Embedding Post-Capitalist Alternatives? The Global Network of Alternative Knowledge Production and Mobilization

Since the 1970s, transnational alternative policy groups (TAPGs) have emerged as a component of global civil society, generating visions and strategies for a “globalization from below” that point toward post-capitalist alternatives. Here, we map the global network of TAPGs and kindred international groups in order to discern how TAPGs are embedded in a larger formation. In this era of capitalist globalization, do TAPGs, like their hegemonic counterparts, bridge across geographic spaces (e.g. North-South) and movement domains to foster the convergence across difference that is taken as a criterial attribute of a counter-hegemonic historical bloc? Our network analysis suggests that TAPGs are well placed to participate in the transformation of the democratic globalization network from a gelatinous and unselfconscious state, into an historical bloc capable of collective action toward an alternative global order. However, there are gaps in the bloc, having to do with the representation and integration of regions and movement domains, and with the salience of post-capitalism as a unifying social vision. Also, our architectonic network analysis does not reveal what the various relations and mediations in which TAPGs are active agents actually mean in concrete practice. There is a need both for closer analysis of the specific kinds of relations that link transnational alternative policy groups to other international actors, including intergovernmental organizations and funding foundations, and for field work that explores the actual practices of these groups, in situ. a component global civil society generating a below” post-capitalist alternatives. Such the Transnational the Global South International on Globalization Francisco) for Civil Society as “collective intellectuals” in facilitating the construction of a fledging historical bloc that transects national borders poses democratic alternatives to neoliberal globalization 2011). This study explores a social network of alternative think tanks that critique capitalist economic and social organization and that might be said in some respects to foster practices of transition from a capitalist present to a post-capitalist future. We proceed from a neo-Gramscian understanding that hegemonic think tanks and TAPGs are embedded in opposing historical blocs, as they develop and deploy knowledge with the intent to make their respective blocs more coherent and effective (Carroll 2007). The hegemonic bloc can be conceptualized as a variegated, multi-tiered network of institutions, organizations, publics and individuals that provide leadership in the ongoing globalization of capitalism (cf. Smith 2008). Such globalization “from above” can incorporate reformist initiatives that address concerns expressed “from below” – as in the World Bank’s opening to civil society in the late 1990s, or more recent advocacy of the “green economy,” but cannot subvert “the decisive function exercised by the leading group in the decisive nucleus of economic activity” (Gramsci 1971: 161). world’s corporations key policy boards

atomism in liberal economics, insisting on "the intrinsically relational nature of all social action" (Krippner and Alvarez 2007: 231).
Both senses of embeddedness are relevant to our study of TAPGs. As for the first, in producing and mobilizing alternative knowledge, TAPGs struggle against the market logic of capitalist disembedding; they offer cognitive resources to re-embed economic relations, under democratic control. Ideologically, Somers and Block (2005) point out that the hegemony of neoliberalism actually embeds the market within structures of common sense, particularly in the domain of political policy but also in wider cultural fields. TAPGs are to a considerable degree engaged in discursive struggles to dislodge neoliberalism from its embeddedness within common sense, but their intellectual production extends beyond critique to the advancement of alternative visions, strategies and policies. To have effect, however, this praxis must produce not simply knowledge of post-capitalist alternatives; it must produce -in collaboration with the agency of allied groups -transnational counter-publics, cultures of solidarity and communities of practice that instantiate that knowledge, embedding it in and beyond their own lifeworlds. This sort of embedding is not centered upon national states (as in Polanyi's formulation from the 1940s); the emphasis is on constructing critical knowledges and solidarities, across borders, within global civil society -understood as a terrain of contention (Conway and Singh 2009).
For the purposes of this paper, we work with Granovetter's network concept of embeddedness. Focusing on a purposive sample of 16 TAPGs -eight based in the global North and eight based in the global South -we ask how these sites of alternative knowledge production and mobilization are embedded in a larger network of international NGOs (INGOs), and what the nature of that positioning might mean for their capacity, as cultural and political agencies, to embed post-capitalist alternatives.

Method
Our analysis centers upon a judgment sample of 16 major TAPGs, each of which satisfies these criteria:  a core function of knowledge production and mobilization that challenges existing political-economic hegemonies and presents alternatives, creates new paradigms, etc.;  a significant part of that cognitive praxis takes up transnational issues and speaks to transnational counter-publics;  the group engages a wide range of issues, i.e., it is not specialized in one domain (such as water, trade, or capital-labor relations).
The groups were selected to achieve geographical representation of every region of the world, with recognition of the fact that Europe contains a particularly large contingent of INGOs, relative to the rest of the world (Smith and Wiest 2012: 62). Table 1 offers a temporally sequenced list of the 16 TAPGs. Four groups formed in the mid-1970s (three of them based in the global North), at the culmination of the 1960s protest wave, and as the crisis of the post-war era set in. Four (all based in the global South) formed in the 1980s. In the past couple of decades, and particularly from the mid-1990s to 2005, TAPGs proliferated as an intellectual aspect of the gathering global democracy movement, but also as critical responses to the crises and contradictions of neoliberal globalization (Carroll 2013). These 16 groups are diverse in organizational form and political priorities, including the extent to which they focus their efforts on prefigurative KPM that is relevant to the construction of alternative futures, as distinct from the critique of existing reality (cf. Coy et al. 2008). Groups such as the Third World Network (TWN), the Tricontinental Centre (CETRI) and Third World Institute/Social Watch (ITeM) focus on critique of current political-economic conditions; CACIM and the Network Institute for Global Democratization (NIGD) emphasize the construction of "open spaces" for political dialogue; groups like Focus on the Global South (Focus), the International Forum on Globalization (IFG) and the Transnational Institute (TNI) consciously strive to construct counter-hegemonic projects that prefigure actual alternatives to neoliberal capitalism, as in Focus' "deglobalization" paradigm, IFG's post-capitalism project and TNI's New Politics initiative. Despite these specificities (explored in more detail in Carroll 2013), the political sensibilities of all 16 groups seem to converge upon a conception of sustainable human development that blends radical-democratic and ecological imaginaries.

