Aroka

It is now common in the literature on transnational social movements and networks to assert that they have expanded dramatically in recent years. But few researchers have provided strong quantitative evidence or analysis of this growth and its relevance for transnational movements and networks. Re- searchers have shown significant growth in international nongovernmental organizations (INGOs) of all types, but only a small portion of these organi- zations are engaged in the kinds of social change activities typical of social movements and networks (see, for example, Skjelsbaek 1971). Jackie Smith presented clear evidence of the growth of INGOs established for the explicit purpose of promoting social or political change. Drawing from the Yearbook of International Organizations for the years 1983, 1988, and 1993, Smith ex- amined the population of what she labeled “transnational social movement organizations” or “TSMOs,” demonstrating their parallels with research on national social movements.1 This chapter presents the results of an expanded collaborative effort at data collection and coding from the Yearbook, using Smith’s coding procedures, to include the years 1953, 1963, and 1973.2 This long-term perspective can offer insights about the nature of the subset of INGOs working across national borders to promote social and political change, and it can tell us about change in this sector over time. This allows us to relate the arguments made in this book derived from case studies of particular transnational social change efforts to a broader understanding of the organizations that help compose such movements.

We cannot measure the growth of transnational movements and net- works using the definition presented in the first chapter of this volume be-

24

2

Infrastructures for Change: Transnational Organizations, 1953–93

Kathryn Sikkink and Jackie Smith

cause there is no source that systematically documents such organizations and networks. Accounting for the extent of contemporary transnational net- works would prove a time-consuming and difficult task, given the informal and fluid nature of these networks. As a proxy for such a measure, however, we use the changing number and characteristics of international nongovern- mental organizations that are explicitly devoted to social and political change goals. For consistency with other chapters and research, we refer to the or- ganizations we examine as “social change INGOs.” A subset of INGOs whose organizational missions involve the promotion of some form of social and/or political change, these organizations serve as social infrastructures for change, and they represent more or less routine efforts to generate resources and collective action for social change efforts. Moreover, they provide op- portunities for individuals to participate in international politics in ways that would otherwise be impossible. International organizations and their member states provide few formal channels of access to international policy- making arenas. But transnational social change organizations have worked to expand citizens’ access to the global polity by creating structures that con- vey information and ideas between individuals and groups within societies and the institutions that structure interstate relations (see, e.g., Smith 1997). We can therefore expect these particular actors—in contrast to actors whose principal objectives, instructions, or routines lie outside the sphere of social change work—to be routine, though by no means the only participants in transnational advocacy networks and transnational social movements. Under- standing the nature of this organizational infrastructure can help us antici- pate the issue arenas most likely to attract transnational advocacy efforts.

As we know from work on the precursors to modern transnational so- cial movements and networks, including the chapter by August Nimtz in this volume, transnational social movements are not a new phenomenon (see also Keck and Sikkink 1998; Chatfield 1997). We find numerous early examples including the transnational campaign against slavery, transnational labor organizing, and campaigns for women’s suffrage. The oldest social change INGO in the Yearbook’s current listing, the Anti-Slavery Society for the Protection of Human Rights, was founded in 1839. The International Working Men’s Association (1864–72), discussed by August Nimtz, was short-lived but was also the forerunner of diverse networks, movements, and political parties. Other early social change INGOs include the World Woman’s Christian Temperance Union (1883), the World Zionist Organiza- tion (1897), and the International Bureau for the Suppression of Traffic in Persons (1899). Nevertheless, the vast majority of social change INGOs emerged in the latter half of the twentieth century, and our data suggest that

infrastructures for change 25

their number and size, as well as the density and complexity of international linkages, have grown dramatically in the last three decades. More than 60 per- cent of all social change INGOs active in 1993 were formed after 1970. This is consistent with the material from our case chapters, where many of the or- ganizations in the networks we discuss were formed in the 1970s and 1980s.

Data

The Union of International Associations (UAI), based in Brussels, produces an account of international nongovernmental organizations as part of its annual Yearbook of International Organizations, published since 1950.3

International organization scholars have long referred to the Yearbook to document trends in international organization and international nongov- ernmental activity (see Skjelsbaek 1971; also Jacobson 1979, appendix B, 435–39). The Union of International Associations was commissioned by a United Nations resolution to keep track of international organizations, which it does by consulting relevant UN records, and by seeking out infor- mation directly from INGOs themselves. The Yearbook is the only source of systematic annual data on international NGOs.

For scholars and activists interested in social movements or networks, however, the Yearbook data poses some problems. The Yearbook only collects information about international nongovernmental organizations, and they use a quite stringent criteria to determine if an organization is international. Members, officers, voting, and substantial budgetary contributions have to come from at least three countries for the organizations to be included. This creates a list of formal and bureaucratized INGOs. But many of the organi- zations that scholars and activists would consider transnational social move- ments or networks would not be listed in the Yearbook. The Yearbook only lists those organizations that are formally international, not all networks with substantial, but informal international linkages. Moreover, the groups most likely to respond to the UAI’s census survey are those that wish to have information about their group made public. This would likely exclude groups working for political and social change that is seen as illegitimate, as well as groups using illicit means to advance their change goals. Thus any data set based on information from the Yearbook will underestimate the ab- solute number of transnational social movement organizations. This sug- gests that data from the Yearbook can provide only a partial picture of trans- national network sectors. Nevertheless, the dynamics and patterns of this particular subset of formally organized actors may be helpful to help us understand general patterns in transnational networks.

