Putting Religion on the Map

In 1961 Wilbur Zelinsky published a lengthy article in the Annals of the Association of American Geographers on the religious geography of the United States. The author came to the subject with no background in religious studies but with a wealth of geographical expertise and a rigorous loyalty to inductive method. During his training in geography at the University of California at Berkeley during World War II, Zelinsky labored as a map draftsman. In 1946, after earning a master’s degree from the University of Wisconsin, he analyzed terrain for the U.S. Army Corps of Engineers in occupied Germany. By the time he trained his sights on religion, he had become a keen observer of culture. Yet he crept toward his object with caution, prefacing his argument with abundant qualifications. American religious life is diverse, he noted, and its most consistent feature is the “constant and rapid change, the restless shifting of forces, the transmutation and re-shaping of regions, and thus the provisional configuration of current patterns.” He acknowledged that the available data were “hopelessly inadequate … even for satisfactory basic description.” Nonetheless, he also recognized that “the temptation to speculate on the interrelationships between religious affiliation and political, social, and economic behavior is admittedly almost irresistible.”[1]

Zelinsky proposed a classificatory scheme of seven distinct “religious regions” and five “subregions,” based on statistics drawn from the census conducted by the National Council of Churches in 1952. His conceptual framework occupied only the final few pages of his article. But it was that hesitantly created rubric, with subsequent glosses and elaborations, that has profoundly influenced efforts to study the interconnection of religion and region in the United States to this day. Utilizing a combination of geographical and denominational indicators, Zelinsky divided the nation into New England, an area dominated by Roman Catholics but inflected by a history of Congregationalism; the Midland, a lateral arc of Methodist strength stretching from the mid-Atlantic states to the Rockies; the Upper Middle Western Region, populated by Lutherans and Catholics but also exhibiting the influence of the westward trek of New England Congregationalism; the Southern Region, a sea of Baptists and slightly fewer Methodists “interrupted by islands of Catholics”; the Mormon Region; the Spanish Catholic Region of the Southwest; and the Western Region, characterized, according to the author, by its lack of recognizable religious personality.[2]

Nearly a half century later, an intrepid set of researchers, led by Mark Silk and Andrew Walsh, have succumbed to similar temptation. The Religion by Region Series consists of eight slim volumes, each coedited by Silk and another prominent scholar of American religion. Silk, director of the Leonard E. Greenberg Center for the Study of Religion in Public Life at Trinity College in Hartford, Connecticut, is committed to the public understanding of religion, a goal that lends a distinctive focus to this project. Silk, Walsh, and their regional teams (each volume includes essays by at least a half dozen scholars) set out to discover how religion works in each of their designated sites, that is, to delineate the connections and interactions between religious variations and the “public cultures” of different parts of the country.[3] The series is intended for a general audience, with a particular eye to educating journalists about the diversity and complexity of American religious life.

The volumes employ numerous resources that Zelinsky lacked: membership statistics from religious bodies collected in the North American Religion Atlas (NARA); the American Religious Identification Survey, which tracked self-identification with religious traditions; and a series of National Surveys of Religion and Politics from 1992, 1996, and 2000.[4] Each volume analyzes data from the designated region in an opening chapter and follows with essays that probe the history and social status of particular traditions, in an effort to explore how religion is expressed and employed in—and how it shapes—political life.

The datasets allow the authors to explore in detail a geographic distinctiveness that Zelinsky could only imagine. At their best, these volumes supply a treasure trove of information and moments of real discovery, not unlike those furnished by the far more massive and expensive New Historical Atlas of Religion in America published in 2001.[5] But rather than presenting data in graphic and relatively undigested form, the Religion by Region volumes lead the reader through the wealth of detail, summarizing critical points and providing reasoned speculation about trends and patterns. This is a wonderful resource for anyone with an eye to the vast geographical and religious diversity of the nation. Silk and Walsh have promised a ninth volume that presumably will summarize and further assess the importance of the data collected. But even without that volume, the existing works offer much information that will be useful to scholars of American life and that is more nuanced than anything the press currently provides about religion in public life.

The individual essays in the volumes are rich and rewarding, offering detailed snapshots of religion in local settings. Moreover, because of the data now available on new immigrants and non-Anglos, we are afforded glimpses into religious worlds left untouched in earlier regional overviews. Native Americans in the Southwest, indigenous Hawaiians, and Asian communities on both coasts take their place alongside Methodists, Baptists, Catholics, and Jews. The data also include those who self-identify as having no religion, a feature that illuminates characteristics of public life that are then explored in the essays. The trait is notable in the Pacific Northwest, where a significant minority of the population claims no religious affiliation.

