Towards a Dynamic Theory of Civil Society: The Politics of Forward and Backward Infiltration

Forward Infiltration: Three Modes

We conceptualize the forward infiltration of civil society into the state and the economy in terms of three modes: the first is to affect state and economic functions through politics of influence (without having personnel or resources enter into either domain); the second is to take over state or economic functions through internal civil society initiatives (also without directly entering either), and the third is to institutionalize movement forces through actors’ active entrance into the state and the economy. We label these three modes of forward infiltration the politics of influence, the politics of substitution, and the politics of occupation, respectively.

The politics of influence encompasses efforts by civil society actors to influence formal political and economic actors through contentious action, such as consciousness-raising, advocacy, and protest, or through lobbying, providing information, and other forms of more cooperative interaction. These forms of contentious politics are a positional project, one in which movements confine their ideals and goals within civil society while passively attempting to affect the direction of the state and economic institutions through the formation of public opinion in dissipated public spheres (Gamson and Wolfsfeld 1993; Habermas 1996; Koopmans 2004). Most contemporary social movements that aim to influence and alter existing rules and policies pursued by formal institutions without altering such institutions belong to this category of infiltration. This form of politics focuses on pressuring incumbent policy-makers and managers of institutions to respond to movements’ claims. Alternatively, movements may attempt to convince and mobilize the general public to adopt their goals and alter their quotidian practices to change the status quo. The majority of “contentious politics” (McAdam et al. 2001) occurring between the state and civil society emerges, flourishes, and wanes as movements achieve their goals. Strikes, demonstrations, and petitions undertaken by old and new social movements rarely go beyond the conventions of the politics of influence, as they aim neither to establish new organizational and institutional forms nor to take over the existing institutions in order to realize their goals.

The concept of the politics of influence found its seminal articulation in Cohen and Arato’s (1992) account of civil society. They worry that when “the new grassroots associations capable of spreading enlightenment lose their rootedness in the lifeworld” (Cohen and Arato 1992:473) they will eventually be co-opted by the intrinsic logics of complex organizations. Instead, Cohen and Arato envision that a politics of influence could “alter the universe of political discourse to accommodate new need-interpretations, new identities, and new norms” (1992:526). Yet through institutionalization, the solidarities generated by movement activities are routinized and absorbed into the existing status quo. Meyer and Tarrow (1998) define such an absorption (a mode of institutionalization) into the existing power bloc as an adaptation or transformation of movement goals so as not to disrupt extant institutionalized procedures. ³ From this perspective, the institutionalization of social movements is indeed another name for defeat, as it means that movement forces have failed to realize their initial goals—an upheaval or fundamental change of the existing status quo. Ultimately, the politics of influence is a political project in which citizens seek to organize and channel their voices through their own collective action rather than through existing political structures.

Antinuclear movements in Western Europe in the 1980s and environmental protests during the last couple of decades in many wealthy and developing democracies are good examples of the politics of influence, in that civil-society forces do not actively attempt to replace state functions or occupy formal institutions. They engage in ideological debate, spontaneous demonstration, and routine lobbying, aiming to challenge the current institutional rationality while limiting their radicalism to the defense of vulnerable, discursively structured lifeworlds (Habermas 1984; Offe 1985; Touraine 1985). Civil society can “have at most indirect effect on the self-transformation of the political system” (Habermas 1996:372). This process is depicted in Figure 1-a(1), in which politics of influence is denoted as the first dashed line (forward) between the state and civil society.

OPEN IN VIEWER

Figure 1.

Two Modes of Infiltration: Three Modes of the State-Civil Society Boundary-Making ⁴

The politics of influence also encompasses a wide range of civil society-economy interactions, as depicted in Figure 1-a(2). Indeed, one classic example of the politics of influence—the environmental movement—was part of an effort, still ongoing, to compel corporate actors and other economic organizations to incorporate the externalities of their economic activities so as to protect the natural preconditions for their production. More recently, civil society actors increasingly target and attempt to influence economic actors directly, such as through anti-sweatshop campaigns, environmental campaigns, and animal welfare campaigns (McDonnell et al. 2015). As Colin Crouch observes the rise of multinational corporations contributes to “a considerable enhancement of current trends towards a displacement of political activity from parties to civil society organizations and social movements” that seek to directly influence the policies of corporate actors (Crouch 2009:397). The politics of influence most preserves the organizational identity of actors in both civil society and the economy, which will enter into only temporary, issue-specific alliances with each other. For theorists who prize differentiation, this is a positive feature of the politics of influence. As a self-limiting politics, it prevents the takeover of civil society by either the state or economic actors. Yet this comes at a theoretical and political cost: It leaves us unable to conceptualize the dynamics through which civil society actors may ally with hybrid economic-civil society actors, like trade unions, to pursue more ambitious democratizations of the state and the economy.

