Commentary Magazine

Bowling with Others

In his celebrated book, Bowling Alone (2000), the political scientist Robert D. Putnam argued that America, and perhaps the Western world as a whole, has become increasingly disconnected from family, friends, and neighbors. We once bowled in leagues; now we bowl alone. We once flocked to local chapters of the PTA, the NAACP, or the Veterans of Foreign Wars; now we stay home and watch television. As a result, we have lost our “social capital”—by which Putnam meant both the associations themselves and the trustworthiness and reciprocity they encourage. For if tools (physical capital) and training (human capital) make the modern world possible, social capital is what helps people find jobs and enables neighborhoods and other small groupings of society to solve problems, control crime, and foster a sense of community.

In Bowling Alone, Putnam devised a scale for assessing the condition of organizational life in different American states. He looked to such measures as the density of civic groups, the frequency with which people participate in them, and the degree to which (according to opinion surveys) people trust one another. Controlling for race, income, education, and the like, he demonstrated that the higher a state’s level of social capital, the more educated and affluent are its children, the lower the murder rate, the greater the degree of public health, and the smaller the likelihood of tax evasion. Nor is that all. High levels of social capital, Putnam showed, are associated with such civic virtues as greater tolerance toward women and minorities and stronger support for civil liberties. But all of these good things have been seriously jeopardized by the phenomenon he identified as “bowling alone.”

After finishing his book, Putnam was approached by various community foundations to measure the levels of social capital within their own cities. To that end he conducted a very large survey: roughly 30,000 Americans, living in 41 different communities ranging downward in size from Los Angeles to Yakima, Washington and even including rural areas of South Dakota. He published the results this year in a long essay in the academic journal Scandinavian Political Studies on the occasion of his having won Sweden’s prestigious Johan Skytte prize.



Putnam’s new essay takes an in-depth look not at social capital per se but at how “diversity”—meaning, for this purpose, racial and ethnic differences—affects our lives in society. Such diversity is increasing in this country and many others, if for no other reason than immigration, and so Putnam has tried to find out how it changes the way people feel about their neighbors, the degree of their confidence in local government, their willingness to become engaged in community-wide projects, and their general happiness.

The ethnic and racial diversity that Putnam examines is widely assumed to be very good for us. The more time we spend with people different from us, it is said, the more we will like and trust them. Indeed, diversity is supposed to be so good for us that it has become akin to a national mandate in employment and, especially, in admissions to colleges and universities. When the Supreme Court decided the Bakke case in 1978, the leading opinion, signed by Justice Lewis Powell, held that although a university was not allowed to use a strict numerical standard to guarantee the admission of a fixed number of minority students, it could certainly “take race into account,” on the theory that a racially diverse student body was desirable both for the school and for society at large.

As a result of this and similar court rulings, not only colleges but many other institutions began invoking the term “diversity” as a justification for programs that gave preferences to certain favored minorities (especially blacks and Hispanics). Opponents of these programs on constitutional and civil-liberties grounds were put in the difficult position of appearing to oppose a demonstrated social good. Did not everyone know that our differences make us stronger?

But do they? That is where Putnam’s new essay comes in. In the long run, Putnam argues, ethnic and racial diversity in neighborhoods is indeed “an important social asset,” because it encourages people to form connections that can reduce unproductive forms of ethnocentrism and increase economic growth. In his words, “successful immigrant societies create new forms of social solidarity and dampen the negative effects of diversity by constructing new, more encompassing identities.”

Whatever his beliefs about the positive effects of diversity in the long run, however—not only does he consider it a potentially “important social asset,” but he has written that it also confers “many advantages that have little or nothing to do with social capital”—Putnam is a scrupulous and serious scholar (as well as a friend and former colleague at Harvard). In the short run, he is frank to acknowledge, his data show not positive effects but rather the opposite. “The more ethnically diverse the people we live around,” he writes, “the less we trust them.”

Diversity, Putnam concludes on the basis of his findings, makes us “hunker down.” Not only do we trust our neighbors less, we have less confidence in local government, a lowered sense of our own political efficacy, fewer close friends, and a smaller likelihood of contributing to charities, cooperating with others, working on a community project, registering to vote—or being happy.

