 45 percent of the least religious. Over three quarters of the most religious reject evolution altogether, and believe instead that God created human beings less than ten thousand years ago. Interestingly, this position is also held by 16 percent of the least religious.27
The most and least religious also differ on ways to spend their leisure time. Both casinos and R-rated movies are apparently more likely to be frequented by secular, not religious, Americans. Sixty-one percent of those at the top of the religiosity index say that gambling is always wrong, compared to only 10 percent of those at the bottom of the index. Roughly the same percentages also believe that it is always wrong to watch movies with “a lot of violence, profanity, or sexuality.”28 But this is not to say that religious and not-so-religious Americans could never find something to do together. Both are equally likely to participate in the two great American pastimes of watching sports and eating out.29 And even if they stay in, they are about equally likely to watch television.30
When it comes to leisure activity of another sort, more and less religious Americans disagree sharply on the propriety of premarital sex. While 4 percent of the least religious portion of the population say that premarital sex is always wrong, only 3 percent of those in the top 20 percent of religiosity say that sex before marriage is never wrong.
Similarly, opinions on abortion vary substantially according to religiosity. Sixty-five percent of the least religious Americans believe in a woman’s unfettered right to choose when it comes to abortion, a position held by only 13 percent of the most religious. Attitudes toward homosexuality differ dramatically as well. Nearly nine out of ten highly religious people say that homosexual activity is always wrong, in contrast with two out of ten of secular Americans. In Chapter 11, we shall see that both abortion and homosexuality have come to be especially salient in contemporary politics, which in turn has led to a religious divide at the ballot box.
Abortion and homosexuality have an unusually strong connection to religiosity. Smaller differences are seen on other matters. While just 6 percent of secular Americans believe that divorce is always wrong, 24 percent of the highly religious believe the same. That is surely a nontrivial gap, but it also means that three quarters of religious Americans approve of divorce in at least some circumstances.
When it comes to the public policy question of how the government spends tax money, religious and nonreligious Americans are more alike than different. Majorities of both want to spend more on conservative issues like fighting crime and protecting the border, but majorities of both also support the liberal position of more spending to help the poor.31
AMERICANS’ RELIGIOUS PROFILE
When we compare the religiosity of Americans to one another, interesting patterns appear. Take, for example, what happens when you map the flavor of religion (religious tradition) against intensity. Just as some flavors are more likely to be intense than others, so are members of some religious traditions more likely to be religiously intense than others. Figure 1.3 shows both flavor and intensity. The more intense flavors—the traditions that are more highly religious—are to the right of the line, while those that are less intense are to the left. The line itself represents the national average of religiosity.
Which religious tradition has the average level of intensity? Catholics, with mainline Protestants coming close. Not surprisingly, nones are the least religious group in the population. Jews are next, yet even though they fall below the average, they still score well above the nones. (Non-Jews may be surprised to find, as our synagogue visit described in Chapter 10 illustrates, that half of all self-identified Jews are not so sure they believe in God.) On the other side of the spectrum, the three most religious groups in America are Mormons, Black Protestants, and evangelicals, in that order. Their shared level of religious intensity means that members of these three traditions have much in common, although we shall see that they do not see eye-to-eye on everything.
Comparisons across religious traditions are merely one way to describe the religious landscape in America. In addition to asking which religious traditions are most, and least, religious, it is also informative—and perhaps illusion-shattering—to see the types of individuals who are, and are not, religious. We describe these comparisons below; you can see them in Figure 1.4.
MORMONS, BLACK PROTESTANTS, AND EVANGELICALS ARE THE MOST RELIGIOUSLY OBSERVANT GROUPS IN AMERICA
Religiosity is standardized with a mean of 0 and a standard deviation of 1
SOURCE: FAITH MATTERS SURVEY, 2006.
First, women are modestly but consistently more religious than men. According to the 2006 Faith Matters survey, women are more likely to say that they consider themselves to be spiritual and to report having experienced the presence of God. And this is only the beginning. More women than men say that right and wrong should be based on God’s laws rather than the views of society; women are more likely to believe that God created the world less than ten thousand years ago. Women more frequently say that there are clear guidelines to good and evil. More women than men believe that the world will end soon, that scripture is the literal word of God, and that everyone will answer for their sins. Women read scripture, talk about religion, and read religious books more than men. You get the point. No matter the specific yardstick, women exhibit a greater commitment to, involvement with, and belief in religion.32
RELIGIOSITY VARIES A LOT BY RACE AND AGE, AND A LITTLE BY GENDER, TYPE OF COMMUNITY, AND INCOME
Vertical axis is religiosity, standardized with a mean of 0 and a standard deviation of 1
RACE AND ETHNICITY
TYPE OF COMMUNITY
SOURCE: FAITH MATTERS SURVEY, 2006.