Data Collection
Our main source of information, for TAPGs and other organizations, was the Yearbook of International Organizations (YIO, online edition, 2012). Following a snowball sampling methodology, we used the YIO to collect the list of all first-and second-order neighbors of the initial 16 TAPGs, as presented in Table 2. First-order neighbors are organizations directly linked to one or more TAPG, and thus in the immediate neighborhood of TAPGs. Second-order neighbors are in the immediate neighborhoods of one or more first-order neighbors, but not directly linked to a TAPG.
The YIO lists a variety of types of relationships between organizations: resource and financial flows (as when an organization supports another by providing funding, office space, or other types of support); information flows (as when an organization acts as a consultant or advisor for another, or is an accredited observer at meetings); partnerships and collaborations taking the form of short term joint projects or long term alliances; coordination of an organization by another one (e.g., representation on the board or coordinating committee); and the various flows and collaborations taking place between organizations and their membership. For the purposes of this study, we assumed that all these links were indicative of a substantial relationship between organizations, and thus included them all in the analyses presented in the latter sections of this paper (see Katz 2006 for a similar use of YIO data). We gathered data on all relationships connecting TAPGs directly to their first-and indirectly to their second-order neighbors, plus all relations among the first-order neighbors. In effect, we used a 1-network sampling strategy to take a core network (the 16 TAPGs) and then to include all nodes linked to the nodes within the core network (Wasserman and Faust 1994), but we supplemented this with relations between the 1-network and all international organizations linked to any of its members. To keep the focus on civil society organizations, international governmental organizations, states and state agencies were excluded, yielding a sample of INGOs, broadly construed. 2 Although the YIO provides extensive listings of INGOs, its coverage is not complete. In the case of TAPGs that did not have an entry in the YIO, we gathered data on links we judged to be equivalent to YIO relationships, from the organizations' websites. For three TAPGs, information was not available from either the YIO or the organizations' websites. They were contacted by email, and two graciously provided a list of the organizations they collaborate or otherwise have substantial relationships with, which we included in our database. The other did not reply to our request and was thus excluded from the study. For first-order neighbors, we limited our data collection to the YIO, or to organizations' websites in the case they were absent from the YIO. 3 As Table 2 and Figure 1 show, TAPGs are indeed embedded in an extensive network. Their intersecting neighborhoods add 247 INGOs to the network. Taking the neighborhood to include second-order neighbors adds another 3714 INGOs. The sociogram shows the configuration as a cloud with TAPGs at or near the center, a result to some degree built into our snowball method. 4 In addition to network data, we browsed each organization's website to gather individuallevel data for all TAPGs and their first-order neighbors. We first geocoded the location of each organization according to the city in which its headquarters are located. Second, we classified each according to a typology of its main activity, such as alternative media, social movement, NGO, foundation, and so on. Finally, we also collected information about the main "movement domain" of each organization, i.e. the way each frames its action in relation to what it sees as the main problem to be addressed: for example, ecology, globalization, capitalism, or patriarchy (see the analysis section for more detail on the coding). Tabulations were produced using UCInet 6 (Borgatti et al. 2002); graphs were produced using ORA, version 2.3.6 (Carley et al. 2012).

Figure 1. The first-order and second-order neighborhoods of TAPGs
Embedding Post-Capitalist Alternatives 218

