Civil Society’s Links to Politics

The Importance of Second Level Political Institutions1


Douglas A Chalmers

Columbia University

March, 2000


The Problem and the Argument

Consider: A small group of newly-graduated aspiring professionals, funded by a Dutch foundation, form a group to promote policies for industrial development in a regional city in Mexico, applying ideas from a course taken at the Kennedy School. Or, an industrial union in South Africa sends young potential leaders to an Ebert Foundation workshop on flexible production, and prepares to lobby for new laws governing retraining. Or, a group of people living on a river in India, having watched UN produced television programs convene to pressure local government to crack down on polluting factories. Activities such as these are multiplying. What will make the efforts of these professionals, workers and neighbors benefit society?

The argument of this paper is that the benefits depend not only on their savvy, skill and luck, but also on whether political institutions encourage their work and constrain it effectively. Where can these groups find the resources to organize their efforts and sharpen their impact? What costs of these resources? What avenues of access are available to these groups to make demands? What rules prevent frivolous claims? Do opportunities exist for these groups to learn about industrial growth, worker retraining or anti-pollution methods? What blocks are there to gaining needed information, and what temptations are there for them deceive or be deceived? And if, as is common in this dynamic arena, the energies dissipate, organizations shrivel or talents find other outlets, what rules and practices facilitate or block their replacement by new, perhaps more innovative groups? The answers to these and many other questions crucial for the future of civil society’s role in the polity are to be found in the political structures that emerge or are imposed as a second level of institutions existing side by side with elections, party systems and parliaments.

In the last decade , responding to the perception that many of their objectives require cooperation from authorities, many in the NGO community have, in fact - on their own, or pressured by donors - developed strategies for working with governments. This is another side of the argument being made here that the impact of civil society organizations depends on how they are linked to the system. But viewing the general issue of civil society’s role as a question of how to improve project performance misses a crucial dimension. The success of an individual project is not important as such (except to the donors and actors). The failure, or even the disappearance of a grassroots neighborhood organization, an economic policy ‘think tank’, or a women’s rights collective, will, in itself, not have much significance on the working of the political system. What counts is whether there are other organizations to take their place, and if perhaps the new organization will bring an improved approach. The question is how the constantly changing performs altogether. As a crucial step in understanding that overall impact, we must look at the institutions which link them to the state.

How does the formal and informal institutional context shape, respond to, and integrate the activity of this collection of groups? Whether its activities promote a stable and effective democratic polity or not will depend not only on whether they are democratic and seek democratic goals, but also on whether the institutions through which they work promote democracy as well.

The questions that are relevant here are a modern form of a classic concern of political theory: what sorts of institutions linking civil society and the state promote democratic values? The principal answer to that question has always been the constitutional institutions of a liberal democracy: parliaments, party systems, elections and an accountable executive, all supported by the rule of law2. Since these institutions make sovereign law and policy, it is obviously crucial for democracy that they should work well, and that the two aspects of these institutions which link the state and polity with the citizens, elections and the protection of individual rights, be effective.

But there is more to democracy than strong constitutional institutions. Elections and the justice system cannot bear the burden as the only link between them and citizens. People commonly say ‘elections are not enough’, but then they often turn to raise questions of social and economic equity. But before leaving institutions for outcomes, one must focus on a second level of institutions that also complete such links. In a complex society the space between constitutional institutions and the citizens is dense with procedures, regulatory practices, and rule-governed arenas. And these other institutions are crucial for democracy, as well.

The most democratically organized and elected parliament will not operate democratically if the only issue brought to its formal attention are those desired by a small elite. Elections are thought to be corrupted by ‘special interests’, yet since purging all group interests from elections is impossible, the key to knowing if elections are democratic is whether the special interests are formed freely enough as to reasonably represent the ‘market of interests’. The operation of the justice system depends not only on good laws and official oversight, but also monitoring from both society and from the legal profession. Whether these things happen in a way conducive to democracy is, in good measure, a function of a second level of institutions, between the citizen and the constitutional institutions.

If constitutional institutions depend on second level institutions, it is also true that civil society organizations depend on them to shape their impact on the system. Civil society’s importance depends, to be sure, on whether the constitutional level is democratic since autonomous groups and private networks need openness and the protection of civil and political rights. But that too is not enough, and institutions at the second level are also crucial for explaining the impact of civil society groups. The persistent question of whether non-governmental organizations good for democracy or not depends for its answer on the effectiveness of second level institutions.

There was a time when grassroots and non-governmental organizations were seen as the solution to the evils of over-centralized states. Like firms in a free market (which had become the chosen instrument for economic growth), civil society organizations are non-governmental, dynamic and adaptive. Like progressive social movements, they are engaged, forward-looking and critical of the existing arrangements that (almost) everyone agreed needed reform. "Social capital" became a measure of success, a strategy and, a buzz word.

Later, as solutions proved elusive and development programs lacked the necessary coordination and regulation to succeed, pessimism set in. The apparent faults -- fragmentation of effort, lack of accountability and excessive influence by ‘outsiders’ -- were more commonly cited. Human rights campaigns and emergency efforts by private groups to deal with poverty, natural disasters or ethnic conflict were seen as positive, but were acceptable only because their role was seen as transitional. States, with their democratic institutions should be pushed to take back their role, not relieved of responsibility. The explosion in the number of NGOs, some said, was merely a make-work job scheme for underemployed intellectuals.

But it is clear by now that civil society organizations are not going away. They have become a part of the political landscape. Individual ones, even whole clusters are transitory, but as a sector they are here for a while. They cannot take the place of government, but they do have a place. The current challenge is not to decide if they are a good or a bad thing, but to see how they fit in.

The reason for the growing visibility of NGOs and private networks is not to be found in time-bound political needs (such as overthrowing authoritarian regimes or handling emergencies). Their presence is the result of the long term spread of communications, the rapid increase in the number of sophisticated activists, leaders and organizers, the globalization of public policy questions, and the complexity of problems and of the programs people devise to deal with them. It not longer seems possible to imagine a reversal of those trends. And it they are the source of the growth of civil society organizations as I believe it to be, then their presence, even their growth, is permanent. The unresolved issue is how the resulting organizations and networks can be, and are being built into the political process, into the ‘state’. The geography of political representation and public action has been changed. The second level institutions I am talking about will be a major part of the conditions under which the explosion of new civil society organizations will make a positive contribution.

In the last section I have listed 4 arenas where second level institutions are found.

