Journal of Democracy 7.3 (1996) 38-52
The "civil society argument," as Michael Walzer calls it, is actually a complex set of arguments, not all of which are congruent. 1 In the rough pastiche that has become the commonly accepted version, a "dense network of civil associations" is said to promote the stability and effectiveness of the democratic polity through both the effects of association on citizens' "habits of the heart" and the ability of associations to mobilize citizens on behalf of public causes. Emergent civil societies in Latin America and Eastern Europe are credited with effective resistance to authoritarian regimes, democratizing society from below while pressuring authoritarians for change. Thus civil society, understood as the realm of private voluntary association, from neighborhood committees to interest groups to philanthropic enterprises of all sorts, has come to be seen as an essential ingredient in both democratization and the health of established democracies.
Thus summarized, the argument leaves many questions unanswered. Some of these are definitional, arising from the different ways in which civil society has been applied in various times and places. Does it, for instance, include business ("the market") as well as voluntary organizations, or does the market constitute a separate, "private" sphere? If we exclude the market, should we nevertheless include economic associations--trade groups, professional organizations, labor unions, and the like? What about political organizations? Does it make sense, following Antonio Gramsci, to distinguish "civil" from "political" society? If so, [End Page 38] how are we to distinguish between political associations per se and the political activities of groups in civil society, from interest groups to religious bodies, which are intermittently mobilized in pursuit of political goals? 2 Just when does the "civil" become the "political"?
Beyond such definitional concerns, there is also the elusive character of the relationship between "civil society" and democratic governance. Just how is it that associations formed among individuals produce the large-scale political and social benefits postulated by the civil society argument? Is the cultivation of "habits of the heart" that encourage tolerance, cooperation, and civic engagement the key? If so, under which circumstances and forms of small-scale interaction are these effects likely to appear? If, as some hold, civil society's chief virtue is its ability to act as an organized counterweight to the state, to what extent can this happen without the help of political parties and expressly political movements? Finally, what prevents civil society from splitting into warring factions (a possibility that theorists since Hegel have worried about) or degenerating into a congeries of rent-seeking "special interests"? What is it about civil society, in other words, that produces the benevolent effects posited by the civil society argument?
In attempting to answer these questions, it might be useful to make a rough distinction between two broad versions of the "civil society argument." The first version is crystallized in Alexis de Tocqueville's Democracy in America, with important antecedents in the work of the eighteenth-century "Scottish moralists," including Adam Smith, Adam Ferguson, and Francis Hutcheson. This approach puts special emphasis on the ability of associational life in general and the habits of association in particular to foster patterns of civility in the actions of citizens in a democratic polity. We shall call this family of arguments "Civil Society I." The second version, articulated most forcefully by Jacek Kuron, Adam Michnik, and their associates in formulating a strategy for resistance to Poland's communist regime in the 1980s, is also evident in recent literature on processes of "redemocratization" in Latin America. This argument, which we call "Civil Society II," lays special emphasis on civil society as a sphere of action that is independent of the state and that is capable--precisely for this reason--of energizing resistance to a tyrannical regime.
It might already be apparent that there is a degree of contradiction between "Civil Society I" and "Civil Society II," for while the former postulates the positive effects of association for governance (albeit democratic governance), the latter emphasizes the importance of civil association as a counterweight to the state. There is no reason in principle why the "counterweight" of civil society should not become a burden to a democratic as well as an authoritarian state. Indeed, there are those, such as the economist Mancur Olson, who perceive the "dense webs of association" praised by the civil society argument as enduring [End Page 39] threats to the smooth and equitable functioning of modern states and markets alike. 3
Perhaps the most persuasive recent version of "Civil Society I" is that of Robert D. Putnam, formulated in part as an empirically grounded answer to the concerns raised above. Putnam's argument gained particular currency in his January 1995 Journal of Democracy essay, "Bowling Alone." It is more richly elaborated, however, in his study of regional governments in modern Italy, entitled Making Democracy Work. 4 In this book, Putnam attributes the superior effectiveness of northern Italy's regional governments to the dense "networks of civic engagement" fostered by "civil associations" of all kinds. "The denser such networks in a community, the more likely that its citizens will be able to cooperate for mutual benefit," he writes. He continues, with explicit reference to Olson's arguments: "social capital, as embodied in horizontal networks of civic engagement, bolsters the performance of the polity and the economy, rather than the reverse: Strong society, strong economy; strong society, strong state." 5
Putnam's analysis in "Bowling Alone" examines trends in the United States over the last 30 years and argues the obverse: weakening society, weakening economy; weakening society, weakening state. According to Putnam, the basis of "civil community" has been eroding in the United States since the 1960s. We have been depleting our national reserves of social capital, social trust, and generalized reciprocity, he claims, and undercutting our capacity for mutually beneficial collective action. This trend is most evident in the decline of "traditional secondary associations" like the Boy Scouts, parent-teacher associations (PTAs), the League of Women Voters, and even weekly bowling leagues. 6
Putnam's argument is stimulating, and his recourse to empirical evidence is refreshing. Nevertheless, his concerns about the health of U.S. civil society and his account of Italy's experiment in regional government point to significant weaknesses in both forms of the civil society argument. These weaknesses raise important questions that must be examined empirically as well as theoretically before we can properly understand the role of civil society in democracy and democratization.
