Non-state Organization

Margaret E. Keck is Associate Professor of Political Science, Johns Hopkins Uni- versity, 338 Mergenthaler Hall, Balti- more, MD 21218, USA, email: mkeckKjhu.edu She is author ofThe Workers’ Party and Democratisation in Brazil (1992) andPT: A Lógica da Difer- ença (in Portuguese) (1991). Kathryn Sikkink is Professor of Political Science, University of Minnesota, 1414 Social Science, 267 19th Avenue South, Minne- apolis, MN 55455, USA, email: KsikkinkKpolisci.umn.edu She is author of Ideas and Institutions: Development- alism in Brazil and Argentina(1991).

Transnational advocacy networks in international and regional politics*

Margaret E. Keck and Kathryn Sikkink

World politics at the end of the twentieth cen- tury involves, alongside states, many non-state actors who interact with each other, with states, and with international organizations. This article considers how these interactions are structured in networks, which are increasingly visible in international politics. Some involve economic actors and firms. Some are networks of scien- tists and experts whose professional ties and ideas underpin their efforts to influence policy (Haas, 1992). Others are networks of activists, dis- tinguishable largely by the centrality of principled ideas or values in motivat- ing their formation. We call these transnational advo- cacy networks.

Advocacy networks are significant transnation- ally, regionally and dom- estically. They may be key contributors to a conver- gence of social and cultural norms able to support pro- cesses of regional and inter- national integration. By building new links among actors in civil societies, states and inter- national organizations, they multiply the oppor- tunities for dialogue and exchange. In issue areas such as the environment and human rights, they also make international resources available to new actors in domestic political and social struggles. By thus blurring the boundaries between a state’s relations with its own nation- als and the recourse both citizens and states have to the international system, advocacy net-

ISSJ 159/1999 UNESCO 1999. Published by Blackwell Publishers, 108 Cowley Road, Oxford OX4 1JF, UK and 350 Main Street, Malden, MA 02148, USA.

works are helping to transform the practice of national sovereignty.

Scholars have been slow to recognize either the rationality or the significance of acti- vist networks. Motivated by values rather than by material concerns or professional norms, they fall outside our accustomed categories. Yet more than other kinds of transnational networks, advocacy networks often reach beyond policy change to advocate and instigate changes in the

institutional and principled bases of international inter- actions. When they succeed, they are an important part of an explanation for changes in world politics. A trans- national advocacy network includes those actors work- ing internationally on an issue, who are bound together by shared values, a common discourse, and dense exchanges of infor- mation and services.1 Such networks are most prevalent in issue areas characterized

by high value content and informational uncer- tainty, although the value-content of an issue is both a prerequisite and a result of network activity. At the core of the relationship is infor- mation exchange. What is novel in these net- works is the ability of non-traditional inter- national actors to mobilize information strategically to help create new issues and categ- ories, and to persuade, pressurize, and gain leverage over much more powerful organiza- tions and governments. Activists in networks

90 Margaret E. Keck and Kathryn Sikkink

try not only to influence policy outcomes, but to transform the terms and nature of the debate. They are not always successful in their efforts, but they are increasingly important players in policy debates at the regional and inter- national level.

Simultaneously principled and strategic actors, transnational advocacy networks ‘frame’ issues to make them comprehensible to target audiences, to attract attention and encourage action, and to ‘fit’ with favourable institutional venues. By framing, we mean ‘conscious strat- egic efforts by groups of people to fashion shared understandings of the world and of them- selves that legitimate and motivate collective action’ (McAdam et al., 1996, p. 6). Network actors bring new ideas, norms and discourses into policy debates, and serve as sources of information and testimony. Norms ‘describe col- lective expectations for the proper behaviour of actors with a given identity’ (Katzenstein, 1996, p. 5; see also Klotz, 1995; Finnemore, 1996).

Shared norms often provide the foundation for more formal institutional processes of regional integration. In so far as networks pro- mote norm convergence or harmonization at the regional and international levels, they are essen- tial to the social and cultural aspects of inte- gration. They also promote norm implemen- tation, by pressuring target actors to adopt new policies, and by monitoring compliance with regional and international standards. As far as is possible, they seek to maximize their influ- ence or leverage over the target of their actions. In doing this they contribute to changing the perceptions that both state and societal actors may have of their identities, interests and prefer- ences, to transforming their discursive positions, and ultimately to changing procedures, policies and behaviour. We thus believe, with Finne- more, that ‘States are embedded in dense net- works of transnational and international social relations that shape their perceptions of the world and their role in that world. States are socialized to want certain things by the inter- national society in which they and the people in them live’ (Finnemore, 1996, p. 2).

Networks arecommunicative structures. To influence discourse, procedures and policy, transnational advocacy networks may become part of larger policy communities that group actors from a variety of institutional and value

 UNESCO 1999.

positions. Transnational advocacy networks may also be understood as political spaces, in which differently situated actors negotiate – formally or informally – the social, cultural and political meanings of their joint enterprises. In both of these ways, transnational networks can be key vehicles for the cultural and social negotiations underpinning processes of regional integration.

