Thirty-odd years ago, Dean M. Kelley of the National Council of Churches wrote a book he intended to call Why Strict Churches Are Strong but that his publisher insisted on titling Why Conservative Churches Are Growing.1 It was a relatively short work, with a relatively simple thesis, but it was clearly the right book on a long-overlooked topic, and it arrived at the perfect moment. Sociology of religion in North America was in the early stages of being born again after a prolonged gestation, and Kelley’s book followed important works by Peter Berger, Robert Bellah, Andrew Greeley, Charles Glock, Rod Stark, and others.
Kelley argued that “the conservative churches, holding to seemingly outmoded theology and making strict demands on their members, have equaled or surpassed in growth the yearly percentage increases of the nation’s population. … It is the sectarian groups that have had most success in attracting new members.” Many people took issue with Kelley, then and later and for numerous reasons, but with that elegant thesis he framed the terms of the debate for three decades.
Recently, however, Michael Hout, Andrew Greeley, and Melissa Wilde have exclaimed “WRONG!” so loudly as to compel a retrospective review of the Kelley thesis and its rivals.2 Hout, Greeley, and Wilde (HGW) don’t merely disagree with Kelley. They want to get his thesis off the table once and for all. Instead of growth resulting from sectarian theology and strictness, they argue for a “demographic imperative” or a “differential natural increase.” Of those who have insisted for 30 years that strictness and switching explain growth, HGW say simply, “They were wrong. … The explanation for the changing shape of U.S. Protestantism is … demographic, not ideological.”
Are the revisionists right? And what is finally at stake in the debate? To answer those questions, we have to make our way through a thicket of sociological controversy.
Dean Kelley, Strictness, and the Subsequent Discussion
The first two chapters of Why Conservative Churches Are Growing sought to answer two questions—Are the churches dying? and Is religion obsolete?—that Dean Kelley used to confront two existing interpretations of religion in America. In American Piety, Rodney Stark and Charles Glock had concluded by asserting, “If there has been an erosion of traditional faith [as they believed there had] we would expect people to be shifting from denominations which have retained unswerving commitment to that faith into denominations with more demythologized positions.”3 A year earlier, Peter Berger had painted an even bleaker picture in The Sacred Canopy, depicting the forces of secularization advancing over the church. Two responses were possible, Berger suggested—accommodation or resistance—with each carrying grave, perhaps fatal, consequences.4
Kelley was familiar with the trends and processes observed by Stark, Glock, Berger, and others. But his conclusion differed radically: the “churches that have not tried to adjust to the times—to ingratiate themselves with the world—in many cases are not declining. In them we see no indication that religion is obsolete, churches outdated, or modernization helpful.” Kelley provided mainly denominationally based statistics as evidence; the big gainers were the Jehovah’s Witnesses, the Seventh-Day Adventists, the Salvation Army, the Nazarenes, the Mormons, and other conservative groups—the “wrong” churches, he suggested tongue-in-cheek. “In or out,” he concluded: “upon this distinction the survival of any serious group depends. … Groups which preserve their seriousness through strictness will not only mediate effective meaning to their members and others, but as a consequence will thrive and grow.”
In a 1978 follow-up article, Kelley restated his two main points from the book. Few people had disagreed with his first point, that “the basic business of religion is to explain the ultimate meaning of life.” But his second point had proven to be the “lightning rod”: that “the quality which makes one system of meaning more convincing is not its content but its seriousness/costliness/strictness.” Nevertheless, Kelley reaffirmed his thesis: “I have been asked if I would make any changes. Essentially, I would not. I am unrepentant and unreconstructed. The curves are continuing much as they were.”5
Serendipitously, Kelley had unwittingly stimulated a new variant of the existing literature on “religious switching” that attempted to measure and interpret the movements of individuals and groups among Protestant denominations. For years, researchers had examined the correlation between Americans’ upward mobility in social class and their movement from conservative sects toward mainline denominations. Individuals moved into higher-class denominations over time, alongside lower-class, sect-like groups moving upward collectively into a more respectable status. In apparently related ways, both trends expanded groups such as the Episcopalians, Presbyterians, and Methodists.