Second, and perhaps more troubling for the social movement theorist,

26 sikkink and smith

the Yearbook does not distinguish between different types of international nongovernmental organizations. It lumps together in a single INGO cate- gory all different types of international NGOs, such as international profes- sional organizations like the International Political Science Association (IPSA) and international medical associations, with what we would call more genuine social movements or networks. Smith overcame this problem by drawing from sociological research on social movement organizations to introduce a “political or social change” criteria in her coding scheme. Coders identified organizations whose specified goals (or “aims” under Yearbook headings) indicated that their principal purpose was to work for some form of social or political change.4 It is this social change subgroup that we refer to as social change INGOs. Coders also excluded the following types of or- ganizations outright: funds, foundations, institutes or organizations primari- ly devoted to research, groups that seek world peace through spiritual trans- formation (e.g., yoga, transcendental meditation), organizations whose primary mission is the promotion of a particular religious tradition, ex- change programs, and service delivery and general education organizations. In addition, because the Yearbook did not indicate whether they were au- tonomous from government control, labor organizations were excluded from the data set. However, labor organizations working principally on workers’ rights and protection from forced labor were included, since their social change emphasis was more apparent. Religious organizations were ex- cluded, except when their principal organizational focus went beyond the propagation of a particular religious ideology to include the promotion of social or political change. The idea here is that churches and other religious institutions—while frequently supportive actors within transnational social movements—have organizational missions that are not primarily for the promotion of social change, and thus their work toward movement goals may vary substantially across time and place.

These distinctions were necessary to exclude the large number of INGOs that do not fit our definition of transnational social movements in any sense. Nevertheless, these criteria do exclude some organizations that are a critical part of some of the networks in this volume. For example, the cod- ing scheme excluded groups whose primary aim is the promotion of reli- gion, yet in her chapter on the debt network, Elizabeth Donnelly describes the significant activity of the hierarchy of the Catholic Church, including the pope himself, on behalf of debt forgiveness. Though not formally a part of the debt networks, the Catholic Church and Catholic bishops were some of the most active and influential actors on these issues. Likewise in his chap- ter on Chile, Darren Hawkins argues that international church organizations,

infrastructures for change 27

like the World Council of Churches (WCC), played a crucial role in the human rights network in Chile, but the WCC is not in our list of INGOs whose specific purpose is promoting social change. In addition, funds and foundations are excluded from the database, yet most of the chapters in the volume point to the extremely important role that some funds and founda- tions play in supporting transnational social movements and network activity. Some of the transnational labor networks discussed by Thalia Kidder and Mark Ritchie would not appear in our database because they are listed pri- marily as labor unions. Finally, sometimes parts of a government become members of a transnational network, as in Daniel Thomas’s discussion of the Helsinki network, where the Congressional Helsinki Commission and even Ambassador Goldberg himself often seemed to be more identified with transnational network principles than with U.S. government policy, eventual- ly succeeding in changing U.S. policy to bring it in line with network goals.

Given these problems, we emphasize that what we present here is a look at the population of international nongovernmental organizations with so- cial or political change aims or both, rather than a comprehensive portrait of transnational social movements or networks discussed in the empirical chap- ters of this book. Nevertheless, since organizations are major components of movements and networks, many of the specific transnational organizations mentioned in the chapters here appear in this data set as well.

In addition to analyzing the trends in the growth of social change INGOs, we examine the data in light of some of the hypotheses proposed in this volume. Authors in this volume make a series of arguments about the explanations for the growth and effectiveness of networks in their issue area. Many of these arguments have to do with the impact of international institutions, including international norms, on social movement activity (see Tarrow 1999). More specifically, as noted in chapter 1, authors discuss how specific aspects of international institutions, or international oppor- tunity structures, facilitate the growth and effectiveness of networks. For ex- ample, a number of authors argue that issue areas where preexisting well- institutionalized norms exist are more likely to lead to network growth and successful network action. For example, in his chapter on Chile, Darren Hawkins argues that “preexisting domestic and international human rights norms acted as a ‘pull factor’ to facilitate the emergence and growth of the transnational advocacy network.” Without such preexisting norms, he says, “it is difficult to imagine the formation and growth of such a dense network in such relatively short time periods.”

