Knowledge Networks/Policy Networks PDF

2 Knowledge Networks/Policy Networks Ideas matter. Although a cliché, access to and control over, knowledge is one key in determining the way power is constituted and used. As authority over political, social and economic activity is diffused glob- ally among a variety of public and private actors,...

2 Knowledge Networks/Policy Networks Ideas matter. Although a cliché, access to and control over, knowledge is one key in determining the way power is constituted and used. As authority over political, social and economic activity is diffused glob- ally among a variety of public and private actors, knowledge networks (KNETs) become crucial arbiters and coordinators in policy formula- tion. Yet, there remain considerable shortcomings in our conceptual and empirical understanding of how knowledge organisations have become transnational, densely networked and interactive with policy communities. This chapter explores the transnational features of knowledge agen- cies. The specific focus is upon the policy-related roles of researchers and other experts who may be based in universities, consultancy firms, philanthropic foundations, independent research institutes and think tanks and who interact in KNETs. The discussion is based on the assumption that there is a dynamic relationship between (global) governance and knowledge. The EU, the World Bank, the WTO, the OECD and the World Health Organisation (WHO) are just some of the international organisations that have become important funders and consumers of research and policy analysis. For organisations that have global or regional remit, networks provide connections between knowl- edge producers and decision makers for evidence-informed policy. In post-modernist perspectives, such networks have become one form of ‘governmentality’ (Sending and Neumann, 2010; Ilcan and Philips, 2008), that is, symptomatic of a ‘capillary’ character of power relations (Walters, 2012: 9). The following section introduces the idea of ‘global knowledge net- works’ and how they connect to, but are different from, related concepts such as ‘transnational advocacy networks’, ‘transnational executive 37 38 Knowledge Actors and Transnational Governance networks’ and ‘global public policy networks’. Rather than forcing a syn- thesis, each of these network types offer different perspectives on why networks emerge and the principles and/or interests around which col- lective action occurs. In short, these are different ‘species’ of networks. The second section addresses global knowledge networks as a contem- porary manifestation of long-standing debates about the link between ideas and politics. If ideas matter, how do they matter and why? Knowledge networks are one mechanism to make ideas matter. However, to provide an explanation of why networks make ideas matter requires recourse to some of the theoretical literature on networks. The epistemic community, discourse coalition and neo-Gramscian network approaches provide different foci of analysis – science, discourse and ideology – for the interpretation of the sources of power of knowledge networks. The final section discusses networks as policy processes and govern- ance structures. As a social technology, the network can be regarded as both organising structures for governance and agents or vehicles for activism and policy advocacy. As will be argued, transnational networks are creating new public spaces in the global agora. Networks in the global agora KNETs – as well as other kinds of network – contribute to the shape, diversity and (in)equality of the global agora. Networks can be thought of as a form of assembly and public action. Networks are potentially a means for civic engagement and a vehicle for expanding participa- tion. This is neatly captured in the social movement character of ‘transnational advocacy networks’ (Keck and Sikkink, 1998; Diani and McAdam, 2003). TANs have become a popular concept within the IR literature, especially on environmental issues and gender studies (see inter alia, Huelshoff and Kiel, 2012; True, 2003), but there are also other network concepts to allow us to identify four main species or types. Global public policy network – GPPN Transnational executive network – TEN Knowledge network – KNET and Transnational advocacy network – TAN By no means is this an exhaustive list of social science thinking on net- works. Older concepts from public policy have been ‘stretched’ to keep pace with supra-national and sovereignty-challenging developments of global governance and transnational regulation. For instance, the  Knowledge Networks/Policy Networks 39 concepts of ‘policy community’ and ‘advocacy coalitions’ have some- times been used to identify transnational action (see the network lit- erature reviews undertaken by Holten, 2008 and by Thompson, 2003). Notwithstanding the diversity of network ideas, this chapter selects four of the better established network concepts for detailed discussion. Each is sufficiently distinct in terms of explaining the composition, sources of power and authority as well as causal impact of networks on policy. Nevertheless, these four concepts are ideal types. Table 2.1 summarises the features of each type. In reality, there can be significant blurring or overlap rather than hard and fast conceptual boundaries between each species. Transnational advocacy networks The first species is ‘transnational advocacy networks’ (TAN – Keck and Sikkink, 1998). TANs are similar to, but more issue focused than social movements. Characteristically, they accommodate a range of NGOs and activists. They are bound together by shared values or ‘principled beliefs’ and a shared discourse where the dominant modality is informa- tion exchange. They are called advocacy networks because ‘advocates plead the causes of others or defend a cause or proposition’ (Keck and Sikkink, 1998: 8). In other words, these networks are norm based and mission oriented in seeking to have their values recognised in policy. Examples include the transnational campaigns surrounding issues like anti-slavery and debt relief. In response to women’s transnational advo- cacy networking, ‘the United Nations has opened spaces for their global influence’ (True, 2007: 378). This has sometimes had a ‘boomerang effect’ back into national policy contexts with re-definitions of state interests and identities as a result of transnational pressures. A TAN emerged around the theme of ‘blood diamonds’ or ‘conflict diamonds’ in part as a response to the covert ‘dark network’ mode of operation of arms traffickers (Raab and Milward, 2003). Other examples include the Cluster Munition Coalition or Development Alternatives with Women for a New Era (DAWN) operating in the ‘Global South’. At a regional level, there are civil liberties advocacy bodies like ILGA-Europe, the International Lesbian, Gay, Bisexual, Trans and Intersex Association. TANs usually have a strong normative basis for any moral judge- ments they make in seeking to shape the climate of public debate and influence global policy agendas. However, compared to other network species, they are not well integrated into policy making and operate more like ‘outsider groups’. They exercise ‘voice’ and seek to sway public opinion. The growth and relatively high degree of public exposure of Table 2.1 Network Species 40 Transnational Advocacy Global Public Policy Transnational Knowledge Networks Networks (TANs) Networks (GPPNs) Executive Networks (KNETS) (TENs) Alternative Terms Global civil society Global programmes, global Governmental Epistemic community networks public private partnerships networks, transnational constitutionalism Participants True believers Stakeholders specific to a Judges, politicians, Scientists, experts Purposive actors policy issue or problem, regulators, officers of tri-sectoral the state Authority Normative, value based Partnership principles of