As a domain-preserving orientation, the politics of influence provides only partial analytical tools to understand the diverse boundary-reformulating processes arising from civil society. Without denying that the politics of influence is a necessary category for analyzing the politics of civil society, we are skeptical of both the analytical potentials and political implications of the predominant focus on the politics of influence, and therefore we propose two additional modes of forward infiltration from civil society to the state and the economy: the politics of substitution and the politics of occupation.

The politics of substitution represents the first mode of forward infiltration we distinguish from the politics of influence. This strategy seeks to bring some central functions of the state or the economy back into civil society or to build the key governing functions within civil society independently of the state’s influence. Here, civil society infiltrates the state (state functions), not by seeking its inclusion in the state but by shrinking the state sector through civil society’s capacity for self-organization and voluntarism. The strategies either can be embedded in and fostered by state actors (Evans 1996) or can reflect an aggressive effort to restrict the activities of the state in favor of private action. Or else civil society groups build economic organizations, such as cooperatives, which fulfill productive tasks typically left to privately owned, hierarchically structured economic institutions. The philosophical origin of this view dates back to Montesquieu’s and Tocqueville’s conservative, communitarian ideal of self-governing local associations (Putnam 1993, 2000; Taylor 1990). More recently, the Third Way vision of the state has sought to either transfer key government services into the voluntary sector or incorporate nongovernment organizations into the management of public services (Donahue and Zeckhauser 2011; Fung and Wright 2001; Giddens 1998; Zeitlin and Trubek 2003).

If the politics of influence is inherently realm-preserving, the politics of substitution either can be realm-expanding at the expense of another (zero-sum) or can lead to the coproduction of goods between the different realms (positive-sum). We distinguish between the scope of each domain and the capacity of actors in each domain. With the politics of substitution, there is always a substitution in terms of the scope of either domain (what goods or practices are carried out by which actors), even as such substitution can lead to simultaneous increases in the capacity of organizations in both domains to successfully implement their agendas. ⁵ The politics of substitution can range from efforts to explicitly substitute civil society for the state with the goal of undermining the capacity of state institutions (exemplified by conservative ideas of subsidiarity) to the Third Way division of labor, where civil society is seen as more flexible than the state, to more complex forms of coproduction where civil society groups take over functional tasks from the state in an effort to bolster the capacity of state institutions. In the latter cases, the politics of substitution will often coexist with the politics of occupation. Yet they remain analytically distinct, insofar as the politics of substitution preserves the organizational demarcation between the state and civil society, while the politics of occupation means that personnel and organization structures from within civil society directly enter the field of formal state or economic institutions.

There are numerous examples illustrating the politics of substitution, albeit with a wide range of political implications. Churches have traditionally run their own welfare programs to help the poor, disabled, homeless, and other minority and disadvantaged populations. In Catholic Europe, Catholic churches have for several hundred years been charged with educating and nurturing children under their own institutions and persistently claim to preserve that role, with important consequences for the broader structure of state-civil society interactions (Manow and van Kersbergen 2009). Conservative regimes seek to relegate welfare programs to these local faith-based organizations by subsidizing their budgets or granting tax exemptions and credits. Churches often even run their own correctional institutions to help inmates experience spiritual or moral changes, which have been legally supported in countries such as the United States, Brazil, and South Korea. In the United States, numerous city-based public services such as park management, emergency medical services, worker training, and preschool education have been delegated to nonprofit or voluntary organizations specializing in such services, although in many cases city governments chose to collaborate rather than completely relinquish their role (Donahue and Zeckhauser 2011; Fung and Wright 2001). We call these modes of substitution Tocquevillian voluntarism.

Elinor Ostrom’s celebrated “self-governing commons,” exemplified by irrigation systems (Ostrom 1990), illustrate how civil society can create its own regulatory, coordinating institutions without the intervention of a remote central authority. In Ostrom’s case, farmers developed their own sanctioning and monitoring institutions to avoid the abuse of the commons. While churches’ role in welfare and correctional institutions and Donahue and Zechhauser’s examples of private organizations in running parks and educational institutions show how civil society could take over traditional state functions, Ostrom’s examples further demonstrate how civil society can regulate complex disputes without the state.

