The following is an essay published in 1995 that became a book. It was recommended by Marguerite Hampton of the Turtle Island Institute.
Bowling Alone
Robert D. Putnam
Many students of the new democracies that have emerged over the past decade and a half have emphasized the importance of a strong and active civil society to the consolidation of democracy. Especially with regard to the postcommunist countries, scholars and democratic activists alike have lamented the absence or obliteration of traditions of independent civic engagement and a widespread tendency toward passive reliance on the state. To those concerned with the weakness of civil societies in the developing or postcommunist world, the advanced Western democracies and above all the United States have typically been taken as models to be emulated. There is striking evidence, however, that the vibrancy of American civil society has notably declined over the past several decades.
Ever since the publication of Alexis de Tocqueville’s Democracy in America, the United States has played a central role in systematic studies of the links between democracy and civil society. Although this is in part because trends in American life are often regarded as harbingers of social modernization, it is also because America has traditionally been considered unusually “civic” (a reputation that, as we shall later see, has not been entirely unjustified).
When Tocqueville visited the United States in the 1830s, it was the Americans’ propensity for civic association that most impressed him as the key to their unprecedented ability to make democracy work. “Americans of all ages, all stations in life, and all types of disposition,” he observed, “are forever forming associations. There are not only commercial and industrial associations in which all take part, but others of a thousand different types–religious, moral, serious, futile, very general and very limited, immensely large and very minute. . . . Nothing, in my view, deserves more attention than the intellectual and moral associations in America.”1
Recently, American social scientists of a neo-Tocquevillean bent have unearthed a wide range of empirical evidence that the quality of public life and the performance of social institutions (and not only in America) are indeed powerfully influenced by norms and networks of civic engagement. Researchers in such fields as education, urban poverty, unemployment, the control of crime and drug abuse, and even health have discovered that successful outcomes are more likely in civically engaged communities. Similarly, research on the varying economic attainments of different ethnic groups in the United States has demonstrated the importance of social bonds within each group. These results are consistent with research in a wide range of settings that demonstrates the vital importance of social networks for job placement and many other economic outcomes.
Meanwhile, a seemingly unrelated body of research on the sociology of economic development has also focused attention on the role of social networks. Some of this work is situated in the developing countries, and some of it elucidates the peculiarly successful “network capitalism” of East Asia.2
Even in less exotic Western economies, however, researchers have discovered highly efficient, highly flexible “industrial districts” based on networks of collaboration among workers and small entrepreneurs. Far from being paleoindustrial anachronisms, these dense interpersonal and interorganizational networks undergird ultramodern industries, from the high tech of Silicon Valley to the high fashion of Benetton.
The norms and networks of civic engagement also powerfully affect the performance of representative government. That, at least, was the central conclusion of my own 20-year, quasi-experimental study of subnational governments in different regions of Italy.3 Although all these regional governments seemed identical on paper, their levels of effectiveness varied dramatically. Systematic inquiry showed that the quality of governance was determined by longstanding traditions of civic engagement (or its absence). Voter turnout, newspaper readership, membership in choral societies and football clubs–these were the hallmarks of a successful region. In fact, historical analysis suggested that these networks of organized reciprocity and civic solidarity, far from being an epiphenomenon of socioeconomic modernization, were a precondition for it.
No doubt the mechanisms through which civic engagement and social connectedness produce such results–better schools, faster economic development, lower crime, and more effective government–are multiple and complex. While these briefly recounted findings require further confirmation and perhaps qualification, the parallels across hundreds of empirical studies in a dozen disparate disciplines and subfields are striking. Social scientists in several fields have recently suggested a common framework for understanding these phenomena, a framework that rests on the concept of social capital.4 By analogy with notions of physical capital and human capital–tools and training that enhance individual productivity–”social capital” refers to features of social organization such as networks, norms, and social trust that facilitate coordination and cooperation for mutual benefit.
For a variety of reasons, life is easier in a community blessed with a substantial stock of social capital. In the first place, networks of civic engagement foster sturdy norms of generalized reciprocity and encourage the emergence of social trust. Such networks facilitate coordination and communication, amplify reputations, and thus allow dilemmas of collective action to be resolved. When economic and political negotiation is embedded in dense networks of social interaction, incentives for opportunism are reduced. At the same time, networks of civic engagement embody past success at collaboration, which can serve as a cultural template for future collaboration. Finally, dense networks of interaction probably broaden the participants’ sense of self, developing the “I” into the “we,” or (in the language of rational-choice theorists) enhancing the participants’ “taste” for collective benefits.