This raises the question of what exactly we mean by norms, and how we know a norm when we see one. In chapter 1 we stressed that although inter-

28 sikkink and smith

national norms are manifest in complex ways, for many research purposes it is possible to operationalize international norms using international treaties, or declarations of “soft law,” such as the Helsinki Final Act. Looking at the entry into force of a treaty as a threshold in norm institutionalization may provide a starting place for talking about the influence of norms on the growth of networks and network success. In this chapter we will briefly look at the relevant treaties in a handful of issue areas and discuss whether there is any apparent relation between treaty ratification and social change INGO growth. In some cases, we also need to look at “soft law.” For example, Daniel Thomas suggests that the preexisting Helsinki norms were essential for the mobilization and strength of that network, and yet Helsinki was not a treaty, but rather soft law. In any case, both Hawkins and Thomas argue that norms legitimate the work of change advocates operating in networks and/or movements and thereby encourage network growth and success.5 We shall briefly explore this in relation to available data.

In the case of human rights, such an argument would suggest that net- work growth would occur after the mid-1970s, because both the Covenant on Civil and Political Rights and the Covenant on Economic, Social, and Cultural Rights entered into force in 1976. Some regional human rights treaties and laws also date to this period: the Helsinki Final Act was passed in 1975, and the American Convention on Human Rights entered into force in 1978. In the women’s rights area, this argument would anticipate social change INGO expansion after 1981, when the Convention on the Elimina- tion of All Forms of Discrimination Against Women (CEDAW) entered into force. The environmental area is more difficult to evaluate because there is not a single treaty, but a set of several specific treaties. Some issue areas we deal with in the volume—for example the area of debt forgiveness—have virtually no strong preexisting or institutionalized treaty norms. If institu- tionalized norms lead to the growth of networks, we would expect slight growth of social change INGOs in this issue area. Elizabeth Donnelly points out that the Catholic Church has a “jubilee” tradition of periodic debt for- giveness, but this has not been translated into any international treaties. Like- wise, in other areas, such as development, Paul Nelson suggests that multiple and contradictory norms exist, some stressing economic rights and basic needs, and others stressing fiscal austerity and the need to balance budgets.

A second hypothesis is that large international conferences created the impetus for network growth. In the area of women’s rights, this would lead us to expect that the large UN conferences associated with the International Women’s Year in 1975 and International Women’s Decade (1975–85) would lead to the growth of networks and movements. In the environmental area,

infrastructures for change 29

it suggests we would see network growth after the 1972 Conference in Stockholm and the UN Conference on Environment and Development (UNCED) in 1992 in Rio de Janeiro. Likewise, we might expect human rights network growth in response to the 1968 International Conference on Human Rights in Tehran and the 1993 World Conference on Human Rights in Vienna.6

Growth in the Number of Social Change INGOs

The data support the general view presented in this book and elsewhere that there has been a significant increase in the total number of international nongovernmental organizations. The total number of social change INGOs listed in the Yearbook increased almost six times between 1953 and 1993. The growth has been particularly dramatic in the ten-year period from 1983 to 1993 (see Table 2.1).

Types of Issues around Which Social Change INGOs Organize

The data also tell us about the kinds of issues around which social change INGOs organize and how this has changed over time. The story here is one of both remarkable stability in some issues and some significant changes. Table 2.1 shows that the number of social change INGOs has increased across all issues, though to varying degrees in different issue areas. This vari- ance across issue areas suggests that social change INGO growth is not just an epiphenomena of more generalized growth of organizations internation- ally. The number of social change INGOs working on some issues has in- creased dramatically, while those working on other issues has stagnated, and in a relative sense, declined.

30 sikkink and smith

Table 2.1. Number of Transnational Organizations by Issue Focus

1953 1963 1973 1983 1993 Issue (N = 110) (N = 141) (N = 183) (N = 348) (N = 685) Human rights 33 38 41 79 190

Peace 11 20 14 22 81

Women’s rights 10 14 16 25 62

Environment 2 5 10 26 123

World order/ 22 23 37 57 80 international law

Development/ 3 3 7 13 47 empowerment

Note: These six issues account for approximately 70 percent of all organizations in each time period.

Human rights has been a predominant focus of transnational social change organizations since the 1950s. The issue has been the focus of rough- ly a quarter of all groups in each decade studied. There were five times as many organizations working primarily on human rights in 1993 as there were in 1953, but human rights groups are about the same proportion of total groups in 1993 as they were in 1953. The number of groups focusing on human rights has grown in each period, although the issue areas declined proportionally in the 1973 and 1983 period, only to recover again by 1993.

Two of the chapters in this volume focus on these human rights organi- zations: the chapter on Chile by Darren Hawkins and the chapter on U.S. policy toward Eastern Europe and the Soviet Union by Daniel Thomas. Both Hawkins and Thomas argue that these networks were relatively more effective because they could identify their case with well-institutionalized international norms. The United Nations General Assembly passed the first clear statement of international human rights norms, the Universal Declaration of Human Rights, in 1948. The relatively large numbers of human rights INGOs in 1953 could reflect this early declaration on emer- gent international norms. Indeed, some human rights groups’ efforts to press for rights protections in the wake of World War II’s atrocities were criti- cal to achieving the Universal Declaration. These norms became increasing- ly detailed and institutionalized in 1976, when the Covenant on Civil and Political Rights and the Covenant on Economic, Social, and Cultural Rights entered into force, and in 1975 when the Helsinki Final Act was signed. The data on the number of human rights INGOs seems to reflect the increasing institutionalization of international norms: between 1973 and 1983 the number of transnational human rights NGOs nearly doubled. The World Conference on Human Rights in Vienna in 1993 also coincided with an in- crease in both the absolute number of human rights INGOs and their pro- portion of all social change INGOs. This might be explained by the conver- gence of this conference with the end of the cold war and the subsequent expansion of the global agenda to include issues and voices that had been ex- cluded in the bipolar struggle that had dominated global affairs.