Politico-legal office, Episteme shared decision-making sovereign authority Capacities Convening power, Force for market deepening, International Evidence providers, persuasive narratives of delegated decision-making, coordination, state interpreters, theorisation change, advocacy implementation efficiencies interests extended and and conceptualisation internationalised Theorisation Neo-pluralist Neo-corporatist Power sharing, Neo-Gramscian inter-state regulation, or post-positivist ‘transnational state’ perspectives Policy Change Agenda-setting Agenda-setting, resource Collaborative Paradigm shift or (in case mobilisation, interest regulation of doxic communities) mediation paradigm maintenance Locus Civil society Between market and state Public sector, state Cross-cutting civil based. society, market and state differentially according to issue Policy transfer Spreads values and Develop best practices and Peer review, Modes of evaluation, modalities norms standards harmonisation assessment frameworks, methods of (social) scientific inquiry Accessing networks Open to those Limited to stakeholders Limited to state Limited to those with subscribing to the with vested interest and authorised actors recognised education, principles sharing the common training and mastery of policy project ‘communication codes’ Network coherence Common cause, shared Shared interests, resource Allegiance to the state, Consensual knowledge, and coordination ideas and values, interdependencies and shared pooled sovereignty or scientific protocols, peer reform-oriented policy responsibilities inter-governmental review and standards of decision-making professional expertise authority Public Action Overtly in public sphere, Creating issue-specific Extension of state Global knowledge, public generally outsiders to public space via network, policy responsibilities goods production policy process global public goods via networks of provision government officials Policy outputs Voice, alternative vision Soft law, policy coordination, Rules, harmonised Codified knowledge, and outcomes shared implementation regulation, com- expert advice and responsibilities mon standards and analysis for policy information sharing formulation and review 41 42 Knowledge Actors and Transnational Governance TANs has been propelled by technological advances in transport and communications (Waddell, 2011). Global public policy networks The second kind of network is the ‘global public policy network’ (GPPN) (Reinicke and Deng, 2000). Sometimes called ‘transnational public- private partnerships’ these networks are tri-sectoral in character. That is, they are alliances of government agencies alongside international organisations as well as corporations and elements of civil society. Official involvement of public actors bestowing governmental patron- age or development assistance gives some ‘insider’ status to official decision-making for these networks. Stakeholders invest in these com- munities to pursue material interests but have in common a shared problem. Their interactions are shaped by resource dependencies and bargaining. They tend to cohere around the international organisations and governments that have entered into a policy partnership for the delivery of global public goods. The transnational character of policy problems establishes rationales for co-operation. Examples include the Global Gas Flaring Initiative, the Global Environment Facility, and one of the oldest, CGIAR. Although technically an inter-governmental organisation, the International Network on Bamboo and Rattan (INBAR) connects a global network of partners from the government, private and not-for-profit sectors in over 50 countries to define and implement a global agenda for sustainable development for bamboo and rattan producers. Over time a network may become institutionalised with the crea- tion of formal arrangements such as advisory committees, consultation procedures and recognition by state and multilateral agencies in the implementation of policies. GPPNs are different in their publicness and sources of authority. Although the term ‘corporatism’ has fallen out of fashion and the ‘operative word today is partnership’ (Ottaway, 2001: 266), a neo-corporatist framework of interpretation has consid- erable applicability in that groups settle problems through negotia- tion and joint agreement. While official actors from governments or international organisations are partners in these networks, they are not ‘inter-governmental’ in the manner of the next type of network species. Instead, GPPNs are quasi-public or semi-private. Transnational executive network The third kind of networks is not based in civil society as is the case with TANs, or overlapping into it as do GPPNs. Instead, ‘transnational  Knowledge Networks/Policy Networks 43 executive networks’ are strategic devices for states to extend their pub- lic authority beyond borders. These networks are almost exclusively composed of ‘internationalised public sector officials’. In this perspec- tive, the state is not disappearing but it is becoming disaggregated and penetrated by horizontal networks existing between ‘high level officials directly responsive to the national political process – the ministerial level – as well as between lower level national regulators’ (Slaughter, 2004: 19). These networks of, for example, judges, legislators or regula- tors such as utilities commissioners are inter-governmental in character, and the state remains core as a sovereign actor. The actors who compose TENs are formally designated power holders and rule makers who derive their authority from their official positions within their nation-state. Examples of such networks include the International Association of Insurance Supervisors and the International Network on Environmental Compliance and Enforcement (INECE). Terrorists, arms dealers, money launderers, drug cartels and human traffickers operate through global networks. In this context, an under- lying logic of some TENs is that networked threats require networked responses (Kahler, 2009). For example, the Financial Action Task Force (FATF) was created as a response to cross-border money laundering. Networks become tools for the maintenance of sovereignty where global problems are solved by ‘networked government’ collaboration. As mech- anisms for the state to re-invent itself, TENs offer a system of ‘checks and balances’ to ensure accountability and public responsiveness (Slaughter, 2004: 29). TENs are the most public type of network discussed in this book. They operate at the behest of national polities who delegate state representatives to develop coordinated responses to shared international policy concerns. The concept also ‘falls back on a Neorealist notion of power based on “state capacities”’ (Scott-Smith and Baumgärtel, 2011: 274) and is considerably more state-centric than the other concepts. Knowledge networks A KNET is a fourth type of network, a system of coordinated research, results dissemination and publication, intellectual exchange, and financ- ing across national boundaries. These networks create and advance knowledge as well as disseminate that knowledge to inform policy and practice (Parmar, 2002). Another definition: Global knowledge networks create and transfer knowledge – scientific, community based and policy relevant – as well as the nec- essary hardware and finances to support knowledge acquisition and 44 Knowledge Actors and Transnational Governance implementation. … such knowledge networks operate within a glob- ally shared system of knowledge creation and transmission, while the practices of individual members are informed by the histories, politics and ecologies of the national and local places in which they work … (Gross Stein et al., 2001: 6–7) KNETs take quite different shape and many are impermanent entities. For example, the ‘networks of excellence’ funded by the European Commission