TAPGs within Civil Society: Basic Patterns and Parameters
The issue of embeddedness is multifaceted. Our first approach takes the perspective of network centrality. On this issue, our snowball method of case selection allows us to compare the centrality of TAPGs with the 247 INGOs in their immediate neighborhoods, which include many well-known organizations. Table 3 lists the organizations in the sample with the highest degree centrality, calculated on the basis of ties involving all 3977 organizations, and shows the centrality ranking of each TAPG within the network. Degree centrality is the sum of all incoming and outgoing ties of an organization (see Freeman 1978Freeman /1979. Given the structure of the data (which excludes relations among second-order neighbors to TAPGs), degree is the only measure of centrality that can be meaningfully computed for the network of TAPGs and their 247 immediate neighbors. This affords us a comparison of the relative centrality of transnational alternative policy groups among the 263 INGOs.
Here, we find the bulk of TAPGs to inhabit spaces near but not at the network center. Two TAPGs -Focus on the Global South and CRID -appear among the top 30, and nine others rank among the top 100 organizations of the 263 for which we collected complete network data. However, five TAPGs, all of them based in the global North, occupy positions more on the margins of the network (though only two are notably marginal). Overall, TAPGs show a slightly higher mean degree than the INGOs in their immediate neighborhoods, with respective means of 36.75 and 29.91 (n=263; eta 2 = 0.0013). When we consider the geographical location of their headquarters, we find that TAPGs based in the global South are particularly well connected within the network, which stands to reason since the five relatively marginal TAPGs are all Northern-based. The mean degree for Southern-based TAPGs, 45.13, is considerably higher than the mean for TAPGs from the North (28.38), (n=16; eta 2 = 0.1575).  When we map the relations between the 44 organizations in Table 3, we arrive at the configuration in Figure 2. This sociogram uses a spring embedded algorithm that projects the nodes in a two-dimensional space so that those with the shortest path lengths also appear the closest in the graph (Hanneman and Riddle 2005). The points representing TAPGs are circled by rings. Even though few of them appear among the 30 most central nodes, TAPGs nonetheless link tightly to the central organizations, with some of them (Focus, Third World Network, DAWN, RosaLux) acting as hubs that pull the network core closer together. Organizations from the global South (in gold) and the global North (in blue) seem to be fairly evenly distributed in the network, with TAPGs headquartered in the South occupying slightly more central positions. 5 The position of the World Social Forum (WSF), ranked 9 th overall in degree centrality, at the center of this subnetwork, is striking. Contributing to its location are ties radiating to 12 TAPGs. Clearly the WSF, as an "open space" for considering the possibility of another world, has been a key site for the efforts of TAPGs. Indeed, TAPGs that are relatively marginal in the network overall, such as NIGD, IFG and CETRI, appear in this sociogram as fairly ensconced among core organizations, in part by virtue of their links to the WSF.
It is not surprising to find TAPGs near but not exactly at the core of the network. The most central positions are occupied by some of the major, well-resourced, international development NGOs and environmental networks with substantial membership bases, and by global forums that bring movements together on a cross-sectoral basis (WSF, Civicus, NGO Forum on ADB). As producers of alternative knowledge, TAPGs connect extensively with what Smith and Wiest (2012) call transnational social movement organizations, but they do not dominate the segment of global civil society in which they are immediately embedded. Rather, as they work interdependently with groups that are more oriented toward movement process (WSF) or political action (EAN, CJN, GCAP, FoEI), they connect with some of which are the most central in the network.  Table 3 except Climate Justice Now! have headquarters in specific cities, some, like DAWN, Alternatives International, and the World Social Forum, are global networks, whose headquarters simply house a secretariat. For these groups especially, the geocoded coordinates do not necessarily represent the center for actual activities.
Embeddedness always has a spatial dimension. To explore this, in Figure 3 we map the entire network of 16 TAPGs and their 247 direct neighbors, according to the location of each organization's headquarters. One can observe extensive traffic across the North Atlantic, and in this respect the network bears some resemblance to the elite network associated with the transnational capitalist class (Carroll 2010). Similarly, organizations based in Europe are particularly prominent. However, Tokyo, a global city for business, is marginal to the network, while Southern cities like Durban, Montevideo and Manila appear as important switch points. Although many ties connect across North and South, there is also extensive networking among Southern-based INGOs, spanning South America, Africa and Asia. Conspicuously absent are organizations based in Russia and China. 6

Figure 3. TAPGs and their immediate neighbors as a global network
Embeddedness can also be appraised as an aspect of geodesic distance. Points can be connected directly, but in large networks most are mutually reachable only indirectly, through paths along which other points serve as intermediaries. The shortest path between a pair of points defines their geodesic distance. For the network of 16 TAPGs and their immediate neighbors, Table 4 provides the distribution of distances in the network.
Although only 2.3% of all organizations are directly connected, another 21% are linked at one remove, while 46% are reachable at two removes, i.e., nearly 70% of the 263 organizations can reach each other via no more than two intermediate network members. The network's diameter is 5 -for three percent of pairs of organizations a path across four intermediates is required. The mean distance in the network is 3.1. These parameters give us a baseline for assessing the integrative contribution that TAPGs make to the segment of global civil society in which they are embedded. In a network organized around policy groups, a basic way of assessing the contribution policy groups make to structural cohesion is to assess the impact that removing the policy groups has on the connectivity of the other organizations. Carroll and Carson (2003) find that, for the transnational hegemonic bloc, removal of five key policy groups has a substantial structural effect, that the mediations of policy groups dramatically reduce distances between corporate boards in the interlock network. When we similarly dis-embed TAPGs from the transnational alternative policy network, the network also loses connectivity. Most obviously, 74 organizations disappear from the connected network (twelve of them form six dyads; the others become isolates). Among the 173 remaining organizations that form a single, connected component, mean distance increases from 2.092 when TAPGs are included as mediating agents, to 3.279 when we dis-embed them. In these comparisons we can see that the impact of TAPGs on the network is twofold: the fringes of their social circles bring 74 organizations into a connected configuration from which they are otherwise isolated; among the other 173 organizations the mediation of TAPGs in connecting across various global civil society groups reduces mean distance by 36.2%. These findings show that TAPGs are embedded in a segment of global civil society which, although not without its own interconnections, derives considerable structural coherence and connectivity from the mediation provided by TAPGs.