Our argument is threefold: First, both Putnam's assessment of the state of "civil community" in the United States and his account of regional government in northern Italy underestimate the ability of newer organizations, and of specifically political associations such as social movements and political parties, to foster aspects of civil community and to advance democracy. Second, talk about "networks of civic engagement" glosses over the real, and often sharp, conflicts among groups in civil society. These conflicts, in the absence of specifically political settlements, may spill over into civil disruption and violence. Third, and most important, to understand any polity we must look first at the political settlements that ground it, and to the effects that such [End Page 40] settlements have on social forces and civil society. Taken together, our objections suggest the problematic character of both Putnam's definition of civil society and the larger civil society argument itself.
The broader civil society argument ascribes a variety of functions to civil associations. Putnam's discussion focuses on a narrow, though seemingly powerful, segment of them. For Putnam, the chief virtue of civil associations lies in their capacity to socialize participants into the "norms of generalized reciprocity" and "trust" that are essential components of the "social capital" needed for effective cooperation. Civil associations provide the "networks of civic engagement" within which reciprocity is learned and enforced, trust is generated, and communication and patterns of collective action are facilitated. These are horizontal networks, as opposed to the vertical networks of patron-client arrangements or of traditional hierarchical organizations such as the Catholic Church. The broader their reach, the more effective they are:
Dense but segregated horizontal networks sustain cooperation within each group, but networks of civic engagement that cut across social cleavages nourish wider cooperation. . . . If horizontal networks of civic engagement help participants solve dilemmas of collective action, then the more horizontally structured an organization, the more it should foster institutional success in the broader community. Membership in horizontally ordered groups (like sports clubs, cooperatives, mutual aid societies, cultural associations, and voluntary unions) should be positively associated with good government. 7
That secondary associations generate "social capital" of the sort highlighted by Putnam is scarcely questionable. Whether this "social capital" is truly a "public good," available to society at large and capable of producing the effects ascribed to it, is another matter. 8 Putnam's caveat is important: "Dense but segregated networks" may or may not contribute to effective democratic governance; at times they become the basis for civil strife. Thus the networks that associations create should "cut across social cleavages" in order to nourish wider cooperation. 9
Putnam's preoccupation here is a familiar one. In order to foster a genuine spirit of "wider cooperation," his argument suggests, such associations must not be "polarized" or "politicized." They must "bridge" social and political divisions and thus, presumably, be autonomous from political forces. These caveats echo a long tradition of "pluralist" analysis. Yet how can such associations shape political participation and "civic engagement" without engaging in specifically political issues and without representing compelling social interests? [End Page 41]
Putnam's formulation is all the more puzzling in the light of his findings, which place the Emilia-Romagna region in the heart of Italy's zone of "civic engagement." Putnam fails to note that most of this region's sports clubs, choral societies, cooperatives, and cultural associations had been organized by and for two major political parties, the Communists and the Christian Democrats. Indeed, one observer argues, "If a contemporary Tocqueville searched for autonomous groups on which to write on `democracy in Italy,' he would produce a thin volume indeed. . . . Parties usurp space that in other advanced industrialized countries is held by bureaucracies and by local grass-roots organizations. It is hardly an exaggeration to say that they pervade all aspects of political, economic, and social life in Italy." 10 If Putnam's "civil community" can coexist with a seemingly politicized civil society, what are we to make of the argument about the necessity for "inclusive" associations that cut across social and political cleavages?