Then as depicted by Stark and Glock, switching to mainline denominations more recently was part of a larger accommodation to the structural forces of a “demythologized modernism.” Or in Berger’s terms, the “plausibility structures” provided by mainline churches were more compatible with the forces of secularization than were the theologies and services provided by conservative Protestant groups. Conversely, if any switching occurred that favored the conservatives, that was simply assumed to be anomalous and also overwhelmed by the switching trends favoring the mainline. For all practical purposes when religious switching occurred, it was really in only one direction.
What Kelley’s editors unexpectedly accomplished by changing the title of the book6 was to provoke an argument among researchers other than, or perhaps in addition to, the two main points that Kelley intended on meaning and strictness. Two new directions of research resulted. One was an empirical variant of the existing literature that examined the likelihood of switching toward conservative churches; the second was less empirical and more interpretive in proposing numerous reasons why people actually switch from one group to another.
In hindsight, we can detect several themes from the studies of religious switching in the following decade. The first was a nearly-total rejection of Stark and Glock’s contention that the liberals were winning . They were not. Second, Canadians Reg Bibby and Merlin Brinkerhoff qualified much of the conservative growth and labeled it as a likely “circulation of the saints.”7 Most conversions were from within groups. Third, Clark Roof and Kirk Hadaway quantified Kelley, setting the total proportion of “switchers” at 35 percent of Protestants.8 The big winners were the “conservative sectarians,” with some liberal groups also gaining. The greatest losers were several liberal and moderate groups. Fourth, Hadaway added a qualitative evaluation of switching, that “converts make better members.”9 Baptists and sectarians “get more out of their members and converts with respect to actual participation than do the more liberal denominations.” And fifth, Roof reiterated that about one-third of actively religious persons switched their denominational identities, adding that about one-third of that group switched more than once.10
Perhaps the best summary statement of where things stood after the first series of responses to Kelley came from Dean Hoge and David Roozen. They conceptualized a four-factor model of church growth—national, local, contextual, and institutional. In relation to denominations, local contextual factors such as the relative affluence and homogeneity of race worked to the advantage of mainline Protestants. By contrast, national contextual trends were working to the advantage of conservative groups; “a broad cultural shift … has hit the affluent, educated, individualistic, culture-affirming denominations hardest.”11 Hoge and Roozen’s summary challenged the Kelley thesis by emphasizing the contextual factors surrounding conservative church growth that Kelley had largely ignored when he argued for a more institutionally based strictness and demand.
Meanwhile, how was the conservative growth being explained theoretically? When Kelley wrote, two alternatives existed. One was Berger’s version of secularization theory that permitted two survival strategies for religious groups—accommodation and resistance. Most conservative groups were in the resistance camp, and Berger did not hold out much promise for their future. A second theoretical interpretation had come primarily from political scientists. They typically assumed that a loss of power was motivating conservatives, with “status discontent” theories explaining the resulting perceptions of and responses to their plight.
What Kelley implied theoretically was a third explanation of conservative growth that unfortunately he did not articulate as fully as he might have. Instead, he devised a cryptic formula: “meaning = concept + demand.” Kelley explained this to mean “the quality that enables religious meanings to take hold is … the demand they make upon their adherents and the degree to which that demand is met by commitment.”
Not only would Rod Stark and colleagues develop this later into a full-blown “rational-choice” theory of religious growth based on strictness, but Stark was also instrumental in influencing a fourth possible explanation of conservatives’ growth—their superior competitive marketing strategies. Thus Stark emphasized both the religious “demands” that were more compatible with rational choice theories and the religious “supply” that fit his competitive marketing explanations.
Berger had already asserted in The Sacred Canopy that unlike monopolistic religious settings (i.e., Europe), religion in America resembles a competitive open-market. Given pluralism and the absence of a state church, “religious groups must organize themselves in such a way as to woo a population of consumers, in competition with other groups. … Religious institutions become marketing agencies and the religious traditions become consumer commodities.” Conservatives were likely better marketing strategists and more in touch with their religious consumers than were the traditional denominations.