First, I will look at this set of institutions and the aspects of representation they deal with. Then I will offer a typology of sets I call ‘participation regimes,’ and then end by describing some second level institutions.


The Relation of Second Level Political Institutions to Democracy

Liberal political thought, on the one hand, and monistic authoritarianism on the other, have assumed (or desired) a direct link between citizens and the institutions of the state. But there have always been intermediary rules, norms, and procedures. Society, of course, has many institutions, (families, the market, churches etc.) which regulate and order individuals and groups. But these, as such, do not mediate between citizens and the state. In most cases, they are not political institutions. Elections, the party system, Congress and the executive, what I am calling the constitutional institutions, are, of course, political, and they do regulate, support and constrain actors in the final, sovereign decision making process.

However, if we look at the space between citizens, groups and these social institutions on the one hand, and the constitutional institutions of the party system, elections, parliament and the executive on the other, we find that this arena is not chaotic, but, for good or ill, is full of practices, laws, rules and norms that regulate, support, and constrain public and private organizations in how they deal with the state. That is, this space is institutionalized by second level institutions which help determine who among the civil society organizations shall be heard in party discussions, elections and parliamentary debates, what weight will be given to their points of view, how conflict will be regulated and how debate will be carried through.

Which people are listened to, and how they are trained and socialized, is not random nor the result of clever politicians. Which policy proposals surface and how (or if) knowledge and information is applied in policy debates does not happen only in Congress. They are shaped by established practices. Certain schools or local machines are part of the network that produces leaders. Established newspapers and electronic media become the 'accepted' place for discussions to take place, and 'trial balloons' to be launched. Certain established leaders are routinely consulted on new initiatives. These are some of the pieces of structures which regulate the second level space.

Second level institutions have not, in fact, been ignored by social scientists and analysts. Any good discussion of policy formation will pay attention to the customs and laws and organizations which affect the clash of parties, lobbyists, expert observers and foreign activists and the like. Thinking about these groups as an ensemble relevant to regime type is less common. But still, for a long time observers have used (rather loosely, to be sure) a few ideal types to describe broad variations in political patterns I will describe as ‘participation regimes’, that is, distinctive patterns of links between civil society and the political decision process3.

In particular it will be useful to discuss those designated ‘clientelism’, ‘corporatism’ and, ‘partisan organization’ as elaborated below4. Although there are elements of clientelism, corporatism and partisan division in every political system, the dominance of one or another of these forms changes, causing the participation regime to alter. In some cases all three of these fade (and I would argue that today in many countries this is such a moment. The standard liberal model postulates that in the absence of these ‘traditional’ links, ’modern’, direct, individualistic connections between citizens and law-making develop. I am arguing instead that new forms of institutions emerge. Out of these, I argue, a new form of participation regime takes shape, which I call an open network participation regime.


The Basic Model of Representation and its Limits

To describe the institutions of a regime, we must know what is happening when this mediation between civil society and the state takes place. The elements of representation may seem straight forward, but there are some complications which are often glossed over. It is crucial to examine these complications to understand the patterns in second level of institutions.

I take the conventional model of representation to be something like the following: The polity is made up of individuals within a territory (citizens), each with interests. Through parties, interest groups or electoral blocs they join together to influence political decisions affecting them. As long as the interests are not too sharply divided, democracy is ensured by the representation of those parties, interest groups and electoral blocs. This happens as party systems, parliaments and executive departments respond to elections and other forms of public opinion formation (polling, the press, etc). Within those institutions, competing demands are brought together in a process of negotiation and bargaining. The tasks taken as crucial for institutions derives from the model – in the terms of an older but still almost ‘common sense’ language, we look for how interests are articulated and then aggregated and converted into authoritative allocations through law or policy5.

I believe we must avoid the temptation to accept this conventional model of representation, even 'for purposes of argument', or as 'simplifications necessary for theorizing'. There are several major limitations on this view of what is going on, each of which is important for the institutions of a participation regime.

First, the interests of each citizen are multiple and constantly changing. There may be enduring principles, goals and rationalizations, but confronted with concrete situations, choices involve different sets of goals and different ways of reaching them. They change over time, and they also change depending on the context. Perhaps it can be said that institutions always represent a way of establishing stability in the face of multiplicity and change, but unless strong ideological parties, ethnic blocs or other forms of long term polarization stabilize (or freeze) the expression of interests, institutions manage (or are strained by) the constant rise and fall and redefinition of the groups and networks. They must be assessed not by whether they eliminate this volatility, but whether they shape it effectively.

Second, the individuals who seek influence are not only citizens and not only from within that unit’s territory. The reality of international actors cannot be doubted, and although their legitimacy is sometimes questioned, over time it has been increasingly accepted. Foreign pressures are multiple and far from being managed through diplomacy, and many actors with the system are non-citizens. Political institutions have to deal with, or be undercut by, an internationalized politics.

Third, the process of transforming interests into policy is often not bargaining and negotiation but, in varying degrees, deliberation. Changes in the definition of the situation, whether from ‘science’ or by social construction, are often as important as compromises or changes in the power of interested parties. The handling of information and decision rules for using evidence and employing theory are as important in political institutions as managing compromise and bargains.

Fourth, the links between the citizens and authoritative centers of policy making are not only a) based on individual acts like elections and polling, and b) collective action through parties and interest groups, but also, c) networks of face-to-face (or screen-to-screen) contacts. Political institutions have to be looked at in how they manage informal and rapidly reconfiguring networks as much as dealing with voting and organized interest groups. Although personal ties are often seen as penetrating and altering institutions, institutions shape them as well. For example, schools often create networks and ambitious actors form networks on the basis of the institutional position of the people they are linked to.

To elaborate briefly on his last point, political analysis and discourse has long turned around individual participation or membership in organized groups. Individual acts of political involvement such as voting, writing letters, informing oneself, even rates of joining groups are the mainstay of studies (and moral assessment) of participation. On the other hand, organizations – parties, interest groups, associations, bureaucracies -- have long received priority attention in studying how decisions are made. At least since Max Weber, organizations have been seen as the direction of change, the ‘modern’ style. Networks of face-to-face contacts were seen as local, traditional, and, perhaps, corrupt. Organizations of one kind or another were seen as more efficient, more in harmony with the rule of law and basically, more modern. In the clientelistic form of networks, backwardness and corruption seems possible, even likely.