We refer to transnational networks (rather than coalitions, movements, or civil society) to evoke the structured and structuring dimension in the actions of these complex agents. By importing the network concept from sociology and applying it transnationally, we bridge the increasingly artificial divide between inter- national relations and comparative politics. Moreover, the term ‘network’ is already used by the actors themselves; over the last two decades, individuals and organizations have consciously formed and named networks, developed and shared networking strategies and techniques, and assessed the advantages and limits of this kind of activity. Scholars have come late to the party.

Our theoretical apparatus draws upon sociological traditions that focus on complex interactions among actors, on the intersubjective construction of frames of meaning, and on the negotiation and malleability of identities and interests. These have been concerns of con- structivists in international relations theory and of social movement theorists in comparative politics, and we draw from both traditions. The networks we study participate simultaneously in domestic and international politics, drawing upon a variety of resources,as if they were part of an international society. However, they use these resources strategically to affect a world of states and international organizations constructed by states. Both these dimensions are essential. Rationalists will recognize the langu- age of incentives and constraints, strategies, institutions and rules, while constructivists and social constructionists will be more comfortable with our emphasis on norms, social relations and intersubjective understandings. We are con- vinced that both matter; whilst recognizing that goals and interests are not exogenously given, we can think about the strategic activity of actors in an intersubjectively structured political universe. The key to doing so is remembering that the social and political contexts within

91Transnational advocacy networks

which networks operate contain contested understandings as well as stable and shared ones. Network activists can operate strategically within the more stable universe of shared under- standings at the same time as they try to reshape certain contested meanings.

Part of what is so elusive about networks is how they seem to embody elements of agent and structure simultaneously. Our approach must therefore be both structural and actor- centred. We address five main questions:

(1) What is a transnational advocacy network? (2) Why and how do they emerge? (3) How do they work? (4) Under what conditions can they be effec-

tive – that is, when are they most likely to achieve their goals?

(5) What are the implications of network activi- ties for the social and cultural processes of regional integration?

Although we had initially expected that transnational networks would function in quite different ways from domestic social movements, we found that many of the characteristic stra- tegies, tactics and patterns of influence resembled those outlined in the literature on social movements. Organizations and individ- uals within advocacy networks are political entrepreneurs, mobilize resources like infor- mation and membership, and show a sophisti- cated awareness of the political opportunity structures within which they operate (Tarrow, 1994). Our emphasis on the role of values in networks is consistent with some arguments contained in the literature on ‘new social move- ments’ (Dalton et al., 1990). Most importantly, however, over the last decade social movement theory has increasingly focused on the inter- action between social–structural conditions and action, on the social context of mobilization, and on the transformation of meanings among activists and among mass publics that make people believe they can have an impact on an issue.

What is a transnational advocacy network?

Networks are forms of organization charac- terized by voluntary, reciprocal and horizontal

 UNESCO 1999.

patterns of communication and exchange. Organizational theorist Walter Powell calls them a third mode of economic organization, dis- tinctly different from markets and hierarchy (the firm). ‘Networks are “lighter on their feet” than hierarchy’ and are ‘particularly apt for circum- stances in which there is a need for efficient, reliable information. . .’, and ‘for the exchange of commodities whose value is not easily meas- ured’ (Powell, 1990, pp. 295–6, 303–4). His insights into economic networks are extraordi- narily suggestive for an understanding of polit- ical networks. Policy networks also form around issues where information plays a key role, and around issues where the value of the ‘com- modity’ is not easily measured.

In spite of differences between the dom- estic and international realms, the network con- cept travels well because it stresses the fluid and open relations among committed and knowledgeable actors working in specialized issue areas. We call them advocacy networks because advocates plead the causes of others or defend a cause or proposition; they are stand- ins for persons or ideas. Advocacy captures what is unique about these transnational net- works – they are organized to promote causes, principled ideas and norms, and often involve individuals advocating policy changes that can- not be easily linked to their ‘interests’.

Some issue areas reproduce transnationally the webs of personal relationships that are cru- cial in the formation of domestic networks. Advocacy networks have been particularly important in value-laden debates over human rights, the environment, women, infant health, and indigenous peoples. These are all areas where through personal, professional and organizational contexts, large numbers of differ- ently situated individuals became acquainted with each other over a considerable period, and developed similar world views. When the more visionary among them proposed strategies for political action around apparently intractable problems, the potential was transformed into an action network.

Major actors in advocacy networks may include the following:

(1) international and domestic NGOs, research and advocacy organizations;

(2) local social movements; (3) foundations;

92 Margaret E. Keck and Kathryn Sikkink

(4) the media; (5) churches, trade unions, consumer organiza-

tions, intellectuals; (6) parts of regional and international

intergovernmental organizations; (7) parts of the executive and/or parliamentary

branches of governments.