If one takes Religion by Region to be the fulfillment of Zelinsky’s work, however, one misses the profound procedural differences between the projects, and it is the assumptions underlying the analytical creation of regions that should be of most interest to scholars. Although Zelinsky broached tentative generalizations about religious regions, he also insisted that attempts to define regions should not merely consist of a series of “logical, but arbitrary, statistical assumptions.” Delineated regions had to matter to those who lived within them; each had to emerge from conscious study and charting of a religious distinctiveness that was evident both to the observer and to the local population. Thus, although Zelinsky’s seven regions encompassed the entire United States, he also admitted that only a few of them could be justified empirically. Aside from the Mormon Region, and perhaps scattered parts of New England and the South, the designation of a “region” depended on only “some moderate degree of areal homogeneity apparent to the map analyst if not to the majority of inhabitants.” Differences between rural and urban areas further confounded the attempt to generalize, since cities such as New York, Miami, and Detroit formed smaller subregions that stood apart, religiously, from their surroundings. Regions, it would seem, were born, not made.[6]

That approach characterizes long traditions of dividing the nation into regions in disciplines other than geography. Some, such as the economic analyses of Joel Garreau’s Nine Nations of North America (1981), posit regions as objectively and quantitatively verifiable entities. Garreau counted nine regions, arguing somewhat apocalyptically that the socioeconomic distinctions among them threatened to tear at the national fabric. The sociologist Raymond D. Gastil, in his 1975 Cultural Regions of the United States, also asserted that measures of homicide rates, infant mortality, and other social statistics create cultural enclaves within the country. Historians and scholars of American studies in recent years, while more ambivalent about the constructed nature of regions, have nonetheless focused attention on the powers of myth, tradition, and even the natural landscape to create regional identity. In all these cases, as in Zelinsky’s pursuit, social and cultural markers, not state boundaries, determine regions. Garreau even stated baldly that his regions (which he hyperbolically referred to as “nations”) “look different, feel different, and sound different from each other, and few of their boundaries match the political lines drawn on current maps.”[7]

The Religion by Region project began from the opposite direction. The project divided the United States into eight regions, but those areas were constructed in advance of the demographic profiles that characterize them. Utilizing state boundaries rather than topographical, ethnic, or class differences, the project joined together places that cultural geographers might have separated. Idaho, for example, becomes part of the Mountain West instead of the Pacific Northwest. The stated rationale for the decision to follow state boundaries is that it shifts the focus away from a religious mapping of an area toward an examination of how politics relates to local religious experience. Since Idaho is a bureaucratic unit, its inhabitants all share a state political ethos that relates to their religious life. States, therefore, were kept intact in order to understand localized dynamics of religious public life. Taken together, study of these eight regions will move us, in Silk’s words, toward a heightened awareness of “the cultural dynamics underlying our national politics in a way that looking simply at what goes on inside the Beltway cannot.”[8]

This tactic presupposes that the United States can be divided into eight discrete entities with definable characteristics. While that is the operating principle of all such projects, in order to do so here and still keep the state borders intact, the editors must anticipate, if only in a preliminary way, what they will find. In some cases, that means settling for received wisdom about religious trends. Why Idaho in the Mountain West? Because populous southern Idaho, being heavily Mormon, is linked more closely to Utah than to its western neighbors. New England consists of the usual six suspects (Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, Vermont), and the South is also easily surmised (Alabama, the Carolinas, Florida, Georgia, Kentucky, Mississippi, Tennessee, Virginia, West Virginia). In other instances, regions seem cobbled together out of disparate parts, such as the sprawling Midwest (Illinois, Indiana, Iowa, Kansas, Michigan, Minnesota, Nebraska, North and South Dakota, Ohio, Wisconsin) and the Pacific (California, Hawaii, Nevada). Yet in at least one case, that of the Southern Crossroads (Arkansas, Louisiana, Missouri, Oklahoma, Texas), the grouping results in interesting findings that might otherwise have been overlooked.

Zelinsky’s skepticism about forced regionalism aside, the question remains whether this process for demarcating regions yields an illuminating analysis. Silk asserts that regional profiles allow us to get at the cultural dynamics of religion in the United States in a nuanced and sophisticated way—far better than reading the public face of religion from the perspective of red/blue states or Beltway politics. Each region identified in this project has a distinctive demographic profile that tells us something important about how religion works there, from which religious institutions are predominant to the idioms of religion in public speech. The volumes explain, for example, why environmentalism is a viable religious issue in the Northwest (because it constitutes what Silk terms the area’s sacred “gospel of biodiversity”).[9] They also give us insight into why the 2004 presidential hopeful Howard Dean, a Vermonter, had so much trouble understanding how to talk about religion to a southern constituency (because New Englanders do not discuss religion in public). In turn, comparing those profiles answers questions about larger cultural patterns, about how people vote, about their differing expectations for the role of religion in the public sphere.