One final piece of this present puzzle was identifying when research on birth rates and fertility encountered the discussion on church growth. Demographers had studied the social factors affecting fertility, assuming that the religious variable was simply one more causal influence on variations in birth rates. The possibility of an opposite relationship—fertility rates “causing” religious trends—or that a single cause such as education or social class might influence both religious participation and fertility rates apparently either never occurred to most demographers, or if it did, it was not relevant for them.
In 1962, Dennison Nash and Peter Berger were two of the first sociologists to “suggest that the upsurge in church membership is due at least partly to an increase in the number of children who enter the church and parents who follow,” as Nash verified in a followup report six years later.12 When Bibby and Brinkerhoff added their “circulation of the saints” qualification of the Kelley thesis, they invoked an existing model of church growth based on three means: birth, reaffiliation, and conversion. Bibby persisted in a 1978 study, asserting that “religious groups … appear to add members through (1) birth, (2) geographical mobility, and (3) proselytism.”13 And Hoge spelled out how the differential birth rates might contribute to the patterns in church growth. He identified three “possible linkages”: higher birthrates producing more babies and eventually higher church membership (a variation on Nash and Berger); families with children participating more greatly for child-rearing support; and an “association between traditional religious commitments and larger desired families.”14
So after a decade following Kelley, what role were birthrate variations playing within the overall explanation of why churches grew? Relating the parallel contexts of huge increases in both birth rates and church attendance between the late 1940s and the early 1960s provided an early first step. But only after Kelley’s book would sociologists such as Bibby and Brinkerhoff and then Hoge begin to put the pieces together. Despite Roof’s quantifying the proportion of religious switchers at one-third of all adherents, apparently no one thought to question whether that relatively small number of people alone could account for the proportionate gains being made by conservative groups. In addition, Bibby’s “circulation” motif qualified the Kelley thesis nicely, but it likely distracted him from pursuing the birthrate variations he also had identified. Bibby had the correct pieces, but did not connect them as effectively as he might have.
Two further contributions came during the 1980s. One was from Jack Marcum, who called for “a longitudinal approach [that] would be useful to untangle the relations among denominational liberalism-conservativism, religious participation, and fertility.”15 Seven years later, Marcum added this helpful interpretation: “religious participation after childbearing is not solely a function of child-rearing demands, but also reflects the influence of religious participation prior to those births. A complete model would include indicators of religious participation before childbearing began, as well as indicators of other a priori factors.”16
Roof and McKinney summarized what sociologists thought were the key components in their influential American Mainline Religion. Two sets of factors determined any religious group’s membership over time: “its natural growth and its net gains or losses from conversion.” The “most significant variable” composing natural growth was birthrate; “factors of theological heritage as well as group experience combine to create differential fertilities, which once in place tend to be perpetuated.” In sum, “demography has proven to be destiny for American Protestantism.”17
Suffice to say as an evaluative response, the Kelley thesis had been enhanced and fine-tuned significantly in the 15 years following its appearance. Kelley was probably correct empirically, as he reasserted in 1978, but in limited ways and for reasons beyond those he articulated originally. Bibby and Brinkerhoff; Roof, Hadaway, and McKinney; and Hoge and Roozen—they were the main figures in the discussions that followed. Then theoretically, four alternatives could explain the Kelley thesis on growth—rational choice-based strictness, resistant enclave persistence, superior competitive marketing, or status discontent—with no single interpretation decisively superior. And so the matter largely rested until Hout, Greeley, and Wilde fired their shot across the bow in 2001.