Inter-regional and North-South Mediation
One of the main questions driving this paper is whether TAPGs bring the alter-globalization network closer together across space, especially between countries of the global North and those of the global South. Table 3 and Figure 3 already offer some indications as to the positions most TAPGs occupy near the center of the network. The analyses we present next explore the differences among TAPGs in their particular patterns of mediation.
Table 5 presents the regional distribution of TAPGs and directly linked organizations, using an eight-fold classification proposed by Smith (1997; also used by Dewitt 2008 andCarroll 2010) that incorporates the core-periphery distinction within the world-system. As already emphasized in Figure 3, many organizations have their headquarters in Western Europe; North America, Asia, and Sub-Saharan Africa (especially South Africa) are also well represented. However, Eastern Europe and the Middle East/North Africa are under-represented in the network. As embedded actors in global civil society, how do specific producers and mobilizers of alternative knowledge connect across spatial difference? At one extreme, that of complete introversion, an organization might maintain its entire network neighborhood within its own region; at the other, that of complete extraversion, its network neighborhood might be located entirely "elsewhere" -beyond its own spatial locale. We compared the extent to which TAPGs link within their own region and outside of it by calculating E-I (E minus I) indices. 7 Positive numbers indicate extraversion (i.e. a predominance of ties that reach beyond the group's own region); negative numbers indicate the group links mostly within its own region. Table 6 shows two E-I indices for each TAPG, the first one calculated relative to the eight-fold categorization presented above (Table 5) and the second one based on a dichotomous North-South divide. In the latter case, a relation is introverted if it connects, say, a Northern TAPGs to another Northernbased organization, and extraverted if it connects, say, a Southern TAPG to an organization based in the North.
First, let us look only at the E-I index for the eight-region classification in Table 5. Overall, the 16 TAPGs are regionally extraverted: 198 of their 336 ties extend beyond a given group's region (E-I = 0.179). However, the patterns of extraversion-introversion vary quite substantially. All the Southern-based TAPGs are extraverted in their network relations, most of them extensively so. Northern-based TAPGs are much more variable along this dimension, and show a slight overall introversion. Among the northern introverts, CRID (E-I = -0.784) participates at the heart of the French development NGO network while People's Plan Study Group in Japan (E-I = -0.647) connects mostly with other Japan-based organizations. The "regionally introverted" TAPGs are all based in the global North. Among the most extraverted TAPGs, the ones with sizeable numbers of ties (Third World Network, DAWN, and PRIA) are all located in the South. Two TAPGs based in Europe (TNI and RosaLux), however, also show extensive ties beyond their region, as does Alternatives International, which is structured as an international network, with its secretariat based in Montreal. Combining the basic finding from Table 6 with our earlier analysis of degree centrality, we see that TAPGs located in the global South tend to engage in more extensive networking and to connect with organizations outside of their region, whereas Northern-based TAPGs network less, and some show an introverted tendency to link within their region.
Comparing the regional and North-South E-I indices provides us with more insight into the patterns of mediation by Northern-and Southern-based TAPGs. On the one hand, TAPGs in the North present almost the same patterns of extraversion-introversion, with a total North/South E-I of -0.323, compared to -0.218 for the eight regions E-I. This indicates that their internal links tend to keep within their regions, whereas their external links connect them mainly to organizations located in the South. On the other hand, the E-I pattern varies much more markedly for TAPGs from the South, especially for the Third World Network, DAWN, and the Third World Forum. These groups are extraverted in their ties across the eight regions, but introverted when it comes to the North/South divide: their ties span across the regions of the global South, but connect less to organizations in the North. Overall, Southern-based TAPGs show a mild tendency to connect with other organizations of the South (E-I = -0.143). Thus, their relatively pronounced extraversion across the eight regions (E-I = 0.438) signifies that they mainly bridge across regions of the global South. Only Focus and PRIA are extraverted toward the North while Centre for Civil Society and ITeM/Social Watch show a balance in their North/South ties.   Table 5) were excluded from this analysis.
Brokerage analysis provides another angle from which to consider how TAPGs mediate relations between the Northern and Southern segments of global civil society. An ego is said to broker between a pair of unconnected alters if it is linked to both of them, thus connecting them at one remove (see Gould and Fernandez 1989;Stovel andShaw 2012). 8 McAdam (2003) and Vasi (2011) show that brokerage plays a major role in spreading contention more rapidly and across lines of fraction within civil society. As well, recent work by Von Bülow (2011) emphasizes how transnational civil society actors deliberately pursue brokerage strategies in seeking to build long-term collaborations between transnational civil society actors: thus the importance of looking at how TAPGs broker relations within the structure of global civil society.
We can parcel each TAPG's set of brokerages into those that mediate between two organizations of the global North (North-North), those that mediate between two organizations of the global South (South-South) and those that mediate between one organization of South and one of North (North-South). By definition, North-South brokerage helps integrate the Northern and Southern segments of global civil society. The significance of North-North and South-South brokerages for transnational historical bloc formation depends on the location of the mediating group itself. TAPGs based in the global North, whose brokerage relations mediate between other Northern-based groups, integrate the Northern region of global civil society. The same intraregional integration occurs when the brokering TAPG and the brokered relations are all contained in the South. Alternatively, however, Southern TAPGs that broker relations between Northern-based groups may be seen as engaging, from a Southern position, with the Northern region, thereby constructing bridges from South to North. Table 7 presents each TAPG's pattern of brokerage within and between the global South and the global North, using undirected absolute brokerage scores.