Beyond the broad brushstrokes, however, the implicitly comparative characterizations of regions, signaled by the subtitles attached to the volumes, are symptomatic of the potential pitfalls of regional generalization. The Pacific Northwest, for instance, is the “none zone,” the Middle Atlantic is dubbed “the fount of diversity,” and the Pacific Region is distinguished by “fluid identities.” I found some of the characterizations helpful and others a bit jarring. Does the software billionaire Bill Gates know that he is living in the “none zone,” and does he feel like a “none”? Do Californians really look to the singer Madonna, with her mix-and-match faith, as their cultural standard-bearer? The regional analyses work better in some places than in others, as the authors acknowledge. New England sees itself as in many respects a region, and it seems to have a stable—if complex—identity. The Midwest does not—in fact, Philip Barlow begins that volume by announcing that the area has an identity problem, in that it lacks natural boundaries, a unified culture, or a defined consciousness. Other regions have multiple identities and cultures that do not quite fit their profiles—indeed, the grouping of states to form some of the regions seems to undermine the use of the category “region.”[10]

The volumes that are most coherent—and therefore most revealing about regional distinctiveness—are those that describe regions birthed “naturally,” according to Zelinsky’s criteria. The Mountain West and New England stand out in this regard because each is a region with a strong sense of religious identity. Yet even here, the most textured essays challenge the very generalizations that the volumes proffer. Jan Shipps, for example, opens the volume on the Mountain West (“sacred landscapes in transition”) by asserting that it is not a cohesive region at all; it must be broken down into three subregions that exhibit greater geographic and cultural coherence. Thus, she ignores the state-based method and zeroes in on the culture regions. That strategy allows essay authors to focus on smaller pieces: Ferenc Szasz examines the many sacred sites in Arizona, New Mexico, and parts of Colorado; Randi Walker focuses on the “Catholic heartland” of Arizona and New Mexico; Kathleen Flake explores the Mormon corridor; and Philip Deloria assays Native American religions in Colorado, Wyoming, and Montana. Not only does this breakdown yield regions that are religiously more interesting than those the state borders provide (in fact, little mention is made of state politics at all), but the authors are not shy about pointing out situations that challenge the statistical categories of the NARA data or the problematic nature of the definitions of religion employed in the survey instruments.[11]

Generalization to the point of stereotype, then, is a potential pitfall of regionalizing religion, and, fortunately, many of the essays in these volumes steer us away from the abyss the structure itself threatens to cause. The implicit goal of regionalization, conversely, is fruitful comparison. How does one place stack up against another, and why are there such profound differences from one end of the country to the other? In this regard as well, Religion by Region promises insights that are fulfilled by the individual essays and the demographic data. The biggest surprise, perhaps, is the volume on the Southern Crossroads of Arkansas, Louisiana, Missouri, Oklahoma, and Texas, edited by Silk and William Lindsey, which explores the effects of major “religious showdowns” among the French, Hispanic, Native American, African American, and Anglo-American communities who historically have met in that region.[12] While the Wild West reduction threatens to impart stereotypes, available data from these states indicate that Pentecostal and restorationist traditions have particularly deep roots there. In fact, the evidence suggests that this region is more conservative on many social and political issues than any other area of the country. That implies, perhaps, that although inhabitants might not claim a regional identity, the pitched religious battles of many decades have shaped belief and behavior in distinctive ways.

Zelinsky aptly identified the two biggest challenges for the scholar of region and religion: diversity and constant change. People do not stay in one place; they tend to move, and as the twentieth century drew to a close, Americans were on the move in unprecedented numbers and directions. Moreover, with the influx of new immigrants into the United States after 1965, religious diversity has increased dramatically. Places are “forever being made, contested, and remade,” as Karen Halttunen recently observed.[13] Similarly, religious traditions are constantly adapting, negotiating, and mutating into new forms. How can one take a snapshot of religion in a particular location that captures that dynamism?

