One Nation under God: Religion in Contemporary Society. - book reviews

ONE NATION UNDER GOD

Religion in Contemporary Society

Barry A. Kosmin and Seymour P. Lachman

Harmony Books, $25, 320 pp.

Sociology is most likely to make it into the media when its findings contradict common sense. Kosmin's and Lachman's findings first appeared in hundreds of newspapers in April 1991. Their spin was that religion was doing pretty well in America and that politicians and others who did not realize this would not succeed. Kosmin and Lachman, researcher and dean, respectively, at City University of New York's Graduate Center, found traditional religion mostly alive and healthy, and a contributor to national unity. This might have surprised editors who, according to other surveys about them, find out about religion mostly from poll data.

The new story about religion in America, presented now formally in One Nation under God, turns out to be that it's pretty similar to the old one. The data seem to break both old and new ethnic and religious stereotypes. Most Americans who claim Irish or French descent are not Catholics, most American Muslims are not Arabs, and most Asians are Catholics. Talk of new religions and New Age religion is much exaggerated. There is even more generational continuity than gap. Among the under twenty-five, young-adult population, 86 percent of women and 81.1 percent of men identify with a Christian church or denomination. The authors conclude: "There is little evidence that Americans have abandoned religion during the course of the twentieth century."

The first social science question we need to ask is methodological. Kosmin and Lachman's first claim on our attention is that theirs is the only large, national sample on religion in the United States that we have. The only national Census Bureau Survey that included questions on religion was conducted in March 1957; all other Census Bureau efforts were stopped by objections from some Jews, Christian Scientists, and First Amendment guardians. The usual national surveys or polls asking about religion have sample sizes too small for much more than generic Catholic, Protestant, or "other" comparisons. Kosmin and Lachman used a much larger sample for their "National Survey of Religious Identification" (NSRI)--113,000 respondents (eighteen years or older). The poll was conducted by the commercial firm of ICR Survey Research Group of Media, Pennsylvania, and entailed selected telephone interviews from April 1989 to April 1990 by random digit dialing to achieve a geographically representative sample.

The authors describe the NSRI sample as "unique." For example, the sample includes 26,000 Catholics, including 900 black Catholics, 600 Asian-American Catholics, and over 3,000 Hispanic Catholics. Table 1-2 displays the number and estimated percentage of the adult population of forty-three distinct Protestant groups. Many Protestants--almost 10 percent, second only to Catholics and Baptists--do not identify themselves as members of a specific denomination but as members of a local church and refer to themselves simply as "Protestant." Samples from seventeen "other" religions (including Jewish, Muslim, Buddhist, Hindu, Native American, New Age) and from agnostics, humanists, and an oxymoronic "religious nones" were also used. Readers are alerted to some sampling problems, such as the lack of telephones among Old Order Amish and the lack of English among newer immigrants. Similarly, Catholic officials tend to overestimate the number of Catholic Hispanics--Kosmin and Lachman found only two-thirds of Hispanics identify as Catholics--and underestimate Catholic success among Asians who now comprise 2 percent of all American Catholics. Non-English-speaking Hispanics might be most likely to be Catholic and non-English-speaking Asians to be other than Catholic.

It should be noted that the interviewees did not know that they were to be interviewed about religion. The questions on religion were mixed in with a longer series of questions about consumer preferences and other social issues. There was only a 2.3 percent refusal rate for the questions on religion, but the overall survey cooperation rate was 50 percent even after four attempts. The authors do not discuss whether those who refused to participate in a consumer preference survey might be different, especially with regard to judgments about religion, than those who were willing. Nor do they speculate about the indirect effects of placing questions about religion alongside questions about consumer preferences. Might this suggest to some willing interviewees that, just like any other commodity, Americans are also expected to have brand preferences about religion?