differ on criteria of legal status, membership, degree of institutionalisation and issue focus when compared to more permanent scientific bodies like CGIAR. KNETs incorporate professional bodies, academic research groups and scientific communities that organise around a special subject mat- ter or issue. Individual or institutional inclusion in such networks is based upon professional or official recognition of expertise such as commitment to certain journals, conferences or other gatherings and organs that help bestow scholarly, ideological and scientific credibility. KNETs are often also practically engaged in ‘capacity building’, that is, mobilising funds and other resources for scholarships and training, supporting institutional consolidation that facilitates both network regeneration and knowledge construction. The primary motivation of such networks is to create and advance knowledge as well as to share, spread and, in some cases, use that knowledge to inform policy and apply it to practice. KNETs are essential for the international spread of research results, scientific practice and what is deemed international ‘best practice’ on matters such as banking standards, gender mainstreaming or corporate social responsibility. International organisations and other multilateral initiatives require policy analysis and research to support problem defi- nition, outline policy solutions, monitor and evaluate existing policy as well as to provide scholarly legitimacy for policy development. In other words, knowledge is a key resource in global public policy development. KNETs are often represented as a form of ‘governmentality’. … one of the key issues of global knowledge networks concerns how the production, collection and movement of knowledge are linked to the technologies of government. From a governmentality approach, technologies of government are not simply mechanical devices; they are assemblages of forms of practical knowledge, with practices of calculation and types of authority and judgements… (Ilcan and Phillips, 2008: 713)  Knowledge Networks/Policy Networks 45 From this theoretical vantage, governance is regarded as a widespread phenomenon and not one limited to the sphere of the state (Walters, 2012: 11). In other words, practical knowledge is mobilised to govern a domain (such as development or security policy), but is linked to theories, programmes and expertise that supply it with policy objec- tives and which can be viewed as an apparatus of rule. Recognising the techniques by which knowledge organisations seek to shape their own conduct (such as via peer review, rigorous methodologies and international standard setting), or that of other groups or organisa- tions, provides insight into the ‘forms of reason’ and ‘regimes of truth’ that operate within institutions and at specific historical junctures (Walters, 2012: 11). Many KNETs are engaged in the so-called ‘disinterested pursuit and exchange of knowledge’, and this is a key feature distinguishing them from ‘transnational advocacy networks’ and ‘global public policy net- works’. However, knowledge production is not divorced from the social and political worlds of the policy process. Whilst at one level this may be obvious, the social practices within KNETs give their product – ideas, publications, analysis – a patina of scientific objectivity and techno- cratic neutrality. Sophisticated computer modelling, positive economic theories or scientific papers published in refereed professional journals create ‘communication codes’ that construct some knowledge as more persuasive or reliable. These codes are not only expensive to reproduce but difficult to access. Another distinction to be made is the degree of policy relevance of these networks. Some are focused primarily on knowledge creation and sharing, for example, the academic ‘networks of excellence’ sponsored by the European Commission. Others like the Asian Fisheries Social Science Research Network operate with an agenda of using knowledge to inform policy and practice in their field of food security, fisheries and rural development. Reprise An important distinction needs to be made between policy networks and KNETs. Knowledge networks are composed of organisations with shared perspectives, joint interests and common scholarly agendas where members are ‘homogeneous rather than heterogeneous in their fundamental views’ (Struyk, 2007: 83). By contrast, policy networks are more heterogeneous and designed to mediate between participants with differing interests. Nevertheless, KNETs can be drawn into policy devel- opment, business-related advocacy and civil society activism. KNETs are not a pure type. Instead, these networks blur and blend with other 46 Knowledge Actors and Transnational Governance network types in a ‘web’ of interactions. Consequently, DAWN has not only the features of both a KNET in sponsoring research and various publications but also the character of a TAN given its advocacy of wom- en’s rights. It is also the case with CGIAR having the character both of a GPPN with long standing support and policy input from governments and international organisations, and a KNET that conducts agricultural research and funds scientific laboratories. Some TENs are strongly focused on information sharing and draw upon research and professional expertise to support their deliberations. Consequently, these network species are fluid conceptual categories. In policy practice, KNETs do not operate in isolation or hermetically sealed from other networks. Moreover, policy-engaged networks are always founded on some form of ‘taken-for-granted’ knowledge and expertise entrenched in the way it approaches economic, political and social issues. Of the four network species, TENs have greatest executive authority where government officials have a dual domestic and international func- tion. Networks become tools for the maintenance of sovereignty where global problems are confronted by governmentally organised networks. What makes TENs ‘public’ is that actors who compose them are formally designated power holders and rule makers who derive their authority from their official positions within their nation-state. Compared to TANS, which tend to be generated by ‘bottom-up’ strategic initiatives with solid foundations in civil society or connected to wider social movements, GPPNs have greater official standing and public authority as they tend to be initiated or convened by international organisations. Despite the blurring that occurs in reality, the power bases and logic of organisation of these species as ‘ideal types’ are distinguishable. GPPNs are defined by shared material interests and partnership princi- ples in delivering public goods; TANS by their normative ambitions and advocacy orientation; TENS by their political–legal character as inter- governmental networks; and KNETs by their claim to knowledge creation and epistemic authority. All species are to be found in the global agora. Explaining networks Just as there are quite different types of networks in operation in the global agora, so too there are competing explanations as to how knowledge influences political thinking, if at all. Where the preceding section identified overlapping types of network that are traversing the global agora – TANs, GPPNs, TENs and KNETs – this section draws on an  Knowledge Networks/Policy Networks 47 additional set of network concepts but to a different purpose, that is, to provide varying conceptions of how the knowledge and norms, the ideas and values of networks might exert influence. Three distinct schools of network interpretation are outlined by dis- cussing first, the ‘epistemic community’; second, post-positivist concepts of ‘discourse coalitions’ or interpretive communities and third neo- marxist and Gramscian thinking on ‘embedded’ or ‘subaltern’ knowl- edge networks. ‘Ideas do matter’, and while these frameworks share the