These examples reflect deliberate efforts at substitution, often presenting the politics of state-civil society interactions as a zero-sum game. Where states are weak or porous, there may be informal modes of substitution whereby civil society groups substitute for the low administrative or enforcement capacity of the state. Here, the scope of activities of civil society groups increases, either at the expense of the state or where the state lacks the organizational capacity to implement existing laws and policies. But that substitution in terms of scope coexists with a mutual increase in terms of capacity. State institutions can now coproduce public goods in cooperation with civil society groups. In these cases, the politics of substitution becomes positive-sum: the activities of civil society groups simultaneously strengthen state institutions. Amengal’s (2016) study of environmental and labor regulation in Argentina illustrates these dynamics. He shows that civil society groups contributed monitoring and compliance capacity to ensure that Argentina’s weak administrative state was able to enforce its existing labor and environmental regulations. Similarly, Agarwala (2013) examines the organization of informal workers in India and found that unions and civil society groups, in the absence of a formal wage relationship that would enable collective bargaining, instead came to facilitate the delivery of social goods to those workers. In both cases, these organizations also engage in the politics of influence—pushing state actors to implement new regulations or social welfare provisions. But the crucial point is that they go beyond the politics of influence to actively fulfill functions that, in other contexts, are exclusively carried out by formal political institutions. While these dynamics are often studied in terms of the politics of incorporation of social movements into the state, we distinguish these forms of incorporation from the politics of occupation, as discussed in the next section, as they have distinct effects on the organizational integrity of civil society groups and on the structure of the field between the state, the economy, and civil society.

The third instance of the politics of substitution occurs at local community levels as a reaction to the repression of civil society by authoritarian states. ⁶ In several authoritarian countries that experienced democratization movements in the 1980s, formal civic spheres were mostly occupied by the authoritarian state, but oppositional segments of civil society developed new, clandestine organizational structures within civil society in order to preserve their organizational identities and resources. Examples of this range from religious organizations like Base communities in authoritarian Brazil to self-help through workers’ associations under a range of authoritarian regimes (SPD in Imperial Germany, Solidarity in Poland). While in some ways a limit case for our theory, as here there is little interaction between the state and civil society organizations, the politics of clandestine free space, such as in Poland, is exemplary of civil society formation and development. It is also not the case that under such conditions civil society is completely cut off from the state. Rather, civil society groups, as part of their resistance to authoritarian powers or pursuit of democratization, also generate alternative collective goods for their participants. These examples are illustrated in Figure 1-b, in which the state (dashed circle) is located in civil society. Table 1 summarizes these three different historical incidences of the politics of substitution in which constitutive, self-coordinating, and ideological solidarity operates under different historical contexts and with different images of the state.

OPEN IN VIEWER

Table 1.

Divergent Cases of Politics of Substitutions (Forward Infiltration).

Cases of Politics of Substitution	Modes and Features of Infiltration	Historical Occurrence	Image of the State
(1) Tocquevillian Voluntarism	Forward, cooperative, constitutive power for community building	In pioneers’ villages, in which new settlers built community-level institutions during colonial times	Potential exploiter
(2) Ostrom’s Self-Governing Commons	Forward, deliberative, coordinative institution for resource-sharing	In rural villages, in which the use of public goods was controlled and coordinated by local independent farmers or fishermen	Unnecessary monitor
(3) Clandestine Free Space	Autonomous constitutive institutions generating public goods	In local informal communities	Despotic oppressor

Just as civil society actors can try to fulfill their own tasks often carried out by the state, so too can they create institutions that are meant to replace for-profit corporate actors and so internalize economic activity within civil society. While these two processes are distinct, they often happen in conjunction. The classic example of civil society infiltration into the economic field through substitution is the cooperative movement. Here, civil society actors attempt to create their own parallel economy, one that operates on the basis of solidarity rather than profit. Countries with strong civil societies and deep civil society-union alliances, such as Italy, have historically had a flourishing cooperative economic sector (Riley and Fernández 2014). In the United States, antimonopoly social movements in civil society have been crucial incubators for farming and insurance cooperatives (Scheinberg et al. 2008). More recently, social movements targeting corporations have produced a host of nongovernment regulatory structures that certify corporations’ commitments to issues such as environmental sustainability (especially in the logging industry) and labor rights (especially in the garment industry). These private regulatory structures, which resemble Ostrom’s self-governing of the commons, are a response to the transnational nature of corporate activities and the limited ability and desire of formal political actors to tackle such transnational issues (King and Pearce 2010). The concept of social enterprise seeks to capture the rise of hybrid business-social movement organizations, organizations that typically have strong ties to other civil society groups.