The second largest proportion of social change INGOs active in 1953 worked to promote the development of international law and institutions. These organizations made up 13 percent of the population in the period 1953 through 1973, but by 1983 the numbers remained steady as the total number of social change INGOs grew. By 1993, this issue focus accounted for only 4 percent of all organizations. This does not mean that international law has become an irrelevant concern for these organizations. Many inter- national human rights, women’s rights, and increasingly environmental

infrastructures for change 31

organizations pay significant attention to international law, but they do so from the vantage point of their particular issue. It is likely that the growth in the quantity and complexity of international law and international institu- tions in the postwar period have made the INGOs that serve them less vi- able as they compete with other, more focused groups for funding and other resources.7 None of the chapters in this volume focuses on transnational movements or networks dedicated to promoting the development of inter- national law or institutions. Indeed, as Paul Nelson’s chapter on the World Bank, Elizabeth Donnelly’s chapter on the International Monetary Fund, and Mark Ritchie’s comments about campaigns against NAFTA and the WTO point out, today’s transnational social movements are more likely to critique the practices of international institutions than they are to promote their development. It is interesting to note that there were no major at- tempts to write treaties or hold conferences to institutionalize or deal with general norms about international law or international institutions.

Groups working on women’s rights accounted for about 9 percent of all groups in 1953 and in 1993. Like human rights, the women’s rights issue area grew very slowly before 1973, but has grown much more rapidly since 1983. This is somewhat surprising, because as Karen Brown Thompson points out in chapter 5, histories of the international women’s movement have signaled the UN International Women’s Year in 1975 and the UN International Women’s Decade (1975–85) as a major growth period for international women’s rights NGOs. Likewise the early 1980s saw the emer- gence for the first time of clear treaty norms on women’s rights. CEDAW was adopted in 1979, and entered into force in 1981. Yet, despite all this ac- tivity around norms and conferences, 1983 was the year when the women’s rights issue accounted for the lowest proportion of the total groups since 1953 (7.2 percent). In some cases there appears to be a significant lag time between the founding of organizations and their appearance in the Yearbook. So, for example, the parallel NGO meeting at the International Women’s Year Conference in Mexico City in 1975 encouraged a group of women to found the International Women’s Tribune Center, which used the mailing list generated at Mexico City to keep in touch with individuals and groups around the globe. By 1990, the Tribune Center was a communication link for sixteen thousand individuals and groups working on behalf of women in 160 countries. The Tribune Center, though founded in the mid-1970s, does not show up on the Yearbook list until 1993. We do not know whether this lag is more characteristic of women’s groups than other groups. It could be that the feminist movements that came to the forefront in the 1970s and 1980s, which eschewed hierarchical forms of organization, avoided setting

32 sikkink and smith

up the kind of formal international groups that met the Yearbook criteria. In general, it appears that the full effect of the Women’s Decade from 1976 to 1985 may not have been reflected in the data until the 1993 Yearbook.

Transnational environmental organizations have grown most dramati- cally in absolute and relative terms. In 1953, only two organizations worked to promote environmental change. By 1973, the year following the UN Conference on the Human Environment in Stockholm, their number rose to ten, and it more than tripled between 1983 and 1993, the year following the Earth Summit, or the UN Conference on Environment and Development in Rio. This growth supports Khagram’s point in chapter 10 that the Stockholm conference in 1972 contributed to the growth of international networks.

Proportionally, environmental INGOs rose from 1.8 percent of total groups in 1953 to 14.3 percent in 1993. Forty-two percent of all environ- mental INGOs active in 1993 were formed after 1985, and 80 percent were formed after 1970. Some of this growth might be the result (or even the partial cause) of UN meetings on the environment in 1972 and 1992 (see Willetts 1999). This dramatic growth parallels the expansion of national and locally based environmental movements. If this is the case, it is perhaps not surprising that some of the more successful social movements and network activity discussed in this volume, such as the anti-dam transnational social movements discussed by Sanjeev Khagram and movements pressuring the World Bank to consider the developmental impact of major infrastructural projects, discussed by Paul Nelson, are linked to international environmen- tal activism.