position that ideas, research and knowledge are endemic to the policy process, they do so from quite different epistemological standpoints. Respectively, the network models posit first, science, objectivity and rationalism as a compelling force that can drive policy; second, the influ- ence of discourse and subjectivity; and third, the role of hegemony and material interests as the sources of power in the global agora. Epistemic communities These communities are ‘scientific’ in composition, founded on ‘codified’ forms of knowledge. Researchers and scientists seek privileged access to decision-making forums on the basis of their expertise and scholarly insight. Yet, epistemic communities also assert their independence from government and other societal interests on the basis of their commit- ment to expert knowledge. These communities are identifiable by: 1. shared normative and principled beliefs which provide the value- based rationales for their action; 2. shared causal beliefs or professional judgements; 3. common notions of validity based on inter-subjective, internally defined criteria for validating knowledge; 4. a common policy enterprise (Haas, 1992). The status and prestige associated with the expertise of epistemic com- munity members and their high professional training and authoritative knowledge regarding a particular problem is politically empowering and provides some communities limited access to the political system. This is especially the case in conditions of ‘uncertainty’ – for example, persistent societal problems such as urban decay or poverty, natural dis- aster or rapid technological advancement – where policy makers cannot make decisions on the basis of existing knowledge or past experience and approach expert groups for assistance. An epistemic community of cetologists experienced policy influence in circumstances where governments around the world were uncertain 48 Knowledge Actors and Transnational Governance regarding whale populations and species, and the impact of the whal- ing industry on this ecology. The scientific knowledge made available by this community during international debates played an important role in the decision of most nations to support the 1982 adoption of a moratorium on commercial whaling (Peterson, 1992). Their scientific consensus also bolstered the TANs agitating to ‘Save The Whales’. An epistemic community is one kind of KNET. But the approach has a specific representation of the role of knowledge or science as being based on facts and empirically discernible realities. Consensual knowl- edge takes the form of concrete knowledge of the physical world, objec- tively beholden by an epistemically privileged Cartesian observer (and collectively the epistemic community) who then turns into a dispas- sionate advisor to the powerful. It is rationalist, technocratic approach to decision-making. Analysis from this network perspective considers that solutions to problems can be found by utilising the correct knowl- edge and evidence. The utility of this approach for understanding the power of KNETs is that knowledge in the form of scientific consensus represents a source of policy change. ‘Truth speaks to power’, so goes the famous phrase coined by Aaron Wildavsky (1987). However, as the discussion below outlines, rarely is knowledge or expertise uncontested; KNETs, or the individuals and organisations within them, are better regarded ‘as “competitive definers” engaged in a contest to define the truth’ (McKewon, 2012: 279). Discourse coalitions and interpretative communities The assumption of a universal and factual body of knowledge grounded in the natural world from which epistemic communities and other bodies of experts can draw from in order to devise directions for policy makers on ‘what to do next’ has been challenged by different perspec- tives of knowledge as being contextually constructed, situational and always bound up with particular practices and powers. Post-positivist and post-modern scholarship on networks is diverse and occasionally dense. The literature is replete with jargon and over-lapping terms that nevertheless signify different assumptions and categories. The lesser known ‘transnational discourse community’ approach developed by Danish public administration scholars identifies symbols, language and policy narrative as a source of power. The framework puts emphasis on the transnational qualities of professional groups, and secondly, the role of discourse. Networks are interactive venues where the national identities of researchers, donors and international civil serv- ants are complicated by the professional commitment to questions of  Knowledge Networks/Policy Networks 49 international development assistance, financial crisis or trade reform that are increasingly less questions of national determination under the forces of globalisation. Public sector professionals, traditionally expected to represent a specific national view on any issues in their international activi- ties, no longer only do that. In fact, by foregrounding their profes- sional identity, they transcend the power of the nation-state system to impose its categories of identity upon them. They also tend to assume a global or regional rather than a national outlook on key issues … (Krause Hansen et al., 2002: 109) Transnational identities are further enhanced by face-to-face communi- cation at international meetings. This frame provides a social construc- tivist view of how transnational policy communities (outlined in the previous chapter) cohere. Second, drawing upon the ideas of social theorist Michel Foucault, the discourse community concept locates discourse at the interface of power and knowledge. Discourses generate ‘effects of truth’, that is, naturalis- ing specific ways of thinking and normalising certain ways of doing things. Discourse is a system that, through language or text, or a set of statements or social interactions, structures the way we perceive reality. Our discourse constrains perceptions. It shapes how groups respond to particular situations and how some things come to be regarded as normal or legitimate – the ‘taken-for-granted’ features of a social order. Discourse institutionalism (Kjaer and Pedersen, 2001; Schmidt, 2008) and the interpretive turn in policy studies (Fischer, 2003) regard proc- esses of meaning-making – deliberation and argumentation – as prior to, and informing, interest formation and institutionalisation. Transnational discourse communities construct identifiable policy narratives which serve to establish the goals of reform, justify the neces- sity of change, describe the means to achieve better results and predict outcomes. This approach is very similar to the idea of ‘interpretive communities’ arising from socio-legal studies. An interpretive commu- nity rests upon ‘professional interpreters’: think tank directors, research fellows, legal experts, scholars and others. All professional interpreters … are situated within an institutional context, and interpretative activity makes sense only in terms of the purposes of the enterprise in which the interpreter is participating. Furthermore, a given text is always encountered in a situation or 50 Knowledge Actors and Transnational Governance field of practice, and therefore can only be understood in light of the position it occupies in that enterprise. … Thus, interpretation is con- strained … by the ‘cultural assumptions within which both texts and contexts take shape for situated agents’. Meaning is produced neither by the text nor by the reader but by the interpretative community in which both are situated. (Johnstone 2005: 189, quoting Stanley Fish) ‘Texts’ is an umbrella term for the web-sites, meetings, publications and policy commentary (briefs, speech writing, etc.) produced by the inter- preters. The situations and ‘field of practice’ are constituted through networks that function as forums where