Such questions affect not only Putnam's analysis, but also the work of those emphasizing the capacity of civil society to resist state repression, or what we call "Civil Society II." A commonplace of this version of the argument is that civil associations must be "autonomous." The argument is aired not just in the scholarly literature, but in political debates over the character and trajectory of civil society in Latin America and Eastern Europe. As Aleksander Smolar put it in a recent Journal of Democracy article, the "society-first" approach of the East European dissidents was "antipolitical." 11 Putnam's evidence, if not his argument, suggests that political autonomy is less important than the fact of association itself. 12 Indeed, Tocqueville himself identified specifically political associations as the key to the rich associational life that he celebrated in the United States of 1832.
Ironically, both "Civil Society I," which emphasizes the political benefits of an apolitical civil society, and "Civil Society II," which focuses on politically mobilized social actors outside customary political associations, tend to marginalize specifically political associations, especially parties. Convergence on this point, however, does not resolve the definitional imprecision raised at the outset. Indeed, the two versions of the argument disagree about the character of the very civil society they juxtapose to political society. The imprecision stems, in part, from the sheer sweep of the argument itself. Civil society has been credited with enhancing democracy or curbing authoritarianism in contexts as varied as Jacksonian America, postWorld War II Italy, Eastern Europe at the end of the Soviet empire, and Latin America under the generals. At times the concept seems to take on the property of a gas, expanding or contracting to fit the analytic space afforded it by each historical or sociopolitical setting.
Both "Civil Society I" and "Civil Society II" reflect the particular contexts to which they have been applied. Proponents of "Civil Society [End Page 42] II" wish to include groups that enable citizens to mobilize against tyranny and counter state power. In doing so, they rightly emphasize the conflictual potential of civil society. They also tend to emphasize new forms of association, because political and traditional associations are often tainted by cooperation with the regime. Proponents of "Civil Society I" likewise define "civil society" in ways that fit their particular context. Thus for Putnam, civil society includes chiefly groups whose activities generate the desired "networks, norms, and trust" at the heart of "social capital." In contrast to "Civil Society II," he emphasizes traditional secondary associations, organizations not noted for their conflictual character or political weight. The two versions of the argument tend to produce two different portfolios of qualified organizations. But if the argument is not to be circular, we must do better than that. The civil society argument hinges on the virtues of association and of organized society per se; it cannot confront the conflictual potential thereof by definitional sleight of hand, but neither can it ignore the conflict at the heart of the modern "organizational society."
Putnam's neglect of political associations is motivated by both empirical and theoretical considerations. On empirical grounds he rejects political parties as significant actors in the transformations he notes in northern Italy because the same parties produce strikingly different results in the dissimilar social contexts of northern and southern Italy. He also rejects American social movements and nonprofit organizations (whose numbers and membership have grown tremendously over the last 30 years) as significant counterweights to the trends he observes in the United States, but with less careful attention to the evidence. 13 His negative assessment of social movements is based on the grounds that membership in groups like the National Organization for Women (NOW) and the Sierra Club is mainly a matter of financial support and scarcely evidence of "civic engagement" in the concrete sense intended by his argument.
Putnam's rather cursory examination of social movements in the United States overlooks both the multifaceted character of many national groups and, more importantly, the grassroots bases of the vast majority of social-movement groups operating nationally. Putnam correctly notes that membership in highly professionalized social-movement groups most often consists of little more than writing a check and perusing the group's magazine. Yet closer examination of the Sierra Club, for example, would have shown that this is not always the case. State chapters and especially local chapters of the Sierra Club routinely sponsor community-service projects like urban river clean-ups and "environmentally aware" outdoor activities broadly similar to those sponsored [End Page 43] by the Boy Scouts, a group close to the heart of Putnam's "civic community."