The percentage share of groups in the development issue area has also grown (from 2.7 percent in 1953 to 5.4 percent in 1993), but not as dramati- cally as the growth in the environmental issue area. Groups focusing on devel- opment still account for a relatively small percentage of the social change INGOs recorded in this data set.8 In this volume, the poverty and structural adjustment networks discussed by Paul Nelson and the debt networks dis- cussed by Elizabeth Donnelly are the groups most engaged in the develop- ment issue area. Some of the major development NGOs (such as Church World Service, Community Aid Abroad, CARE, Christian Aid, Oxfam International, Save the Children Federation, and World Vision International) not only advocate policy positions, but also carry out significant development projects on their own. To carry out these projects, they may seek funding from the very institutions they are pressuring for policy changes. Paul Nelson suggests that this creates tensions within some major development NGOs be- tween “their desire to advocate for new policy and practice” and “their domi- nant organizational need to secure funding from major aid donors.”

infrastructures for change 33

Other groups that might be considered “development” organizations have found that they are more effective if they “frame” their demands in human rights or environmental terms. Thus a number of the organizations discussed by Jim Riker, Sanjeev Khagram, Paul Nelson, and Thalia Kidder might be characterized as “development” NGOs but also frequently use the frames of the environment or human rights (and workers’ rights in the case of labor networks) because they have found that these frames have more resonance than the development frame. These are examples of strategic venue shift, as activists frame issues in certain ways in the search for more support and a more receptive political venue (Baumgartner and Jones 1991; Keck and Sikkink 1998). Even more far reaching than simple venue shift is what occurs, as Sanjeev Khagram describes, when new norms on indigenous peoples, human rights, and the environment “restructur[e] the global poli- tics of development.”

The percentage share of groups in such issue areas as peace, ethnic unity, and Esperanto has declined. None of the chapters in this volume looks at international organizing in these issue areas. To the initial U.S. observer, the inclusion of Esperanto organizations in social change INGOs and the high number of Esperanto organizations in our database may seem puzzling. But for many internationalists around the world, the desire to learn a “world lan- guage” such as Esperanto was an expression of a desire for an international identity that transcended national identities. In parts of the world where more explicit international political activity was limited, Esperanto organiza- tions became the depository of internationalist norms and aspirations, as well as an opportunity to join together with organizations elsewhere. These or- ganizations were frequently organized along the lines of occupation or interest/ identity, so we find such organizations as the International Federation of Esperantist Railwaymen, the League of Homosexual Esperantists, and the World Esperantist Vegetarian Association. None of the chapters in this vol- ume examines such organizations, but the steady presence of such inter- national language organizations in our data set provides support for the argu- ment we present in the conclusions that transnational social movements and networks are harbingers of the emergence of international identities that co- exist with, and in some cases transcend, national identities.9

Although the networks discussed in this book represent only a subset of the total number of networks, these include the issue areas around which the largest number of social change INGOs have organized, and/or the ones that are currently experiencing the greatest growth. Together, groups work- ing on human rights, women’s rights, the environment, and development account for well over half of all total social movement and network activity.

34 sikkink and smith

As pointed out above, many labor organizations have not been included in the data set. In summary, the issues dealt with in this volume are not an un- usual or marginal form of social movement activity, but a sample of the most important issues to which the majority of modern social change INGOs de- vote themselves.

Finally, although coders of the data set were obliged to choose the “major focus” of an organization, in the cases in this book many social movements and campaigns work simultaneously on a number of different issues, or they work on issues that are at the intersection of issue areas. This may be increasingly true as economic and political globalization make the connections among issues more obvious, as with the NAFTA and WTO ef- fects on environmental and human rights concerns. Perhaps the clearest ex- ample of this is described in Sanjeev Khagram’s chapter on the Narmada Dam campaign, where activists worked on issues involving the protection of indigenous or tribal people, human rights, environmental preservation, and development. Paul Nelson also points to the overlapping of three networks, poverty, the environment, and structural adjustment, in his focus on the World Bank.

Geographic Dispersion of Social Change INGOs and Networks

One common criticism of transnational social movements and networks is that they tend to be disproportionately based in the countries of the North. Critics claim that this physical location in the most developed countries has led these social change INGOs to reflect primarily the sensibilities of citizens or governments in the developed world.

Our data supports the general understanding that the great bulk of so- cial change INGOs are based in the developed world and primarily have branch offices and membership from the North (see Tables 2.2 and 2.3). This is the closest indicator in the data set to what we have referred to as “asymmetries” of power and influence within the networks and trans- national social movements. Of course, the physical location of an INGO is only one possible indicator of the points of view held by network members. Hosting the secretariat of an international NGO is a costly and complex ad- ministrative operation. There are examples of efforts made to locate secre- tariats of social change INGOs in Southern countries that failed because Southern NGOs were hesitant to take on the large financial and administra- tive burden of serving as the secretariat.

Another explanation for the geographic distribution of social change INGO headquarters is that it makes practical sense for these groups to base their organizational operations near the political targets they seek to influence.

infrastructures for change 35

Locating in sites near international organizations facilitates access to infor- mation on international negotiations and it enables organizational staff to monitor negotiations, to lobby government delegations and members of international organization secretariats, and to otherwise facilitate commu- nications between geographically dispersed memberships and global insti- tutions. Thus, it is not surprising that we found some of the largest concen- trations of social change INGOs near major IGO headquarters in Brussels and Geneva. This pattern holds throughout the time periods we examined.