A Demographic Imperative Explaining Church Growth: What Its Proponents Mean and How Others Respond
When Hout, Greeley, and Wilde responded to the Kelley thesis, they entered the conversation with little apparent investment in the previous discussion. All three are eminent sociologists who more often have studied American Catholics. In retrospect, Greeley’s contribution may have been anticipated in his 1989 book, . Ironically, then he believed that mainliners were converting to fundamentalism in part because the “religious conservatism of the fundamentalist and evangelical denominations makes them appealing to those who still want ‘religion’ from their religion. The available data provide some support for such an explanation.”18
Even so, HGW’s total rejection of switching to conservative groups as having any effect at all on their relative growth was surprising indeed. Could the previous 30 years of research interpreting a phenomenon that now approximates a “taken-for-granted” in American sociology of religion have been so thoroughly misguided?
To set up their interpretation, HGW dichotomized five possible means by which religious groups can grow. On the one hand, growth can come from a “natural increase” which combines birth rates with the timing of childbearing. On the other hand, more sociological forms of change take four possible forms, all implied above (although with various labels). HGW identified these as “conversion” to conservative groups, “switching” to mainline groups, “apostasy” by leaving one’s current (conservative) group; and “inflow” from outside any existing Protestant attachments. Among these four explanations, HGW especially examined “conversion” to conservative groups, the most-cited explanation of conservative growth over the previous 30 years. In sum, they sought to demonstrate how fundamentally mistaken the debate since Kelley has been.
Their plan of analysis was “to develop a demographic simulation model that combines observations on religious origins, current religion, and fertility (both the amount and timing of it) to predict the proportion of Protestants in mainline and conservative denominations for each cohort born between 1900 and 1973.” That became their “first scenario,” followed by four alternatives scenarios “in order to isolate the contributions of fertility and denominational switching” to explain religious growth trends over time. To do so, they combined data from two sources: (1) the General Social Survey (GSS) from 1974-98 for data on peoples’ current religion, religious origins, fertility, and mortality; and (2) birth registry data compiled in 1976. None of these data sets was available to Kelley in 1972, and HGW’s use of current GSS data facilitated a significant updating of patterns observed earlier by other scholars.
Their findings follow, by scenario. From scenario one, all the relevant factors of demographic natural increase, conversion, and apostasy were combined to predict nearly exactly what HGW had hypothesized. In other words, these three variables were all present, and together they accounted for the changes among religious groups born between 1903-73. In scenario two—isolating the natural increase variable—equal conversions among groups were assumed and so cancelled each other out. What remained was that 76 percent of the religious change could be attributed to the single source of differential fertility rates. This was obviously the major finding for HGW.
In scenario three, switching, apostasy, and conversion were combined. What HGW discovered was that apostasy from the mainline to “no religion” has increased, while the rate of conservative switching to the mainline has decreased. By contrast, “switching from mainline to conservative denominations has not changed significantly over 70 cohorts,” and so explains only 4 percent of the total change, having “no effect on the proportion … in a mainline denomination,” another key finding.
In scenario four—conservative to mainline switching—HGW tracked an overall decrease over time, totally 36 percent of the observed change. Two things are important here. First, the age-old practice of upwardly mobile switching has declined greatly. Second, the net effect is that much of the oft-lamented mainline decline, then, is the apparent result of those groups’ no longer attracting higher status-seeking (ex-) conservatives. The further implication is that the mainline has declined not primarily because of losing its adherents (especially to conservative churches), but because of attracting far fewer switchers from other (often conservative) groups. When HGW further contrasted the effects of the third and fourth scenarios, they discovered that while the rate of switching from the mainline to the conservative groups has been steady (about 13 percent), the historic upwardly mobile pattern of switching from conservative to mainline has declined from 15 to 10 percent recently.
Finally for scenario five, the three forms of switching combined to explain 43 percent of the mainline decline. But the mainliners who do convert are more likely to convert to Roman Catholicism than to some form of conservative Protestantism. The mainline should be concerned, then, not about losing adherents to conservative groups but about no longer gaining converts from the conservatives plus now losing them to Catholicism.