Table 7. Relations brokered by TAPGs between North and South (n=253) a Organization
Total brokerage  a Organizations that do not have a headquarter in a single location or for which no location could be determined (see Table 5) were excluded from this analysis.

North-South
We find a remarkable amount of variation among TAPGs in the overall volume of brokerage. Four groups -two in the South (CCS and Focus) and two in the North (RosaLux and CRID) -account for 72% of all the relations brokered by the 16 TAPGs. Among these four highvolume brokers, CRID mediates many Northern-based relations and extremely few Southernbased ones. It also links at one remove 57 Southern-based organizations with Northern-based ones, consistent with its project of building North-South solidarity relations around development issues. The other three show more even brokerage profiles, mediating within and across the two zones. TAPGs with few network connections are, of course, not heavily involved in brokerage, although the People's Plan Study Group mediates 208 North-North relations and 29 North-South ones. Alternatives International (which functions as an international network, with secretariat in Montreal) is the second most active South-South broker among the Northern-based TAPGs. Still in the North, the Transnational Institute and RosaLux display particularly balanced brokerage profiles, mediating between organizations based in and across the global North and South. Overall, TAPGs based in the North mostly broker relations between Northern INGOs, yet hundreds of these mediated relations involve pairs of Southern INGOs, or link South to North. These latter, cross-regional brokerages are especially profuse among Southern TAPGs. Overall, the profile for Southern TAPGs is quite balanced. Whereas among the six Northern TAPGs 63% of brokerage is contained within the North, among the six Southern TAPGs 35% of brokered relations are South-to-South, with the North-South and North-North categories claiming close to a third of all brokerage each. PRIA (whose project to proliferate knowledgedemocratizing practices in Asia and more globally links it to partners like UK-based Institute of Development Studies) is the only group for which a majority of mediations run North-North. DAWN, CACIM, and Third World Forum primarily broker between Southern-based INGOs, although DAWN and CACIM also participate in dozens of North-South mediations. The overall brokerage analysis shows Southern-based TAPGs to be involved in complex mediations that contribute to North-South solidarities as well as to the formation of a Southern network. As noted above, for Southern-based TAPGs, North-North brokerages are themselves instances of North-South mediation. Thus, while most of the relations brokered by Northern TAPGs contribute to a Northern NGO network, the tendency for Southern-based TAPGs is to participate in (and mediate) that network while also building relations in the South and between Southern and Northern segments of global civil society.
To illustrate patterns of brokerage between North and South, Figure 4 presents sociograms of the immediate neighborhood of four TAPGs, two located in the global South and two in the global North. The patterns seem congruent with each group's specific project. In the North, CRID, a coalition whose purpose is in good part to build unity among progressive development NGOs, is embedded within a Western European network (in blue), which links to Coordination Sud and Ritimo, two major French NGO networks; hence, its brokerages run North-to-North. RosaLux, with a strong commitment to building a social base for democratic socialism in Germany and internationally, is well embedded within Western Europe, where it links with global trade union organizations such as the International Trade Union Confederation (ITUC), its European counterpart (ETUC), and the Trade Union Advisory Committee to the OECD (TUAC). But RosaLux, through its Center for International Dialogue and Cooperation, also has several neighbors in Sub-Saharan Africa (in yellow, including the Durban-based Centre for Civil Society) and in Latin America, Eastern Europe, and Asia. This combination of ties to Northern and Southern INGOs, which are not directly linked to each other, positions RosaLux as a broker within and across North and South.
In the South, consistently with the general observation that Southern-based TAPGs bridge between regions of the South, DAWN links extensively with organizations from Asia (in red), Sub-Saharan Africa (in yellow), and Latin America (in purple); it also collaborates with two North American development feminist NGOs: the Association for Women's Rights in Development (AWID) in Toronto and the Women's Environment and Development Organization (WEDO) in New York. DAWN's membership base (feminist scholar-activists based in the global South) and its explicit commitment to fostering a Southern feminism has led it to link across the global South, and thus to broker a good many South-South relations. Unlike DAWN, Focus, whose project is to "Focus on the Global South" in a way that strengthens North-South solidarity while building capacity in the South, links not only across the South but with all three regions of the North, thus bridging between, on the one hand, the Asian NGO network and, on the other hand, the networks of North America (in light blue), Western Europe, and Japan (in green). This pattern positions Focus as a North-South broker, but also means that it brokers across the regions of the North and, to a lesser extent, the South.

