Delivered October 4, 2007

I appreciate the opportunity to be here and to speak with you today about American religion. The myths I want to talk about today are the old-fashioned kind: falsehoods, errors of fact, misinterpretations. I want to correct those errors, if I can, or at last cast enough doubt on them that if you happen to have believed them, you will now have reason to question them.

I grew up in Kansas where my parents and I attend­ed a little Baptist church in our rural community. Plain speaking was all we knew. I hope to engage in some plain speaking today to correct some mistaken im­pressions of American religion.

I have spent more than 30 years studying American religion. I've designed and conducted more than a dozen major national studies on various aspects of American religion, analyzed dozens of other national surveys, and collected in-depth interview transcripts in which more than a thousand people talked at length about their faith journeys. I have written books and articles based on this information and have assisted journalists and other writers who were committed to getting accurate information to the public.

But there is a dynamic in the modern media that too often gets the story wrong. Sometimes it starts with a hunch, which a journalist seeks evidence to support and an editor turns into a headline, and some­times it starts with academics themselves, running with an idea that gets more attention than it should. There is an enormous amount of misinformation out there about American religion, and this is unfortunate, especially because it misleads the public and our pol­icymakers, but also because there is usually good solid evidence that should be brought to light.

The particular falsehoods I am going to talk about today are ones I happen to believe are especially mis­leading. They have been recounted again and again in the media, and yet they are easily refuted.

Myth #1: America is in the midst of a religious and spiritual awakening.
The specific formulation of this notion can be traced to a single book by the distinguished econo­mist Robert Fogel,[1] but it was an idea that became popular among writers who had never heard of Robert Fogel. Among journalists, it started to circu­late soon after the Christian conservative movement began gaining headlines in the 1980s, which was understandable but in another sense curious because Fogel had actually been talking about a presumed awakening in the 1960s and '70s, what others had termed the Age of Aquarius, or simply New Age religion.

Among academics, the idea gained popularity among scholars of immigrant churches, Pentecostal movements, and megachurches. They saw vitality everywhere.[2] It was easy to imagine that American religion went through cycles, through ups and downs. If the 1960s had witnessed the death of God movement, the 1980s would be what none other than the distinguished scholar Daniel Bell had anticipated in 1976 as a "return to religion."[3] The search for God would find new leaders and new followers, just as it did when Jonathan Edwards and George Whitefield led the first awakening in the 1740s and when subsequent revivals occurred in the 1830s and again at the start of the 20th century. And it was not an argument about some vague inner working of the Holy Spirit that could not be seen or would not be observed for years to come. The idea was that this new awakening would be evident in a return to institutional religion itself and, specifically, in rising church attendance.

There was some evidence, of course, which seemed to support the idea. On several occasions, the Gallup Organization reported that Americans believed the importance of religion in America was increasing.[4] A study of baby boomers showed that many of them were returning to the houses of wor­ship they had abandoned when they were younger.[5] In self-report questions, many Americans said they were more interested in religion than they had been five years previously.[6]

In recent years, journalists have sometimes con­tinued to entertain the idea of an upsurge in Amer­ican religion. It makes for a good news story if attendance at a local megachurch is shooting through the roof, for example. But this idea is pretty easily disproved simply by looking at trends in church attendance. Of course, those figures have been questioned on grounds that people may be over-reporting attendance, but never with evidence to suggest that more people are attending than say so to pollsters.

The best data, which come from the General Social Surveys conducted by the National Opinion Research Center at the University of Chicago, show that there has been no growth in church attendance over at least the past three decades, if not longer. These surveys, which ask not simply whether a per­son has attended in the past week but how often the person has attended in the past year, and which have response rates anywhere from three to five times higher than the usual Pew or CNN poll, show that between 30 and 35 percent of the adult U.S. public claim to attend religious services nearly every week, every week, or more than once a week, and these surveys show no indication of a religious revival anytime since the surveys began in 1972. Indeed, if one relies on Gallup data to look at trends before this, the high point in weekly church atten­dance occurred in 1958 during the Cold War and has never reached that level again.