HGW concluded with a relatively optimistic assessment of the long-term prospects of the mainline, while suggesting that conservatives will have to learn to live with reduced expectations. As they tell the story, something approaching equilibrium is now in place, and the net gains or losses from religious change are about over. Conservatives have lost their earlier demographic advantage, now that their birthrates approximate those of the mainline. Switching in either direction has about stabilized: “exhaust both sources of change and change will stop unless and until a third source comes along.”
During the summer of 2002, I chatted with a number of sociologists—especially those keeping current in sociology of religion—in an effort to assess the value and implications of the HGW article. (Unfortunately, Dean Kelley died five years earlier.)
First, the sociologists I talked with were singularly complimentary of HGW for their methodological sophistication in addressing why some religious groups have grown. Dean Hoge of Catholic University, for example, found their analysis “goes far beyond anything I have seen before, certainly far beyond ours. It’s quite a thorough job.” Joe Tamney from Ball State evaluated the HGW article as something to deal with, something that “will become a basic piece in the field,” although he was prone to “accept the methodological stuff on faith.”
Two methodological cautions were sounded, however, by Alvaro Nieves from Wheaton College. One problem was with “time series data” in which possible “multi-colinearity” exists; that is, Nieves asserted that variables known to be associated with fertility and birth rates such as education and social class may still be exerting an influence that HGW have not controlled for apart from the fertility rates. As a result, whether birth rates by themselves were explaining 76 percent of the change is less certain than one might conclude. Similarly, Brad Wilcox of the University of Virginia suspected that “it is hard to disentangle fertility patterns as a consequence of social class and region.” Nieves was also concerned with the two different time sets used by HGW. The GSS data on religious affiliation began in 1974, whereas the fertility data began much earlier. With these two different time lines, comparing them may be “comparing apples with oranges.” Despite these reservations, Nieves was impressed with the methodology.
Second, these sociologists were uniformly appreciative of HGW’s revisiting a thesis that had become a commonplace. For Tamney, one strength of the article was that it demonstrated that the gains made by the conservatives are “clearly smaller than was portrayed by the popular press,” although many sociologists have “long believed the two reasons they have pointed out.” In Hoge’s words, they “have done something nobody has done before.” For others, the “degree” of what HGW claimed to have discovered as new was unclear. Christian Smith of the University of North Carolina was “pretty friendly to the article,” despite HGW’s criticism of his recent “subcultural identity” theory that explains apparent evangelical growth. What concerned Smith was that “they overplay the novelty of their theory, since scholars have been saying for a few decades now that differential birthrates explain a lot.” Similarly, Robin Perrin from Pepperdine University (whose 1997 study of evangelical growth HGW cite approvingly) was concerned that the role of birth rates “seems to be presented as a ‘novel idea,’ which obviously is not the case. Could they have proceeded more cautiously in presenting their argument in a more ‘relative’ way?”
Third, although these sociologists conceded the importance of fertility and birth rates, they had reservations about the narrow scope of the “demographic imperative” as understood by HGW. For Nieves, whose expertise lies in ethnicity and immigration studies, the problem was partly that GSS data are based only on interviews with English-speaking adults; even more important, the impact of immigration is completely ignored. Nieves’ points were seconded by Steve Warner from the University of Illinois at Chicago, who has specialized in immigrant religions during the past decade. Warner observed that of the million or more immigrants arriving in the United States each year, most are Christian and many of them conservative Christians. According to Warner, “We cannot overstate the significant effect these first- and second-generation immigrants are having on conservative churches.”
Nieves and Warner are on to something here. Some new immigrants of the past 30 years—particularly Spanish-speaking and Asian groups—have probably contributed a “demographic boost” to the growth of some conservative denominations. Moreover, because the GSS counts only English-speakers over 18 years old, some conservative religious immigrants have been omitted systematically from HGW’s view. Both Smith and Tamney added that the emphasis demographically should also be on “retention” alongside fertility. For Smith, “Some of [the growth] is explained by better conservative Protestant retention.” (On a related theme, Steve Warner was also concerned with how easily HGW dismissed the advantages conservatives enjoy in the “retention” of their adherents, especially youth. Socialization and retention are vital factors for any group. Therefore, the decline in upwardly mobile switching among conservatives interests Warner, but not for the narrowly “demographic” reasons that HGW allowed. He suggests that the fact that switching among conservatives is no longer motivated by the desire for higher social status deserves more attention than HGW gave it.) So while HGW have focused their demographic argument narrowly on fertility, they have neglected other key aspects of a demographic influence.