These examples illustrate the capacity of civil society groups to cooperatively create or maintain (alternative) public goods and services in contexts where the influence of existing state institutions is minimal, nonexistent, or in retreat. In many circumstances, the politics of substitution is less a replacement of or infiltration into the former state realm than a restoration of the capacity of traditional civil society that was earlier confiscated by an overwhelming nation-state system. And in authoritarian regimes, the politics of substitution can then help foster spaces of oppositional freedom and political organization, creating a parallel society to official, repressive, state-led politics. Alongside these state-like functions, civil society actors often try to create crucial economic institutions on their own, especially when the state is not receptive to their claims.

The politics of occupation is defined by either (1) the (systematic) entrance of the representatives of civil society and civic associations into critical state or economic realms or (2) the institutionalized participation of civil society in the administration of state programs or economic institutions (participatory and industrial democracy). The first pathway can occur in the course of the maturation process of social movements in civil society. After vociferous and contentious confrontations with the state and existing powerful stakeholders, new social movement actors attempt to stabilize their organizational foundations and resources and then carve out a space for their agenda in formal institutional politics. With the politics of occupation, movement leaders within civil society decide to enter into formal political arenas by engaging in democratic electoral competitions, creating new institutions for their claim-making, and occupying (Gamson 1990) key executive positions in the state.

We distinguish the politics of occupation from the passive institutionalization of social movements, which, from the perspective of the politics of influence, only appears in terms of moderation or co-optation. The politics of occupation, instead, points to a more aggressive and significant replacement of status quo institutions. Under the politics of occupation, challengers reach a stage in which they become new institutional actors replacing old, established actors and their rules. The emergence and entrance of the Green Party leaders and supporters into parliaments and bureaucratic positions in Western Europe illustrates this mode of forward infiltration. The gradual occupation of public offices by feminist politicians and bureaucrats through the introduction of quotas and social policies advocating wage-earner women’s interests also testifies to this mode of forward infiltration (Banaszak 2010). Through this process, new forms of institutional actors appropriate the privileged positions of the state and shift the existing social and political orders to new equilibria, so that incumbent power-holders cannot help but adapt their preferences and strategies to new rules and logics created by novel institutional structures. These processes are depicted in Figure 1-c.

The second, more aggressive mode of the politics of occupation (participatory democracy) has been highlighted recently in several developing countries (Heller 2012). Participatory budgeting (PB) introduced by the Brazilian Labor Party (the PT) illustrates how civil society is capable of infiltrating the state field through participatory institutions established by the incumbency of a social movement–based party. PB and similar initiatives invite formerly excluded, less privileged social groups and their community-level representatives into the policy-making process. A series of meetings and councils scaling up from the grassroots to the municipal government provide citizens and their local leaders with ample opportunity to access necessary information on budgeting and participate in actual decision-making processes through open discussion, deliberations, and negotiations. PB ultimately “created a formal channel of interaction between civil society and the state with clearly defined and publicly known rules that broke with the practice of discretionary demand-making that had fueled clientelism” (Baiocchi et al 2011:114). Participatory governance (Fung and Wright 2001) is in a sense a second step of the politics of occupation, one that is enabled by the first step—the occupation of public offices by key social movement actors (Figure 1-c(2)). Grassroots civil society actors are then invited to occupy the new participatory roles and tasks under participatory budgeting. Through these two steps of the politics of occupation, civil society actors build their institutional channels to the state by dispatching their own representatives, while such representatives in the state continue to provide their grassroots with participatory spaces and channels. ⁷

In its most ambitious iterations, the politics of occupation often extends beyond the state and into the economic realm. In the politics of occupation, labor and its allies in civil society attempt, often in conjunction with their occupation of formal state institutions, to directly alter the ownership structure of the economy so as to occupy key decision-making positions. Rather than just influence their decisions from the outside, unions or civil society associations will try to get a direct voice within the organization of major economic structures. While relatively rare, this forward infiltration into the state and the economy has occurred in historical situations where the power of social-democratic movements was at its peak. Article 165 of the German Weimar Constitution enshrined the principle codetermination—active participation of workers and their representatives in the administration of economic structures—in an effort to head off the more radical move to form fully autonomous workers’ councils (Rogers and Streeck 1995). In the 1970s, the Meidner Plan, advanced by the Swedish Social Democrats, called for the creation of wage-earners funds that would, over time, enable labor unions to gain ownership of major economic firms and control their boards (Pontusson 1992). These both reflected a process of political radicalizations taking aim at preexisting corporatist structures that can channel civil society voices into the decision-making process of economic institutions. In both cases, ambitious civil society–labor union alliances sought to enter into the ownership and management of private economic firms. This possibility is illustrated in Figure 1-c(3).