Not all these will be present in each advo- cacy network. Initial research suggests, how- ever, that international and domestic non- governmental organizations (NGOs) play a cen- tral role in most advocacy networks, usually initiating actions and pressuring more powerful actors to take positions. NGOs introduce new ideas, provide information, and lobby for pol- icy changes.

Social scientists have barely addressed the political role of activist NGOs assimultaneously domestic and international actors. There is a literature on NGOs and networks in specific countries (Fru¨ hling, 1991; Scherer-Warren, 1993). Much of the existing literature on NGOs comes from development studies, and either ignores interactions with states or spends little time on political analysis (see, for example, Korten, 1990). Examining their role in advocacy networks helps both to distinguish NGOs from, and to see their connections with, social move- ments, state agencies and international organiza- tions.

Groups in a network share values and fre- quently exchange information and services. The flow of information among actors in the net- work reveals a dense web of connections among these groups, both formal and informal. The movement of funds and services is especially notable between foundations and NGOs, but some NGOs provide services such as training for other NGOs in the same, and sometimes other, advocacy networks. Personnel also circu- late within and among networks.

Relationshipsamong networks within and between issue areas are similar to those that scholars of social movements have found in the case of domestic activism. Individuals and foundation funding have moved back and forth among them. Environmentalists and women’s groups have looked at the history of human rights campaigns for models of effective inter- national institution-building. Because of these interactions, refugee resettlement and indigen- ous peoples’ rights are increasingly central

 UNESCO 1999.

components of international environmental activity, and vice versa; mainstream human rights organizations have joined the campaign for women’s rights. Some activists consider themselves part of an ‘NGO community’. This convergence highlights important dimensions that these networks share: the centrality of values or principled ideas, the belief that indi- viduals can make a difference, creative use of information, and the employment by non- governmental actors of sophisticated political strategies in targeting their campaigns. Besides sharing information, groups in networks create categories or frames within which to organize and generateinformation on which to base their campaigns. The ability to generate information quickly and accurately, and deploy it effec- tively, is their most valuable currency; it is also central to their identity. Core campaign organizers must ensure that individuals and organizations with access to necessary infor- mation are incorporated into the network; differ- ent ways of framing an issue may require quite different kinds of information. Thus, frame dis- putes can be a significant source of change within networks.

Why and how transnational advocacy networks have emerged?

The kinds of groups characteristic of advocacy networks are not new; some have existed since the nineteenth century campaign for the abol- ition of slavery. Nevertheless, their number, size, professionalism, and the density and com- plexity of their international linkages have grown dramatically in the last three decades, so that only recently can we speak oftransnational advocacy networks (Keck and Sikkink, 1998).

International networking is costly. Geo- graphical distance, nationalism, the multiplicity of languages and cultures, and the costs of fax, telephone, mail, or air travel make the prolifer- ation of international networks a puzzle that needs explanation. Under what conditions are networks possible and likely, and what triggers their emergence?

Transnational advocacy networks appear most likely to emerge around those issues where:

93Transnational advocacy networks

(1) channels between domestic groups and their governments are hampered or severed where such channels are ineffective for resolving a conflict, setting into motion the ‘boomerang’ pattern of influence character- istic of these networks;

(2) activists or ‘political entrepreneurs’ believe that networking will further their missions and campaigns, and actively promote them;

(3) international conferences and other forms of international contacts create arenas for forming and strengthening networks.

The boomerang pattern

It is no accident that ‘rights’ claims may be the prototypical language of advocacy networks. Governments are the primary ‘guarantors’ of rights, but also among their primary violators. When a government violates or refuses to recog- nize rights, individuals and domestic groups often have no recourse within domestic political or judicial arenas. They may seek international connections to express their concerns and even to protect their lives.

Many transnational advocacy networks link activists in developed countries with others in or from less developed countries. These kinds of linkages are most commonly intended to affect the behaviour ofstates. When the links between state and domestic actors are severed, domestic NGOs may directly seek international allies to try to bring pressure on their states from outside. This is the ‘boomerang’ pattern of influence characteristic of transnational net- works where the target of their activity is to change a state’s behaviour. This is most com- mon in human rights campaigns. Similarly, indigenous rights campaigns, and environmental campaigns supporting the demands of local peoples for participation in development pro- jects that would affect them, frequently involve this kind of triangulation. Where governments are unresponsive to groups whose claims may none the less resonate elsewhere, international contacts can ‘amplify’ the demands of domestic groups, pry open space for new issues, and then echo these demands back into the domestic arena. Needless to say, in such cases the use of a boomerang strategy is politically sensitive, and is subject to charges of foreign interference in domestic affairs.

 UNESCO 1999.

Linkages are important for both sides. For the less powerful Third World actors, networks provide access, leverage and information (and often money) they could not expect to have on their own. For northern groups, they make cred- ible the assertion that they are struggling with, and not only ‘for’, their southern partners. Not surprisingly, such relationships can produce considerable tensions. It is not uncommon to see reproduced internally the power relations that the networks are trying to overcome. Increasingly, network members are forced to address this problem.