new social realities can be constructed, debated and interpreted as individuals came into contact and interact. Discourse contributes to the formation of inter-subjective understandings operating as the social glue of networks/coalitions/ communities. Power and knowledge also operate through the social and institu- tional practices of the network and the ‘boundary drawing discourses’ that create distinction between experts and lay people, between good and bad professionals, and thus who is inside or outside the com- munity. ‘The specific vocabulary and jargon, the speech and meeting rituals etc. create possibilities for the professionals who master them’ (Krause Hansen et al., 2002: 111). Boundary-drawing helps a community to canonise certain viewpoints at the cost of others, elevating them to unquestioned status and superior position. In this perspective, multiple truths are constructed by persuasive or credible expert narratives about social and economic realities. Truth is created. A related concept developed by a Dutch environmental policy scholar is of discourse coalitions (or network), which makes useful distinctions between different stages of discursive influence starting with the forma- tion of coalitions, then the extent to which a coalition shapes public understanding of a problem and, finally, governmental responses in the form of public policy and law (Hajer, 1993: 47). Discourse coalitions seek to impose their ‘discourse’ in policy domains. If their discourse shapes the way in which society conceptualises the world or a particular problem, then the coalition has achieved ‘discourse structuration’ and agendas are likely to be restricted to a limited spectrum of possibilities. If a discourse becomes entrenched in the minds of many as the dominant mode of perception, it can become distilled in institutions and organisa- tional practices as the conventional mode of reasoning or ‘global space characterised by regimes of truth’ (Prügl, 2004: 72). This latter process is ‘discourse institutionalisation’. The framework captures how discourses  Knowledge Networks/Policy Networks 51 are transformed, or discarded, in their articulation through the policy cycle. There is a greater degree of indeterminacy compared to rational- ist epistemic community approaches that regard evidence and scientific consensus as implacable and stable truths. The classic example of a discourse coalition is the manner in which air- borne pollutants were re-cast as ‘acid rain’ (Hajer, 1993). Another exam- ple of the power of narratives shaping public and political consciousness is the shift in terminology from ‘female circumcision’, which implies a minor surgical procedure, compared to the emotionally loaded term of ‘female genital mutilation’, which has strong connotations of human rights abuse. Likewise, the TANs concerned with illegal trading of dia- monds found power in the portrayal of their cause with emotive terms such as ‘blood diamonds’ or ‘dirty diamonds’. In this latter case, the discourse community, or coalition, achieved discourse structuration in popular imagination with Hollywood films starring Leonardo DiCaprio, and extensive media coverage, as well as discourse institutionalisation with the creation of the Kimberley process of diamond certification. The substantive content of policy discourses is important, but so too is the interactive processes by which ideas are spread. The mode of dis- course can be ‘coordinative’ or ‘communicative’. Coordinative discourse refers to the creation, elaboration and justification of policy and pro- grammatic ideas among transnational policy communities. By contrast, communicative discourse is concerned with the relationship between policy makers and the public (Schmidt, 2008: 310). In general, discourse approaches allow scope for ideas to have inde- pendent force and inherent power in policy. Even so, discourses are not stable or uncontested and can be transformed by the institutional con- text into which they are propelled. Discourse coalitions and commu- nities stress agency; by contrast discourse institutionalism has greater recognition of pre-existing structural constraints regarding how any discourse is received in society or politics. To understand the politics of discourse is to understand a key element of how knowledge in the form of research, professional codes and expert advice gets translated into policy and then moulded by that context (Schmidt, 2008). But, meaning-making has structural consequences in shaping or limiting the frame of reference for policy making or what is considered politi- cally viable. Power and capacity for change comes from the idea itself, irrespective of who, or what, articulates that discourse. This set of concepts with post-modernist or social constructivist sensi- bilities are in distinction from both the epistemic community framework and neo-marxist instrumentalist accounts of the next section which 52 Knowledge Actors and Transnational Governance portray ideas, norms or expert knowledge as a resource utilised by an agent. Power or change capabilities are seen to reside in the individual or institutional agent that advocates on behalf of state, communal or corporate interests. Both the first and the next approach presuppose agents who create shared identities and common interests through instrumental actions and coalition building. This is not to negate the role of those actors or institutions that articulate the ideas. But repre- senting KNETs as one important interpretative mechanism in the global agora provides a structural explanation of the power of ideas. Over time and through multiple discourses and venues, specific policy ideas become an organising logic or coordinative paradigm. As discussed in later chapters, expert discourses, such as the ‘ASEAN Way’, become a structuring force. (Dis)Embedded knowledge and subaltern networks The ‘embedded knowledge network’ framework stresses the role of ideas being connected and subsidiary to interests (Sinclair, 2000). This perspective is crystallised in the oft-quoted phrase: ‘theory is always for someone, and for some purpose’ (Cox, 1981: 128). From this theo- retical vantage, KNETs represent a means for sustaining the neo-liberal capitalist order through the reproduction of ideas supportive of it. That is, scientific expertise is used for ideological purposes of ‘paradigm maintenance’ and the normalisation of dominant discourses of power (Bull and McNeill, 2007). For instance, within the World Bank, the researchers in the Development Economics Research Group and the other academic and professional economists with whom they interact created, disseminated and broadcast the neo-liberal market principles of the (post-)Washington Consensus: privatisation, deregulation and liberalisation (Broad, 2006). Consequently, policy becomes a battle of ideas, networks the battlegrounds and knowledge a weapon in the service of material interests. Not dissimilar to institutionalised discourses, ‘embedded knowledge networks’ are ‘ostensibly private institutions that possess authority because of their publicly acknowledged track records for solving prob- lems, often acting as disinterested ‘technical’ parties in high-value, high-risk transactions, or in validating sets of norms and practices for a variety of service-provision activities’ (Sinclair, 2000: 488). The approach emphasises the importance of authoritative judgement making, built and sustained through trade journals, professional associations and research departments (of investment banks) or consultancies. Credit- rating agencies such as Moody’s or Standard and Poor are one example.  