Participatory budgeting and the occupation of economic institutions both reflect ambitious and aggressive social movements forcing their way into the state and other domains of structured power relationships. However, civil society actors are also invited into the realm of the state without the preexisting occupation of the state by social movement actors. A reformist, inclusive state may consciously mobilize civil society actors to the state policy decision-making process. Bloemraad’s (2006) analysis of the role of the Canadian state fits this scenario: In this case, the state provided the necessary resources, guidance, and space to immigrant communities to help ordinary immigrants and their leaders learn how to participate in politics and eventually run for office. This case of structured mobilization by state actors and operating through formal political institutions is different from the participatory budgeting case in the sense that the initiators of the forward occupation were state actors. The results of such processes may be quite ambiguous, as reflected in Paschel’s (2016) analysis of how the Columbian state invited black activists to formalize their organizational structure and participate in state-initiated institutions. She shows that this politics of occupation often left activists feeling isolated from their broader societal constituencies and entrapped in token consultation. In these examples, the direction of occupation, whether it is forward or backward, is not unidirectional but circular (Figure 1-c(1) vs. Figure 1-c(2), respectively). This circular feature of the state-driven mobilization of civil society blurs the line between the politics of occupation perspective and more traditional state-centric theories of the state–civil society relationships. For instance, Skocpol’s theory of social policy emphasizes the role of “the institutional arrangements of the state and political parties” and their influences on “the capabilities of various (social) groups to achieve self-consciousness, organize, and make alliances” (1992:47). The blurred boundary between the state-centric approach and the politics of forward infiltration is, to an extent, inevitable, as the very relationship between the state and civil society is often a feedback process in which actors in the two realms mutually penetrate over time. Yet our approach differs from the traditional state-centric view in that we emphasize the circular dynamics of state–civil society interactions: Civil society actors may mobilize themselves and participate in the newly institutionalized spaces but may also be guided toward reinforcing the legitimacy and performance of preexisting state actors.

Overall, the two submodes (civil society initiated and state initiated) of the politics of occupation posit a challenging question: To what extent and under what conditions can civil society actors successfully bring their own agendas intact into state institutions? How much should the state intervene in and control these processes? Such questions will become thornier in light of our discussion of backward infiltration, where the second, circular mode of the politics of occupation becomes more problematic when viewed in light of the top-down mobilization of civil society in authoritarian or totalitarian contexts.

Backward Infiltration: Three Modes

We define backward infiltration as the conscious strategies on the part of incumbent actors embedded in state and economic fields to mobilize or reshape civil society either to solidify the existing regime’s legitimacy or to promote actors’ political and economic agendas. While backward infiltration deploys many of the same mechanisms as forward infiltration, backward infiltration is different from civil society’s forward infiltration, as the former purports to control and regulate civil society in order to achieve political and economic actors’ agendas. Similarly, while we distinguish the same three modes, the exact mechanisms deployed in processes of backward infiltration will not be identical to those used in forward infiltration. While state institutions are more likely to deploy coercion rather than persuasion in pursuit of their ends, we are interested in the strategies that state and economic actors use that go beyond coercive power to try to restructure civil society itself so as to privilege certain issues, identities, and interests. Backward infiltration is conceptually different from mere suppressive strategies (although it could be pursued in conjunction with coercion), in that infiltration activities ultimately attempt to create a form of quasi-voluntary legitimacy or compliance from civil society or else among supporters who could carry and spread such legitimacy on behalf of elites. We conceptualize backward infiltration using our three subcategories: the politics of influence, the politics of substitution, and the politics of occupation. However, note that, despite the identical form of classification (three modes), the implication of each mode is not at all the same, as each mode in forward infiltration: The three modes of backward infiltration ultimately aim to achieve either right-wing or left-wing versions of domination, regulation, and discipline rather than influence, independence, or democratization, respectively.