I do not intend here to survey (much less contribute to) the development of the theory of social capital. Instead, I use the central premise of that rapidly growing body of work–that social connections and civic engagement pervasively influence our public life, as well as our private prospects–as the starting point for an empirical survey of trends in social capital in contemporary America. I concentrate here entirely on the American case, although the developments I portray may in some measure characterize many contemporary societies.
Whatever Happened to Civic Engagement?
We begin with familiar evidence on changing patterns of political participation, not least because it is immediately relevant to issues of democracy in the narrow sense. Consider the well-known decline in turnout in national elections over the last three decades. From a relative high point in the early 1960s, voter turnout had by 1990 declined by nearly a quarter; tens of millions of Americans had forsaken their parents’ habitual readiness to engage in the simplest act of citizenship. Broadly similar trends also characterize participation in state and local elections.
It is not just the voting booth that has been increasingly deserted by Americans. A series of identical questions posed by the Roper Organization to national samples ten times each year over the last two decades reveals that since 1973 the number of Americans who report that “in the past year” they have “attended a public meeting on town or school affairs” has fallen by more than a third (from 22 percent in 1973 to 13 percent in 1993). Similar (or even greater) relative declines are evident in responses to questions about attending a political rally or speech, serving on a committee of some local organization, and working for a political party. By almost every measure, Americans’ direct engagement in politics and government has fallen steadily and sharply over the last generation, despite the fact that average levels of education–the best individual-level predictor of political participation–have risen sharply throughout this period. Every year over the last decade or two, millions more have withdrawn from the affairs of their communities.
Not coincidentally, Americans have also disengaged psychologically from politics and government over this era. The proportion of Americans who reply that they “trust the government in Washington” only “some of the time” or “almost never” has risen steadily from 30 percent in 1966 to 75 percent in 1992.
These trends are well known, of course, and taken by themselves would seem amenable to a strictly political explanation. Perhaps the long litany of political tragedies and scandals since the 1960s (assassinations, Vietnam, Watergate, Irangate, and so on) has triggered an understandable disgust for politics and government among Americans, and that in turn has motivated their withdrawal. I do not doubt that this common interpretation has some merit, but its limitations become plain when we examine trends in civic engagement of a wider sort.
Our survey of organizational membership among Americans can usefully begin with a glance at the aggregate results of the General Social Survey, a scientifically conducted, national-sample survey that has been repeated 14 times over the last two decades. Church-related groups constitute the most common type of organization joined by Americans; they are especially popular with women. Other types of organizations frequently joined by women include school-service groups (mostly parent-teacher associations), sports groups, professional societies, and literary societies. Among men, sports clubs, labor unions, professional societies, fraternal groups, veterans’ groups, and service clubs are all relatively popular.
Religious affiliation is by far the most common associational membership among Americans. Indeed, by many measures America continues to be (even more than in Tocqueville’s time) an astonishingly “churched” society. For example, the United States has more houses of worship per capita than any other nation on Earth. Yet religious sentiment in America seems to be becoming somewhat less tied to institutions and more self-defined.
How have these complex crosscurrents played out over the last three or four decades in terms of Americans’ engagement with organized religion? The general pattern is clear: The 1960s witnessed a significant drop in reported weekly churchgoing–from roughly 48 percent in the late 1950s to roughly 41 percent in the early 1970s. Since then, it has stagnated or (according to some surveys) declined still further. Meanwhile, data from the General Social Survey show a modest decline in membership in all “church-related groups” over the last 20 years. It would seem, then, that net participation by Americans, both in religious services and in church-related groups, has declined modestly (by perhaps a sixth) since the 1960s.