Just as injustice and oppression may not produce movements or revolutions by them- selves, claims around issues amenable to inter- national action need not produce transnational networks. Activists are ‘people who care enough about some issue that they are prepared to incur significant costs and act to achieve their goals’ (Oliver and Marwell, 1992, p. 252). They form networks when they believe it will further their organizational missions – by shar- ing information, attaining greater visibility, gaining access to different publics, multiplying channels of institutional access, and so forth.

Networks are normally formed around particular campaigns or claims. Networks breed networks; as networking becomes a repertoire of action that is diffused transnationally, each effort to network internationally is less difficult than the one before. Over time, in these issue areas, participation in transnational networks has become an essential component of the collective identities of the activists involved. The political entrepreneurs who become the core networkers for a new campaign have often gained experi- ence in earlier ones.

Opportunities for network activities have increased over the last two decades, in part through the efforts of the pioneers among them. Network activists have been creative in finding new venues in which to pursue claims – a process we discuss in the next section. The proliferation of international organizations and conferences has provided foci for the contacts. Cheaper air travel and new electronic and com- munication technologies speed information flows and simplify personal contact among them.

Underlying the trends discussed here, how- ever, is a broader cultural shift. The new net-

94 Margaret E. Keck and Kathryn Sikkink

Greenpeace in action.Guelbiz/Action Press

works depended on creating a new kind of glo- bal public (or civil society), which grew as a cultural legacy of the 1960s. The activism that swept western Europe, the United States, and many parts of the Third World during that dec- ade contributed to this shift, alongside the vastly increased opportunities for international contact.

Obviously, internationalism was not invented in the 1960s. Several long-standing ethical traditions have justified actions by indi- viduals or groups outside the borders of their own state. Broadly speaking, we could designate these as religious beliefs, the solidarity tra- ditions of labour and the left, and liberal inter- nationalism. While many activists working in advocacy networks are from one of these tra- ditions, they no longer tend to define themselves in terms of these traditions or the organizations that carried them. This is most true for activists on the left, for whom the decline of socialist organizations capped a growing disillusionment with much of the left’s refusal to address seri- ously the concerns of women, the environment,

 UNESCO 1999.

and human rights violations in eastern bloc countries.

Advocacy networks in the north often func- tion in a cultural milieu of internationalism that is generally optimistic about the promise and possibilities of international networking. For network members in developing countries, how- ever, justifying external intervention or pressure in domestic affairs is a much trickier business, except when lives are at stake. Linkages with northern networks require high levels of trust, because arguments justifying intervention on ethical grounds often sound too much like the ‘civilizing’ discourse of colonial powers, and can work against the goals they espouse by producing a nationalist backlash.

How do transnational advocacy networks work?

Transnational networks seek influence in many of the same ways that other political groups or

95Transnational advocacy networks

social movements do, but because they are not powerful in the traditional sense of the word, they must use the power of their information, ideas and strategies to alter the information and value context within which states make policies. Although much of what networks do might be considered persuasion, the term is insufficiently precise to be of much theoretical use. We have developed a more nuanced typology of the kinds of tactics that networks use. These include:

(a) information politics, or the ability to move politically usable information quickly and credibly to where it will have the most impact;

(b) symbolic politics,or the ability to call upon symbols, actions or stories that make sense of a situation or claim for an audience that is frequently far away (see also Brysk, 1994, 1995);

(c) leverage politics, or the ability to call upon powerful actors to affect a situation where weaker members of a network are unlikely to have influence; and

(d) accountability politics, or the effort to oblige more powerful actors to act on vaguer policies or principles they formally endorsed.

The construction of cognitive frames is an essential component of transnational networks’ political strategies. David Snow has called this strategic activityframe alignment– ‘by render- ing events or occurrences meaningful, frames function to organize experience and guide action, whether individual or collective’ (Snow et al., 1986). Frame resonanceconcerns the relationship between an organization’s inter- pretive work and its ability to influence broader public understandings. The latter involves both the frame’s internal coherence and its fit with a broader political culture (Snow and Benford, 1988). In recent work, Snow and Benford (1992) and Tarrow (1992), in turn, have given frame resonance a historical dimension by join- ing it to Tarrow’s notion of protest cycles. Struggles over meaning and the creation of new frames of meaning occur early in a protest cycle, but over time, ‘a given collective action frame becomes part of the political culture – which is to say, part of the reservoir of symbols from which future movement entrepreneurs can choose’ (Tarrow, 1992, p. 197).

 UNESCO 1999.

Network members actively seek ways to bring issues to the public agenda, both by fram- ing them in innovative ways and by seeking hospitable venues. Sometimes they create issues by framing old problems in new ways; occasionally they help to transform other actors’ understandings of their identities and their inter- ests. Land-use rights in the Amazon, for example, took on an entirely different character and gained quite different allies when viewed in a deforestation frame than in either social justice or regional development frames.