The politics of influence as a mode of backward infiltration was initially theorized by neo-Marxist intellectuals in Western Europe. The idea of the backward politics of influence has some affinity to the concept of hegemony as originally suggested by Antonio Gramsci (1971): Hegemony implies a cultural and moral leadership that generates and justifies the dominance of one social class over the rest of society. Gramsci did not dissect the specific processes by which a dominant class is able to establish and exert such a cultural leadership, but the later generations of neo-Marxists identified the ideological (Abrams 1988; Althusser 1970; Wedeen 1999) and discursive (Laclau and Mouffe 1985) processes by which individuals are produced as compliant subjects though ritualized social practices in state and cultural institutions. In this framework, civil society is infiltrated (or even fully constituted) by ideological or hegemonic projects enacted by state institutions that coalesce with the dominant class interests. Furthermore, individuals in such a society, interpellated as particular subjects who comply with the dominant system through ideology, cooperate with the hegemonic projects of elites even if such projects are not necessarily beneficial to them.

In the orthodox Gramscian formulation, economic actors played a central role in the politics of influence. Struggles over hegemony were struggles over the production of a common sense that made the particular interests of dominant classes appear as the interests of all social groups. Similarly, in Althusser’s structuralist framework, the function of ideological interpellation is always the reproduction of class relationships within capitalism. We can reject these deterministic Marxian schema while still examining the dynamics of backward infiltration of dominant economic interests into civil society. For example, the rise of what scholars call the “civil society agenda” reflects a process of backward infiltration through the politics of influence (Alvarez et al. 2017). The civil society agenda refers to the shift, especially in the Global South, from a more critical, movement-oriented civil society to a focus on the role of civil society groups in promoting good governance and other development goals. This shift has occurred in part because donors and powerful economic actors have used their material influence to reshape the agendas of civil society organizations. Here, then, the politics of backward infiltration operates through the subtle monetization of civil society through civil society promotion.

As productive as this theorization of state-initiated projects at forming hegemony is, we view the politics of influence (by the state) tradition as limited and propose two additional modes of infiltration that more fully capture the penetration efforts of the modern states. First, in contrast to neo/post-Marxist version of politics of hegemony/ideology, we neither assume that the state is a mere agent of a dominant class nor insert any scope or strength of unilateral domination into the notion of influence. In addition, we do not presuppose that the state or ideological state apparatuses are such omnipotent structures that they are capable of reaching and completely ideologically controlling (or producing) individuals or social groups in civil society. Second, we contend that in light of core characteristics of electoral politics in modern representative democracies, the politics of influence tradition fails to fully capture the institutional processes of penetration initiated by the state. Instead, we argue that theorists need to examine the relatively fragile efforts of state actors to reorganize the will and interest of civil society by intervening directly in civil society’s institutional arrangements (the politics of substitution) or its personnel arrangements (the politics of occupation), rather than just through its framing projects via discourse or cultural institutions.

In the politics of substitution, state actors replace the functions or institutions of traditional civil society with state-driven policies and organizations, or economic actors attempt to reshape the interests and organizational structure of civil society by substituting their own activities for tasks traditionally carried out by civil society, thereby crowding out civil society groups. This mode of backward infiltration resembles a “revolution from above” (Moore 1966) in the sense that incumbent state actors initiate the transformation of society as well as the state itself. However, it differs conceptually from co-optation (Malloy 1979) because the politics of substitution by the state seeks the nationalization of social institutions rather than the incorporation of social forces and classes into the state (which is discussed in the next section in terms of the politics of occupation). In the course of constructing national administrative structures, state actors encounter patchwork networks of local powers and voluntary associations—civil society actors that fulfill many of the functions that states seek to monopolize. The relative success of this infiltration through substitution of civil society will then play a key role in forming the infrastructure of state power (Mann 1984).

The politics of substitution is particularly relevant to the formation of modern social welfare institutions. Here, central state actors seek to undercut or incorporate local, private, and scattered welfare institutions through the formation of national benefits that cover all members of the political community. In general, societies have developed their own safety nets at the familial, village, and community levels by pooling the resources of kin, village, or community members for protection from the vicissitudes of life risks. Modern welfare states establish their institutional control by eliminating these diverse kinship or community-level welfare institutions. In this manner, as the scholarship of institutionalism has repeatedly pointed out, the modern (welfare) state structures its own distinctive interest-channeling procedures within civil society (Hall and Taylor 1996) and creates social bases to sustain its interest-representation system (Pierson 1996). Both authoritarian Bismarckian and social-democratic welfare reforms substitute the universal social provisions of welfare for formerly private, scattered, and voluntarist social welfare institutions (Steinmetz 1993). As with forward infiltration, the effects of backward infiltration through substitution can range from zero-sum substitution to more composite forms of coproduction. While in all cases, the scope of the state field is expanding to take over tasks that were previously carried out by organizations within civil society, this can be done with or without undermining the organizational capacity of civil society groups to carry out their other goals or tasks. Thus, one crucial difference is whether the national institutions seek to fully supplant these local institutions (Bismarckian reforms) or integrate them into cooperative governance structures (social-democratic reforms). For instance, Bismarck’s social insurance laws sought to supplant local welfare provisions—most importantly, health and accident insurance institutions run by workers’ associations and unions—that were central recruitment tools for oppositional civil society movements (Tennstedt 1983). Fragmented welfare states, such as the United States, reflect the persistence of diverse civil society and nongovernmental social provisions, which then become part of the private welfare state or submerged state (Hacker 2002; Mettler 2011).