For many years, labor unions provided one of the most common organizational affiliations among American workers. Yet union membership has been falling for nearly four decades, with the steepest decline occurring between 1975 and 1985. Since the mid-1950s, when union membership peaked, the unionized portion of the nonagricultural work force in America has dropped by more than half, falling from 32.5 percent in 1953 to 15.8 percent in 1992. By now, virtually all of the explosive growth in union membership that was associated with the New Deal has been erased. The solidarity of union halls is now mostly a fading memory of aging men.5
The parent-teacher association (PTA) has been an especially important form of civic engagement in twentieth-century America because parental involvement in the educational process represents a particularly productive form of social capital. It is, therefore, dismaying to discover that participation in parent-teacher organizations has dropped drastically over the last generation, from more than 12 million in 1964 to barely 5 million in 1982 before recovering to approximately 7 million now.
Next, we turn to evidence on membership in (and volunteering for) civic and fraternal organizations. These data show some striking patterns. First, membership in traditional women’s groups has declined more or less steadily since the mid-1960s. For example, membership in the national Federation of Women’s Clubs is down by more than half (59 percent) since 1964, while membership in the League of Women Voters (LWV) is off 42 percent since 1969.6
Similar reductions are apparent in the numbers of volunteers for mainline civic organizations, such as the Boy Scouts (off by 26 percent since 1970) and the Red Cross (off by 61 percent since 1970). But what about the possibility that volunteers have simply switched their loyalties to other organizations? Evidence on “regular” (as opposed to occasional or “drop-by”) volunteering is available from the Labor Department’s Current Population Surveys of 1974 and 1989. These estimates suggest that serious volunteering declined by roughly one-sixth over these 15 years, from 24 percent of adults in 1974 to 20 percent in 1989. The multitudes of Red Cross aides and Boy Scout troop leaders now missing in action have apparently not been offset by equal numbers of new recruits elsewhere.
Fraternal organizations have also witnessed a substantial drop in membership during the 1980s and 1990s. Membership is down significantly in such groups as the Lions (off 12 percent since 1983), the Elks (off 18 percent since 1979), the Shriners (off 27 percent since 1979), the Jaycees (off 44 percent since 1979), and the Masons (down 39 percent since 1959). In sum, after expanding steadily throughout most of this century, many major civic organizations have experienced a sudden, substantial, and nearly simultaneous decline in membership over the last decade or two.
The most whimsical yet discomfiting bit of evidence of social disengagement in contemporary America that I have discovered is this: more Americans are bowling today than ever before, but bowling in organized leagues has plummeted in the last decade or so. Between 1980 and 1993 the total number of bowlers in America increased by 10 percent, while league bowling decreased by 40 percent. (Lest this be thought a wholly trivial example, I should note that nearly 80 million Americans went bowling at least once during 1993, nearly a third more than voted in the 1994 congressional elections and roughly the same number as claim to attend church regularly. Even after the 1980s’ plunge in league bowling, nearly 3 percent of American adults regularly bowl in leagues.) The rise of solo bowling threatens the livelihood of bowling-lane proprietors because those who bowl as members of leagues consume three times as much beer and pizza as solo bowlers, and the money in bowling is in the beer and pizza, not the balls and shoes. The broader social significance, however, lies in the social interaction and even occasionally civic conversations over beer and pizza that solo bowlers forgo. Whether or not bowling beats balloting in the eyes of most Americans, bowling teams illustrate yet another vanishing form of social capital.
Countertrends
At this point, however, we must confront a serious counterargument. Perhaps the traditional forms of civic organization whose decay we have been tracing have been replaced by vibrant new organizations. For example, national environmental organizations (like the Sierra Club) and feminist groups (like the National Organization for Women) grew rapidly during the 1970s and 1980s and now count hundreds of thousands of dues-paying members. An even more dramatic example is the American Association of Retired Persons (AARP), which grew exponentially from 400,000 card-carrying members in 1960 to 33 million in 1993, becoming (after the Catholic Church) the largest private organization in the world. The national administrators of these organizations are among the most feared lobbyists in Washington, in large part because of their massive mailing lists of presumably loyal members.
These new mass-membership organizations are plainly of great political importance. From the point of view of social connectedness, however, they are sufficiently different from classic “secondary associations” that we need to invent a new label–perhaps “tertiary associations.” For the vast majority of their members, the only act of membership consists in writing a check for dues or perhaps occasionally reading a newsletter. Few ever attend any meetings of such organizations, and most are unlikely ever (knowingly) to encounter any other member. The bond between any two members of the Sierra Club is less like the bond between any two members of a gardening club and more like the bond between any two Red Sox fans (or perhaps any two devoted Honda owners): they root for the same team and they share some of the same interests, but they are unaware of each other’s existence. Their ties, in short, are to common symbols, common leaders, and perhaps common ideals, but not to one another. The theory of social capital argues that associational membership should, for example, increase social trust, but this prediction is much less straightforward with regard to membership in tertiary associations. From the point of view of social connectedness, the Environmental Defense Fund and a bowling league are just not in the same category.