Authors in the Religion by Region volumes, as one might anticipate, struggle with how to characterize religious diversity, contact, and conflict. The choices raise intriguing questions about the multiple ways variety can be construed. While the volume on the Southern Crossroads chooses the “showdown” metaphor, religious variation in the Pacific region (“fluid identities”) seems decidedly more exotic. The appellation “fluid identities” is undoubtedly a great improvement over the older stereotypes of the West Coast as a site for crackpots, hippies, and cults, yet some of the authors of that volume use terms (“easy-going religious environment,” “peculiar religious economy,” “mutable,” “loose cultural establishment of mystical religious sensibilities,” “bizarre”) that indicate not simply movement away from previous patterns, but a diminution of religious piety and practice.[14] The Middle Atlantic region (Delaware, Maryland, New Jersey, New York, Pennsylvania, Washington, D.C.), while clearly just as diverse in religious tenor, is interpreted as a “proving ground” for pluralism, an area where diversity tethers communities to ethnic roots rather than forces them into perpetual itinerancy. Finally, Philip Deloria’s essay on Native American religions in the Mountain West represents a fourth (and quite favorably rendered) form of pluralism; here religious diversity embraces “openness and syncretism,” “tolerance,” and “religious cosmopolitanism.”[15]

One hopes that the final volume will address those characterizations and explain the divergences, particularly because descriptions of pluralism are never value neutral. As qualitative indicators, then, such choices merit explanation. It is possible that the Middle Atlantic “fount of diversity” has a very different feel from the conflicted Southern Crossroads, but why? What factors make one form of religious diversity positive and another problematic? The more belligerent religious style of the Southern Crossroads, especially when juxtaposed to the fluid, tolerant mode of the Pacific region, indicates that something more than sheer variety affects these patterns. Several intriguing possibilities suggest themselves, most notably class differences, neighborhood patterns, or something more specific to religious traditions, for example, varying teachings about relations with outsiders. Another possibility is that religious intensity—and therefore the way difference is expressed—varies by region. That need not mean that some people are less religious than others (although that, too, is possible), but it could signal that individuals in different regions are socialized to behave religiously in ways that have more to do with local practices than with the specifics of religious ritual. In an intriguing 1976 study, James R. Shortridge observed that the built environment, the density of social options available, ethnic heritage, and political beliefs have all been linked to religious commitment, but that those factors correlate in very different ways depending on where one looks. So, for example, South Dakota, with its relative dearth of church buildings, nonetheless maintains a high level of religious intensity because of strong ethnic ties and a need for sewing circles and youth programs where few social options exist.[16] Manifestations of diversity are themselves diverse, in other words.

It is also possible, of course, that religious diversity is a matter of perspective. Scholars are also inhabitants of regions, and so it is fair to wonder whose perspectives are being expressed in these volumes. Those of cultural insiders? Or those of a relatively more mobile, affluent, and highly educated knowledge class? Many of the authors, to their credit, pay attention to both insider and outsider perceptions of a region. They examine how it might feel to be a religious adherent in a particular place, how that setting might shape one’s beliefs and behavior, and also how a particular regional culture is assessed from the outside—how it is read culturally. Some of the essays shift readily between external and internal vantage points and employ metaphors of cultural identity that might encapsulate both perspectives. But even if one brackets the question of whether it is important that the characterization of a region be recognizable to its inhabitants, we are still left to query the normative position of the observer. If region shapes religious outlook, doesn’t it shape the outlook of scholars themselves and affect their perceptions of other places?

As the issues of diversity and mobility suggest, this set of volumes introduces a wide range of observations that open into broader questions. Is a region constituted by physical space? By the people who live there? By a particular constellation of institutions? Region is all these things, obviously, and employing it as a stable category allows us to explore the many factors that shape religious behavior. But unless we are to believe that human beings are chameleons who change color as soon as they enter a new place, we must surely take into account mobility and transplantation within the nation as well. If one is raised in the Pacific Region and moves to the South, what most centrally shapes one’s religious sensibilities and practices? Is it the person’s region of origin? The place of settlement? If it is a combination of those things, what does that tell us about the importance of mobility across regions? Or about the shaping power of class identities (since mobility itself is often a function of socioeconomics)? Do those with the power to move have a different perspective on religion? What about family patterns and how they shape religious identity? Is religious identity necessarily shaped by the same set of factors in different places? The volumes suggest that it probably is not; instead, environment might be more formative in the Northwest, for example, migration more formative in southern California.

None of those questions take away from the many achievements of the Religion by Region project. To juggle the variables of shifting places and mutable religious identities is an ambitious task, one that is bound to lead to spectacular insights and, if not noble failures, at least to the admission that in some places and at some times, the search for a regional religious personality may result in an identity crisis. As in the very best projects, though, the diligent and insightful work of the authors outruns the capacity of the paradigm to account for its findings.