But clientelism is not the only form of network. Given the rapidity and multiplicity of communication, it is quite possible to argue that formal organizations are either changing to more flexible forms or being left behind. In studies of business and bureaucracies, getting personal networking and organizational goals in sync is a major thrust of modern management techniques.

In politics, the presence of personal networks is obvious, but too often it is viewed from the perspective of the older which shows (or hints at) the way these personal networks undercut everything from the rule of law to the fidelity to the ‘principles’ of the people being represented in the political process. But recent research on political realities have been generating studies showing the importance of ‘policy networks’ in influencing what policies are considered. And ‘embeddedness’ has become a new focus, referring – imprecisely, but certainly – to the way in which process are supported (or undercut) by networks of personal ties.

Scott Martin, Kerianne Piester and I have offered a template for a politically significant set of networks we call associative networks6. Their distinctive characteristics are the following: First, any single associative network is likely to be characterized by a diversity of organizations and individuals, very often including people from political parties, government agencies, non-governmental organizations and international officials and activists. Second, any particular network is likely to be reconfigured over time as issues, decision making rules, participants and opportunities change. Third, the associative network entails strong emphasis on what we have called cognitive politics, involving debate and discussion of preferences, understandings, and claims, in addition to æ and potentially transforming æ more conventional bargaining over demands and interests. We called them ‘associative’ because they include many associations at nodes on the web of links, and because they are formed more with an eye to particular tasks, not built out of the multiplex personal family-like ties of a clientelistic relationship. In its politically relevant form, personal contacts are often crucial, but they are specialized and are likely to link individuals, organizations, specialized media such as journals or newsletters, officials in government bureaus and international organizations. They are usually held together by some sort of task or common interest. This unifying factor might be promoting a particular policy, securing a particular advantage or in finding a solution to a particular problem. Being task focussed rather than clientelistic, they are also much more subject to rapid change. This template offers a point of departure for identifying new forms and understanding the challenges of institutionalizing them.

Assessing whether the institutions of a particular country are producing effective, stable, peaceful and just governance will require changing some assumptions. Stability, for example, needs to be assessed not in terms of the continuity of the organizations of civil society, but in terms of the stability (or ‘metastability’) of representation as a whole in the face of rapidly changing individual elements. A just outcome will depend not so much on whether particular set of NGO’s with their associated programs are effective, but whether changing coalitions of interests find appropriate and effective champions at the right time. And effective decision making will depend not only on the ability to bargain effectively, but on whether information and analytical ability is mobilized to make ‘smart’ decisions7.

That is to say, the impact of fluidity, internationalization, deliberation and networking, the four amendments to the standard model of representation, is to call attention to rapid change in people’s interests, in their understandings of what the world is like and in reconfiguring networks.

One might object that most of these principles are familiar. We all know that politics is fluid, internationalized, with complex infrastructure, and that policy discussions involve arriving at an understanding of the problems, as well as bargaining among interests. It is the nature of politics. One tends to think of these aspects of politics not as something which needs to be institutionalized, but simply as the activity that institutions like parliaments, executives and parties manage. But in a world in which policy making is very complex in substance and process, a world in which internationalization is increasing and information is more widely available, the signs are clear -- people have sought to create some order. Politics is not simply a buzzing confusion skillful politicians manage to order though elections, public opinion polls and hearings with constituents. Organizations, laws or sanctioned practices are, of course, adopted or mandated to impose order. But also, patterns are emerging from the 'market' of political activity in a 'self-organizing' fashion. From both sources, we need to look for patterned, valued behavior (i.e., institutions) implicated in:

Of course, the classic constitutional institutions of the political system also address these problems. Legislatures structure conflict and manage deliberation, elections order recruitment, and basic laws establish the rights and obligations of ‘citizens’ and ‘foreigners’. But for a long time, parliaments have ceased to be more than a capstone of the legislative process, elections recruit only a small percentage of leaders, and basic rules governing executives are a very small part of the emerging practices and institutions which govern the modern state establishment. Any examination of a polity has to start with the constitutional institutions. Their operation is the exercise of sovereignty, i.e., making decisions having the force of law in a territory. However it would be a mistake to believe that because of their centrality they are comprehensive.

Three Classic Forms of Participation Regimes

Examining conventional usage, three types of second level institution-sets are conventionally identified. I will call them participation regimes since they structure the way civil society participates in the political process8. They are ‘regimes’ because even though they are not at the constitutional level, they still integrate some aspect of the political process for the entire population, not just some subset.

The three types are clientelism, corporatism and partisan division. These are ideal types, in that each articulates a common logic of a wide range of structures. In the negative sense, they are ideal types because each one is hardly ever the exclusive pattern in an real case. Since the degree to which they dominate particular systems varies considerably, there is a danger in using them loosely. For example, people sometimes use ‘corporatism’ or ‘clientelism’ as characterizing a regime, when they really mean to point to a few prominent instances. Here, I will try to confine their meaning to the dominance of their characteristic forms of links, sometimes at the top, constitutional level, but specifically in this intermediate, second level of institutions. I use them essentially as counterweights, to highlight the distinctiveness of the open network regime.


In a clientelism regime the multiplicity of links are organized into the familiar patron-client networks of domination and subordination linked through personalistic contacts. Often face-to-face and involving a personal relationship, it is a type of network formation. But it is a special form, institutionalizing representation by infusing the links in the networks with multiple personal exchanges creating a broad based dependency. In a system with law and formal organizations, it changes their meaning and formal shape, undercutting the associational aspects in favor of ties of strong dependency. The actors’ expectations (or, from the negative side, the transaction costs of enforcement) are managed through a system of mutual dependency even including (often fictive) family or ethnic community ties.

Although clientelism is now used almost as a synonym for inefficiency, corruption and injustice, there are many forms and contexts in which it inefficient, stable, and just, even when they do not loosen and change shape to become associative networks. Personal following of populist leaders may bring oppressed groups into the political system, the 'teams' of political figures around prominent political leaders (e.g., the 'people around Clinton’) can make the management of public affairs smoother, and ‘boss’ run political machines provided a channel for incorporation of immigrant populations. Broad personal trust enhancing ‘thin’ contractual arrangements can be positive as well as corrupting.

Whether the democratic potential for patron-client relationships is weak or strong at the micro level, most seem to agree that if expanded to the level of organizing the 'whole' system of political relationships between the citizen and the state, clientelism becomes a rationale for excessive concentration of power in the hands of a few ‘patrons’. The rigidity of the multiple ties in the relationship may be useful in a specific case, but it is likely to inhibit the flexibility required to deal with reconfiguring networks with changing purposes. In a small-scale situation, clientelism might help stabilize a situation, but as the field becomes larger, the chances for factional conflict seem enhanced by the strength of the competing personal loyalties in a clientelist framework.