Figure 4. Embeddedness of four key TAPGs within their respective neighborhoods
Embedding Post-Capitalist Alternatives 230

Embeddedness and Mediation Among Movement Domains
The formation of a global counter-hegemonic bloc requires reaching out across different regions of the world, but also across different locations within the conceptual space of alternative and oppositional politics. Movements distinguish themselves in part by frames they use in analyzing political issues and in constructing strategic pathways toward change (Snow and Benford 1992). To assess whether TAPGs do in fact bridge across this differentiated space, and the shape this mediation takes, we studied the websites of all TAPGs and their immediate neighbors and coded each organization into a domain category representing the predominant framing of its political project, including how each group defines the main problem it addresses (see the data and methods section above). The main categories are presented in Table 8 together with the associated frequencies and a listing of where we placed each TAPG in the schema. This tabulation offers a rough picture of the political composition of the segment of global civil society in which TAPGs are immediately embedded. It is in the first place interesting to note the ideological diversity of the formation. Nearly one-fifth of the 263 organizations (including three TAPGs) cast their projects within multi-issue frames: they do not focus on a single concern but on a range -without, however, providing a strong analysis of how those issues may be interrelated. 9 "Critical liberal" groups, comprising 16% of the formation (including two TAPGs), emphasize substantive human rights issues without articulating a critique of relations of production. Alter-globalization -focusing on a strong critique of transnational neoliberalism and its institutions -account for another 14% (and four TAPGs). Liberal humanitarianism, a frame that promotes international charity and falls well within hegemonic discourse, claims 8% of the network; intersectional political projects that provide an explicit analysis of the inter-relations between multiple issues, and ecological political projects that focus on impacts of humans on their environment also each represent 8% of the network. Frames that emphasize global solidarity, feminist, and anti-capitalist politics account respectively for 6%, 5%, and 2% of the 263 INGOs (and five TAPGs in total).
The classification is admittedly imperfect; for instance, our categorization of DAWN as feminist, while accurate, ignores its deep and longstanding commitment to "interlinkage analysis," a strong form of intersectionality. And although RosaLux is the only TAPG that articulates a clear, anti-capitalist (and democratic socialist) perspective, critiques of neoliberal policies and corporate power, and advocacy of alternatives, are the stock-in-trade of all 16 TAPGs. In any case, what this analysis of the frames that predominate among TAPGs and in their immediate neighborhoods indicates is considerable diversity. The range represented among the neighbors is particularly broad, extending from the liberal mainstream to radical anti-capitalism. The question is, how are these perspectives socially organized into a global network within which the alternative policy groups are active, and how do the latter mediate relations in the former?
In the introduction, we suggested that TAPGs' contribution to transnational historical bloc formation might involve their bridging across political frames to foster a "convergence across difference." From the above categorization of TAPGs and their neighbors, we can pursue a network-based approach to grappling with this issue. The question can be put as follows: how extensively do TAPGs connect with INGOs in ways that bridge across movement domains, to connect movement actors whose self-understandings differ, into articulated ensembles of communication and practice? Figure 5 plots in a bipartite sociogram the relationships between TAPGs and neighboring organizations, aggregating the latter according to the main political frames to which they subscribe. By aggregating TAPGs' network neighbors into meta-nodes representing predominant domains of movement politics, we are able to observe how alternative policy groups connect with the broad segments of global civil society in which they are embedded. TAPGs are represented by dots (gold indicating South-based, blue indicating North-based); aggregations of INGOs categorized within each major movement domain are represented by black squares. The configuration is arranged according to a spring embedded layout. Line thickness indicates the density of links between nodes, and the graph leaves out links with density lower than 0.1 so as to capture only the stronger relations between TAPGs and domains.
The figure indicates that TAPGs do occupy strategic positions between movement domains, and that their relations are not limited to groups associated with their own domain.

Figure 5. TAPGs mediating between movement domains
Focus and RosaLux each mediate among six different domains (including the liberalhumanitarian sector), with RosaLux showing relatively extensive ties to anti-capitalist groups. Third World Forum maintains ties to five movement domains. The Centre for Civil Society, also with extensive links to anti-capitalist INGOs, mediates among four different domains (including critical liberal groups), as does CRID, whose connections with global solidarity groups are particularly profuse. Only three TAPGs, DAWN, 10 IFG, and Alternatives International, maintain substantial ties to just one of the major domains. 11 Viewing the sociogram from the perspective of movement domains, eight TAPGs connect with the alter-globalization domain, six connect to the ecological domain, five are tied to groups with multi-issue frames, and five are tied to the critical-liberal segment. Additionally, four TAPGs link to groups whose political projects can be described as anti-capitalist, intersectional, or liberal-humanitarian. This configuration suggests that, as they produce and mobilize alternative knowledge, TAPGs mediate among a plurality of movement sectors and political projects. Their cognitive praxis positions them to speak to multiple counter-publics, with the possibility of fostering a convergence across difference. The ideological frameworks with which TAPGs engage extend from liberal-humanitarian to radical anti-capitalist.
A final measure of the extent to which TAPGs mediate among groups within global civil society can be seen in the heterogeneity of each group's immediate neighborhood, as presented in Table 9. A group whose neighborhood, or social circle, is heterogeneous has direct ties to a diverse array of other groups. The heterogeneity index (which varies from 0 to 1) shows that, overall, TAPGs have much more diversified social circles than the other organizations, across all three variables -region, North/South, and movement domain. Despite great differences in the size of their social circles (already evident above in Table 3), all TAPGs maintain social circles that take in a diverse range of movement domains. And, with the exceptions of CRID and PPSG (whose social circles are nationally-focused), TAPGs show much higher levels of diversity in bridging regional and North/South differences.