Transnational networks normally involve a small number of activists in a given campaign or advocacy role. The kinds of pressure and agenda politics in which they engage rarely involve mass mobilization, except at key moments, although the peoples whose cause they espouse may engage in mass protest (for example, the expelled population in the Narm- ada Dam case). Boycott strategies are a partial exception. Instead, network activists engage in what Baumgartner and Jones (1991), borrowing from law, call venue shopping: ‘This strategy relies less on mass mobilisation and more on the dual strategy of the presentation of an image and the search for a more receptive political venue’ (p. 1050). The recent coupling of indigenous rights and environmental struggles is a good example of a strategic venue shift by indigenista activists, who found the environ- mental arena more receptive to their claims than had been human rights venues.

Information politics

Information binds network members together and is essential for network effectiveness. Many information exchanges are informal – through telephone calls, e-mail and fax communications, and the circulation of small newsletters, pam- phlets and bulletins. They provide information that would not otherwise be available, from sources that might not otherwise be heard, and make it comprehensible and useful to activists and publics who may be geographically and/or socially distant.

Non-state actors gain influence by serving as alternative sources of information. Infor- mation flows in advocacy networks provide not only facts, but alsotestimonies– stories told by people whose lives have been affected.

96 Margaret E. Keck and Kathryn Sikkink

Moreover, they interpret facts and testimony; activist groups frame issues simply, in terms of right and wrong, because their purpose is to persuade people and stimulate them to take action.

How does this process of persuasion occur? An effective frame must show that a given state of affairs is neither natural nor accidental, identify the responsible party or parties, and propose credible solutions. This requires clear, powerful messages that appeal to shared prin- ciples, and which often have more impact on state policy than the advice of technical experts. An important part of the political struggle over information is whether an issue is defined prim- arily as technical, subject to consideration by ‘qualified’ experts, or as something that con- cerns a much broader global constituency.

Even as we highlight the importance of testimony, however, we have to recognize the mediations involved. The process by which tes- timony is discovered and presented normally involves several layers of prior translation. Transnational actors may identify what kinds of testimony would be valuable, then ask an NGO in the area to seek out people who could tell those stories. They may filter through expatri- ates, through travelling scholars, through the media. There is frequently a huge gap between the story’s telling and its retelling – in sociocul- tural context, in instrumental meaning, and even in language. Local people, in other words, sometimes lose control over their stories in a transnational campaign.

Non-governmental networks have helped to legitimize the use of testimonial information along with technical and statistical information. Linkage of the two is crucial: without the indi- vidual cases, activists cannot motivate people to seek to change policies. Increasingly, inter- national campaigns by networks take this two- level approach to information. In the 1980s even Greenpeace, which initially had eschewed rigor- ous research in favour of splashy media events, began to pay more attention to getting the facts right. While testimony does not avoid the need to manage technical information, it helps to make the need for action more real for ordi- nary citizens.

A dense web of north–south exchange, aided by computer and fax communication, means that governments can no longer mono-

 UNESCO 1999.

polize information flows as they could a mere half-decade ago. These technologies have had an enormous impact on moving information to and from Third World countries, where mail services are often both slow and precarious. We should note, however, that this gives special advantages to organizations that have access to such technologies.

The central role of information in all these issues helps to explain the drive to create net- works. Information in these issue areas is both essential and dispersed. Non-governmental actors depend upon their access to information to help make them legitimate players. Contact with like-minded groups at home and abroad provides access to information necessary to their work, broadens their legitimacy, and helps to mobilize information around particular policy targets. Most NGOs cannot afford to maintain staff in a variety of countries. In exceptional cases, they send staff members on investigation missions, but this is not practical for keeping informed on routine developments. Forging links with local organizations allows groups to receive and monitor information from many countries at low cost. Local groups, in turn, depend on international contacts to get their information out, and to help to protect them in their work.

The media are essential partners in network information politics. To reach a broader audi- ence, networks strive to attract press attention. Sympathetic journalists may become part of the network, but more often network activists culti- vate a reputation for credibility with the press, and package their information in a timely and dramatic way to draw press attention.

Symbolic politics

Activists frame issues by identifying and pro- viding convincing explanations for powerful symbolic events, which in turn become catalysts for the growth of networks. Symbolic interpret- ation is part of the process of persuasion by which networks create awareness and expand the constituency. Awarding the Nobel Peace Prize to Rigoberta Menchu, during the Inter- national Year of Indigenous People, heightened public awareness of the situation of indigenous peoples in the Americas. The ability of the indigenous people’s movement to use 1992, the

97Transnational advocacy networks

500th anniversary of the voyage of Columbus to the Americas, to raise a host of indigenous issues revealed the ability of networks to use symbolic events to reshape understandings (Brysk, 1994).

The coup in Chile played this kind of cata- lytic role for the human rights community. Often it is not one event, but the juxtaposition of disparate events that makes people change their minds and take action. For many people in the US, it was the juxtaposition of the coup in Chile, the war in Vietnam, Watergate, and civil rights that gave birth to the human rights movement. Likewise, the juxtaposition of the hot summer of 1988 in the US with dramatic footage of the Brazilian rainforest burning may have convinced many people that global warm- ing and tropical deforestation were serious and linked issues. The assassination of Chico Mendes at the end of that year crystallized the belief that something was profoundly wrong in the Amazon.