Of course many of these traits can reflect just the characteristics of the people Putnam happened to interview, rather than some underlying condition. Aware of the possibility, Putnam spent a great deal of time “kicking the tires” of his study by controlling statistically for age, ethnicity, education, income or lack of same, poverty, homeownership, citizenship, and many other possible influences. But the results did not change. No matter how many individual factors were analyzed, every measure of social well-being suffered in ethnically diverse neighborhoods—and improved in ethnically homogeneous ones.



“Shocking” is the word that one political scientist, Scott Page of the University of Michigan, invoked to describe the extent of the negative social effects revealed by Putnam’s data. Whether Putnam was shocked by the results I cannot say. But they should not have been surprising; others have reported the same thing. The scholars Anil Rupasingha, Stephan J. Goetz, and David Freshwater, for example, found that social capital across American counties, as measured by the number of voluntary associations for every 10,000 people, goes up with the degree of ethnic homogeneity. Conversely, as others have discovered, when ethnic groups are mixed there is weaker social trust, less car pooling, and less group cohesion. And this has held true for some time: people in Putnam’s survey who were born in the 1920’s display the same attitudes as those born in the 1970’s.

Still, Putnam believes that in the long run ethnic heterogeneity will indeed “create new forms of social solidarity.” He offers three reasons. First, the American military, once highly segregated, is today anything but that—and yet, in the Army and the Marines, social solidarity has increased right alongside greater ethnic diversity. Second, churches that were once highly segregated, especially large evangelical ones, have likewise become entirely and peaceably integrated. Third, people who once married only their ethnic kin today marry across ethnic and religious (and, to a lesser degree, racial) lines.

I can offer a fourth example: organized sports. Once, baseball and football teams were made up of only white or only black players; today they, too, are fully integrated. When Jackie Robinson joined the Brooklyn Dodgers in 1947, several teammates objected to playing with him, and many fans heckled him whenever he took the field. Within a few years, however, he and the Dodgers had won a raft of baseball titles, and he was one of the most popular figures in the country. Today such racial and ethnic heckling has virtually disappeared.

Unfortunately, however, the pertinence of the military, religious, or athletic model to life in neighborhoods is very slight. In those three institutions, authority and discipline can break down native hostilities or force them underground. Military leaders proclaim that bigotry will not be tolerated, and they mean it; preachers invoke the word of God to drive home the lesson that prejudice is a sin; sports teams (as with the old Brooklyn Dodgers) point out that anyone who does not want to play with a black or a Jew is free to seek employment elsewhere.

But what authority or discipline can anyone bring to neighborhoods? They are places where people choose to live, out of either opportunity or necessity. Walk the heterogeneous streets of Chicago or Los Angeles and you will learn about organized gangs and other social risks. Nor are these confined to poor areas: Venice, a small neighborhood in Los Angeles where several movie stars live and many homes sell for well over $1 million, is also a place where, in the Oakwood area, the Shoreline Crips and the V-13 gangs operate.

In many a neighborhood, ethnic differences are often seen as threats. If blacks or Hispanics, for whatever reason, are more likely to join gangs or commit crimes, then whites living in a neighborhood with many blacks or Hispanics will tend to feel uneasy. (There are, of course, exceptions: some, especially among the well-educated, prefer diversity even with all its risks.) Even where everyone is equally poor or equally threatened by crime, people exhibit less trust if their neighborhood is ethnically diverse than if it is homogeneous.

Of Putnam’s three or four reasons for thinking that ethnic heterogeneity will contribute to social capital in the long run, only one is compelling: people are indeed voluntarily marrying across ethnic lines. But the paradoxical effect of this trend is not to preserve but to blunt ethnic identity, to the point where it may well reduce the perception of how diverse a neighborhood actually is. In any case, the fact remains that diversity and improved solidarity have gone hand in hand only in those institutions characterized by enforced authority and discipline.