Fourth, the disproportionate emphasis HGW made on birth rates as “cause” also suggests a need to return to Jack Marcum’s concerns that religion be examined as both a “cause” as well as an “effect” of differences in birth rates. When I chatted with Marcum, he was pleased that HGW had made an important point, but was concerned that “prior motivations that different groups have for having children” still have not been addressed. For Marcum, conservative Protestants do “share a common set of beliefs that result in having larger families,” although he doubted they “sit around talking about birth rates.” This is an important gap in the HGW model; decisions about how many children to have are not made in a moral or religious vacuum. What Marcum was writing about in the 1980s is still true; HGW appear not to have been adequately concerned.
Warner echoed Marcum’s concern when emphasizing “religious pro-natalism.” For Warner, having children is “not merely a demographic factor.” Whether to have children and how many to have are decisions that couples make in contexts that include religiously based concerns. Conversely, Warner told of hearing a mainline pastor semi-facetiously tell his flock, “We’re gonna die unless you have more kids.”
A fifth concern occurred to me as I reflected on my own experiences in largely conservative churches the past 40 years. HGW based their study on Tom Smith’s denominational classification.19 That is a valid approach, but Smith also has some notable biases, including those found in his article debunking “conservative church growth.”20 So it is at least possible that Smith’s classificatory model helped HGW “stack the deck” in favor of their assumption that switching matters little.
But a more salient concern I also had was that “being a conservative Protestant” did not mean the same thing Christianly and culturally in 2002 as it did in 1972 or surely in 1952. HGW assumed (perhaps as non-Protestants; who knows?) that the categories distinguishing Protestants such as “mainline” and “conservative” have essentially the same meaning that they did 30 or 40 years ago. They also tended to conflate terms such as “conservative” and “evangelical” without any definitional rationale.
Sixth, Brad Wilcox offered an analysis of the declining birth rates that intrigued me. Wilcox, who has tracked the relationship between religion and family life, argued that trends in evangelical family life demonstrate an “accommodation” to the fertility practices in America over the past 30 years, not incidentally the same time frame under review here. In contrast to higher fertility rates earlier, conservative Protestants now “mirror” the national rate. Furthermore, Wilcox suggested, this may also be due to a growing emphasis in the family advice literature that stresses a therapeutic, even sentimental, view of the family and children. “Stressing the emotional quality of the parent-child relationship tends to limit the number of children, while focusing on the intensity of relationships within the family.”21
Finally, Tamney has recently published a book entitled The Resilience of Conservative Religion that bears on this discussion.22 Tamney pays his dues early in the book to differences in birth rates and effectiveness of retention. He used Greeley’s 1987 data to make the point that 1 of 5 mainliners switch to conservative groups, while the conservative-to-mainline rate of switching is 1 of 7, asking “Why the difference? The purpose [here] … is to further our understanding of what people like about conservative Protestant congregations,” assuming that people join congregations, not denominations.
Tamney’s theoretical explanation differs greatly from the HGW thesis, and it also takes issue with Kelley’s emphasis on strictness. Perhaps as a variation of my own earlier observations about how the “meaning” of being a religious conservative has shifted, Tamney theorizes that today’s conservative congregations have learned how much they can modernize, accommodate, and adjust to the emerging cultural realities.
When I chatted recently with Tamney, he stated his “main criticism” of the HGW article. As had Warner, Tamney emphasized the matter of retention that he “located” in terms of his “growth-by-modernization” theory. “The key is when they [HGW] talk about how the conservative churches have been better at retaining their members, there are lots of reasons. But especially the roles of having become more educated and doing better socialization are important. It’s also important that although the labels are the same, the substance of what the groups—especially the conservatives—believe and are doing now is not the same as in the 1950s and ’60s.” In short, Tamney allowed for birth-rate differences but emphasized retention; retention advantages, in turn, he explained as adaptation to changing cultural realities.