Myth #2: There is no secularization.
This view has been the more sober-minded argu­ment by writers who doubt that there is a religious revival but who nevertheless see a lot of vitality in American religion. It has been an easy argument to make based on the same church attendance figures that disprove the myth about a religious revival. For example, in the General Social Surveys, 35 percent of the American public reported attending religious services "nearly every week" or more in 1973; in 1983, that figure had risen slightly to 37 percent; and in 1993, it was still 35 percent. Looking at some of these data, two prominent sociologists wrote in the American Sociological Review in 1987 that there was no evidence to suggest that secularization, meaning a decline in the strength of American churchgoing, was taking place.[7]

However, closer inspection of these and more recent data reveals that religious attendance has not remained stable. When the political scientist Robert D. Putnam began examining trends in religious par­ticipation in the late 1990s as part of his massive compendium of evidence about civic involvement, he came to the conclusion that there had been "a sharp rise in church attendance in the first several decades after World War II, followed by a decline in church attendance of roughly one-third between the late 1950s and the late 1990s."[8] Putnam also concluded that evidence about religious member­ship and attendance at Sunday school showed a similar decline.

Not everyone was convinced. Perhaps Putnam's argument had more to do with decline in the 1960s than later, or perhaps he was intent on finding evidence that fit his larger thesis about the collapse of community.

Subsequent research, though, shows that there has been decline. In the General Social Survey data, there has actually been a slow but definite long-term slide in churchgoing of about a quarter of a percentage point each year. Overall, between 5 and 6 percent fewer Americans participate regularly in religious services now than in the early 1970s. There are about 200 mil­lion adults in the United States. So that is a loss, con­servatively, of 10 million regular churchgoers. There are somewhere between 300,000 and 350,000 con­gregations in the United States. So that is a loss per congregation of about 30 members.

And yet, if one talks to clergy or reads stories about religion in the newspapers, the headline is all about growth, not decline. So what is going on? Well, for one thing, the total population of the Unit­ed States has grown by almost 50 percent since 1970, so even though a smaller proportion of the public is attending religious services regularly, the absolute numbers are larger. It's just that they would have been considerably larger if the rate of church­going had held steady.

What else is going on is that older people attend more often than younger people, and Americans are living longer and healthier, which means there are more old people to attend. In fact, among Ameri­cans in their 70s and 80s, a higher percentage are regular attendees than was true a generation ago.[9]

But if that is the case, it means that the decline among younger adults is more severe than we might have supposed. Consider the following: Among adults in their 30s who were surveyed between 1972 and 1976, 35 percent attended services regu­larly, but among adults in the same age group sur­veyed between 2002 and 2006, only 28 percent did so. Those younger adults in the 1970s were the baby boomers that we heard so much about. We worried then that boomers were going to church less. But now the next wave-the boomers' chil­dren-are attending even less, so that does not bode well for the future.
Do they eventually come back to the church as they mature? Yes, among baby boomers, about half of those who dropped out eventually came back. That is one thing if it means that someone drops out at 18 and comes back when they marry and start having children in their early 20s, but young adult­hood now lasts longer. Thirty is the new 20, as the saying goes, so people in their 30s are not attending as often as those in their 30s did a generation ago. Neither are people in their 40s, or even in their 50s.

Myth #3: Politics is driving people from the church.
This one has probably been articulated less often than the first two. It goes something like this: Amer­ican religion was hijacked in the 1980s by the Reli­gious Right. Jerry Falwell, Pat Robertson, and others politicized religion to the point that people who dis­agreed with them-political moderates and liber­als-eventually said a plague on all of it. They quit attending religious services as often-which offers an explanation for the downward trend we have just considered. They also quit identifying themselves with a religious tradition and instead described themselves as religiously nonaffiliated.[10]

The growth in identifying oneself as religious nonaffiliated is, in fact, one of the more dramatic changes in American religion in recent decades. In the early 1970s, between 5 and 7 percent of the public in General Social Surveys said they were nonaffiliated; by 2006, that figure had risen to 17 percent-more evidence, seemingly, that there is a secularizing trend.

No affiliation, like nonattendance, is especially prominent among younger adults. For instance, among those in their 20s surveyed between 2002 and 2006, nearly a quarter (23 percent) said they were nonaffiliated. And among those in their 30s, almost as many (19 percent) gave the same response.

However, those who resist interpreting any trend as evidence of secularization might say this is just a short-term blip that can be blamed on conten­tious politics. As people forget about Jerry Falwell and Pat Robertson, they will gradually feel more comfortable identifying themselves with religion once again.