Discussion
From several vantage points, our exploration of the embeddedness of transnational alternative policy groups suggests that they are indeed positioned to mediate across differences in global civil society, and thus to facilitate the formation of a transnational counter-hegemonic bloc.
 While not the most central organizations within their segment of GCS, most TAPGs are relatively central, and connected to such highly central actors as the Climate Justice Network, the Global Campaign Against Poverty, Friends of the Earth International, and of course the World Social Forum. As producers of alternative knowledge, TAPGs connect extensively with what Smith and Wiest (2012) call transnational social movement organizations, but they do not dominate the segment of global civil society in which they are immediately embedded. Rather, their centrality stems in part from ties to key groups that are oriented toward political process (WSF) or political action (EAN, CJN, GCAP, FoEI).  TAPGs have a demonstrable, structurally integrative impact upon GCS, as indicated by the effects of removing their mediating ties from the network of INGOs in which they are embedded. That segment of global civil society derives considerable structural coherence and connectivity from the mediation provided by TAPGs.  Closer examination of the role TAPGs play in integrating INGOs across space shows them to be regionally extraverted in their ties to other GCS actors. However, patterns of extraversion-introversion across the eight regions vary quite substantially, and there is a sharp difference between Northern and Southern TAPGs in their regional mediations. Northern-based TAPGs tend to be more regionally introverted, and show less of a tendency to link North/South, compared to Southern TAPGs, which are extraverted both in North/South ties and in linking across the regions of the South.  TAPGs vary greatly in the extent to which they broker ties between other INGOs, and in the extent to which such brokerage mainly reinforces intra-regional relations or contributes to inter-regional, North-South integration. In constructing an historical bloc for alter-globalization, all three forms of regional mediation are strategically important. And given the legacy of Eurocentric capitalist development, it is not surprising that Northern-based groups mediate relations between many Northern groups, as an aspect of a Northern left that has been relatively well-resourced and, since the inception of the First International (1864), in some measure internationalist. Building solidaristic relations among civil society actors in the global North has been an important facet of recent alterglobalization activism. But if Northern-based TAPGs are strongly inclined to mediate relations between Northern organizations, TNI and RosaLux show more balanced profiles involving many Southern-based groups. For Southern-based TAPGs, all three kinds of brokerage may be seen as contributing importantly to historical bloc formation. North/South mediations help bridge that divide, as do North/North mediations. South/South mediations help solidify a Southern network within GCS, which can be taken as integral to any post-colonial global left. We find that Southern TAPGs mediate extensively between global North and South, and even between pairs of Northern INGOs, and that several of them are heavily involved in mediating South-South relations.  As for movement domains and the framing of political projects, the segment of GCS in which TAPGs are ensconced contains a great diversity of perspectives on contemporary relations of domination, from mainstream to radical anti-capitalist, and TAPGs play a definite, though again uneven, role in mediating across domains. All TAPGs contain, within their social circles, heterogeneous ensembles of political frames. In this way, the relations TAPGs maintain with other organizations of global civil society "connect the dots" between movement domains that are sometimes understood as free-standing entities, as in single-issue politics -perhaps most significantly, ecological and social justice concerns.
In all these respects, transnational alternative policy groups are socially embedded to communicate with and mediate between a diverse array of collective actors, North and South. The practices that define TAPGs as distinct organizations within global civil society implicate them in these relations. To produce and mobilize critical knowledge for social change requires that these groups form such relations with change agents within global civil society, just as elite policy-planning groups, to be efficacious, need to maintain extensive networks that reach into and enable dialogue with the top tier of the transnational capitalist class and its intellectuals. The political content of the transnational blocs, however, differs starkly; alternative knowledge producers seek to transform existing social relations, while the knowledge issuing from hegemonic policy groups is intended to manage and reproduce the capitalist status quo in ways that combine efficiency with securing popular consent (Carroll 2007). If with Howarth (2010: 318) we view counter-hegemony as involving "the linking together of disparate demands to forge projects or 'discourse coalitions' that can contest a particular form of rule, practice or policy," it is evident that transnational alternative policy groups are important elements in counterhegemony, across a wide range of political domains.
Indeed, it may be partly their wide-angle view of contemporary capitalism and its maladies (as given in our selection criteria, above) that enables TAPGs to act as network integrators. Here, McPherson's (1990) distinction between specialist organizations (those with narrowly defined goals) and generalist organizations is helpful. TAPGs fall into the latter category. In organization studies, generalists have been found to play a broader role in interorganizational collective action (Barnett, Mischke, and Ocasio 2000;Shumate and Lipp 2008). The work of TAPGs, again, parallels that of the transnational policy groups that lend cohesion to the global corporate elite. In both cases, the groups produce and mobilize knowledge that is framed beyond particular and immediate interests, whether of civil-society organizations or of global corporations whose primary concern is their own profitability. But the historical blocs that TAPGs and hegemonic policy groups help organize are set against each other. In the case of TAPGs, their social ties position them to participate in the transformation of the democratic globalization network from a gelatinous and unselfconscious state (cf. Smith 2008: 226), into an historical bloc capable of collective action toward an alternative global order.