If the growth of tertiary organizations represents one potential (but probably not real) counterexample to my thesis, a second countertrend is represented by the growing prominence of nonprofit organizations, especially nonprofit service agencies. This so-called third sector includes everything from Oxfam and the Metropolitan Museum of Art to the Ford Foundation and the Mayo Clinic. In other words, although most secondary associations are nonprofits, most nonprofit agencies are not secondary associations. To identify trends in the size of the nonprofit sector with trends in social connectedness would be another fundamental conceptual mistake.7
A third potential countertrend is much more relevant to an assessment of social capital and civic engagement. Some able researchers have argued that the last few decades have witnessed a rapid expansion in “support groups” of various sorts. Robert Wuthnow reports that fully 40 percent of all Americans claim to be “currently involved in [a] small group that meets regularly and provides support or caring for those who participate in it.”8 Many of these groups are religiously affiliated, but many others are not. For example, nearly 5 percent of Wuthnow’s national sample claim to participate regularly in a “self-help” group, such as Alcoholics Anonymous, and nearly as many say they belong to book-discussion groups and hobby clubs.
The groups described by Wuthnow’s respondents unquestionably represent an important form of social capital, and they need to be accounted for in any serious reckoning of trends in social connectedness. On the other hand, they do not typically play the same role as traditional civic associations. As Wuthnow emphasizes,
Small groups may not be fostering community as effectively as many of their proponents would like. Some small groups merely provide occasions for individuals to focus on themselves in the presence of others. The social contract binding members together asserts only the weakest of obligations. Come if you have time. Talk if you feel like it. Respect everyone’s opinion. Never criticize. Leave quietly if you become dissatisfied. . . . We can imagine that [these small groups] really substitute for families, neighborhoods, and broader community attachments that may demand lifelong commitments, when, in fact, they do not.9
All three of these potential countertrends–tertiary organizations, nonprofit organizations, and support groups–need somehow to be weighed against the erosion of conventional civic organizations. One way of doing so is to consult the General Social Survey.
Within all educational categories, total associational membership declined significantly between 1967 and 1993. Among the college-educated, the average number of group memberships per person fell from 2.8 to 2.0 (a 26-percent decline); among high-school graduates, the number fell from 1.8 to 1.2 (32 percent); and among those with fewer than 12 years of education, the number fell from 1.4 to 1.1 (25 percent). In other words, at all educational (and hence social) levels of American society, and counting all sorts of group memberships, the average number of associational memberships has fallen by about a fourth over the last quarter-century. Without controls for educational levels, the trend is not nearly so clear, but the central point is this: more Americans than ever before are in social circumstances that foster associational involvement (higher education, middle age, and so on), but nevertheless aggregate associational membership appears to be stagnant or declining.
Broken down by type of group, the downward trend is most marked for church-related groups, for labor unions, for fraternal and veterans’ organizations, and for school-service groups. Conversely, membership in professional associations has risen over these years, although less than might have been predicted, given sharply rising educational and occupational levels. Essentially the same trends are evident for both men and women in the sample. In short, the available survey evidence confirms our earlier conclusion: American social capital in the form of civic associations has significantly eroded over the last generation.
Good Neighborliness and Social Trust
I noted earlier that most readily available quantitative evidence on trends in social connectedness involves formal settings, such as the voting booth, the union hall, or the PTA. One glaring exception is so widely discussed as to require little comment here: the most fundamental form of social capital is the family, and the massive evidence of the loosening of bonds within the family (both extended and nuclear) is well known. This trend, of course, is quite consistent with–and may help to explain–our theme of social decapitalization.