The argument is similar to one made about baby boomers in the 1970s. Their apostasy, critics said, was sort of a wild-oats thing, brought on by the Vietnam war and student protests; once those were over, people would rediscover their faith. That argu­ment proved to be at least half wrong, and the cur­rent successor argument probably is too. The reason is that religion, then and now, is influenced more by demographics than by politics. It is especially influ­enced by marrying, settling down, having children, and raising them.

Elsewhere, with more statistical detail than I am able to present here, I showed that nearly all the decline in church attendance among young adults could be explained by the fact that they were mar­rying later. In fact, it exaggerates only a little to say that Americans in their 20s and early 30s divide into two groups of about equal size: those who are mar­ried, the majority of whom participate in religion, and those who are not married, the majority of whom do not participate.

Marriage, it turns out, has a stronger statistical effect on churchgoing than having no children or some children, having more children, or having school-age children. Those matter, but not as much as might be expected, and possibly because the growing number of people who have children out of wedlock tend not to attend church, while those who have children after marriage do. The important point is that younger adults' decisions about attend­ing church are influenced by family lifestyle- which of course may also be related to politics- but they are not staying home simply because they dislike the Religious Right.[11]

The evidence on religious identity also bears the imprint of delayed marriage. A generation ago, young married adults were far less likely to be reli­giously nonaffiliated than young single adults, but being married was the norm, so relatively few young adults said they were religiously nonaffiliat­ed. Currently, young single adults are even more likely to be religiously nonaffiliated, and there are far more young single adults than was true a gener­ation ago, so it has become more common to be nonaffiliated. Furthermore, those who are married have married later, so they have also had time to become nonaffiliated, and more of them remain that way even after they marry.

A pastor who has worked with young adults for several generations explained it in a way that is per­haps clearer than citing statistics. With only a slight overstatement, he put it this way: It used to be that boy and girls got confirmed at 13, and you didn't see them again until they were 18 and came back to get married. That was five years, and maybe they were still living at home most of the time. Nowadays, you don't see them again until they are 30 and come back to get married. That's 17 years. It's a long time.

And one other thing: It used to be that most young adults who were religiously nonaffiliated had grown up in a religious tradition. Nowadays, a much larger percentage have grown up without a religious affiliation. This means they are more likely to stay nonaffiliated. They are more likely to marry someone who is also nonaffiliated, and they are less likely to have parents putting pressure on them to become affiliated again.

It is true that the religiously nonaffiliated still have ties to religion. At least half, if not more, be­lieve in God. As many as a quarter attend religious services, often at churches that do not require mem­bership or formal affiliation. And about a quarter fall into that much-discussed category who say they are "spiritual but not religious," meaning that they have some interest in prayer, meditation, and spiri­tuality. But all of these aspects of religion are driven mostly by the demographics of growing up, getting married, and settling down-all of which are hap­pening later-more so than by disaffiliating from religion because of politics.

Myth #4: Membership in evangelical denominations is growing.
We move now from perceptions of American religion that are hardly credible at all to ones that are partly true but require some modification. The idea that evangelicalism, or conservative Protestantism as it is sometimes called, is flourishing is a very popular idea. It makes intuitive sense. News stories about church growth are always about conservative congregations.

This perception of growth fits well with argu­ments about the rising influence of evangelicals in politics; and in the academic literature, there have been arguments for a long time that conservative churches grow and can be expected to grow precise­ly because they offer strict teachings, require a lot from their members, and give people something steady to cling to in an uncertain world.

But whether evangelicalism is actually flourish­ing is harder to determine than might be imagined, for two reasons: vagueness about "evangelical" and vagueness about "flourishing." In some accounts, evangelical means anyone who tells a pollster that they are evangelical or have ever had a born-again experience or something like it-questions that pretty easily evoke assent from 40 to 50 percent of the population; and in some accounts, flourishing is inferred from the fact that James Dobson claims to have the ear of the President or that atten­dance at Willow Creek or Saddleback or Preston­wood is rising.

The historian of American religion, Mark A. Noll, who identifies as an evangelical and until recently taught at Wheaton College, has argued that evangel­icalism was far stronger on the eve of the Civil War than it has been anytime since. In the 1850s, he esti­mates, evangelical denominations, which at the time included Methodists, Baptists, and Presbyterians, accounted for 85 percent of all congregations. These churches had enough seating capacity to include nearly all of the nation's population, and even if churchgoing was not as common then as now, the preponderance of religious influence in local com­munities and in national politics was in the hands of evangelicals.[12] If we take the 1850s as a benchmark, then, we may not be as impressed with recent evan­gelical flourishing as pundits often are.