Laurie F. Maffly-Kipp is associate professor of religious studies at the University of North Carolina at Chapel Hill.
Readers may contact Maffly-Kipp at [email protected].

1 Wilbur Zelinsky, “An Approach to the Religious Geography of the United States: Patterns of Church Membership in 1952,” Annals of the Association of American Geographers, 51 (June 1961), 139–93, esp. 149, 165.

2Ibid., 164. National Council of Churches of Christ in the United States of America, Churches and Church Membership in the United States: An Enumeration and Analysis by Counties, States, and Regions (80 vols., New York, 1956–1958).

3 For an example of the ubiquitous references to “public cultures,” see “Preface” in Religion and Public Life in the Midwest: America’s Common Denominator?, ed. Philip Barlow and Mark Silk (Walnut Creek, 2004), 7.

4 The North American Religion Atlas (NARA) is an online interactive multimedia tool using data based on research from the 2000 Glenmary Religious Congregations and Membership Survey, available at Polis Center, North American Religion Atlas, http://religionatlas.org/. The American Religious Identification Survey, produced in 2001 by the Graduate Center of the City University of New York, is online at Barry A. Kosmin, Egon Mayer, and Ariela Keysar, American Religious Indentification Survey, http://www.gc.cuny.edu/faculty/research_studies/aris.pdf. The National Surveys of Religion and Politics were produced by Dr. John C. Green at the Ray C. Bliss Institute of Applied Politics at the University of Akron. He can be reached at [email protected].

5 Edwin S. Gaustad, Philip L. Barlow, and Richard W. Dishno, eds., New Historical Atlas of Religion in America (New York, 2001).

6 Zelinsky, “Approach to the Religious Geography of the United States,” 162, 163.

7 Joel Garreau, The Nine Nations of North America (Boston, 1981), esp. 2; Raymond D. Gastil, Cultural Regions of the United States (Seattle, 1975); Edward L. Ayers, Patricia Nelson Limerick, Stephen Nissenbaum, and Peter S. Onuf, All over the Map: Rethinking American Regions (Baltimore, 1996).

8 Mark Silk, “Religion and Region in American Public Life,” Journal for the Scientific Study of Religion, 44 (Sept. 2005), 265.

9 Bruce Murray, “How Region Influences Religion in America,” FACSNET, http://www.facsnet.org/issues/faith/region.php.

10 Patricia O’Connell Killen and Mark Silk, eds., Religion and Public Life in the Pacific Northwest: The None Zone (Walnut Creek, 2004); Randall Balmer and Mark Silk, eds., Religion and Public Life in the Middle Atlantic Region: The Fount of Diversity (Walnut Creek, 2006); Wade Clark Roof and Mark Silk, eds., Religion and Public Life in the Pacific Region: Fluid Identities (Walnut Creek, 2005); Andrew Walsh and Mark Silk, eds., Religion and Public Life in New England: Steady Habits, Changing Slowly (Walnut Creek, 2004); Barlow and Silk, eds., Religion and Public Life in the Midwest.

11 Jan Shipps, “Religion in the Mountain West: Geography as Destiny,” in Religion and Public Life in the Mountain West: Sacred Landscapes in Transition, ed. Jan Shipps and Mark Silk (Walnut Creek, 2004), 9–14. For challenges to statistical categories and definitions, see, for example, Kathleen Flake, “The Mormon Corridor: Utah and Idaho,” ibid., 92; and Philip Deloria, “Polarized Tribes: Colorado, Wyoming, and Montana,” ibid., 120–23.

12 William Lindsey and Mark Silk, eds., Religion and Public Life in the Southern Crossroads: Showdown States (Walnut Creek, 2004).

13 Karen Halttunen, “Groundwork: American Studies in Place—Presidential Address to the American Studies Association, November 4, 2005,” American Quarterly, 58 (March 2006), 10.

14 Wade Clark Roof, “Religion in the Pacific Region: Demographic Patterns,” in Religion and Public Life in the Pacific Region, ed. Roof and Silk, 41, 50; Douglas Firth Anderson, “Toward an Established Mysticism: Judeo-Christian Traditions in Post–World War II California and Nevada,” ibid., 58; Phillip E. Hammond, “Introduction: Religion in the Pacific Region,” ibid., 15.

15 Randall Balmer, “Introduction: The Proving Ground for Pluralism,” in Religion and Public Life in the Middle Atlantic Region, ed. Balmer and Silk, 9; Deloria, “Polarized Tribes,” 120.

16 James R. Shortridge, “Patterns of Religion in the United States,” Geographical Review, 66 (Oct. 1976), 420–34.