At any rate, Kosmin and Lachman found that most Americans not only had religious identifications but kept the ones they first found at home. Well-educated males were the most likely to switch brands, and the Pacific region had the highest proportion of religious switchers (36.5 percent) and New England (14.9 percent) the lowest. The most diverse religious markets are the Western states and they reported the most "no religion" answers. One-third of all Catholics live in the four states of New York, California, Pennsylvania, and Texas; only in Delaware do Methodists comprise as much as one-fifth of the population; and Jews show the most geographical mobility. Market researchers for religious merchandise will appreciate learning that the two states with a religious profile that most closely approximate the total United States are Missouri and Florida. Those looking for quick surprises will note that 90 percent of New Yorkers identify with a religious background and that one-third attend a public worship service in any given week. Those looking for quick solutions will be disappointed to learn that religious practice and belief are much more prevalent in high-crime areas, such as Brooklyn and the Bronx.

While the NSRI sample is larger by far than others. the overall findings basically corroborate what smaller samples have already told us. When asked, Americans tell surveyers that they are a religious people. Only 7.5 percent said they had "no religion." There are few grassroots American humanists, only .7 percent to be exact, and a few more agnostics, only 1.7 percent of them. But there's lots of religion. More than 86 percent of those polled identified with a Christian denomination, and 2 percent with a Jewish tradition. More specifically, Roman Catholics are the largest American religious group, about 26.2 percent. Baptists come next, with 19.4 percent of the population. While most of the mainline Protestant groups have lost about a quarter of their membership since the 1960s, many nonmembers maintain their religious identification. For example, the Presbyterian Church U.S.A. lists only 3 million members but the NSRI estimates affiliation at nearly 5 million. Unfortunately for multicultural enthusiasts, most of the Asian-Americans said they were not members of Asian-based religions, such as Hinduism, Buddhism, or Islam, but that they were Catholics or Protestants. Native Americans follow the same pattern.

The authors caution against "an overemphasis on the differences between groups of people without equal attention to the common denominators that bind together all Americans." They disparagingly note the "premium on claims for special status." If you're looking for American success, look at black Catholics. Just about 9 percent of blacks are Catholic, representing about 5 percent of Catholics in general, but they are more likely to be college graduates than other Americans and 40 percent more likely than other black Americans. The religious affiliation of Asian-Americans in the 1990s is basically the same as that of other Americans--Catholics and Baptists are the largest groups. "Surprisingly few" New Age "devotees," Scientologists, or followers of Eckanar were found, and most were in California.

My review copy did not contain an appendix with the questions actually asked but the introduction explains that the questions simply asked about affiliation, not behavior. A book about "religion and contemporary American society" based on a few self-report answers to religious identification questions must be largely inferential. Some of the chapters are good summaries of available data. I liked especially chapter 5, "The Politics of Religion and the Religion of Politics," where, among other things, the authors report that evangelical Protestants now comprise 20 percent of the nation's population, that Catholics are still more likely to identify as Democrats (38 percent) than Republicans (27 percent) and are more likely than white Protestants to support an expanded role for government. National Opinion Research Center data show that 34 percent of Catholic college graduates believe that government should do more to improve society, whereas only 16 percent of white Protestant college graduates hold that opinion. But the explanations for Kosmin's and Lachman's findings in the other six chapters are mostly familiar and largely derivative. For example: "To some degree, all faiths in America have been Protestantized." And: "Religion in the United States frequently sanctifies the goals of a basically secular society, and the secular society affects and influences the very meaning of religious identification and association. It is therefore not surprising that America appears to be growing more secular precisely at a time when religious identification is highly pronounced."

The individualism of American religion and the religion-secular correlation is old stuff, and the authors simply rehash the discussion here. There's little additional freshness or energy beyond their survey findings. The authors limply conclude that "predictions are not always wise or safe," and their last sentence reads: "The important thing to realize is that religion in America is dynamic, creative, and relevant, continuously changing and developing."

If you read the conclusion first, you'd be tempted to turn right to the survey data and then skip all the commentary. I'd guess the book is most likely to be enthusiastically cited by those desirous of showing that secularist and multiculturalist "knowledge-class" agendas are highly unrepresentative of even minority American sentiment.

COPYRIGHT 1994 Commonweal Foundation
COPYRIGHT 2004 Gale Group