Furthermore, large groups with a national media profile are simply the most visible promontories in the broader landscape of contemporary social movements. Grassroots groups form the vast majority of social-movement organizations nationally and, compared to Washington-based groups, are far more engaged in the communal life and civic networks of local communities. For example, a recent directory of national environmental organizations listed 645 groups, while the Citizen's Clearinghouse for Hazardous Wastes, a national clearinghouse for grassroots environmental groups, claims to be in contact with between seven thousand and nine thousand groups nationwide. 14 The demography of the peace movement evinces the same pattern. The 1987 edition of the Grassroots Peace Directory listed 7,700 groups working for peace, but less than 300 claimed to be national in scope and only about half of those were located in Washington. To assess the capacity of the women's movement to promote "civic engagement" solely on an examination of NOW paints a refracted portrait of the efforts of the thousands of rape-crisis centers, battered-women's shelters, women's health clinics, bookstores, theater groups, women's studies programs, credit unions, reading circles, clinic-defense teams, and other "nontraditional" women's groups in localities across the United States. 15 The social ties and trust among members of community-based movement organizations should not be equated with those typical of "mass-mediated" membership organizations. 16
These observations would come as no surprise to Tocqueville, who saw specifically political associations as the "great free schools" of democracy in the United States of the 1830s. Tocqueville regarded political associations as dangerous to any regime, including an open one. Yet despite such reservations, he ends by endorsing free political associations on the grounds that only such freedom can prevent "either despotism of parties or the arbitrary rule of a prince," and that political association is in practice the mother of civil association, not the other way around: "So one may think of political associations as great free schools to which all citizens come to be taught the general theory of association." Where political associations are forbidden, Tocqueville continues, "civil associations will always be few, feebly conceived, and unskillfully managed and either will never form any vast designs or will fail in the execution of them." 17
Theoretically, the short shrift that Putnam gives to political associations and social movements seems to stem from concerns, similar to those raised by Tocqueville, that are rarely addressed by the proponents of "Civil Society II." For Putnam correctly perceives that not all organized groups contribute positively to "effective governance." In this respect, he echoes the fears not only of the pluralists but of Tocqueville [End Page 44] as well, who treated "political associations" as a different species and wrote, "One must not shut one's eyes to the fact that unlimited freedom of association for political ends is, of all forms of liberty, the last that a nation can sustain. While it may not actually lead it into anarchy, it does constantly bring it to the verge thereof"--a passage that Tocqueville pointedly quotes in his later discussion of the relations between political and civil associations.
The doubts that Tocqueville raised were grave ones, and were soon enough confirmed by the outbreak of the Civil War in 1861. In full-blown political associations, Tocqueville worried, dissidents can league together to "form something like a separate nation within the nation and a government within the government." Were political passions to be raised to such a pitch, could it then be supposed "that in the long run such a group will just talk and not act?" 18 In such circumstances, the "dense networks of civil engagement" generated by associations of all kinds would provide apt recruiting grounds for rival bands bent on destroying their opponents, a probability ably documented by contemporary students of the recruitment strategies of revolutionary movements.
Against such a possibility Putnam erects a twofold defense: First, he downplays or rejects the role of specifically political associations and movements in his portrait of the "civil community." Second, in describing civil society itself, he adopts the pluralist restriction that only broad, horizontally structured groups capable of "cutting across" salient social cleavages are likely to achieve the effects more generally attributed to civil society. According to the pluralist argument, if citizens belong to a variety of organizations whose memberships "cut across" rather than track "salient social cleavages," such associations will weaken those cleavages and thereby make society more governable. But this is equivalent to arguing that "the weaker civil society, the stronger the state," which is clearly not Putnam's point. Nor is it the civic utopia envisioned by proponents of "Civil Society II," for whom the precise function of civil society is to serve as a counterweight to the state.
Proponents of the civil society argument cannot have it both ways. If civil society is to be "strong," it must be strong in defense of citizens' interests, whether those spring from "salient social cleavages" or mere personal taste. 19 Within the broader civil society argument, as we have seen, there seem to be two centers of gravity concerning civil society's relationship to the state and influence upon it. In both cases proponents of a strong civil sector emphasize its autonomy from partisan or electoral politics, though such autonomy is said to be crucial for opposite reasons. The version of the argument that we have called "Civil Society II" focuses on civil society as a counterweight to the state. In [End Page 45] contexts of democratic transition, in particular, where established political parties have been repressed, weakened, or used as tools by the authoritarian state, autonomy from traditional politics seems to be a prerequisite for oppositional advocacy. In such contexts, civil society is treated as an autonomous sphere of social power within which citizens can pressure authoritarians for change, protect themselves from tyranny, and democratize from below.
If civil society is a beachhead secure enough to be of use in thwarting tyrannical regimes, what prevents it from being used to undermine democratic governments? To the extent that civil associations are strong, they challenge governing institutions to meet particular needs, aspirations, and conceptions of the common good. Established interests may lock up social resources and block society's ability to meet the demands of the dispossessed (as in the southern United States up to the advent of the civil rights movement in the 1960s); social blocs may form, each with its own panoply of associations, to battle one another for control of the state (as in contemporary Lebanon or the "plural states" of Western Europe before the political settlements of the nineteenth century); political forces may forge powerful ties with community organizations and civil associations, polarizing society and at times threatening the "order" that incumbents so cherish (as in postWorld War II Italy and contemporary El Salvador). To understand the role of civil society in the modern world, we must discern how and under what circumstances a society's organized components contribute to political strength or political failure.