Conclusion
All this being said, there are significant gaps in the network we have explored, and thus in what may be a nascent historical bloc for alter-globalization. Entirely or largely absent from the network are organizations based in Russia and China. Europe, on the other hand, is especially central, as a site for many INGOs. We have noted a tendency among the Northern-based TAPGs to integrate a Northern network, while Southern TAPGs are heavily involved both in North-South and South-South relations. The heavy involvement of Southern TAPGs in the global network contrasts with findings from other studies of NGOs, which have discerned a North-South divide whereby Southern NGOs tend not to form relationships beyond the local level while Northern NGOs are extensively connected internationally (Shumate and DeWitt 2008: 419). Among the TAPGs we have studied here, the Southern groups tend to be more cosmopolitan, both in their direct relations and in the relations they mediate. Again, their commitment to wide-ranging rather than single-issue analysis pushes them to connect the local and the global, and thus to mediate relations trans-locally.
Other gaps in the bloc, and limits in political framing, also merit discussion. Although in the Gramscian tradition, an historical bloc organizes the political project of a "fundamental class" and its allies, labor is rather weakly represented in the TAPG-centered network. Its main international organization, the ITUC, is tied only to the Rosa Luxemburg Foundation, and none of the political frames that inform the work of INGOs in the network directly takes up labor issues. Similarly, although we have featured "post-capitalist alternatives" in this paper's title, our analysis of political frames revealed that only 2.3% of organizations in the TAPG-centered network defined their project primarily as anti-capitalist -suggesting a reluctance of these groups to articulate a definite post-capitalist social vision. Here again, the Rosa Luxemburg Foundation is an exception, with its socialist commitment (expressing the political project of Germany's Left Party, Die Linke) and its extensive inter-continental ties spanning North and South. In general, TAPGs seem to operate pragmatically within what Unger (2009: xxi) has called the "adjacent possible": The possible that counts is not the fanciful horizon of possibilities but the adjacent possible: what is accessible with the materials at hand, deployed in the pursuit of movement in the desired direction.
However, principled pursuit of the adjacently possible, in a way that connects across political domains, has transformative potential as a transitional program that sets in motion changes pointing to a post-capitalist future.
The limits of our method must also be acknowledged. We have presented an architectonic analysis of the region of global civil society in which transnational alternative policy groups are embedded. Although the groups are positioned in ways that mediate and integrate, the substantive content of the relations that these centers of knowledge production and mobilization maintain with other organizations has not been examined here. Although the structure of relations is consistent with the inference that TAPGs are active agents of counter-hegemony, one could also draw a different conclusion.
It could be that TAPGs (or some of them), despite their political ambitions, are caught up in a process of "NGO-ization," as astutely analyzed by Alvarez (1999). Focusing on feminist politics, she noted that trends within NGOs toward professionalization and institutionalization, together with the retreat of neoliberal states from direct social programming, gave rise in the 1990s to a situation in which neoliberal governments and IGOs have come to view NGOs "as surrogates for civil society, assuming they serve as 'intermediaries' to larger societal constituencies," while NGOs are increasingly subcontracted to advise on or deliver government programs (1999: 181). The danger in this "neoliberalization of civil society" (Goldman 2005: 270-1) is that NGOs come to serve as gatekeepers for national and global governance, acting "as brakes on more radical and exceptional ideas emanating from the developing world" (Bob 2005: 194). On this interpretation, the role NGOs, including potentially transnational alternative policy groups, play as "intermediaries" (Choudry 2010) may have more to do with maintaining the hegemonic bloc than with constructing a counter-hegemonic alternative. 12 What limited research exists on the practices of transnational alternative policy groups suggests that their efforts center upon weaving specific and localized concerns together "around the theme of resistance to neoliberal globalization and the need for social justice" (Caouette 2007: 163). Such weaving, however, is not unproblematic. Thus, for instance, Brown's (2003) study of the Third World Network, while acknowledging its impressive scope and range of alternative knowledge production, also asked whether the group's tendency to eschew strong relations with alter-globalization actors, while cultivating ties to intergovernmental organizations, has imposed limits upon its capacity to promote transformation: There is a puzzling lack of any sense of radicalism by which is meant here a failure to provide a sustained alternative manifesto for the replacement, rather than just the reform, of the institutions of global governance. In some respects, it could be suggested that TWN has become institutionalized into the very system of global governance that it seeks to condemn. Its continuous ride on the merry-goround of intergovernmental conferences and ministerial meetings arguably lends these conventions a further degree of legitimacy. (76) Further research should examine how transnational alternative policy groups are embedded in networks of intergovernmental organizations (in which they might come to play a gatekeeping function) and how their funding relation to foundations, UN agencies, and the like may constrict their range of action and expression, pulling them onto the terrain of liberal humanitarianism. Beyond such structural analyses, there is a need for in-depth field work to explore just how these groups produce and mobilize alternative knowledge, and what impact their efforts have upon movement activism, intergovernmental and state policies, and (trans)national political cultures.