So What Have We Learned?
As I was putting the finishing touches on this essay, the Atlanta-based Glenmary Research Center—a reputable group—released the results of its “Religious Congregations & Membership: 2000” study that included lots of numbers summarizing trends from the 1990s. Included in one journalistic version of the findings was this statement from the Center’s director, Kenneth Sanchagrin, “The churches that are demanding in some way—that expect you to come two or three times a week, or not wear lipstick, or dress in a certain way—but at the same time offer you great rewards … those are the churches that are growing.” I can only imagine that Professors Hout, Greeley, and Wilde rolled their eyes and lamented in unison, “Here we go again.”
One initial interpretive response is that the “myth” of conservatives’ growing be-cause of strictness (and subsequent switching) will not die easily. Dean Kelley and his apologists of the past three decades have done their job and done it effectively. By contrast, the sum of what the eight sociologists told me suggests that HGW have made a powerful counterargument but have not made the case for a paradigm shift.
Here are some of the issues that scholars will have to wrestle with as they test HGW’s revisionist thesis:
(1) The HGW methodology is sound and superior to much of what preceded it, but the numbers in themselves do not tell the whole story. The numbers alone do not tell us about causality in a clearly definitive way. Even if they did, they do not allow for religious influences upon the demographic choices that eventuate in some churches apparently growing more than others. Poor Jack Marcum, along with others, has been trying to call attention to the fact that religious values and change provide a context for the changing birth rates that in turn influence religious growth patterns.
(2) When HGW cast their argument as a “demographic imperative,” birth rates in and of themselves are not the only component of that imperative. Demographically, other things are occurring as well. Ironically, given their impressive time lines that go back to 1903 in the case of births, HGW as sociologists appear less sensitive than they might be to aspects of the current demographic scene. Part of this results from their use of GSS data that neglects recent immigrant groups; part of it results from overlooking the disproportionately conservative Christian presence among many of those immigrants.
(3) Why conservative groups have lower birth rates that now more closely resemble those of the nation-at-large is an intriguing matter. Wilcox provided one interpretation of that shift, with the demographic advantage that conservatives once had now declining. When HGW emphasized the long-term consequences of those shifts, however, without considering the parallel changes that have occurred culturally and theologically among conservative Christians, in their churches, and within their families, at best they described only part of a larger picture.
(4) Even though the “switchers” are likely less numerous than we had been led to believe, they are still not an inconsequential number. HGW have nicely updated the patterns of conversion and switching. But they have not told us anything about the underlying motivations and shifting preferences among either the switchers or the non-switchers. Nor have they examined as carefully as they might the labels—conservative, liberal, mainline—that are utilized. Just as the motivations for switching have changed, so the composition of the groups themselves has shifted over the past 30 years.
(5) Finally, the jury is still out on the validity of the “strictness” argument per se. In his excellent American Evangelicalism, Christian Smith accepted some of Kelley’s (and Larry Iannoccone’s supporting) arguments, while making two significant qualifications: strictness is more likely associated with conservative growth than causing it; and a strictness more characteristic of fundamentalism does not explain the increasing vitality of evangelicalism.23 Indirectly, Smith also supported my sense that HGW have incorrectly equated the “conservative” churches of Kelley’s era with those of the early 21st century. As best, that is a debatable assumption worth further study.
So was Dean Kelley correct, or have Michael Hout, Andrew Greeley, and Melissa Wilde debunked him and his strictness allies as they had claimed? Clearly, they have raised a powerful “reasonable doubt” about the validity of explaining conservative religious growth merely in terms of strictness and switching. Ironically, simultaneously they have provoked additional concerns with their seemingly deterministic case for a demographic imperative. Let the debate continue, and stay tuned!