But, of course, pundits seldom have the long view in mind. If they do, it is a long view dating per­haps to the early 1980s when Jerry Falwell's broad­casts from Lynchburg began reaching a national audience and when the first signs that conservative religious values might be making a comeback were observed. If evangelicalism is flourishing, its growth should be evident since the early 1980s.

Evident how? Well, among other things, in belief. Evangelicals have always held to the Bible as their standard of belief and have distinguished themselves from their more liberal brethren by arguing that the Bible is divinely inspired, inerrant, effective for salvation, and to be taken literally as God's truth and moral law.
Since researchers have been conducting surveys, some question or another aimed at tapping this view of the Bible has been regarded as a measure of evangelicalism. Since 1984, the General Social Sur­vey has included a question asking respondents which statement "comes closest to describing your feelings about the Bible," one option of which is "the Bible is the actual word of God and is to be taken lit­erally, word for word."[13] In 1984, 38 percent of the public selected this response. By 2006, the number had fallen to 33 percent. No evidence of evangelical growth there.[14]

The other way in which researchers have tried to measure change in evangelicalism is by tracking membership in conservative Protestant denomina­tions. This approach is generally favored because conservative views of the Bible and other doctrines are indeed more common in these than in other denominations and because participation in these denominations is an indication of their collective strength in attracting members, exposing them to church teachings, and putting them in contexts where conservative moral values are reinforced.

What counts as a conservative Protestant de­nomination is not always clear, but researchers have gone to great lengths to sort out the distinctions between, say, Southern and American Baptists, Missouri Synod and ELCA Lutherans, the Presbyte­rian Church in America and the PCUSA, and so on, and thus to classify respondents accordingly. In two important ways, these data suggest that mem­bership in evangelical denominations has in fact been growing.

First, as a proportion of the entire U.S. public, evangelical Protestant affiliation grew from around 17 to 20 percent in the early 1970s to between 25 and 28 percent in the most recent surveys.

Second, because affiliation with the more liberal or moderate mainline Protestant denominations was declining during this period, the relative strength of conservative Protestantism was even more evident. For example, conservative Protestantism had been only about two-thirds as prominent as mainline Protestantism in the early 1970s but outstripped it by a margin of 2 to 1 in some of the most re­cent surveys.

Caution is required, though, in interpreting even the seeming straightforward growth of conservative Protestantism as measured by denominational affil­iation. One reason is that the growth appears to have slowed considerably among younger adults. For example, adults currently in their 20s are only 3 percentage points more likely than adults of the same age in the early 1970s to be affiliated with evangelical Protestantism, and among adults in their 30s, only 2 percentage points more likely.

A second reason is that this apparent slowing of growth is evident in statistics from large conser­vative denominations themselves. For instance, growth rates in the Southern Baptist Convention and Assemblies of God, both touted for exception­ally high growth in the 1970s, have slowed dramat­ically to the point of barely keeping up with natural increase in the overall population. This tapering off, incidentally, flies in the face of a popular argument based on the earlier growth that suggested conser­vative denominations would always grow because of their strict teachings and ability to extract resources from members.

A third qualification is that the best research on mainline decline shows it mostly to have resulted from demographics, especially low birth rates and greater intergenerational spacing.[15] That trend now seems to be catching up with conservative Protes­tants as well, especially as members become better educated and postpone marriage and childbearing.

Finally, recent research confirms that evangelical Protestants' growth had come more from natural increase than from drawing members away from mainline churches, although some of the latter was evident. In the recent research, though, most of the net growth among conservative Protestants that is not explained by natural increase appears to be from larger numbers of former Catholics becoming evangelical and Pentecostal Protestants, especially among Latinos.[16]

None of this is to suggest that conservative Protestantism is declining or is not a vibrant factor in American religious life. It does, however, chal­lenge observers to be more nuanced in their inter­pretations. Evangelicalism is not experiencing the huge growth suggested by figures from stories about megachurches. As a proportion of the popu­lation, it remains at only about 25 percent, which gives it legitimate reason to feel embattled by the larger culture despite its growth, and it remains denominationally divided and located in many small congregations about which little is heard as well as in the few large ones about which much is written.