36 sikkink and smith

Table 2.2. Regions of TSMO Membership

1953 1963 1973 1983 1993 Region (91) (117) (134) (214) (525) Europe 92% 91% 89% 87% 84%

North America 69 73 72 64 68 (except Mexico)

Latin America 59 65 68 61 61 and Caribbean

Asia 55 67 63 59 60

Africa 53 61 61 56 59

Oceania 45 51 55 52 53

USSR/Eastern 49 46 53 43 50 Europe

Middle East 32 55 57 50 43

Note: The numbers in parentheses after each year indicate the number of organizations able to provide valid data.

Table 2.3. Locations of TSMO Secretariats

1953 1963 1973 1983 1993 Region (N = 110) (N = 141) (N = 179) (N = 343) (N = 679) Western Europe, 83% 77% 73% 68% 58%

Canada

United States 9 8 10 11 14

USSR and Eastern 2 5 4 3 4 Europe

Any global south 5 6 12 17 23 country

Latin America 3 1 2 5 6

Another consideration is that a transnational organization is likely to choose a site for the international secretariat that dispersed members can readily communicate with—electronically, by telephone, or in person. Cities with good physical infrastructures and frequent nonstop airline service facilitate global networking and thus make logical candidates for headquarter siting. Figure 2.1 shows the percentage of groups whose headquarters were located in one of four key cities of Europe, cities that are among what Sassen calls “global cities,” or centers of commercial and professional activity that enjoy high concentrations of specialized service industries and supportive infra- structures (e.g., telecommunications and transportation), which reinforce their centrality as global centers (Sassen 1991).10

Despite the overrepresentation of developed countries among social change INGO memberships and among the secretariat headquarters, these

infrastructures for change 37

1953

60

50

40

30

20

10

0 1963 1973 1983 1993

Figure 2.1. TSMOs headquartered in “global cities.” The Y axis indicates the percentage of all TSMOs headquartered in Brussels, Geneva, London, or Paris. Source: Saskia Sassen, The Global City: New York, London, Tokyo (Princeton: Princeton University Press, 1991).

data show a trend over the past forty years away from this predominance of Northern and European representation and toward more geographically dis- persed membership. Developing areas are somewhat better represented among the memberships of social change INGOs today than they were in the past. Of the developing areas, Latin Americans and Asians are most networked with social change INGOs, with Africans running a close third. While Europeans are most likely to have memberships with social change INGOs, their relative proportions have declined somewhat (from 92 percent in 1953 to 84 percent in 1993), reflecting the more rapid growth in participation from other regions. The proportion of social change INGOs reporting members in the United States and Canada remained fairly constant at around 70 percent. Middle Eastern countries were least likely to have repre- sentation in social change INGOs in most years of the study. Participation in the sector by Eastern European and former Soviet countries was generally low, although it expanded between 1983 and 1993, and may continue to grow as the cold war recedes further into history. Much more dramatic is the decline in the percentage of social change INGOs based in global cities, from more than half of all groups in 1953 to roughly one-quarter of all groups in 1993. While the absolute numbers of groups based in global cities have grown, these groups represent a smaller proportion of the transnational social movement sector than they once did.

With this shift in membership, the location of social change INGO sec- retariats has also shifted to reflect greater participation from developing countries. While only 5 percent of all secretariats were based in developing countries in 1953, this figure rose to 23 percent by 1993. Of these, nearly one-fourth were in Latin America. To a certain extent, modern communica- tion technologies have made secretariat location a less constraining issue than it once was, because all network members can receive information rap- idly through electronic mail. Nevertheless, the kinds of geographic concen- tration shown in the data set were also present in the case studies. Both Paul Nelson and Elizabeth Donnelly, writing on network pressures on the World Bank and the IMF, respectively, report that groups based in Washington, D.C., have had disproportionate influence in campaigns. This led in turn to complaints that the Washington-based activists were not more representa- tive or accountable, but had more influence because they had “loud voices and resources.” Likewise, the close cooperation between some networks, es- pecially environmental and human rights networks, and the U.S. govern- ment strengthens the perception that networks are guided by an “essentially Northern agenda” (Nelson). When the World Bank set up the standing NGO–World Bank Committee it attempted to circumvent the advantages

38 sikkink and smith

of geographical concentration by mandating representation from each of the five major global regions. Other Southern-based networks have proposed es- tablishing their own Washington offices, but the chapters in this volume provide only one example where such an office has been created (an African consortium cited in Nelson).

The Yearbook does not provide significant data on the financial re- sources of organizations. We suspect that such data would suggest that the process of geographic dispersal of membership and of secretariats has not been accompanied by substantial dispersal of the sources of funding for social change INGOs. Many of the authors in this volume point to the problems that funding creates for their movements, which suggests that asymmetries in transnational networks and movements is more a result of concentration of funding sources than of the location of secretariats.