The legal scholar Peter H. Schuck has written an important book on this issue. In Diversity in America (2003), he examines three major efforts by judges and government officials to require racial and income diversity in neighborhoods. One of them banned income-discrimination in the sale and rental of housing in New Jersey towns. Another enabled blacks who were eligible for public housing to move into private rental units in the Chicago suburbs. In the third, a federal judge attempted to diversify residential patterns in the city of Yonkers, New York by ordering the construction of public housing in middle-class neighborhoods selected by him.

Although the Chicago project may have helped minorities to enter communities where they had never lived, the New Jersey and Yonkers initiatives had little effect. As Schuck writes, “Neighborhoods are complex, fragile, organic societies whose dynamics outsiders cannot readily understand, much less control.” A court can and should strike down racist public policies, but when it goes beyond this and tries to mandate “diversity,” it will sooner or later discover that it “cannot conscript the housing market to do its bidding.”



Taking a different approach, Thomas Schelling, a Nobel laureate in economics, has shown in a stimulating essay that neighborhood homogeneity and even segregation may result from small, defensible human choices that cannot themselves be called racist. In fact, such choices can lead to segregation even when the people making them expressly intend the opposite. Suppose, Schelling writes, that blacks and whites alike wish to live in a neighborhood that is (for example) half-white and half-black. If one white family should come to think that other white families prefer a community that is three-fourths white, and may move out for that reason, the first white family is itself likely to move out in search of its own half-white, half-black preference. There is no way to prevent this.

Schelling’s analysis casts a shadow of doubt on Putnam’s own policy suggestions for reducing the disadvantages and stimulating the benefits of ethnic heterogeneity. Those suggestions are: investing more heavily in playgrounds, schools, and athletic fields that different groups can enjoy together; extending national aid to local communities; encouraging churches to reach out to new immigrants; and expanding public support for the teaching of English.

The first recommendation is based on the implicit assumption that Schelling is wrong and on the even more dubious assumption that playgrounds, schools, and athletic fields—things Putnam did not measure in his survey—will increase the benefits of diversity even when age, income, and education do not. The second is empty: Putnam does not say what kind of aid will produce the desired effects. If he is thinking of more housing, Schuck has already shown that providing this usually does not increase diversity. If he is thinking of education, in the 1970’s federal judges imposed forced busing in an effort to integrate schools; it was an intensely unpopular strategy, both among those whose children were being bused and among those whose neighborhoods were being bused into.

The third proposal, encouraging outreach by churches, might well make a difference, but how do we go about it? Require people to attend an evangelical church? Would Robert Putnam attend? I suspect not. And as for the final recommendation, teaching English at public expense to everyone, it is a very good idea—provided one could break the longstanding attachment of the education establishment to bilingual instruction.



Whether we should actually seek to transform the situation described by Putnam’s data is another question. I do not doubt that both diversity and social capital are important, or that many aspects of the latter have declined, though perhaps not so much as Putnam suspects. But as his findings indicate, there is no reason to suppose that the route to the latter runs through the former. In fact, strong families living in neighborhoods made up of families with shared characteristics seem much more likely to bring their members into the associational life Putnam favors. Much as we might value both heterogeneity and social capital, assuming that the one will or should encourage the other may be a form of wishful thinking.

That is because morality and rights arise from different sources. As I tried to show in The Moral Sense (1993), morality arises from sympathy among like-minded persons: first the family, then friends and colleagues. Rights, on the other hand, grow from convictions about how we ought to manage relations with people not like us, convictions that are nourished by education, religion, and experience.

People who celebrate diversity (and its parallel, multiculturalism) are endorsing only one part of what it means to be a complete human being, neglecting morality (and its parallel, group and national pride). Just as we cannot be whole persons if we deny the fundamental rights of others, so we cannot be whole persons if we live in ways that discourage decency, cooperation, and charity.

In every society, people must arrange for tradeoffs between desirable but mutually inconsistent goals. James Madison, in his famous Federalist Number Ten, pointed to just this sort of tradeoff when he made the case for a large national government that would ensure the preservation of those individual rights and liberties that are at risk in small communities. When it comes to the competing values of diversity and the formation of social capital, as when it comes to other arrangements in a democracy, balance is all.

About the Author