Like the analysts of the "Civil Society II" camp, however, Robert Putnam wants to make generalizations about civil society as a whole and under any circumstances. In his analyses of Italy and the United States, Putnam argues that civil society augments, rather than offsets, the state's capacity to govern. Consistent with the "Civil Society I" argument, Putnam focuses attention on the cultural and organizational benefits that citizen participation in civil associations holds for a democratic state. Implicit in his account, however, is the fear that if such associations follow too closely the pattern of divisive political solidarities, they may well sharpen social cleavages and actually undermine the capacity for effective governance. Consequently, he appears reluctant to count among his "civil associations" any that advance a cause, pursue policy change as their central vocation, or provoke conflict. In so doing Putnam, too, seems to want it both ways. He clearly wants an activated and engaged populace, and he argues that the socialization performed by civil associations is vital to the creation of such an engaged citizenry. Yet in the end only those associations qualify that invoke a civic transcendency whose spirit claims to "rise above" the divisiveness of protracted sociopolitical and cultural conflict. This solution to the dilemma threatens to render the argument circular, but it is troubling in more [End Page 46] than theory, for it seems to fly in the face of Putnam's own evidence and of the experience of Western democracies with social movements. Indeed, the arguments presented by Sidney Tarrow and by Dietrich Rueschemeyer, Evelyne Huber Stephens, and John D. Stephens make a forceful case for viewing social movements--that is to say, the organized mobilization of groups along social cleavages--as the central bearers of democratizing pressures within Western democracies. 20
Our purpose here is not to repeat those arguments. Instead, we maintain that the civil society argument as it is commonly presented is partial at best and seriously misleading at worst. In many respects, it presupposes precisely the sort of political peace that it imagines civil society providing. Where emphasis is placed on the ability of civil society to oppose a tyrannical state, its ability to oppose a democratic one is either ignored outright or countered with qualifications that themselves undermine the power of the civil society argument gener-ally. 21 When emphasis is placed on the formation of "habits of the heart" conducive to cooperation and collective action, as it is in Robert Putnam's argument, the mechanisms by which such "microsocial" effects translate into "macropolitical" outcomes are weakly specified or contradictory or both. 22
What is missing in both cases is the political variable. At a minimum, this "political variable" must include both the political associations that play important roles in any society and the work of political compromise, restraint, and accommodation necessary for reconciling competing interests in a peaceful and more or less orderly way. More generally, the political variable includes the prevailing "political settlement" that governs who plays, the rules of the game, and acceptable outcomes. Numerous examples from recent history show that such settlements are the work of political parties and of the best-financed, and often best-armed, elements of civil society.
The arrangements forged in such political settlements do not come easily, and when they do come they may well represent a betrayal of the trust that civil society placed in politicians and political parties, and in a deliberate or circumstantial dampening of social demands and expectations, as many of the recent experiments in "democratization" demonstrate. Here social movements may play a crucial role, and not only in "new" democracies, by taking up neglected or repressed demands and pushing the political system to engage forgotten or marginalized sectors and issues. [End Page 47]
That role, of course, is by no means unambiguous, any more than is that of the rest of civil society. 23 Social movements may represent an armed and paranoiac vision of civic responsibility (the militia movement in the United States is an example) or a retreat from social responsibility beyond the immediate group (as in the case of certain religious movements). Their efforts may start or break down in violence as frustration mounts in the face of an unresponsive political system or a repressive state. Yet they also build trust and habits of cooperation and civic action among their members. Where the political system is even minimally responsive, they can boost the vitality of civil and political society by mobilizing people and stimulating debate. In short, decidedly political associations may well play the roles attributed to civil associations in the civil society argument, and may play them better. 24
What role organized groups in civil society will play, we would argue, depends crucially on the larger political setting. As Michael Walzer puts it, "there is no escape from power and coercion, no possibility of choosing, like the old anarchists, civil society alone." What Walzer calls "the paradox of the civil society argument" is that a democratic civil society seems to require a democratic state, and a strong civil society seems to require a strong and responsive state. The strength and responsiveness of a democracy may depend upon the character of its civil society, as Putnam argues, reinforcing both the democratic functioning and the strength of the state. But such effects depend on the prior achievement of both democracy and a strong state. 25