Second, African Americans are far more religious than whites, or any other ethnic or racial group in America. Nearly 60 percent of blacks report attending religious services “nearly every week,” compared to 39 percent of whites; 84 percent of blacks say that religion is very or extremely important to them, while 56 percent of whites do. Seven in ten African Americans report that their religion is very important to them when making personal decisions, twice the level for whites (35 percent). Eighty-two percent of blacks in America report saying grace at least daily, compared to 38 percent of whites. As with comparing men and women, we could go on, but the pattern is clear. Religion infuses the lives of African Americans in a way it does not for most whites. By nearly every indicator, Latinos are also more religious than whites. Yet lest one think that religiosity is simply equated with minority status, Asian Americans are less religious than whites. We shall have more to say about race, ethnicity, and religion in Chapter 9.
Age matters a lot too, as the old are more religious than the young. There can be different reasons for such variation by age, including natural variation in the life cycle—people who are closer to the grave tend to be more religious—as well as generational differences that are frozen in place as people age. The explanations for these differences are treated in more detail in Chapter 3, but for now note that, as a descriptive matter, being older means a higher likelihood of being religious.
Religiosity also varies by the size of the community in which one lives. John Mellencamp sings that he “was taught the fear of Jesus in a small town” and it appears he is not alone. People who live in rural communities are more religious than city folk, although the difference is modest.
Furthermore, Southerners are more religious than the rest of the country.33 As can be seen in Figure 1.5, Arkansas, Louisiana, Mississippi, and Alabama are the most religious states in the union, with the bordering states just slightly less so. Utah is also a highly religious state, nearly matching the Bible Belt for religiosity. The patch-work nature of America’s religious quilt is underscored by the fact that Utah is bordered by Colorado, one of the least religious states. Colorado’s secularism, though, is less than that of the states of the Far West and the Northeast, which have the lowest religiosity of all.
THE DEEP SOUTH, UTAH, AND THE MISSISSIPPI VALLEY ARE THE MOST RELIGIOUS REGIONS
SOURCE: RELIGIOUS LANDSCAPE SURVEY, PEW RESEARCH CENTER.
Income has a more complex relationship to religiosity. At the extremes, the very poor are somewhat religious while the very rich are somewhat secular. But those right in the middle of the income scale (earning between $40,000 and $50,000 per year) are about as religious as the very poor. Further complicating matters, attendance at religious services is not related to income. No matter their income, roughly two out of five Americans report attending religious services weekly. And when it comes to education, another measure of social class, having more education corresponds to a higher level of attendance at religious services.
Comparisons across subgroups within a population are always tricky to interpret, because one characteristic can actually be standing in for another. Consider the claim that iPhone users have a low level of religiosity (which may or may not be true, but works as an illustration). Is it because they own an iPhone? Or is it because iPhone owners tend to be young, and young people are (on average) less religious than their elders? The latter seems more likely. Like-wise, is the South more religious than (most of) the rest of the country because more African Americans live there than in the rest of the country? Or because more Southerners live in small towns?
The answer to such questions lies in testing the impact of many characteristics—age, gender, income, and so forth—on religiosity simultaneously using the statistical method of multiple regression, which will reemerge often in the subsequent chapters. For the statistical novice, this type of analysis enables us to see whether each of these demographic characteristics continues to be a predictor of religiosity, even when accounting for all of the other characteristics at the same time. That way, if one characteristic is really just standing in for another—if the South is serving as a proxy for living in a small town, say—its statistical connection to religiosity will disappear. It will have been revealed as a substitute for something else. Such a statistical analysis reveals that gender, age, race/ethnicity, size of community, and region all have an independent connection to religiosity. Income, however, does not.
Given all this, who personifies the most religious type of American? An older African American woman who lives in a Southern small town. And the least religious? A younger Asian American man who lives in a large Northeastern city.