James A. Mathisen is professor of sociology at Wheaton College. With Tony Ladd, he is the author of Muscular Christianity: Evangelical Protestants and the Development of American Sport (Baker).
1. Dean M. Kelley, Why Conservative Churches Are Growing (Harper & Row, 1972).
2. Michael A. Hout, Andrew Greeley, and Melissa J. Wilde, “The Demographic Imperative in Religious Change in the United States,” American Journal of Sociology, Vol. 107 (Sept. 2001), pp. 468-500.
3. Rodney Stark and Charles Y. Glock, American Piety: The Nature of Religious Commitment (Univ. of California Press, 1968), p. 206.
4. Peter L. Berger, The Sacred Canopy (Doubleday, 1967).
5. Dean M. Kelley, “Why Conservative Churches Are Still Growing,” Journal for the Scientific Study of Religion, Vol. 17 (1978), pp. 165-72.
6. Kelley, Why Conservative Churches Are Growing, p. 167.
7. Reginald W. Bibby and Merlin Brinkerhoff, “The Circulation of the Saints: A Study of People Who Join Conservative Churches,” Journal for the Scientific Study of Religion, Vol. 12 (1973), pp. 273-83.
8. Wade Clark Roof and C. Kirk Hadaway, “Shifts in Religious Preference—the Mid-Seventies,” Journal for the Scientific Study of Religion, Vol. 16 (1977), pp. 409-412.
9. C. Kirk Hadaway, “Denominational Switching and Membership Growth: In Search of a Relationship,” Sociological Analysis, Vol. 39 (1978), pp. 321-37.
10. Wade Clark Roof, “Multiple Religious Switching: A Research Note,” Journal for the Scientific Study of Religion, Vol. 28 (1989), pp. 530-35.
11. Dean R. Hoge and David A Roozen, eds., On Understanding Modern Church Growth and Decline: 1950-1978 (Pilgrim Press, 1979), pp. 315-33.
12. Dennison Nash and Peter Berger, “The Child, the Family, and the ‘Religious Revival’ in Suburbia,” Journal for the Scientific Study of Religion, Vol. 2 (1962), pp. 85-93; Dennison Nash, “A Little Child Shall Lead Them,” Journal for the Scientific Study of Religion, Vol. 7 (1968), pp. 238-40.
13. Reginald W. Bibby, “Why Conservative Churches Really Are Growing,” Journal for the Scientific Study of Religion, Vol. 17 (1978), pp. 129-137.
14. Hoge and Roozen, eds., On Understanding Modern Church Growth and Decline, p. 118.
15. John P. Marcum, “Explaining Fertility Differences Among U.S. Protestants,” Social Forces, Vol. 60 (1981), pp. 532-43.
16. John P. Marcum, “Religious Affiliation, Participation, and Fertility: A Cautionary Note,” Journal for the Scientific Study of Religion, Vol. 27 (1988), pp. 621-29.
17. Wade Clark Roof and William McKinney, American Mainline Religion: Its Changing Shape and Future (Rutgers Univ. Press, 1987), pp. 157, 182.
18. Andrew Greeley, Religious Change in America (Harvard Univ. Press, 1989), p. 36.
19. Thomas W. Smith, “Classifying Protestant Denominations,” Review of Religious Research, Vol. 33 (1990), pp. 225-45.
20. Thomas W. Smith, “Are Conservative Churches Really Growing?”, Review of Religious Research, Vol. 33 (1992), pp. 305-29.
21. This argument is developed more fully in W. Bradford Wilcox, Soft Patriarchs, New Men: How Christianity Shapes Fathers and Husbands (Univ. of Chicago Press, 2004).
22. Joseph B. Tamney, The Resilience of Conservative Religion: The Case of Popular, Conservative Protestant Congregations (Cambridge Univ. Press, 2002).
23. Christian Smith, American Evangelicalism: Embattled and Thriving (Univ. of Chicago Press, 1998).
Copyright © 2004 by the author or Christianity Today/Books & Culture magazine.Click here for reprint information on Books & Culture.