The key point is that the modern welfare state can achieve this dependency structure (although the universality and accountability of that structure vary widely across societies) only through the politics of substitution. Welfare institutions replace with pensions, medical insurance, and long-term care systems the functions of traditional family organizations. They also replace with child-care systems and family allowances the traditional care labor of stay-home mothers, thereby encouraging women to find economic roles in modern market economies (Esping-Andersen 2009; Orloff 1993). Finally, welfare institutions replace civic and community associations aimed at helping the poor, the unemployed, the disabled, and the homeless with public social assistance programs. In highly developed universal and generous welfare states, churches and welfare societies that were traditionally responsible for the rejuvenation of marginalized populations increasingly found those functions fulfilled by the welfare state. It may not be coincidental that secularization has been radically under way in Western Europe, where the welfare states have effectively substituted their institutions for the traditional functions of churches, while such secularization trends have been less radical in America, where the development of modern welfare institutions has been relatively minimal. Overall, the politics of substitution by the state is double-edged: While it may increase social protection for many, from the perspective of conservative communitarianism, it may enervate the voluntary self-help capacity of civil society, especially when the state preempts society’s own opportunities to rejuvenate less efficient sectors or marginal populations.

Organized economic actors may also engage in the politics of substitution, with circular effects of the structure of state-initiated backward infiltration. Both labor unions and corporations have historically created institutions that fulfill many of the mutual aid functions of civil society groups. Vernon Lidtke describes the German socialist labor movement as forming an alternative culture—ranging from reading groups to aid for funerals—that meant a worker could live his or her entire life with minimal interaction with state institutions and mainstream culture (Lidtke 1985). Conversely, some German corporations sought to create internal welfare institutions that would both bind workers culturally to their workplace and ensure the health and strength of the labor force (Sweeney 2009). The structure of such backward infiltration by economic actors will then interact with the politics of substitution carried out by the state. For example, America’s fractured welfare state is a product of the early politics of substitution carried out by economic actors, a politics that preempted the formation of national welfare programs.

While corporate actors may resist the politics of substitution by the state so as to protect their own projects, from the perspective of the left-wing (especially social-democratic) segments of civil society, the politics of substitution by the state is an opportunity to cure the ills of civil society (the fragmented and conditional nature of social support). Labor unions and their allied civic associations will work with social democratic parties to replace traditional forms of support with more generous and universal welfare institutions that help middle- and working-class workers to manage life cycle risks such as aging, illness, injury, unemployment, and childrearing (Huber and Stephens 2001).

In this process, sympathetic leftist organizations in civil society are eagerly willing to cede the traditional functions of civil society to state institutions. They assume that the risks of modern capitalism cannot be handled adequately from within civil society, as they believe that the capacity of traditional civil society is neither efficient nor fair enough to manage the increasing market-based inequalities produced in the course of capitalist development. Social democrats, therefore, occupy the state and demand that it substitute bureaucratic capacity and efficiency for the role of a (malfunctioning or nonegalitarian) civil society (Huber and Stephens 2012). The successful substitution of state action for some civil society functions then reshapes the sorts of actors and issues that arise within civil society, marked by the transition to the postmaterial concerns represented by the new social movements that arose following the completion of these earlier substitution processes in Western Europe (Habermas 1989; Touraine 1985). We may call this, then, a process of horizontal mutual infiltrations between the state and civil society, one in which the left segments of civil society, in alliance with unions and the social democratic party, will simultaneously pursue the forward infiltration of the state through the politics of occupation and the backward infiltration of civil society through the politics of substitution.