Corporatism is a participation regime where second level institutions are built around formal organizations. These entities, ranging from guilds to Fascist chambers of corporations or formally established ‘tripartite’ management schemes in industry, absorb (or seek to absorb) the complexities of civil society into the procedures and hierarchies of organizations. In recent corporatism, these formal institutions are very likely to be sanctioned in law and become part of the state.

In corporatism, networks are sorted out according to a predetermined set of representable interests, and these are given formal status. One common set of interests were those identified in the early twentieth century from a division between classes, either reflecting Marxist thought or more likely for purposes of fighting it. But there are many possible definitions of the interests. A Mexican form built organizations around labor, peasants, government employees and the military, tying them to the official party from the 1930s until quite recently. In some conceptions, going back to the earliest guild days, the professions have often been put forth as the essential building blocs of the corporations.

Corporatism can refer to central constitutional institutions, as in the fascist experience, with its chambers of labor and business. But except as a façade for state dominance, these were very rarely significant as coordinators of policy making in the state. As second level institutions (not at the constitutional level, but still as a regulator of system wide political activity) corporatism embraces the wide range of efforts to grant recognition (and establish control) over interests through formal membership in advisory councils, incorporating labor unions into government ministries, or, more recently, granting certain NGOs and special interest groups a formally specified place in the councils deciding policy relevant to their interests.

Corporatism also offers ways to deal with the complexity of representation. Competing and changing interests are managed (in part) through the discipline of bureaucratic organizations. Conflict between interests is converted into bureaucratic negotiation (or infighting). Managing the boundary between national and international actors was largely done by excluding the latter. State organizations are, after all, national institutions and managing deliberation was a process of turning open discussion into staff discussions.

In the present climate, the rigidity of corporatist structures in the face of changing challenges, and the violation of standards of justice by entrenching certain interests are considered sufficient reasons for to reject this type of participation regime. Even if one eschews any form of constitutional corporatism in order to concentrate on the second level, using the instruments of formal organizations, with their hierarchies, procedures and rules, to organize fluid and changing networks, the inflexibility with regard to the new global influences suggests inefficiency, and the reliance on bureaucratic structures is often stultifying. Corporatism had its democratic moments, in many countries. It was the means through which organized labor was brought into the political system. But its role now seem confined to specific problems, not to the regime level, even the second level regime.

Partisan organization

By 'partisan participation regime' I simply mean the institutionalization of the networks and multiple organizations by means of tying them to or building them into political parties. This is often accepted as the modern, democratic standard. It has seemed to many that the disciplining of politically relevant civil society in a democratic fashion requires, or at least is efficiently accomplished, by means of 'aggregating' its interests into party programs and policies. Parties are often seen as the appropriate way of extending the constitutional institutions into the civil society.

The obvious advantage of partisan representation, from an institutional point of view, is that it ties an organized version of the confusions of popular demands directly to what have been taken as the central institutions of democracy, elections and the legislature. Further, parties have long been considered (especially in European-centered political science) to be the ideal way of incorporating information, knowledge, theory – all the cognitive elements noted above, into politics. Programmatic parties, parties with distinctive world-views , parties with well thought out platforms, all these have been seen as a way, perhaps the way of bringing ideas into politics. Finally, as the name implies, parties do something that corporatism is often designed to avoid, building orderly conflict into the system by channeling it into party competition at elections and parliamentary debates. Given a good (i.e., working, representative) party system, popular choice at elections is transmitted immediately into teams of leaders in authoritative positions. Of course, the institution might not work, and concentrating everything into the political parties will be negative if legislative processes are blocked, elections fail to justly represent interests, programs are lowest common denominator compromises, or polarization of one kind of another leads to instability. But given the strength of the assumption about the desirability or inevitability of partisan representation, these are analyzed as problems to be overcome, not reasons to question the centrality of partisan ordering of political society.

However we evaluate the ‘weakness’ or ‘strength’ of parties as the organizers of constitutional institutions, the concern here is with the role of partisan organization as second level institutions. Aside from appealing to voters and representing them (well or badly) in the legislatures, how do they ‘penetrate’ the society?

In the US, many have noted, sometimes lamented, the failure of political parties to organize society. In fact, political parties did seem to play that role at the end of the 19th century in some cities which were dominated by the ‘machine’ and also during the long reign of the Democratic party in the post Reconstruction South. But in the latter two, at least, parties were probably patron-client relationships masquerading as political parties. A variety of theorists have argued that the American system contrasts, and may improve on one built on parties in society, being built on a complex of plural ‘interest groups’.

In much of the rest of the world, the link between parties and civil society was usually spelled out quite explicitly as an ideal, if not necessarily a reality. The most influential formulation is some variant of the Marxian one, but there are others. Parties articulated fundamental cleavages in society which did, or should, aggregate the interests of the people, not through the implied balancing and negotiation implied in the American model, but by bringing them in line with dominant visions of society and its goals. In many socialist visions parties managed the networks and organizations of civil society by educating the people into their 'real' interests, and absorbing the 'spontaneous' structures into a complicated web of solidarity, more or less subordinated to the political party which articulated that interest.

Generated from a Michelian logic of tight organization to face confrontation, the same logic of partisan organization of civil society has been applied by Catholic parties, ethnic/linguistic parties and even regional parties. For many, this image of a partisan representative regime has defined modernity. Although its hold was already weakening in Europe, when I first went to Latin America in the early 1960s, the implicit definition of a 'modern political party' was one which had a highly developed programmatic stance (our 'program', their 'ideology'), an extensive network of solidary organizations such as unions and professional associations (our solidarity, their conspiracy), and considerable discipline (our unity, their rigid conformity.)

The Open Network Participation Regime

In some periods in most countries, and perhaps especially right now in many of them, clientelistic, corporatist and partisan patterns become inappropriate, or perhaps, too expensive to maintain with any claim to be stable, efficient, peaceful or just in the face of such temporary or long trends as :

It is tempting to argue that these complexities are unilinear and that clientelism, corporatism and party dominated societies are a thing of the past. But this is probably wrong. A sort of corporatism might well reemerge if centralizing demands are high, to fight a war, deal with an environmental crisis or a major economic depression. Partisan solidarity could be a response to a new, or newly dramatic cleavage in society. The possibilities of partisan division around ethnic cleavages seems logical in the face of current struggles, although weak – perhaps because the enduring ‘consociational’ harmony a long terms partisan division implies seems a fragile and difficult thing to achieve. Clientelism, always present, might become a dominant form again in the face of state decay – certainly not an impossible development in many countries. In any case, whether more open regimes are more likely in the future, there are many situations in which we need to look beyond these formulas.