Knowledge Networks/Policy Networks 53 This approach treats knowledge, discourses or ideas as a tool of power used by dominant interests in maintaining the capitalist order. Knowledge networks are viewed as part of the micro-politics of con- temporary hegemony and symptomatic of the ‘war of position’. Think tanks, foundations, consultants and research institutes are one compo- nent of globalising elites, that is, a ‘directive strategic element within globalising capitalism’ (Stephen Gill quoted in Sinclair, 2000: 494). Ideas do not have independent power (as implied in discourses approaches) but are tied, and usually subservient to, social and political interests. What becomes considered to be the truth involves gaining control over material resources such as the media. This includes gaining control of knowledge networks. The emphasis is on ‘organic intellectuals’, play- ing a central role in hegemonic projects where specific sets of ideas are funded, generated and disseminated by foundations, think tanks, pub- lishing houses and NGOs. Consequently, global knowledge networks can be viewed as an evolving contemporary social mechanism to make certain ideas – put in league with particular social forces – more power- ful and hegemonic. Instead of truth speaking to power, power decides what is true. Analysis of the role of American philanthropic founda- tions funding educational programmes, think tanks and international exchange to promote the idea of US global power and sustain ideologi- cal hegemony is indicative of this school (Parmar, 2002). However, it is often evident that networks are often composed of contradictory knowledge or reflect discursive competition. Hegemony is incomplete and partial. Neo-Gramscian approaches posit a degree of intentionality or purposive strategy to knowledge agents and networks that is not necessarily the case. A grid-like complex of ideas shaping consciousness and dominating the global order gives little credence to alternative world-views and sites of intellectual resistance. A further approach drawing upon subaltern studies and the critical feminist literature sees knowledge-makers not only as elite actors occu- pying command posts of capitalism, but also as ‘those in defiance of dominant epistemological flows of power’ (Rai, 2005: 124). Subaltern studies loosen the hegemonic grid-like power of the neo-Gramscian approach but continues to ask how knowledge networks legitimise and/or challenge flows of power. The expansion of knowledge networks as ‘sites of authority’ potentially accelerates the ‘normalisation of the dominant discourses of power’. Nevertheless, it also brings some new opportunities as networks give activists, critics and feminists some access to expert communities and increased attention within the institutions of global governance (Prügl, 2004: 79). 54 Knowledge Actors and Transnational Governance Network disaggregation The different understandings of networks outlined above provide quite different conceptual tools to address the policy relevance of empirical examples of KNETs and their influence (or not) with other policy insti- tutions. To summarise, the ‘embedded knowledge network’ framework stresses the role of ideas being connected and subsidiary to interests. KNETs represent a means for sustaining the neo-liberal capitalist order through the reproduction of ideas supportive of it. Consequently, policy becomes a battle of ideas and knowledge is a weapon in this battle. By contrast, the ‘transnational discourse community’ perspective posits that ideas have independent force and inherent power. Presenting sci- ence or consensual knowledge as objective truths, epistemic communi- ties claim the power of ‘evidence’ and represent a technical rationality with a bias towards technocratic policy making. Of the KNETs discussed in later chapters, some can be described as having epistemic-like characteristics or aspiring to be perceived as epis- temic communities. The label has some caché. The Global Development Network has epistemic community characteristics; however, it is a very broad and loose coalition of social scientists. Although the develop- ment discourse of economists is dominant, other expert narratives are audible. In another frame of analysis, given the nature of its sponsor- ship and political support from the World Bank and other significant international organisations in the field of development, GDN might be considered as embedded. The ASEAN-Institutes of International and Strategic Studies is an epis- temic-like network given its orientation towards regional cooperation and institutionally embedded given the nature of its official patronage. Over more than two decades, ASEAN-ISIS developed a relationship of trust with governments of the region through the processes of ‘track two diplomacy’. The Open Society Foundations has the features of a TAN with its discourse on the ‘open society’ and direct funding of policy advocacy and civil liberties groups in transition states as well as international NGOs like International Crisis Group. Yet, it also has the character of a KNET given its sponsorship of academic research, scholarship programmes, scientific research and strong organisational orientation towards policy research and analysis. Yet, the capacities to set policy agendas and structure public discourse are highly unstable and mediated by the considerable scientific competition within these networks as well as outside them. None of the case study networks of later chapters could be described as ‘disembedded’ or subaltern. The case studies organisations are all engaged  Knowledge Networks/Policy Networks 55 with the political mainstream and interact with policy makers. While the Overseas Development Institute (ODI) and GDN are synchronised with the bureaucratic interests of UK Department for International Development (DfID) and the World Bank, respectively, it must be said that these network organisations are neither puppets nor unthinking mouthpieces for their funders and patrons. Networks such as DAWN are subaltern KNETs or TANs choosing to chal- lenge dominant policy discourses. Notwithstanding their lack of policy or political influence, these networks perform wider societal roles of knowl- edge creation and capacity building. Their ‘struggles’ – the ‘underside in the play of dominations’ – helps explain the normalisation of ‘winners in history’ (Walters, 2012: 113–34). Networks that appear to have little pol- icy impact or to be espousing unorthodox policy perspectives are neither completely ineffectual nor hopelessly marginalised. Instead, subaltern KNETs, and the TANs they interact with, are symptomatic of how domi- nated groups ‘representing counter-discourses and oppositional constitu- encies’ form identities through common language and understanding. These TANs mobilise resources around alternative definitions of reality, potentially to create ‘competing public spheres’ (Eckersley, 2007: 333). By drawing the above analytical distinctions it is possible to bet- ter understand the ways in which power relations are disturbed and then reconstituted at different governance levels. Networks are politi- cal spaces to re-invent the processes of policy design, implementation and evaluation. Coordination and regulation of the global agora via networks represents regime adaptation. But no one network is alike. Where certain networks perpetuate power constellations, other types can help confront and sometimes undermine it. There is a dual dynamic of counter-public spheres. Noting network power Three further points on network power are worth consideration. First, there is normative and ideational power of networks as convenors of sets of actors who create and broadcast the ideas that inform problem perception and set policy agendas. The content of policy knowledge, and its credibility, is also an important consideration in networks becoming a site of authority. Second, networks function as structures that can either exclude/include, co-opt/induct, legitimise/revoke or accept/deny various perspectives and participants. That is, networks create boundaries. Third, networks become a new locus or a ‘site’ for the enactment of pub- lic authority, albeit a locus