A second aspect of informal social capital on which we happen to have reasonably reliable time-series data involves neighborliness. In each General Social Survey since 1974 respondents have been asked, “How often do you spend a social evening with a neighbor?” The proportion of Americans who socialize with their neighbors more than once a year has slowly but steadily declined over the last two decades, from 72 percent in 1974 to 61 percent in 1993. (On the other hand, socializing with “friends who do not live in your neighborhood” appears to be on the increase, a trend that may reflect the growth of workplace-based social connections.)
Americans are also less trusting. The proportion of Americans saying that most people can be trusted fell by more than a third between 1960, when 58 percent chose that alternative, and 1993, when only 37 percent did. The same trend is apparent in all educational groups; indeed, because social trust is also correlated with education and because educational levels have risen sharply, the overall decrease in social trust is even more apparent if we control for education.
Our discussion of trends in social connectedness and civic engagement has tacitly assumed that all the forms of social capital that we have discussed are themselves coherently correlated across individuals. This is in fact true. Members of associations are much more likely than nonmembers to participate in politics, to spend time with neighbors, to express social trust, and so on.
The close correlation between social trust and associational membership is true not only across time and across individuals, but also across countries. Evidence from the 1991 World Values Survey demonstrates the following:10
- Across the 35 countries in this survey, social trust and civic engagement are strongly correlated; the greater the density of associational membership in a society, the more trusting its citizens. Trust and engagement are two facets of the same underlying factor–social capital.
- America still ranks relatively high by cross-national standards on both these dimensions of social capital. Even in the 1990s, after several decades’ erosion, Americans are more trusting and more engaged than people in most other countries of the world.
- The trends of the past quarter-century, however, have apparently moved the United States significantly lower in the international rankings of social capital. The recent deterioration in American social capital has been sufficiently great that (if no other country changed its position in the meantime) another quarter-century of change at the same rate would bring the United States, roughly speaking, to the midpoint among all these countries, roughly equivalent to South Korea, Belgium, or Estonia today. Two generations’ decline at the same rate would leave the United States at the level of today’s Chile, Portugal, and Slovenia.
Why Is U.S. Social Capital Eroding?
As we have seen, something has happened in America in the last two or three decades to diminish civic engagement and social connectedness. What could that “something” be? Here are several possible explanations, along with some initial evidence on each.
The movement of women into the labor force. Over these same two or three decades, many millions of American women have moved out of the home into paid employment. This is the primary, though not the sole, reason why the weekly working hours of the average American have increased significantly during these years. It seems highly plausible that this social revolution should have reduced the time and energy available for building social capital. For certain organizations, such as the PTA, the League of Women Voters, the Federation of Women’s Clubs, and the Red Cross, this is almost certainly an important part of the story. The sharpest decline in women’s civic participation seems to have come in the 1970s; membership in such “women’s” organizations as these has been virtually halved since the late 1960s. By contrast, most of the decline in participation in men’s organizations occurred about ten years later; the total decline to date has been approximately 25 percent for the typical organization. On the other hand, the survey data imply that the aggregate declines for men are virtually as great as those for women. It is logically possible, of course, that the male declines might represent the knock-on effect of women’s liberation, as dishwashing crowded out the lodge, but time-budget studies suggest that most husbands of working wives have assumed only a minor part of the housework. In short, something besides the women’s revolution seems to lie behind the erosion of social capital.
Mobility: The “re-potting” hypothesis. Numerous studies of organizational involvement have shown that residential stability and such related phenomena as homeownership are clearly associated with greater civic engagement. Mobility, like frequent re-potting of plants, tends to disrupt root systems, and it takes time for an uprooted individual to put down new roots. It seems plausible that the automobile, suburbanization, and the movement to the Sun Belt have reduced the social rootedness of the average American, but one fundamental difficulty with this hypothesis is apparent: the best evidence shows that residential stability and homeownership in America have risen modestly since 1965, and are surely higher now than during the 1950s, when civic engagement and social connectedness by our measures was definitely higher.
Other demographic transformations. A range of additional changes have transformed the American family since the 1960s–fewer marriages, more divorces, fewer children, lower real wages, and so on. Each of these changes might account for some of the slackening of civic engagement, since married, middle-class parents are generally more socially involved than other people. Moreover, the changes in scale that have swept over the American economy in these years–illustrated by the replacement of the corner grocery by the supermarket and now perhaps of the supermarket by electronic shopping at home, or the replacement of community-based enterprises by outposts of distant multinational firms–may perhaps have undermined the material and even physical basis for civic engagement.