When clientelism, corporatism and partisan organizations weaken, the pattern of institutions which regulate and manage the congeries of networks and organizations takes place in a more open ended way – what I call an ‘open network’ type of participation regime. We are looking for a participation regime which links (well or badly) the organizations and networks with the polity as other regimes do. But in contrast to corporatist and partisan regimes, in which either state or party organizations 'take command', there is no dominance of an imperative, hierarchical organizational form, and although networks share a degree of personal connection that is characteristic of clientelism, they lack the multiplex, stable, hierarchical ties of the latter. A degree of autonomy and fluidity will be the hallmarks.

Open network participation regimes differ from the other participation regimes we have discussed, because there is no single type of formation that dominates (as in the clientelistic regime) and no single set of national institutions (as in the corporatist or partisan regimes)9. As a ‘regime’ it refers to a nation wide system, and it is a (potentially) coherent complex of emerging and imposed rules, norms, practices and organizations. But it is complex. The complexity comes from two factors. On the one hand, regulating debate (for example) will be specific to particular issues. Sets of actors and specialized media or locations for discussing health policy, say, may develop. So the first type of complexity comes from the specialization of institutions around many different policy areas.

A second form of diversity comes from the development of institutions around different functions. The people who work in these fields of public action will be educated, recruited, and manage their affairs in different institutions than those which are devoted to setting up seminars for the discussion of policy issues and review of current research. In a partisan system, these are both likely to be incorporated within the party, and in a corporatist one, both take place within the official organizations. A major feature of the open network regime is the breaking apart of these agglomerating institutions.

Admittedly I am approaching the definition of an open network participation regime in an negative way. Such a regime is not dominated by either bureaucratic or party style organizations. The term ‘networks’ is used to suggest that the elements are not just individual citizens (as in the idealized liberal world) nor just associations and organizations. The term ‘open’ suggests that it has many institutions, and there is inevitably a certain amount of reconfiguration constantly occurring. The emphasis so far is on diversity and complexity. Once we can see the institutions that emerge to manage this complexity, we will need to make finer distinctions.


Are open network regimes more democratic?

Before discussing the nature of the institutions in an open network regime. I want to address a problem which often emerges in discussions about complex civil societies, that is, are open network regimes more democratic than others?

There is no reason to believe that open network regimes are or are not more democratic merely because they are open and networked. Just as the clientelistic, corporatist and partisan regimes have their democratic and non-democratic forms, so do open network regimes. In the recent past, the breakdown of authoritarian corporatist regimes, the liberalization of corrupt clientelistic regimes and the weakening of ideologically divided and stalemated or violent partisan regimes has led many to look to ‘civil society’ and ‘social capital’ as a means for bringing about democracy. For this reason it is often supposed that moving towards a different kind of participation regime will necessarily be more democratic. But the institutions of the open regime, as we shall see, can be elitist, irresponsible, non-accountable, conflict ridden, inefficient, and inegalitarian, just as those in other types of participation regime may be.

The point here is that to decide whether the growth of citizens movements and non-governmental organizations and social movements are making the polity more democratic, one has to understand and assess the relevant second level institutions. An open, fluid and complex regime can obviously make room for all kinds of elitism, exploitation, inefficiency and instability. On the other hand, as we know in the ongoing arguments about the market in the very different field of the economy, an open, complex system at least has the capacity to be more efficient, more stable and more just than a bureaucratically organized one. Given the right patterns and opportunities, the energies, knowledge, adaptability, and public commitment found in much, if not all, of civil society organizations and networks can be harnessed. Just as we look to the operation of the constitutional institutions make a first judgment as to whether a country is democratic or not, we need to identify the second level institutions in order to complete that analysis. The growth of civil society and social capital may or may not be democratizing. It will depend on the way the institutions linking the associations and states work10.

Institutions in the Open Network Regime

As in any collection of real institutions, we expect to find ‘formal’ ones (legislated, explicitly imposed) and ‘informal’ (emerging norms and procedures from many interactions) patterns. Focusing on the institutions in an open network regime that are outside of the constitutional ones, we expect informal origins to have an especially important role, in view of the autonomy and fluidity of the actors. For example, policy activists, through repeated planned or unplanned meetings begin to identify each other and establish informal standards on who qualifies as an expert and who doesn’t. Or, after many discussions and meetings, certain journals, meetings or websites may become the ‘established’ location for recognizing and distributing best practices or alerts about impending legislation. As any market-like activity, a major set of controlling norms will emerge from interactions like this.

But that by no means suggests that the ‘formal’, whether in the form of government legal regulation or explicit negotiated agreements, will not play a role. As in the shift from statist to market-driven economies, the role of the formal changes, but doesn’t disappear. In the sphere of second level institutions, government would not force policy discussions into a government arena, but it might help shape one in civil society by mandating that there be a regular series of forums on some policy before it comes before Congress or a ministry. Formal arrangements also might shape institutions, for example through restrictive legislation on what sorts of groups in civil society can be recognized in presenting material to Congress, or simply what liability to taxes they have. The formal and the informal will always be present, however much the informal institutions emerging from civil society appear the distinctive aspect of the open network participation regime institutions.

What sort of institutions do we find in this more open participation regime? The central constitutional institutions are complemented by what set of rules, procedures, and norms? Since this set is not set out in constitutional documents, and since there are many overlapping spheres in which these institutions take shape, it is difficult to suggest anything like a definitive list. One objective of this essay is to promote reflection leading to more consensus on such a list. As noted before, my tentative list of 'places to look' is the following: the legal framework, professionalization, the political service sector and consultative channels.

1) The legal and normative framework

The legal framework is the most basic institution supporting and shaping any broad, dynamic social activity. It is plausible, in fact, to consider the rule of law fundamental to civil society itself, and, conversely, to look to the law in order to understand the most basic way it impacts on politics and how the state is being shaped. This is a complex and wide ranging topic, but let me suggest a few ways normative/legal patterns shape how civil society organizations function in the public sphere.