U.S. and European foundations increasingly provide very significant fi- nancial support for social change INGOs, including those based in and drawing their membership from the developing world. Funding from inter- national organizations and agencies is also increasingly important. Not sur- prisingly, financial difficulties were by far the most significant obstacle reported by Southern (and Northern) groups in recent surveys of trans- national human rights groups and affiliates of EarthAction, a large trans- national organization working for environment, peace, and human rights (Smith, Pagnucco, and Lopez 1998; Smith 1999). However, most groups (in both surveys) still report that a substantial portion (approximately half of operational budgets, on average) of their funding is raised from internal sources, such as dues or fees paid for services or materials.

Despite this concentration in the developed world and the great depen- dence on financing from U.S. and European foundations, it is interesting to note that the issues around which social change INGOs organize do not simply reflect the issues that are most prominent in domestic social movement organizations in Western countries. For instance, while human rights is the most predominant issue for social change INGOs, human rights organiza- tions are a relatively small part of the social movement sector in Western coun- tries. Likewise, women’s organizations are important both domestically and internationally, but the issues they focus on differ (Keck and Sikkink 1998).

Network Density: Countries of Membership, Linkages among Social Change INGOs, and between Social Change INGOs and IGOs

Chapter 1 of this volume argues that to understand transnational social movement activity, one must look not only at single social change INGOs and social movements, but also at campaigns organized by loosely affiliated

infrastructures for change 39

networks, coalitions, and movements. In general, we and others have argued, denser networks and coalitions are likely to be more effective in achieving their goals, especially if they have significant network, coalition, or move- ment participants within the country or countries targeted by the campaign. So, for example, Darren Hawkins argues that the density of the transnational networks working on human rights in Chile was an important part of the explanation for human rights change in that country. Especially important, however, was that social movement activity was often initiated and sustained by significant domestic human rights organizations within Chile and by net- works of Chilean exiles working abroad.

The best measure of network scope and density that one can glean from the Yearbook data is the number of countries reported among social change INGO membership and their reported links with IGOs and other NGOs.

Table 2.4 shows that there has been some growth in the geographic expansion of social change INGO memberships. The number of groups re- porting members in twenty-five or fewer countries remained fairly constant, at least between 1963 and 1993. But between 1953 and 1993, we see some clear growth in the scope of social change INGOs as the proportion of groups reporting fewer than twenty-five countries of membership declined from 68 to 55 percent while the numbers reporting more than forty mem- bership countries grew from 15 to 28 percent. Of these, 8 percent of the organizations now report members in more than eighty countries, up from 2 percent in 1953. The average number of countries of membership for all the social change INGOs in the data set increased from twenty-six in 1953 to thirty-four in 1993.

40 sikkink and smith

Table 2.4. Geographic Dispersal of Members

Number of Countries of 1953 1963 1973 1983 1993 Membership (N = 90) (N = 125) (N = 153) (N = 233) (N = 528) 3–10 19% 14% 12% 20% 17%

11–25 49 38 32 34 38

26–40 17 23 24 21 18

41–80 13 21 27 20 20

More than 80 2 4 4 5 8

Average number of 26 (21) 30 (24) 34 (28) 31 (23) 33 (23) member countries (median)

Social change INGO contacts and formal relations with IGOs and NGOs remained constant between 1953 and 1983. By 1993, however, groups reporting any ties with other intergovernmental or nongovernmental organizations were more likely to have broader ranges of ties. The propor- tion of groups reporting having ties with more than three IGOs more than doubled from 17 percent in 1953 to 37 percent in 1993. The expansion of ties with NGOs is most dramatic after 1983. In 1983, 90 percent of the groups reporting any NGO ties said that they had fewer than four links to other NGOs. By 1993, only 53 percent of such groups had fewer than four links to other NGOs. The proportion of social change INGOs reporting links with four or more other NGOs grew nearly fivefold, from 10 percent in 1983 to 47 percent in 1993. This pattern clearly supports our claim that network density, especially among NGOs, has increased significantly. Also, we would argue that these figures still significantly underestimate the num- ber of network links, since many informal ties exist and these might not be reported on the UAI questionnaire that asks about significant organizational contacts.

infrastructures for change 41

Table 2.5. Contacts with IGOs and NGOs

IGOs Number of 1953 1963 1973 1983 1993 links with IGOs (N = 48) (N = 60) (N = 80) (N = 126) (N = 360) 1–3 83% 76% 70% 78% 62%

4–10 17 24 30 21 30

More than 10 0 0 0 0 7

Average — — — 1.3 2.3

(Standard deviation) — — — (1.9) (4.4)

NGOs Number of 1953 1963 1973 1983 1993 links with NGOs (N = 61) (N = 48) (N = 68) (N = 123) (N = 489) 1–3 88% 87% 91% 90% 53%

4–10 10 10 9 8 32

11–20 0 2 0 2 10

More than 20 2 0 0 0 5

Average — — — 0.98 4.88

(Standard deviation) — — — (1.8) (7.28)

Organizations were not asked to report on their funding links to foun- dations, nor on kinds of linkages to governments, so the total number of network linkages is not represented in the data set. A number of chapters suggest that informal linkages, based on personal trust and ties among indi- viduals in different groups, may be key linkages in some cases, but these would not be reported here.