While there is every reason to think that race, age, and geography have long been related to religiosity, the religious changes we describe later on highlight that all three have become more strongly predictive of religiosity in recent years. Chapter 4 details a general drop-off in religiosity, but it is concentrated among whites. Accordingly, over the last thirty years, the gap between black and white religious observance has widened. Furthermore, that drop-off in whites’ religiosity is also more pronounced among young people and Northerners. And while there is no overall trend for gender, the demographic slice of the population that is most rapidly turning away from religion is young men. In short, the most religious social categories in America are becoming even more religious, and the least religious are becoming even less religious.
CONGREGATIONS MATTER TOO
While individuals’ religious involvement and commitment are obviously a vital component of the American religious landscape, neither are they the whole story. Americans generally do not worship alone, but instead gather in congregations. The importance of the congregation is made clear by the extent of congregational involvement within the American population. As shown in Figure 1.6, more Americans are involved in a religious congregation than in any other type of association, group, or club. The 2006 Faith Matters survey asked respondents to indicate whether they belong to a wide array of groups: from hobby groups to professional associations to self-help programs. Three in five Americans (62 percent) have a particular place of worship where they attend services. The next most popular group is the extremely broad category of “hobby, sports, arts, music, or other leisure activity”; about half of all Americans are involved in a group of this type.
Many Americans have a level of involvement in their congregation that exceeds mere membership. Thirty-six percent of the total population report participating in either Sunday School or another form of religious education, while a quarter participate in prayer or other small groups associated with their congregation (13 percent do so monthly or more frequently). Fourteen percent of all Americans have served as an officer or committee member within a congregation.34 A lot of Americans apparently like their congregation enough to invite others to attend it as well. Over half of all churchgoers, 55 percent, have invited someone to visit their congregation.35
CONGREGATIONS ARE THE MOST COMMON FORM OF ASSOCIATION IN AMERICA
SOURCE: FAITH MATTERS SURVEY, 2006.
To Americans reading this book, an emphasis on congregations as the primary organizational locus for religion undoubtedly seems familiar, as this is the most prevalent form of religious organization within the United States. Indeed, many Americans have probably never thought much about the alternative organizational forms for religion found in other societies. However, the congregation as an all-purpose association with members who choose it, belong to it, and make contributions to it is actually a very Protestant model of religious organization.36 This form and function of the typical American congregation—of whatever religious tradition—is thus a consequence of America’s Protestant heritage. The United States may not be a Protestant nation in law, but its Protestant legacy shapes the contours of the religious landscape.37
The centrality of the congregation, and its Protestant influence, can be seen in how immigrants’ religions adapt to the American religious ecosystem upon arriving in the United States. Even faiths that are not organized around the congregation in other nations come to adopt a congregation-based structure here in the United States. From there, it is a small step to adopting many of the same practices as American, especially Protestant, congregations. For example, Islamic mosques in the U.S. often hold Sunday school, or provide a social hall for community events—not what they typically do in other nations. In the U.S., imams are frequently called upon to serve as counselors and to engage in public relations, responsibilities outside the purview of imams elsewhere, but common for many congregational leaders in the United States. Alan Wolfe, a keen observer of religion in America, describes the contrast between mosques in America versus those in Muslim-majority nations:
Without intermediaries that stand between the believer and God, Islam has not traditionally had churches in the way Christians understand that term. Rather than a congregation with a fixed membership, mosques in Muslim societies were—and continue to be—convenient places into which one steps in order to pray, depending on where one is in the course of the day. . . . But in the United States, mosques inevitably come to resemble churches.38
Congregationalization is not limited to Muslims. For example, Hindu temples in the United States also have a churchlike feel to them, even though Hinduism, like Islam, is not typically organized around a local congregation. Nor is this a new phenomenon. Writing back in 1948 a sociologist described American Buddhist communities as having “congregational bodies analogous to those which appear in contemporary Christian, and particularly, Protestant churches.” Furthermore, these individual congregations were “in no way reminiscent of the temple structure of the Japanese homeland.”39
With congregations as the dominant mode of religious organization, religious communities are a common nexus for friendships—whether because one becomes a member of a congregation and finds friends there, or one makes friends and then joins their congregation. A majority of Americans, 56 percent to be precise, have at least one close friend in their religious congregation.
The prevalence of friends made at one’s place of worship serves to illustrate the social significance of America’s congregations. Faith-based social networks tend to keep people from switching congregations, foster good citizenship