First, the status of civil society organizations as legal entities is clearly crucial for the way in which they operate. Their tax status, financial reporting requirements and rules for registration are among the provisions of new laws in many countries in Latin America governing non-profit organizations or civil associations. These can be very restrictive or very open.

Laws regulating forms of mobilization of human, financial, informational and other resources is crucial. Some countries in Latin America seek to limit foreign fund raising. Some, like China, try to legally restrict the use of the internet. Others like the United States and Britain, seek to limit the role of non-profit organizations in political campaigns. Legal structures here will directly affect the impact of civil society on the polity.

It is important to explore not only ‘static‘ legal limits on actions, but the rules governing the emergence – and disappearance – of associations. From this dynamic point of view, one thinks of social movements. Social movement discourse moves back and forth between the model of (non-institutionalized) transformative organizations and (potentially institutionalized) organizations advocating the rights of an oppressed group or an insufficiently recognized public good. Even looking just at those associations that seek an on going representation rather than revolution, it is still quite possible to have a considerable turnover in the specific organizations and associations which claim this goal.

There are two stages of the process of group formation when legal structures may influence the operation of the system. The first occurs when groups achieve the legal right to organize a class of people. These struggles have been part of the history of movements for women, racial groups, minority religions, indigenous populations, homosexuals and many other groups. One groups like these are organized, the law may or may not facilitate more specialized ones within the existing grid of identities, or uniting existing ones into larger or cross-cutting formations. Not all women, but those who are single mothers, a broad ‘Indian’ group formed from many village based indigenous associations, a bio-diversity advocacy group out from under a broad environmental organization, and many other more specific groups are constantly forming. The law influences not only how new groups are formed, and also conditions how likely that older organizations disappear. Gaining legal personality and official recognition is obviously shaped by existing laws. But becoming the ‘official’ representative of some community, with a seat on policy councils and official status as representative of the country in international meetings may make it difficult to dissolve some organization and move on. Legal provisions can inhibit change as well as encourage it.

It is by no means always the case, of course, that the more turn over in civil society organizations is always good. Change may weaken the representation of groups as much as strengthen it, depending on a lot of factors beyond the law. The question of what legal provisions promote a just, stable, peaceful and effective polity cannot be answered at this abstract level. But it should be clear that the legal structure does play an important role.

A different level of institutionalization concerns not the law, but normative attitudes towards civil society organizations and networks. Beyond the obvious impact of ‘prejudices’ against poor people, racial or ethnic groups, which could easily distort the political impact of civil society, there is the question of public, or elite readiness to accept the legitimacy of civil society activism at all. One cannot move much in Latin America, for example without encountering a strong sentiment among many that ‘civil society’ (by which is meant chiefly non-governmental organizations) may have had an important role in overthrowing authoritarian regimes, but that they have no place in a well ordered state. Some will argue that they are inappropriate, for example, because they require a willingness of citizens to support with volunteer contributions and work – and that those are not part of the ‘Latin culture’.

But this is not a situation confined to statist cultures. In the US, arguably the country with the strongest tradition of politically relevant non-profit organizations and 'interest groups', the worries embodied in the terms 'special interests' and 'lobbies' indicates ambivalence, and periodic efforts to regulate and constrain. What ever the ideal of democratic practice is (and it will no doubt vary by context) this is an arena of institutionalization in the open network regime which warrants attention.

2) Professionalization of policy elites11

Professionalization takes different forms. For example, when we think of doctors and lawyers we think of the organizations they form which set standards, and control admission to the group, among other things. The terms also refers to the development of standards for the civil service, by people who have sought to implement merit as the criteria for public employment. A kind of professionalization was noted by Max Weber for political party cadres, where he seemed to be thinking mostly of their move from amateur to full time, remunerated employment status. In all these cases, a tie is established between people in different organizations doing similar things involving the application of some set of skills or use of a particular body of knowledge. As a device for building links among the professionals, it is straightforward. Here, the emphasis is on the no less important way in which it ‘filters’ demands and knowledge in the political process. In medicine, for example, the profession exerts an important influence on policy relevant to its field.

The argument here is first that a form of professionalization takes place among groupings of 'experts' or 'specialists' in many policy arenas. The jobs the ‘members’ have may be quite varied. For example, activists, consultants, bureaucrats and scholars concerned with health care policy have come to know each other, trade jobs sometimes, meet in conferences, read similar journals and perhaps may form an association. The argument is that similar developments in many countries, in varying degrees, take place in fields within the environment concerns, human rights, public security, judicial reform, social policy and many other fields. And secondly, I argue, the norms and standards, the commonly agreed upon latest information or best practices, acceptable schools for training, and many other aspect, become the ‘filter’ through which certain people, demands, policy ideas and information are shaped into the political process. Professionalization is not only a way of establishing standards for treating each other and 'clients', but is also becomes an institution shaping the way that civil society organizations and networks link to politics.

For the individual, professionalization involves heightened skill levels, the conversion of amateurism and service seen as an obligation of social status to paid careers, and a premium placed on mastery of a specific body of knowledge. From the system level, professionalization entails an ordering of a set of people with certain skills - the profession - and its relation to society. A strongly organized profession develops norms and practices which control access to the field, exerts discipline over those who fail to conform to minimum standards, and determines what set of principles will be accepted as governing, or what disagreements and debates will be accepted and made central to debates in the profession.

A profession requires a body of knowledge forming the core expertise, schools to train adepts, professional associations to call meetings and debate standards, and some capacity to control entry and enforce standards, formally or informally. Social policy experts or environmental economists may not reach the levels of self-conscious professionalization reached by doctors and lawyers, but through regular meetings the actors recognize each other, and develop codes of conduct. Conferences are held in which the body of knowledge is reviewed and members are brought up to date with the current acceptable state of knowledge. Specific schools become identified for approved training, and networks of older graduates recruit younger ones, sometimes from competing groups with divergent points of view organized into different ‘schools of thought’ on a particular policy question.

Thinking of professionalization as a set of second level political institutions requires looking at these arrangements as rules and procedures shaping the relations between citizens and the political process. It involves looking beyond the focus on professionalization in its 'internal' aspect, beyond the relation of the expert to their clientele, and to the way rules and expectations of a professionalized group of policy experts shape its political significance. Like the constitutional institutions, these rules will shape the way that preferences are formed and interests identified. They will influence who among the civil society organizations shall be heard, what weight will be given to their points of view, how conflict will be regulated and how debate will be carried through. Professional reputation will influence which interested parties will be heard by government policy makers, gain access to Congressional committees or be quoted in the press. "Peer review" will influence which projects and individuals get funding. Clientelistic links shaping access to powerful offices involving "extraneous" factors such as family ties or personal loyalties are replaced, for good or ill, with professional status and the support of 'those knowledgeable in the field."