Conclusions

This chapter outlines the broad shifts in the organizational infrastructure that supports global social change efforts. While these organizations are not the only actors behind important global changes, the fact that their principal reason for existing is to promote some social or political change means that they are likely to be key figures that activate and sustain transnational advo- cacy networks. Unlike governments and foundations, these organizations are collections of individuals who care deeply about the social change goals of their organization and who spend large amounts of their time and energy working to promote these goals. Thus, understanding this infrastructure helps us better understand the changes in and potential for transnational networks, coalitions, and movements.

The data we have analyzed suggest that the population of transnational social movement organizations is expanding both numerically and in geo- graphic scope. The number of organizations grew more than sixfold between 1953 and 1993. While there are still gaps between the participation of citi- zens of less industrialized countries and their Northern counterparts, this gap appears to be shrinking as more developing countries are represented among social change INGO memberships and as more social change INGOs locate their organizational headquarters outside the global North and its global cities. These trends suggest that transnational advocacy work is more likely now than in the past to incorporate more globally representative voices. They also suggest that such networks are more likely to be account- ing for the needs and interests of people from regions outside the industrial- ized nations as they advocate for global change.

In addition to noting expanding size and scope of the transnational social movement infrastructure, we found a trend toward greater density in the linkages among social change INGOs and other actors, such as other NGOs and international agencies. This reflects the success of past work by social change INGOs to defend and expand their access to IGOs and it demonstrates the extent to which these groups have become routine actors in international affairs. Moreover, it shows that through decades of work in the international system, social change advocates have learned to cultivate

42 sikkink and smith

networks among themselves in order to share information and coordinate strategies.

We did not find unambiguous support, however, for either of the two arguments about the impact of international institutions on the growth of social movement organizations. In none of the cases did the entry into effect of the relevant treaty appear to contribute to a significant increase in the number of social change INGOs, at least in the short term. Preexisting inter- national norms do seem important for the effectiveness of networks and movements, but they do not lead immediately to the growth of movements. In some cases, such as the two environmental conferences and the 1993 World Conference on Human Rights in Vienna, large international confer- ences appear to contribute to the growth in the number of social change INGOs in the same issue area. In other issues, such as women’s rights and the 1968 International Conference on Human Rights in Tehran, inter- national conferences did not contribute immediately to growth in the num- ber of social change INGOs organized on the issue.

In summary, parallel to the expansion of economic and political global- ization we find that nongovernmental actors have worked to expand and de- velop global social infrastructures. Seeking to shape the character of global- ization and its policies are a larger number of more geographically diverse organizations with stronger and denser communications networks than were present at the founding of the United Nations. This sector of society has proven influential in shaping global changes in the past, and it is likely to continue to be a voice (however discordant) for principled change.

Notes

1. For other discussions of the population of transnational social movement organizations, see Smith 1995, 1997, and 1999. For discussions of the applications of sociological research on social movements for the study of international NGOs, see Smith, Chatfield, and Pagnucco 1997, and Smith 2000.

2. As far as we are aware, this data set spanning fifty years is the only effort to provide quantitative historical data on social change INGOs from the same source.

3. The data in the Yearbook is the most complete census of international or- ganizations, which included information on organizations’ founding dates, goals, memberships, and interorganizational ties. The Union on International Associations uses UN records on nongovernmental organizations (NGOs), self-reports, referrals, and the media to identify organizations and compile organizational profiles for the Yearbook. Editors consistently check this reported information against other sources, such as periodicals, official documents, and the media. We have confidence in this data because it has been gathered by the same organization using the same methods

infrastructures for change 43

over time. Nevertheless, some of the increases we see in the total numbers of social change INGOs may be the result of more complete methods of identifying and sur- veying international nongovernmental organizations in the current period.

4. For more details on coding, see Smith 1995. 5. Transnational movements and networks have also been catalysts for the es-

tablishment and elaboration of international norms; see, e.g., Risse-Kappen 1995a, Finnemore 1996, and Chatfield 1997.

6. On NGOs at these meetings, see Clark, Friedman, and Hochstetler 1998. 7. Many private and government funders require groups seeking resources

to propose specific projects. This may help explain the pattern we find, as organizers are encouraged to specialize their groups’ work around key issue areas or legal/ institutional arenas, such as the Human Rights Sub-Commission or the Commission on Sustainable Development. Also, during the latter half of the twentieth century, the number of United Nations functional agencies and treaty bodies grew tremen- dously, enabling more specialized work in very focused arenas.

8. Many of the larger international relief organizations that would be consid- ered important players in transnational networks (e.g., CARE) are excluded from this data set because they are not international in structure or because their organiza- tional missions are focused more on relief or charity than on social change.

9. For a more detailed discussion of Esperanto groups, see Kim 1999. 10. Unfortunately, we were not asking this question when we recorded infor-

mation on these groups and we did not code separate cities within the United States. We are certain, however, that a substantial portion—if not a majority—of these groups are located in New York or Washington D.C., for the same reasons outlined above.

44 sikkink and smith