Where the state is unresponsive, its institutions are undemocratic, or its democracy is ill designed to recognize and respond to citizen demands, the character of collective action will be decidedly different than under a strong and democratic system. Citizens will find their efforts to organize for civil ends frustrated by state policy--at some times actively repressed, at others simply ignored. Increasingly aggressive forms of civil association will spring up, and more and more ordinary citizens will be driven into either active militancy against the state or self-protective apathy. The breakdown of the tutelary democracies and authoritarian states of Latin America in the 1970s and 1980s attests to what more than one observer has euphemistically called "the dangers of excluding reformists from power." In such settings, all of civil life may become polarized, as Samuel P. Huntington pointed out long ago (though the solutions he advocated proved elusive); and even Putnam's choral societies and bowling leagues--even nuns and bishops!--may become "subversive." 26
These observations do not close the question. If we are to retrieve what is most useful from the civil society argument, we must pursue [End Page 48] nagging empirical questions about the ways in which social power is constituted, distributed, and managed in contemporary societies. One avenue, suggested by the doubts that we have uncovered at the center of the civil society argument, would be to inquire more deeply into the sorts of associational life likely to produce the "social capital" on which Putnam has put so much weight. Sweeping, mutually exclusive ideal types like "political society" and "economic society," the "public," "private," and "nonprofit" sectors, and even "civil society" itself are unlikely to capture the range of associational forms that prepare citizens to engage in collective action for mutual benefit. The role of less formal but more numerous groups, moreover, needs to be taken into account in any such analysis. 27 We are likely to find that social-movement organizations, grassroots interest groups, and grassroots political associations of all sorts are far more likely to generate Putnam's activated citizenry than the choral societies, birdwatching clubs, and bowling leagues he is so fond of citing.
A second avenue of inquiry might start with the demography of "social capital" production in different societies, seeking clues about the ways in which the larger political context shapes the relations between civil society and the state. If the argument developed here is correct, the key to the success or failure of democratic institutions will lie not in the character of civil society but in their responsiveness as institutions--in their ability to mediate conflict by hearing, channeling, and mediating the multiple citizen demands that modern societies express through civil and political associations alike.
Finally, the civil society argument raises intriguing and pressing questions about the nature and political import of the relations that associations of all sorts might forge with the state and with one another. What is the sense, for instance, and what is the nonsense in the frequent demand that civil associations be "nonpartisan"? How might civil society act to advance citizens' interests in the absence of effective political representation, and what are the limits to such action? Is there a viable distinction between the multiple "special interests" of civil society and the "public interest"? If there is, how and by whom shall this public interest be safeguarded? Where civil associations are closely tied to parties, what sorts of relations between them are conducive to greater internal democracy and fuller representation of citizens' interests? Is greater intraparty democracy (and therefore openness to "civil society") correlated with democratic governance, or is there no very clear relationship between the two? And if not, just how might vigorous representation of the interests and demands of civil society be reconciled with the norms of compromise and accommodation that govern political settlements? These are classic questions of political philosophy, but they are also empirical questions that the civil society argument, however faulty as a general account of democratic governance, forces upon us. [End Page 49] Such questions cannot be answered in the abstract but only through careful empirical work.
Michael W. Foley is associate professor of politics at the Catholic University of America in Washington, D.C. He has previously published research on civil society and the emergence of new social movements in Mexico and El Salvador.
Bob Edwardsis assistant professor of sociology at East Carolina University in Greenville, North Carolina, where he focuses on social movements and social change in the United States. Notes1. Michael Walzer, "The Civil Society Argument," in Chantal Mouffe, ed., Dimensions of Radical Democracy: Pluralism, Citizenship, Community (London: Verso, 1992), 89107.
2. Most important are "religious special-purpose groups," which tend to be relatively small organizations formed to fulfill specialized functions often oriented toward the reform of a larger religious body or society as a whole. Such groups have a long history in the United States, but their numbers and membership seem to have grown tremendously in recent decades. For detailed analysis see Robert Wuthnow, "The Growth of Religious Reform Movements," Annals of the American Academy of Political and Social Science 480 (July 1985): 10616; and The Restructuring of American Religion: Society and Faith Since World War II (Princeton: Princeton University Press, 1986).
3. Mancur Olson, The Rise and Decline of Nations: Growth, Stagflation, and Social Rigidities (New Haven: Yale University Press, 1982). Olson's argument is but the most ambitious of attacks on the "rent-seeking" behavior of our "interest-group society" emanating from the public-choice school of political economy.