The origin of this professionalization is no doubt varied. Perhaps mandated by law, but much more likely through a sort of self-organizing process of strong networking, mutual support around the 'latest approach' or the 'best practices', an effort to exclude 'oddballs' and disreputable opportunists, through maneuvers to gain primacy for one school of thought or style of action or monopolize funding from some external source, or for many other reasons, a more or less cohesive group forms. It controls access and identify the relevant specialized knowledge. These strengthened networks form among sets of people working in broad or narrow policy areas (e.g., social policy, or economic development or environmental concerns). The participants move back and forth in public and private organizations, national and international settings, between political and technical roles. What may start as simply 'networking', can become a powerful institution shaping what policy is acceptable to ‘the experts’, and what voices will be heard.

All of these functions of professionalized groupings might be performed in other ways in other participation regimes. Political parties as organizers of state-society relations would develop different means to screen participants in policy discussions and identify acceptable areas of expertise and directions for policy. Corporatist regimes would, in the pure form, at any rate, accomplish the same ends within bureaucratic organizations. Clientelist ones would presumably break up the groups which might form around policy specialties, since the clientelist links are built on other bases.

As I have emphasized, identifying second level institutions does not mean arguing that they always produce good results. Constitutional institutions, after all, may be designed to do good things (ensure a "more perfect union"), but in fact produce gridlock, injustice and fail to prevent instability or violence. Professionalization does have potential qualities which might make it an ideal institution to handle the contemporary characteristics of representative systems. Professionalization easily absorbs international actors, (and domestic ones now operating internationally). The fluid sort of associative networks can be accommodated easily in a professional framework. The deliberative aspect with the increasing prominence of information and debate in policy questions is, clearly central to professions, which are usually built around a body of knowledge.

But accommodating them does not mean that they produce uniformly good results along the dimensions of justice, stability, efficiency and peace. Professionalization can encourage meritocracy and innovation, but also they may lock in favoritism or repetition of old approaches to changing social problems. A professional association, however formal, may be elitist or open. Linking into foreign schools with their own curricula, social and cultural attitudes, may either open up policy making to the best thinking or be an instrument of a kind of intellectual imperialism, or maybe simply inappropriate technologies. The argument here is not that these secondary institutions make everything right. It is only that without looking at them, we will never be able to tell what works and what doesn’t.

3) A political service sector

Another place to look for the institutions which manage an open network participation regime, is among the agencies and firms servicing the organizations and networks that make up civil society. An institutionalized regime will be rich in organizations which are not active political agents themselves, but provide (sometimes sell) their services to those who are. 'Consulting' or 'service' organizations provide the expertise to set up an NGO, to streamline an office, to show new leaders the way to contact relevant government officials, to train grassroots activists, to set up websites, to conduct public opinion polls or focus groups around planned initiatives, to secure funding, or provide access to best-practice policies from around the world. The sector seems to grow with the specialization of functions, the diversity and turnover of the organizations of civil society, and the apparent legitimacy of spending money on such 'consultation and advice'.

In corporatist institutions such services are built into the official organizations. In a partisan system, parties develop such tools. In an open network participation regime, the demand exists and constraints on supply are low. Not only is ‘technical support’ not built into organizations into the relatively smaller organizations or the networks of an open regime, but the problem of finding the skills to create organizations are in much greater demand. In this open system, with its constant reconfiguration of networks and organizations, the most distinctive challenge from the point of view of society, is to make the process of creating new organizations and reorganizing old ones an orderly one which effectively adapts to new challenges, without instability or violence. Having savvy activists and leaders is no doubt very important in making this work (or in undermining it), but to institutionalize this sort of regime, the constellation of private and public organizations which provide these 'political services' are crucial.

One especially important part of this service sector are the funders and those that help groups find funding, whether it is skills in grassroots contributions, philanthropy, international organizations, private foundations or from local government itself. In other regimes, the issue of both project and organizational support is provided by tax monies or party revenues. In the open network regime, with its many non-profit and non-governmental organizations, funding is problematic, and the demand for assistance considerable.

In what sense is this set of service organizations an institution? The ones giving advice and technical support, such as the World Bank or private and foreign government supported foundations, will probably think of themselves as service providers, and so they are. Consulting firms and individual consultants that do polling or project management services or project evaluation, no doubt think of themselves as simply part of a new, but recognizable commercial market, and so they are. Funders think of themselves as promoting on particular projects, and so they do. But if we stand back and look at this group as a whole, they form a structure with general procedures and norms. As such they serve as another filter, or set of constraints on the way citizens and civil society groups interact with the state. Providing management services may be open to any organization that can pay - but that itself is a constraint. The availability of polling technology provides a way of cutting through presumptions about public opinion, but privileges a certain level of surface attitudes.

The international involvement in this sector, particularly in funding, is particularly controversial. As we have said, a major characteristic of contemporary politics in almost all countries is the international presence. In this service area, it has been pronounced. Given the important role of international donors in shaping the sorts of projects and the conditions under which they are undertaken, a lively yet inconclusive discussion has taken place. To what extent have they shaped politics and policies? Do they distort the process of development, or judicial reform or protecting biodiversity? International actors are important in more than funding. Management services, project evaluation, and survey techniques, to name a few, have been largely imported in areas such as Latin America. Although there are signs that this is changing as the skills for these services are found locally (still often among people trained abroad), the significance of meshing with international norms, styles and interests remains a question.

Let me emphasize that I am not trying to offer answers to these questions, which are, in any case, much too broad to admit of any single answer, but rather, to point out that in an open network regime, to assess whether or not the institutions are promoting democracy one must analyze the institution inherent in the pattern of the political service sector.

4) Consultations and policy forums

Finally let us examine how members of civil society are heard in authoritative decision making. Patterns in the conduct of lobbying, legislative hearings, commissions, forums, roundtables, and workshops are relevant if they involve the policy makers and representatives of civil society. The rules that govern these institutions have a lot to do with filtering out or enhancing connections. At stake, however, is not only who gets heard, but how the debate is structured. What sort of information is brought to bear on policy questions, what approaches are considered important, and what decision rules are adopted. Who gives advice, who has a vote, whether a consensus needs to be reached, whether factions are encouraged or discouraged by the rules – these are all consequential aspects of second level institutions in this arena.