The technological transformation of leisure. There is reason to believe that deep-seated technological trends are radically “privatizing” or “individualizing” our use of leisure time and thus disrupting many opportunities for social-capital formation. The most obvious and probably the most powerful instrument of this revolution is television. Time-budget studies in the 1960s showed that the growth in time spent watching television dwarfed all other changes in the way Americans passed their days and nights. Television has made our communities (or, rather, what we experience as our communities) wider and shallower. In the language of economics, electronic technology enables individual tastes to be satisfied more fully, but at the cost of the positive social externalities associated with more primitive forms of entertainment. The same logic applies to the replacement of vaudeville by the movies and now of movies by the VCR. The new “virtual reality” helmets that we will soon don to be entertained in total isolation are merely the latest extension of this trend. Is technology thus driving a wedge between our individual interests and our collective interests? It is a question that seems worth exploring more systematically.
What Is to Be Done?
The last refuge of a social-scientific scoundrel is to call for more research. Nevertheless, I cannot forbear from suggesting some further lines of inquiry.
- We must sort out the dimensions of social capital, which clearly is not a unidimensional concept, despite language (even in this essay) that implies the contrary. What types of organizations and networks most effectively embody–or generate–social capital, in the sense of mutual reciprocity, the resolution of dilemmas of collective action, and the broadening of social identities? In this essay I have emphasized the density of associational life. In earlier work I stressed the structure of networks, arguing that “horizontal” ties represented more productive social capital than vertical ties.11
- Another set of important issues involves macrosociological crosscurrents that might intersect with the trends described here. What will be the impact, for example, of electronic networks on social capital? My hunch is that meeting in an electronic forum is not the equivalent of meeting in a bowling alley–or even in a saloon–but hard empirical research is needed. What about the development of social capital in the workplace? Is it growing in counterpoint to the decline of civic engagement, reflecting some social analogue of the first law of thermodynamics–social capital is neither created nor destroyed, merely redistributed? Or do the trends described in this essay represent a deadweight loss?
- A rounded assessment of changes in American social capital over the last quarter-century needs to count the costs as well as the benefits of community engagement. We must not romanticize small-town, middle-class civic life in the America of the 1950s. In addition to the deleterious trends emphasized in this essay, recent decades have witnessed a substantial decline in intolerance and probably also in overt discrimination, and those beneficent trends may be related in complex ways to the erosion of traditional social capital. Moreover, a balanced accounting of the social-capital books would need to reconcile the insights of this approach with the undoubted insights offered by Mancur Olson and others who stress that closely knit social, economic, and political organizations are prone to inefficient cartelization and to what political economists term “rent seeking” and ordinary men and women call corruption.12
- Finally, and perhaps most urgently, we need to explore creatively how public policy impinges on (or might impinge on) social-capital formation. In some well-known instances, public policy has destroyed highly effective social networks and norms. American slum-clearance policy of the 1950s and 1960s, for example, renovated physical capital, but at a very high cost to existing social capital. The consolidation of country post offices and small school districts has promised administrative and financial efficiencies, but full-cost accounting for the effects of these policies on social capital might produce a more negative verdict. On the other hand, such past initiatives as the county agricultural-agent system, community colleges, and tax deductions for charitable contributions illustrate that government can encourage social-capital formation. Even a recent proposal in San Luis Obispo, California, to require that all new houses have front porches illustrates the power of government to influence where and how networks are formed.
The concept of “civil society” has played a central role in the recent global debate about the preconditions for democracy and democratization. In the newer democracies this phrase has properly focused attention on the need to foster a vibrant civic life in soils traditionally inhospitable to self-government. In the established democracies, ironically, growing numbers of citizens are questioning the effectiveness of their public institutions at the very moment when liberal democracy has swept the battlefield, both ideologically and geopolitically. In America, at least, there is reason to suspect that this democratic disarray may be linked to a broad and continuing erosion of civic engagement that began a quarter-century ago. High on our scholarly agenda should be the question of whether a comparable erosion of social capital may be under way in other advanced democracies, perhaps in different institutional and behavioral guises. High on America’s agenda should be the question of how to reverse these adverse trends in social connectedness, thus restoring civic engagement and civic trust.