disaggregated from the traditional Weberian 56 Knowledge Actors and Transnational Governance understanding of a public sector bureaucracy in sovereign states. Instead, public authority is re-invented in semi-private or quasi-official policy networks in a part privatisation of governance in the global agora. KNETS as expert agents and authority structures Governments as well as international organisations require the crea- tion and widespread acceptance of persuasive accounts of ‘public policy problems’ as the foundations for making legitimate policy and just laws. KNETs not only provide expert interpretations and scientific narratives, they also create self-supporting structures of authority to incarnate as ‘neutral’ research brokers and advisors. The legitimacy and credibility of a KNET’s expertise is drawn through a circular process between the knowledge it produces and the audiences that use and thereby legiti- mise that knowledge. In sum, KNETs do not simply revolve around centres of power waiting for their products to be used by more powerful actors; instead, the network is itself one site and form of power with capacities, of varying degree, to (re)produce dominant knowledge and discourses that define fields of action. Think tanks, law firms and university institutes are recognized as centres for expert, scientific and authoritative advice not simply because of the scholarly credentials of these organisations (and their self-referential habits). It also occurs because of the relationships with policy makers and donor groups that a network structure facilitates. Through their club-like tendencies, networks both accrue and accredit authority through the collective policy entrepreneurship of their members. KNETs draw together their power and policy influence strategically by combining epistemic, discursive and ideological practices. Networks are not mechanically linking knowledge and governance arenas, that is, ‘bridging research and policy’. Instead, the interaction of knowledge and policy is one of mutual construction. As will be discussed in Chapters 3 and 4, the routine activities of research and analysis within the ODI and in the Open Society Foundations network are recursive in a dialogue with governmental consumers and patrons. Inclusion and exclusion in the privatisation and pluralisation of policy Unlike most liberal and pluralist analyses which see the rise of networks of non-state actors as a progressive contribution to a global civil society, this book is more cautious. Organisational density and diversity can  Knowledge Networks/Policy Networks 57 disrupt hierarchies and disperse power, but networks can also represent new constellations of privatised power. Instead of being civil society manifestations of bottom-up, non-statist globalisation, some networks are better viewed as mutually imbricated in the affairs of states and international organisations. This is clearly the case with not only TENs, but also GPPNs given the neo-corporatist inclusion of either interna- tional organisations or governments as sponsors who also delegate representatives as decision-making participants. For example, as a global network of environmental compliance and enforcement practitioners, INECE was launched in 1989 by the Netherlands Ministry of Housing, Spatial Planning and the Environment and the US Environmental Protection Agency to link more than 4,000 practitioners – inspectors, prosecutors, regulators, parliamentarians, judges, international organi- sations and NGOs from 120 countries.1 The parentage of the INECE network is within government. By contrast, the governance functions of KNETs and TANs are obscured by their non-governmental status and often by their critical or adver- sary stance. Nevertheless, their policy advocacy, agenda-setting efforts and expert commentary on transnational problems embed them in a variety of relationships with official decision-making forums (Sending and Neumann, 2010: 110-31). Consider, for instance, the Evian Group. Founded in 1995, the Group is a coalition of opinion leaders from academic, corporate and government opinion leaders committed to fostering an open global market economy and a rules-based multi-lateral framework. It has a strong research identity given the calibre of its members as leading trade economists and lawyers, and its role in higher education and professional training, and is hosted at IMD Business School in Lausanne. It also operates like a TAN in its advocacy of trade liberalisation to achieve growth and sus- tain the momentum of globalisation. But this kind of advocacy closely concords with the policy agendas of the WTO, the G20 and many governments, whereby Evian claims that has played a key role preparing decision makers for international meetings.2 Such networks can be a force for democratisation by creating a venue for representation of ‘stakeholder’ interests, a means for wider partici- pation in modes of global governance and a venue for societal voices. In this sense, networks are ‘gateways’ to new policy spaces. This entails new arenas for alternative politics, new actors and audiences, and new kinds of strategising. However, these same networks can also be exclu- sive, elite and closed to deliberative decision-making. But networks can also be ‘gatekeepers’. ‘Gate-keeping’ can occur in policy networks that 58 Knowledge Actors and Transnational Governance operate on the basis of specialised technical knowledge or privatise decision-making to stakeholders. As has already been noted, networks systematise the knowledge gen- erated by diverse individual and organisational knowledge actors and impose a rationality that gives precedence to a particular conception of knowledge – usually of a codified, technocratic, secular, Westernised and gendered variety. The exclusionary expert language of scientific technique, economic modelling or professional standards enhances the status of KNETs as ‘sites of authority’ with particular ‘communication codes’. Once mastering such ‘codes’, inclusion into such networks can be a double-edged sword for those seeking policy reform. They run the risk of absorption into dominant practices and co-option into net- work ‘truths’, as well as the cost of some estrangement with grassroots groups and activists not so well versed in the ‘communication codes’ and jargonised argot of, for instance, neo-liberal agendas and acronyms on matters as varied as TRIPS, PRSPS, GATS or GPGs.3 The participants in the Evian Group are well versed in the professional language of the WTO. Indeed, some of the leading Evian members could be argued to be epistemic creators of the codes and jargon of neo-liberal trade theory (Higgott and Erman, 2010: 463). In their professional context, specific groups of knowledge producers – whether they are economists, anthropologists or statisticians – have a cognitive interest in the selective use of their mode of problem defini- tion, methodological approaches and policy solutions. It becomes a self-reinforcing dynamic that encourages resistance to other perspec- tives or disciplinary approaches. Consequently, a network can develop a carapace, sometimes in the interests of internal network cohesion and unity but also to exclude those who do not speak the same specialised language. Policy debate is not taken out of the public domain, but it is cordoned off from those not deemed to be so-called ‘stakeholders’ or those without mastery of the communication code. The barriers to participation in policy networks are not restricted to expert credentials or professional conformity to the norms or ideology of its community. Sustaining a presence in, or monitoring a myriad of, transnational regulatory ‘coalitions’ or ‘policy alliances’ is highly resource intensive for individuals or their organisations. Accessing participation in GPPNs requires time, commitment and funds. Many developing countries, and most ordinary citizens, do not have