4. Robert D. Putnam, "Bowling Alone: America's Declining Social Capital," Journal of Democracy 6 (January 1995): 6578; and Making Democracy Work: Civic Traditions in Modern Italy (Princeton: Princeton University Press, 1993).
5. Putnam, Making Democracy Work, 173 and 176.
6. Putnam elaborates the argument in his 1995 Ithiel de Sola Pool Lecture, where he focuses on explaining "the strange disappearance of social capital in America." See "Tuning In, Tuning Out: The Strange Disappearance of Social Capital in America," PS: Political Science and Politics 28 (December 1995): 66483.
7. Putnam, Making Democracy Work, 16775.
8. There are, in fact, what statistical analysis describes as "ecological correlation problems" with Putnam's data interpretation in some instances (see William S. Robinson, "Ecological Correlations and the Behavior of Individuals," American Sociological Review 15 [1950]: 35157). The co-occurrence of civic associations and effective government in the same region does not demonstrate that the people in the associations are the ones making the government work. To address that question, one must systematically introduce a mediating level of analysis that can demonstrate the actual avenues of influence among civically engaged people, civic associations, and institutional performance. Were the civic organizations party-organized, however, it would be easier to demonstrate the links among party membership, associating in party civic associations, parties, interest articulation or aggregation, and institutional performance. In fact, parties apparently play an important role in northern Italy's civic life.
9. As Putnam puts it in a more recent article, "To the extent that the norms, networks, and trust link substantial sectors of the community and span underlying social cleavages--to the extent that the social capital is of a `bridging' sort--then the enhanced cooperation is likely to serve broader interests and be widely welcomed." See "Tuning In, Tuning Out," 665.
10. Ted Perlmutter, "Italy: Why No Voluntary Sector?" in Robert Wuthnow, ed., Between States and Markets: The Voluntary Sector in Comparative Perspective (Princeton: Princeton University Press, 1991), 157.
11. Aleksander Smolar, "From Opposition to Atomization," Journal of Democracy 7 (January 1996): 26. Václav Havel continues to represent this stance, characterized by a discourse that is primarily moral. See Václav Havel and Václav Klaus, with commentary by Petr Pithart, "Rival Visions," Journal of Democracy 7 (January 1996): 1223.
12. For a striking look at the advantages--and dangers--of political alignment for social organizations, see Cathy Lisa Schneider, Shantytown Protest in Pinochet's Chile (Philadelphia: Temple University Press, 1995). Schneider shows that Chilean Communist Party socialization in shantytowns that it controlled led to significantly higher levels of social organization, community solidarity, and protest than elsewhere, but that the militancy of these towns, and the Party's isolation once it turned to armed struggle, left them broken by repression and without political resources after Chile's limited democratic transition.
13. To our knowledge, the proliferation of social-movement and citizen-advocacy organizations was first discussed by John D. McCarthy and Mayer N. Zald in The Trends of Social Movements in America (Morristown, Pa.: General Learning Press, 1973) and demonstrated empirically in the case of Washington-based groups by David King and Jack L. Walker, Jr., "An Ecology of Interest Groups in America," in Jack L. Walker, Jr., ed., Mobilizing Interest Groups in America: Patrons, Professions and Social Movements (Ann Arbor: University of Michigan Press, 1991), 5775; and for women's and racial- or ethnic- advocacy groups generally by Debra Minkoff, Organizing for Equality: The Evolution of Women's and Race-Ethnic Organizations in America, 19551985 (New Brunswick, N.J.: Rutgers University Press, 1995). For similar accounts of nonprofit groups see P.D. Hall, Inventing the Nonprofit Sector (Baltimore: Johns Hopkins University Press, 1992) and Michael O'Neill, The Third America: The Emergence of the Nonprofit Sector in the United States (San Francisco, Calif.: Jossey-Bass, 1989).
14. Only about one in four of these national groups is based in the Washington, D.C., metropolitan area. See John C. Brainard, ed., The Directory of National Environmental Organizations, 4th ed. (St. Paul, Minn.: Environmental Directories, 1992).
15. Topsfield Foundation, Grassroots Peace Directory (Pomfret, Conn.: no pub., 1987); Brainard, ed., Directory of National Environmental Organizations; Myra Marx Ferree and Patricia Yancy Martin, eds., Feminist Organizations: Harvest of the New Women's Movement (Philadelphia: Temple University Press, 1995).