Leverage politics

Activists in advocacy networks are concerned with political effectiveness. Their definition of effectiveness often involves some policy change by ‘target actors’ which might be governments, but might also be international financial insti- tutions like the World Bank, or private actors like transnational corporations. In order to bring about policy change, networks need to both persuade and pressurize more powerful actors. To gain influence the networks seekleverage– a word that appears often in the discourse of advocacy organizations – over more powerful actors. By exerting leverage over more powerful institutions, weak groups gain influence far beyond their ability to influence state practices directly. Identifying points of leverage is a cru- cial strategic step in network campaigns. We discuss two kinds of leverage: material leverage and moral leverage.

Material leverage usually takes the form of some kind of issue-linkage, normally involving money or goods (but potentially also including votes in international organizations, prestigious offices, or other benefits). The human rights issue became negotiable because other govern- ments or financial institutions connected human rights practices to the cut-off of military and

 UNESCO 1999.

economic aid, or to worsening bilateral diplo- matic relations. Human rights groups obtained leverage by providing US and European policy- makers with information that persuaded them to cut off military and economic aid. To make the issue negotiable, NGOs first had to raise its profile or salience, using information and symbolic politics. Then more powerful members of the network had to link cooperation to some- thing else of value: money, trade or prestige. Similarly, in the environmentalists’ multilateral bank campaign, linkage – of environmental pro- tection with access to loans – was very power- ful.

Moral leverage involves what some com- mentators have called the ‘mobilisation of shame’, where the behaviour of target actors is held up to the bright light of international scru- tiny. Where states place a high value on inter- national prestige, this can be effective. In the baby-food campaign, network activists used moral leverage to convince states to vote in favour of the WHO/UNICEF Codes of Conduct. As a result, even the Netherlands and Switzer- land, both major exporters of infant formula, voted in favour of the code.

Although NGO influence often depends on securing powerful allies, making those links still depends on their ability to mobilize the soli- darity of their members, or of public opinion via the media. In democracies, the potential to influence votes gives large membership organi- zations an advantage in lobbying for policy change; environmental organizations, several of whose memberships number in the millions, are more likely to have this added clout than are human rights organizations.

Accountability politics

Networks devote considerable energy to con- vincing governments and other actors to change their positions on issues. This is often dismissed as inconsequential change, since talk is cheap – governments change discursive positions hoping to divert network and public attention. Network activists, however, try to make such statements into opportunities for accountability politics. Once a government has publicly committed itself to a principle – for example, in favour of human rights or democracy – networks can use those positions, and their command of infor-

98 Margaret E. Keck and Kathryn Sikkink

mation, to expose the distance between dis- course and practice. This is embarrassing to many governments, who may try to save face by closing the distance.

Under what conditions do advocacy networks have influence?

To assess the influence of advocacy networks we must look at goal achievement at several different levels. We identify the following types or stages of network influence:

(1) issue creation and attention/agenda setting; (2) influence on discursive positions of states

and regional and international organizations; (3) influence on institutional procedures; (4) influence on policy change in ‘target actors’

which may be states, international or regional organizations, or private actors like the Nestle´ corporation;

(5) influence on state behaviour.

Networks generate attention to new issues and help to set agendas when they provoke media attention, debates, hearings and meetings on issues that previously had not been a matter of public debate. Because values are the essence of advocacy networks, this stage of influence may require a modification of the ‘value con- text’ in which policy debates take place. The theme years and decades of the United Nations, such as International Women’s Decade and the Year of Indigenous People, were international events promoted by networks that heightened awareness of issues.

Networks influence discursive positions when they help to persuade states and inter- national organizations to support international declarations or change stated domestic policy positions. The role that environmental networks played in shaping state positions and conference declarations at the 1992 Earth Summit in Rio de Janeiro is an example of this kind of impact. They may also pressurize states to make more binding commitments by signing conventions and codes of conduct.

At a more concrete level, the network has influence if it leads to changes in policies, not only of the target states, but also of other states

 UNESCO 1999.

and/or international institutions. These changes are easier to see, but their causes can be elusive. We can speak of network impact on policy change where human rights networks have pressured successfully for cut-offs of military aid to repressive regimes, where repressive prac- tices diminish because of pressure, or even where human rights activity affects regime change or stability. We must take care to dis- tinguish between policy change and change in behaviour; official policies may predict nothing about how actors behave in reality.

We speak of stages of impact, and not merely types of impact, because we believe that increased attention and changes in discursive positions make governments more vulnerable to the claims these networks raise. This is not always true, of course – discursive changes can also have a powerfully divisive effect on net- works, splitting insiders from outsiders, reformers from radicals. None the less, a government that claims to be protecting indigen- ous areas or ecological reserves is more vulner- able to charges that such areas are endangered than one that makes no such claim. Then, the effort is no longer to make governments change their position, but to hold them to their word. Meaningful policy and behavioural change is thus more likely when the first three types or stages of impact have occurred.