sufficient resources to devote time to following international policy delibera- tions. When developing countries are stretched significantly to deliver adequate representation in official venues such as WTO negotiations or  Knowledge Networks/Policy Networks 59 treaty discussions, developing effective strategies for their engagement with the more informal transnational policy processes of networks may remain elusive. The dominance of OECD and G20 actors in regional and global policy debates is notable. Accordingly, ‘openness’ and ‘closure’ is not an evenly balanced dynamic across networks. Instead, access and exclusiveness varies considerably across networks, over time and according to issue area or policy field. Networks as public places and public actors A third point on the power of networks revolves around their often ambiguous public or private status. The networks that populate the international political economy often have fluid and interchangeable characteristics, private but undertaking public action and vice versa. The re-creation of the ‘public’ in the global agora is fundamentally different from the public spaces associated within sovereign states. There is no equivalent of public institutions such as the ministries, executive agen- cies and other public bodies with legal remit to deliver public services or impose sanctions. It is less feasible to oppose one category of the public against the other category of the private realm when governance becomes a partnership (McLaughlin, 2004: 163), that is, a mix of private and public. In the interests of legitimacy or to secure resources and official patron- age, network executives sometimes stress their public orientation. The discourse of ‘public goods’ provides one. For example, KNETs represent a means to protect and preserve the public status of knowledge, that is, a means to deliver a public good as will be discussed in Chapter 7 on the GDN (Stiglitz, 2000). The web-sites, publications and data of knowledge groups and networks provide a wealth of information. More tangible training programmes, public events, mentoring and fellowships also represent a public service. Whether it be actively or reluctantly, governments are devolving pub- lic responsibilities and authority. This is a double devolution: first, to domains beyond the nation-state in global and regional domains, and second, a devolution of authority to private networks and non-state actors. Nevertheless, global or regional networks are not public entities – that is, accountable to formally elected representatives of the public or a sov- ereign authority. A network may be accountable to network members, but these member organisations and individuals cannot be considered as representatives of the ‘global public’. Notwithstanding their public sources of support from governments, or their production of public goods, these (semi-) private networks are not subject to the usual 60 Knowledge Actors and Transnational Governance reporting and accountability requirements of public bodies in liberal democracies. The public, even the well-informed OECD countries, are still largely unaware of the roles, reach and influence of global networks. Newspapers and the electronic media do not carry news reports on the Evian Group or INECE. Combined with the technocratic character of many such networks, the potential dynamic is for the public, or the citizenry of nations, to become passive and thereby become excluded and politi- cal responsibility to be undermined. As a consequence of the lack of transparency and mechanisms for public representation, and lack of knowledge about them, these networks operate with relative autonomy and in some anonymity. In any event, they are more able to thwart challenges to their activities or call for transparency by emphasising their non-state, private status or delegated authority. This tendency is compounded in KNETs that also stress their disinterested, scientific and politically neutral endeavours. Consequently, this chapter ends on a cautionary note regarding the democratic potential and deliberative capacity of global (knowledge) networks. They may well be sites of stakeholder engagements, collective action and joint deliberation, but globalisation does not deliver a level- playing field for networks. It is characterised by an uneven distribution of resources and a hierarchy of discourses where ‘subaltern knowledge networks see the spaces for negotiations and deliberations leading to radical outcomes decreasing’ (Rai, 2005: 127). The encroachment into policy making and policy delivery by policy networks and their increasing complexity and autonomy make it difficult, particularly for developing economies, marginalised communities and local activists, to become fully embedded in these transnational policy processes. That is, ‘global spaces and places are increasingly integrated and deterritorial- ized, even as they remain stratified’ (McLaughlin, 2004: 166). The global agora is composed of public spaces, but these are spaces where relatively few can be public actors. The extent to which global and regional net- works become a focal point of public affairs has meaning primarily, or only, for those who have the resources, patronage or expertise to enter and traverse the agora. Conclusion Global policy networks set in motion a structural dynamic that both excludes and opens up policy making to certain groups. In principle, and in popular mythology, the evolution of policy networks offers a  Knowledge Networks/Policy Networks 61 flatter and more horizontal structure (compared to public sector hier- archies) for policy coordination and public service delivery that are porous to participation of private and civil society actors. Yet, networks also privatise decision-making. It is now well accepted that networks provide governance support: networks are viewed as more flexible than the often rule-bound and more rigid bureaucracies; they can promote wider policy participation and build additional routes to decision makers; they can draw together the multiple resources, skills, expertise of actors in corporations and civil society; they perform aggregative and brokerage functions to negotiate and build compromises or boost coordination and compli- ance around rules, as well as performing a socialisation role (Börzel and Heard-Lauréote, 2009). Yet, as this and the previous chapter have suggested, a further critical role of networks is knowledge creation and policy definition. A seventh role, building on all the above, is that authority construction is undertaken via the professional relations within the network that create networks as a new social technology of legitimate and credible governance, that is, a new policy space. Networks are becoming a mode of governance whereby the patterns of linkages and interaction are the means through which joint policy is organised. In short, there is a functional interdependence between public and private actors so that networks allow resources to be mobi- lised towards common policy objectives in domains delegated or delinked from the more traditional hierarchical control of governments. Furthermore, the network logic itself is being diffused by international organisations with their advocacy of partnership and multi-stakeholder policy coalitions as a method to deal with global problems. Networks become platforms of ‘interoperability’ (Walters, 2012: 112). This logic and social technology can promote flexibility and efficiency in dealing with relatively intractable cross-border policy issues. Although these arrangements can become effective mechanisms of transnational policy coordination and partnership, they can also fracture and fragment. These quasi-public policy networks can be ‘under the radar’ of public interest and public scrutiny and represent multiple nodes of policy autonomy and mutated authority that makes them difficult to track, monitor, engage and reform.