Myth #5: The culture war is over-or never happened.
This is an issue that extends well beyond reli­gion, including broader questions about values, morality, politics, education, and the arts, and yet religion is at its core. In 1988, when I published The Restructuring of American Religion, I documented with both quantitative and historical evidence what I termed a growing fracture or division between self-identified religious conservatives and self-identified religious liberals.[17] I did not term the division a cul­ture war but wrote that there were developments in religion, politics, and the media that were reinforc­ing it in ways that had not been present during the 1950s and 1960s.

Three years later, at about the same time that Pat Buchanan delivered his widely viewed address about culture wars, the sociologist James Davison Hunter published the widely read book arguing that there was indeed a culture war between those with orthodox religious worldviews and those with pro­gressive views.[18] Other books and popular articles made a similar point, and ongoing debates within denominations about abortion, homosexuality, and related issues underscored the point that the nation was divided culturally.

However, the image of a culture war never sat comfortably with many observers who saw the debate being fomented more by special-interest groups than at the grassroots and who thought there was a large segment of the population in the middle who either did not care or held less polarized views. By the late 1990s, the pendulum among scholarly experts had swung decidedly in the other direction to the point that it became popular to argue that there was no culture war at all. If one appeared to exist, the critics said, it was either concerned with a very narrow range of issues-perhaps only one- and probably was being exploited cynically by political operatives.[19] In the past year or two, a new line of argument has become popular; namely, that there was a culture war in the 1980s and 1990s, but that was the divisive politics of the baby-boom gen­eration and is now being transcended by a new gen­eration of political and religious leaders.[20]

Forecasting is always risky, but there are reasons to question the view that culture wars are simply a relic of the past. For one thing, some of the data col­lected during the 1980s and 1990s did show increasing polarization. This was the case for opin­ions about abortion. It was also the case on self-report questions about religious identity in which more people identified themselves on the extreme right or extreme left than identified themselves in the middle.

A second reason to suggest that culture wars may continue is the evidence we just considered about religious affiliation: Evangelicals, who are most like­ly to espouse conservative moral views, are at least holding their own; the nonreligious, who are most likely to hold liberal moral views, are becoming a larger segment of the society; and mainline Protes­tants, who have held moderate or mixed views, are declining. Catholics are holding their own as a per­centage of the population, but traditional Catholics hold quite different views from liberal Catholics.

A third consideration is that evangelical Protes­tants in their 20s and 30s, who are relatively mod­erate or mixed in their views about homosexuality, have become more conservative in their views about abortion.

Finally, data on the two largest immigrant groups-Latinos and Asian Americans-suggest that they are religiously divided on moral issues in much the same way that white European Americans are.[21] Naturally, much about how salient these issues will be is likely to depend on who the candidates are in 2008 and what other issues are on the table, but it is quite reasonable to imagine that religious divi­sions about moral issues will still be important.

Conclusion
I have argued that some of the common notions about American religion are wrong. The idea of a religious revival seems not to be supported by any data. The notion that American religion is stable, holding its own, though, also seems not to be sup­ported. There are signs of serious erosion in such standard measures of religious vitality as church attendance and religious affiliation.

The reason for these declines seems not to be that people are leaving because of religious involvement in politics. The more likely reason is that younger Americans are marrying later, having fewer chil­dren, and having them later-all of which means that far more younger Americans are single and childless than was true a generation ago and that these same younger Americans are not settling into religious congregations at the same rate as their par­ents did in the 1970s. The growth of evangelical Protestantism is a reality but is not as strong as many observers have suggested and may be weakening, again as a result of demographic changes. The view that the culture wars never happened or have ended seems questionable.

Time has not permitted an extensive examina­tion of the arguments or evidence. As always, there are different ways of defining terms and assessing data. But there is no reason why we should be con­tent with simplistic arguments about American reli­gion. It is, after all, a complex phenomenon, and even describing the broad contours requires being careful about notions that seem plausible but are not supported by the available research.

Robert Wuthnow, Ph.D., is Professor of Sociology and Director of the Center for the Study of Religion at Princeton University. These remarks were delivered as part of a conference on "Religious Practice and Civic Life: What the Research Says," sponsored by The Heri­tage Foundation, with research partners Child Trends and the Baylor Institute for Studies of Religion and fund­ing provided by the John Templeton Foundation, held in Arlington, Virginia, on October 4, 2007.