In other participation regimes, government ministries and political parties may provide arenas - conferences, working groups, conventions, meetings, ‘smoke filled rooms’ - where policy deliberation among relevant actors take place outside of the legislature. In the open regime, similar discussions are held, bringing together government officials, businessmen, neighborhood or local representatives, non-profit or NGO representatives, international experts, 'think-tank' representatives and others without the benefit of an overarching organization within which to conduct such events. This form of communication might involve regular consultations of 'leading experts' by the governmental officials who have responsibility for decisions in a certain area. It might consist in legislative hearings. It might take the form of official commissions including members of civil society appointed for specific questions. Holding conferences or conducting a forum, even some academic conferences offer other ways of organizing access to decision making.

From the outside, these consultations and meetings seem relatively ad hoc and episodic, developed for specific purposes, as when a government agency calls a meeting of experts on poverty to meet a threat of violence or possible sanctions from international donors. Anyone involved in the meetings and conferences – for example the convention at which this paper is being presented - knows that there are rules and rituals involved. There is also a question of who is invited and why, and with what degree of respect and voice. If the issue demands some sort of action, the manner of shaping the sense of the meeting and the recommendations is relevant to what impact it will have. The insistence upon conformity to a particular approach (or credit given to innovation) is an other one of the many ways in which outcomes are shaped. To the extent that such regular patterns occur, perhaps even regulated by legal or administrative arrangements, they constitute the institutionalization of state-civil society relations in an open network regime, and ought to be analyzed for the impact on democracy.

I have tried to emphasize the shaping of the deliberative process over the (in the end equally important) question of which people, which interests are invited and given a voice. The patterns of inclusion and exclusion are patently relevant to evaluating the impact of these second level institutions on democracy. The impact of the rules governing deliberation are nowhere near so obvious, but none the less significant for that. Much analysis is needed to assess them.


The proliferation of non-governmental organizations less controlled by traditional patrons and bosses, that is in part maintained by the internationalization of politics, sets challenges for those who would understand the institutional requirements for democracy. A society with free elections, a working legislature, and a president who submits to constitutional limits and the rule of law are all crucial. But we know all too well that they may all exist without democracy. They are necessary elements, but not sufficient to ensure it. Society must be democratic, too – more equal distribution of income, protection of individual rights, freedom of expression. But stating these truths avoids another set of problems, that the central constitutional institutions only cover a small part of the political processes going on. The second level of institutions that I have been discussing have a large share in deciding who is to be heard, what weight is to be given to opinion, theory and facts, who shall lead and with what authority. As I have said, these institutions, depending on how they are configured and what the conditions are, can have both positive and negative effects for democracy. But unless we ‘complete’ the analysis of political institutions, we will not be in a position to assess whether the political structure is democratic or not.

So I return to the question I started with. What institutions, under what conditions produce democratic outcomes in this kind of participation regime? I have tried to indicate the road that will have to be taken in order to answer it.



Prepared for the XXII International Congress of the Latin American Studies Association, Miami, March 2000. I want to thank the participants in the Faculty Comparative Politics Seminar at Columbia University and those at the conference on "International And Comparative Politics: New Perspectives On Brazil" held in Brasilia, Nov. 1999, for comments and suggestions on an earlier version.

2 I call these ‘constitutional institutions’ for the obvious reason that they are typically specified by constitutional documents. Sometimes they are called the ‘formal’ institutions of democracy, although formality is a characteristic of all institutions. I include party systems in the category of constitutional institutions, even though they are often not included in constitutional documents. I do so because, and to the extent that, they organize elections, parliamentary discussions and legislative-executive relations. As I will indicate later, political parties as organizers of political society, so to speak, should be included in second level institutions.

3 Patterns such as these are the working tools of those who would describe 'power elites'. The case for such a 'hidden elite' does, I think, turn on very strong, oligarchic second level institutions, but there are many other possible configurations, as we will see below.

4 These are not only second level institutions. Clientelism, and especially corporatism and partisan organization is often used to designate a ‘regime type’ at the top level, as when corporatism is defined in terms of a specific way of organizing the legislature. Here I am referring to its use at this ‘secondary level’. See below for further elaboration of this point.

5 The reference is to the model of the polity in the ‘structural functionalism’ presented in Almond, Gabriel and James Coleman, The Politics of Developing Areas, Princeton, N.J., Princeton University Press, 1960. ‘Functionalism’ is treated here, one might note, not as an explanatory strategy, but as a guide for identifying important dynamics in society. Although theorists are now rather careful to avoid explaining why structures like institutions exist in terms of the functions they serve, they are still likely to accept some model of functions in order to identify what institutions are worth studying. My intention here is to do just that, but to criticize the existing functional model before doing so.

6 Chalmers, Douglas A,. Scott B. Martin, and Kerianne Piester "Associative Networks: New Structures of Representation for the Popular Sectors?" from Chalmers, et al, The New Politics of Inequality in Latin America, Oxford University Press (Oxford, 1997) pp543-582. While associative networks can and often do involve actors with sharply unequal resources, there are typically more chances to escape or shift the ground to avoid a direct test of strength with an unequal competitor. These chances derive, we believe from the lesser importance of rigid, hierarchical authority relations (compared with party or corporatist forms), the shifting and multiple patterns of identity (compared with clientelism and populism) and the more open-ended character of cognitive politics. This results not so much in more equality as less rigid inequality among the participants.

7 See my paper, "What Is It About Associations in Civil Society That Promotes Democracy?", an occasional paper from the Institute of Latin American and Iberian Studies, at

8 We could alternatively call these types 'intermediation' or ‘representation’ regimes. I will use 'participation regimes’ because it keeps our eye on the citizen.

9 It may be that just as clientelistic structures are highly varied, the structures of open network regimes may be varied and we have only to perceive the unifying aspect of all such second level institutions in them as we believe we have done with ‘clientelism’. It would be a theoretical advance to replace ‘open network regime’ with something more positive. But, for now, it would be premature.

10 This addresses one of the controversies surround Robert Putnam’s work on social capital. Increasing social capital can, under the right conditions, make fascism work, too, as it did in Northern Italy. Democratically operating second level institutions, I argue, are part of those ‘right conditions’.


1 Monique Segarra’s work on the importance of professionalization is key here. In her forthcoming work she describes the professionalization of social policy experts in Ecuador.