16. Putnam dismisses thousands of community-based nonprofit service organizations in similar fashion by equating their potential for producing "social capital" with that of the Ford Foundation, the Mayo Clinic, and the Metropolitan Museum of Art. Only if the tremendous growth in the "nonprofit sector" over recent decades were attributable primarily to growth among the large national-level groups specifically mentioned by Putnam would his assessment of the sector as a whole be well founded. His argument appears to be on sounder footing in "Tuning In, Tuning Out," where he appeals to evidence from a variety of national studies, including the General Social Survey, to show that participation in organizations has been declining since the late 1960s. Nevertheless, the use of participation in general as a stand-in for "social capital" makes even this data questionable, as Putnam himself argues that "who benefits from these connections, norms, and trust--the individual, the wider community, or some faction within the community--must be determined empirically, not definitionally" (p. 665)--essentially the point we make below.
17. Alexis de Tocqueville, Democracy in America (New York: Doubleday, 1969), 193, 524, and 192. It is curious that Larry Diamond, in his synthesis of ideas on the role of civil society in democratic consolidation, "Rethinking Civil Society: Toward Democratic Consolidation," Journal of Democracy 5 (July 1994): 8, should quote the "free schools" passage as referring to civil, rather than political, associations. Like Putnam, Diamond wishes to exclude political parties from "civil society," and he would dole out similar treatment to civil associations that refuse to be "civil" in their behavior (p. 11).
18. Tocqueville, Democracy in America, 19091.
19. Larry Diamond, more explicitly than Putnam, would simply exclude from "civil society" those groups whose ends or behavior might threaten democratic governance. See Diamond, "Rethinking," 11. Such an expedient makes the civil society argument nicely circular, and it denies us the opportunity to understand how real societies work.
20. Sidney Tarrow, Power in Movement: Social Movements, Collective Action and Politics (New York: Cambridge University Press, 1994); and Dietrich Rueschemeyer, Evelyne Huber Stephens, and John D. Stephens, Capitalist Development and Democracy (Chicago: University of Chicago Press, 1992).
21. Others solve this dilemma by positing conditions that must be met for civil society to work its magic. Groups must be "moderate" and restrained in their demands; they should be democratic themselves or at least support democracy; they should be institutionalized and have a stake in the system; they should not reinforce social cleavages, but cut across them; and so on. The circularity of such condition-setting should be apparent.
22. There appears to be no relation, for example, between the changes in the character of Italian regional politics and politicians that Putnam's research uncovered and the "civicness" of the regions. Rather, Putnam ascribes the increased moderation and efficacy that he found to the new institutions (the regional governments), not to the civic culture in which some of them were embedded. The increased responsiveness of government represented by the new institutions was the key to their success, though Putnam shows that success varied in character and degree from region to region.
23. There are, of course, versions of the civil society argument in which social movements and social activists play a central role. Michael Walzer's article, cited in note 1 above, is one of them. Jean L. Cohen and Andrew Arato, in Civil Society and Political Theory (Cambridge: MIT Press, 1992), present an extensively elaborated version. In neither case, however, do the authors take sufficiently seriously the difficulty of divided or "dangerous" civil societies.
24. There is generally a close correlation between individual participation in any association and broader civic engagement; see Samuel Barnes and Max Kaase, Political Action: Mass Participation in Five Western Democracies (Beverly Hills, Calif.: Sage, 1979). Nonetheless, it seems likely that specifically political associations (whether social-movement organizations, interest groups, or political parties) are more conducive to promoting civic engagement than many other sorts of association. That would certainly be in keeping with Tocqueville's argument. Like many other aspects of the civil society argument, however, it remains an intriguing empirical question.
25. Walzer, "Civil Society Argument," 1023. See also Diamond, "Rethinking," 1516.
26. Samuel P. Huntington, Political Order in Changing Societies (New Haven: Yale University Press, 1968). Huntington's notion of parties as necessary channels for political participation in modern and modernizing societies relied heavily on the ordering and controlling functions of parties. His favorite example, the Leninist communist parties of the time, proved a fickle choice, because they failed precisely to incorporate the responsiveness that makes successful democracies work but which Huntington systemati-cally ignored in his account of the sort of "institutionalization" necessary to achieve political "order."
27. Given the social significance of bowling in the United States as documented in "Bowling Alone," informally organized groups of nonleague bowlers merit some consideration. On recent excursions to AMF East Carolina Bowl one of the authors of the present essay observed dozens of people bowling without benefit of league sponsorship. Contrary to Putnam's "whimsical, yet discomfiting" example, however, no one was bowling alone.
http://muse.jhu.edu/demo/journal_of_democracy/7.3foley.html