Both issue characteristics and actor characteristicsare important parts of our expla- nation of how networks affect political out- comes and the conditions under which networks can be effective. Issue characteristics like sali- ence and resonance within existing national or institutional agendas can tell us something about where networks are likely to be able to insert new ideas and discourse into policy debates. Success in influencing policy depends on the strength and density of the network, and its ability to achieve leverage.

As we look at the issues around which transnational advocacy networks have organized most effectively, we find two characteristic issues that appear most frequently:

(1) those involving bodily harm to vulnerable individuals, especially when there is a short and clear causal chain (or story) about who bears responsibility;

(2) issues involvinglegal equality of opport- unity.

99Transnational advocacy networks

The first responds to a normative logic, and the second to a judicial and institutional one. Issues involving physical harm to vulnerable or innocent individuals appear more likely to resonate transnationally. Of course, this alone does not ensure the success of the campaign, but is particularly compelling. Nor is it straight- forward to determine what constitutes bodily harm, and who is vulnerable or innocent. Both issues of ‘harm’ and ‘innocence’ or vulner- ability are highly interpretive and contested. Nevertheless, we argue that issues involving bodily harm to populations perceived as vulner- able or innocent are more likely to lead to effective transnational campaigns than other kinds of issues. This helps to explain why it has been easier to work on torture or disappear- ance than some other human rights issues, and why it has been easier to protest against torture of political prisoners than against torture of common criminals or to abolish capital punish- ment. It is also useful for understanding that those environmental campaigns that have had the greatest transnational effect have been those that stress the connection between protecting environmentsand the (often vulnerable) people who live in those environments. We also argue, following Deborah Stone (1989), that in order to campaign on an issue it must be converted into a ‘causal story’ – establishing who bears responsibility or guilt. But in addition to the need for a causal story, we argue that the causal chain within that story needs to be sufficiently short and clear to make a convincing case about responsibility or guilt. The second issue around which transnational campaigns appear to have greater effectiveness is that of greater legal equality of opportunity. Notice that we stresslegal equality of opport- unity, not of outcome. One of the most success- ful international campaigns was the anti-apart- heid campaign. What made apartheid such a clear target was the legal denial of the most basic aspects of equality of opportunity.

Transnational networks and regional integration

Many scholars now recognize that the state no longer has a monopoly over public affairs and are seeking ways to describe the sphere of inter-

 UNESCO 1999.

national interactions under a variety of names: transnational relations, international civil society, and global civil society (Lipschutz, 1992; Peterson, 1992). In their views, states no longer look unitary from the outside. Increas- ingly dense interactions among individuals, groups, actors from states and regional and international institutions appear to involve much more than re-presenting interests on a world stage.

Recent empirical work in sociology has gone a long way towards demonstrating the extent of changes ‘above’ and ‘below’ the state. The world polity theory associated with John Meyer, John Boli, George Thomas and their colleagues, conceives of an international society in a radically different way. For these scholars, it is the area of diffusion of world culture – a process that itself constitutes the characteristics of states (Thomas et al., 1987; Boli and Thomas, in press). The vehicles for its diffusion become global intergovernmental and non- governmental organizations, but neither the sources of norms nor the processes through which global cultural norms evolve are adequately specified (Finnemore, 1996). Pro- ponents of world polity theory present inter- national organizations and NGOs as ‘enactors’ of some basic cultural principles of the world culture: universalism, individualism, rational voluntaristic authority, human purposes, and world citizenship. There is thus no meaningful distinction between those espousing norms that reinforce existing institutional power relation- ships, and those that challenge them.

We argue that different transnational actors have profoundly divergent purposes and goals. To understand how change occurs in the world polity we have to unpack the different categor- ies of transnational actors, and understand the quite different logic and process in these differ- ent categories. The logic of transnational advo- cacy networks, which are often in conflict with states over basic principles, is quite different from the logic of other transnational actors who provide symbols or services or models for states. In essence, world polity theorists elimin- ate the struggles over power and meaning that for us are central to normative change.

Our research suggests that many trans- national networks have been sites of cultural and political negotiation rather than mere

100 Margaret E. Keck and Kathryn Sikkink

enactors of dominant Western norms. Western human rights norms have indeed been the defining framework for many networks, but how these norms are articulated is transformed in the process of network activity. For example, issues of indigenous rights and cultural survival have been at the forefront of modern network activity, and yet they run counter to the cultural model put forward by the world polity theorists.

In other words, as modern anthropologists realize, culture is not a totalizing influence, but a field that is constantly changing. Certain dis- courses – like that of human rights – provide a language for negotiation. Within this language certain moves are privileged over others; human rights is a very disciplining discourse. But it is also a permissive discourse that allows different groups within the network to renegotiate mean- ings. The success of the campaign for women’s rights as human rights reveals the possibilities

within the discourse of human rights. We believe that studying networks is extraordinarily valuable for tracking and ultimately theorizing about the emergence of shared norms and cul- tural meanings underpinning processes of regional and international integration.