The politics of occupation by the state is often accompanied by semi-authoritarian, authoritarian, or even totalitarian politics. State elites dispatch allies to take over executive or leadership positions of important civil-society organizations, thereby managing civil-society organizations under the tutelage of the state power (Riley and Fernandéz 2014). Authoritarian regimes occupy the key formal civic associations such as unions and parties or fill rank-and-file positions with their allies. In totalitarian regimes, these organizations are fully incorporated into the state, and there is no independent civil society. In authoritarian states, these hierarchical associations serve as virtual organs of the state, mobilizing support for the state, spreading the state’s agendas to their members and clients, and monitoring surrounding associations’ interests and behaviors. They coexist with a variety of softer forms of occupation through which the state polices and infiltrates the activities of formally autonomous civil society organizations (Lee and Zhang 2013; Lewis 2013; Spires 2011).

In these circumstances, while the state does not send its bureaucrats directly to the positions, powerful political actors nonetheless indirectly influence the nomination and appointment of the chief executives of associations such as unions, media companies, state-owned firms, and even local community organizations. For instance, the state sends signals to influential members of such organizations that it prefers particular allies within or outside relevant organizations. Or the state may directly co-opt the influential members or leaders of those organizations with the promise that they will exercise more influential power in alliance with state. In more direct forms of occupation, the state may itself create civil society groups (quasi-government organizations), such as the Nashi youth movement in authoritarian Russia, which then crowd out and overpower other, more autonomous civil society actors. Finally, when pursuing the politics of occupation, authoritarian states (in contrast to totalitarian regimes) may also tolerate a limited range of autonomous civil society groups but only under the leadership of the state (Spires 2011). In such cases, these formerly “civil society” organizations are relegated to a subsidiary system of the state bureaucracy.

Outside of authoritarian contexts, the targets of social movements in civil society—state bureaucracies and corporations—may also pursue these tactics, creating a variety of organizations that resemble civil society associations but that are designed to blunt contentious politics. This is the inverse of how civil society forces may pursue ambitious participatory-democratic projects. These forms of backward infiltration through occupation can range from deliberative and consultative forums that co-opt civil society leaders and align their interests with the interests of organizations (Lee and Romano 2013) to the formation of competing civil society–like organizations, such as sweatshop monitoring groups controlled by corporate or state actors (Scherer and Palazzo 2011).

Economic actors both pursue the politics of occupation and are its targets. Many authoritarian states seek to produce integrated forms of corporate economic governance. The end result of this politics of occupation is a form of the “corporatism” (Schmitter 1974; Streeck and Kenworthy 2005) or, more narrowly, state-led corporatism ⁸ that was prevalent in many semi-authoritarian polities (especially in East Asia, Latin Europe, and Latin America). Here, economic actors are brought into a wage-bargaining/economic-policy system organized and controlled by the state. In these circumstances, capital will also act alone or with the state to create its own economic organizations—most centrally, friendly labor unions—to compete with or replace more autonomous labor unions that could then ally with civil society associations, such as democratization movements.

One of the best examples of this politics of occupation are the trade unions created by the authoritarian regimes in Spain, Taiwan, South Korea, China, and Brazil. These trade unions (and their leaders) served the regime through their superficial bargaining with employers and the state, by conforming to the guidelines proposed by the state, channeling state propaganda down to grassroots members, and acting as the political allies of the incumbent authoritarian party. Chinese trade unions (工会) during the 1990s and the 2000s illustrate this politics of occupation by the state, as they were organized and controlled by the communist party to stabilize and manage society (Chan 1993). During the earlier periods of economic growth, trade unions functioned primarily for the purpose of persuading workers to follow the state’s developmental agendas (Baek 2000). The authoritarian states and their intelligence offices in Spain, Taiwan, and South Korea managed these unions by providing leaders with positions in the government and congress, distributing monetary benefits and clientelistic goods through such networks.

While the politics of occupation at the civil society–economy nexus is more overt in authoritarian contexts, there are analogous processes even in democratic regimes. Especially where national-level or cross-industry unionism has been weak, there is a long history of corporate or “yellow” unions that are meant to supplant the autonomous self-organization of workers. And as cooperation between nongovernmental organizations and other civil society groups and corporate actors seeking public legitimacy via corporate social responsibility has become more prevalent, so too has the risk that such nongovernmental organizations will end up being formed by the very corporations they are meant to monitor and hold accountable. Groups such as the Fair Labor Association, a U.S.-based sweatshop monitoring organization created and funded primarily by the corporations that it is meant to monitor, then compete with autonomous civil society groups by providing the appearance of forward infiltration.