Network theory can thus provide an expla- nation for transnational change, a model that is not just one of ‘diffusion’ of liberal institutions and practices, but one through which the prefer- ences and identities of actors engaged in trans- national society are sometimes mutually trans- formed through their interactions with each other. Because networks are voluntary and hori- zontal, actors participate in them to the degree that they perceive mutual learning, respect and benefits. Modern networks are not conveyor belts of liberal ideals, but vehicles for communi- cative and political exchange, with the potential for mutual transformation of participants.

 UNESCO 1999.

Baumgartner, F.; Jones, B., 1991. ‘Agenda dynamics and policy subsystems’,Journal of Politics53, pp. 1044–74.

Boli, J.; Thomas, G., in press. World Polity Formation since 1875: World Culture and International Non-Governmental Organizations. Stanford.

Brysk, A., 1994. ‘Acting globally: Indian rights and international politics in Latin America’. In D. Van Cott (ed.),Indigenous Peoples and Democracy in Latin America, New York: St Martins/ Inter- American Dialogue, pp. 29–51.

Brysk, A., 1995. ‘Hearts and minds: bringing symbolic politics back in’, Polity 27 (Summer), pp. 559–85.

Dalton, R., Kuechler, M.; Burklin, W., 1990. ‘The challenge of new movements’. In R. Dalton, M. Kuechler (eds),Challenging the Political Order: New Social and Political Movements in Western Democracies, Cambridge: Polity.

Finnemore, M., 1996. National Interests in International Society. Ithaca: Cornell.

Frühling, H., 1991. Derechos Humanos y Democracia: La contribución de las Organizaciones Nogubernamentales. Santiago: IIDH.

Haas, P. (ed.), 1992.Knowledge, Power and International Policy Coordination. International Organization, vol. 6.

Katzenstein, P. (ed.), 1996.The Culture of National Security: Norms and Identity in World Politics. New York: Columbia.

Keck, M.; Sikkink, K., 1998. Activists beyond Borders: Advocacy Networks in International Politics. Ithaca: Cornell.

Klotz, A., 1995. Norms in International Relations: The Struggle against Apartheid. Ithaca: Cornell.

Korten, D., 1990. Getting to the 21st Century: Voluntary Action and the Global Agenda. Hartford, CT: Kumarian.

Lipschutz, R., 1992. ‘Reconstructing world politics: the

References

* This article is based on our book Activists Beyond Borders: Advocacy Networks in International Politics (Ithaca: Cornell University Press,

1998). The Editor-in-Chief wishes to thank Cornell for permission to publish material drawn from the book.

1. We developed this definition based on a discussion in Mitchell (1973, p. 23).

Notes

101Transnational and advocacy networks

emergence of global civil society’. Millennium 21, pp. 389–420.

McAdam, D., McCarthy, J.; Zald, M. (eds), 1996.Compar- ative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings. New York: Cambridge.

Mitchell, J., 1973. ‘Networks, norms, and institutions’. In J. Boissevian, J. Mitchell (eds),Network Analysis, The Hague: Mouton.

Oliver, P.; Marwell, G., 1992. ‘Mobilizing technologies for collective action’. In A. Morris, C. Mueller (eds),Frontiers in Social Movement Theory, New Haven, CT: Yale University Press.

Peterson, M. J., 1992. ‘Transnational activity, international society, and world politics’, Millennium 21.

 UNESCO 1999.

Powell, W., 1990. ‘Neither market nor hierarchy: network forms of organization’,Research in Organizational Behaviour12, pp. 295–336.

Scherer-Warren, I., 1993. Redes de Movimentos Sociais. São Paulo: Loyola.

Snow, D.; Benford, R., 1988. ‘Ideology, frame resonance, and participant mobilisation’. In B. Klandermans, H. Kriesi, S. Tarrow (eds),From Structure to Action: Comparing Social Movement Research across Cultures, Greenwich, CT: JAI, pp. 197–217.

Snow, D.; Benford, R., 1992. ‘Master frames and cycles of protest’. In A. Morris, C. Mueller (eds),Frontiers in Social Movement Theory, New Haven, CT: Yale University Press, pp. 133–55.

Snow, D., et al., 1986. ‘Frame alignment processes,

micromobilisation, and movement participation’, American Sociological Review51.

Stone, D., 1989. ‘Causal stories and the formation of policy agendas’,Political Science Quarterly 104, pp. 281–300.

Tarrow, S., 1992. ‘Mentalities, political cultures, and collective action frames: constructing meanings through action’. In A. Morris, C. Mueller (eds),Frontiers in Social Movement Theory, New Haven, CT: Yale University Press.

Tarrow, S., 1994. Power in Movement. New York: Cambridge University Press.

Thomas, G.; Meyer, J.; Ramirez, F.; Boli, J. (eds), 1987. Institutional Structure: Constituting State, Society, and Individual. Newbury Park, CA: Sage.