Transnational Philanthropy, Policy Transfer Networks and the Open Society Institute

Professor Diane Stone1, University of Warwick

GARNET Working Paper No: 41/08
April 2008

ABSTRACT

The Open Society Institute (OSI) is a private operating and grant-making foundation that
serves as the hub of the Soros foundations network, a group of autonomous national
foundations around the world. OSI and the network implement a range of initiatives that aim
to promote open societies by shaping national and international policies with knowledge and
expertise. The OSI provides an excellent case study of the strategies of transnational
activism of private philanthropy. It is an institutional mechanism for the international
diffusion of expertise and ‘best practices’ to post communist countries and other
democratizing nations. This paper avoids assumptions that civil society is an entirely separate
and distinguishable domain from states and emergent forms of transnational authority.
Focusing on the ‘soft’ ideational and normative policy transfer undermines notions of clear
cut boundaries between an independent philanthropic body in civil society and highlights the
intermeshing and mutual engagement that comes with networks, coalitions, joint funding,
partnerships and common policy dialogues.

Keywords:- transnational networks, philanthropy, civil society, policy transfer.

Correspondence for address

Diane Stone
Department of Politics and International Studies
University of Warwick
Coventry Cv4 7AL
United Kingdom
[email protected]

1
Acknowledgements: A good many people from the Center for Policy Studies at the Central European
University, and in the Local Government and Public Sector Initiative of the Open Society Institute shared their
thoughts and views with me since 2002. They know who they are. Any mistakes or misinterpretations remain
my own.

1
Introduction
Founded in 1993 by the billionaire philanthropist George Soros, the Open Society Institute
(OSI) is a private operating and grant-making foundation based in New York City that serves
as the hub of the Soros foundations network, a group of autonomous foundations and
organizations in more than 60 countries. OSI and the network implement a range of
initiatives that aim to promote open societies by shaping national and international policies
with knowledge and expertise. This paper evaluates its roles and activities as a transnational
policy actor through the analytical lens of firstly, policy transfer and norm brokerage and
secondly, transnational networks.

OSI provides an excellent case study of the strategies of transnational activism of private
philanthropy. OSI is an institutional mechanism for the international diffusion of expertise
and ‘best practices’ to transition countries of Central and Eastern Europe (CEE) and the
former Soviet Union (fSU). The ‘open society’ discourses of transition and reform is multi-
faceted. On a local level, OSI implements a range of initiatives to support the rule of law,
education, public health, and independent media. At the same time, OSI works to build
alliances across borders and continents on issues such as combating corruption and rights
abuses. The idea is to give ‘voice’ to communities, and emerging policy elites, in transition
countries through capacity building, the spread of ‘best practices’ and country-specific
translation of ‘open society’ values.

The discussion will draw upon the related literatures of policy transfer and policy diffusion
(Ladi, 2005; Levi Faur, 2005; Simmons, Dobbin & Garret, 2006). These ideas will be
connected to recent scholarship on knowledge networks (Parmar, 2002; Stone & Maxwell,
2004). In undertaking this conceptual synthesis, this paper is concerned to widen our
understanding on two fronts:

First, an objective is to broaden cognition of the potential domains where policy transfer
takes place from its horizontal intergovernmental focus to vertical supra-national policy
venues. In this regard, this paper is distinctive from some international relations analyses
operating within a frame of methodological nationalism that explain norm diffusion in terms
of its impacts only upon domestic politics (Checkel, 1997). The focus here will be on a
transnational actor seeking to inform and give shape to the domains of global and regional
governance in addition to national and sub-national venues of policy making. This is in line

2
with the argument that civil society actors represent a new logic of governmentality (Sending
& Neumann, 2006). The involvement of non-state actors, and specifically transnational
philanthropy, in certain fields of policy making and policy delivery can promote the
‘transnationalization of policy’. The spread of policy and practice does not always occur in a
simple bilateral exchange between sovereign states but can be complemented or by-passed by
transnational policy networks.

Second, the aim is to extend the range of who (or what) engages in policy transfer and the
diffusion of international norms to include transnational non-state actors such as OSI and
including the various academics, specialists and consultants engaged by OSI in knowledge
networks or in ‘global public policy networks’ (Reinicke & Deng, 2000). Rather than
treating such arrangements only as extensions of state action, networks can develop their own
interests and objectives as well as powers and influences, allowing them to transcend the
state. Transnational policy communities of experts and professionals share their expertise and
information and form common patterns of understanding regarding policy through regular
interaction via international conferences, government delegations and sustained e-
communication (Bennett, 1991: 224-25); that is, an international policy culture. By focusing
on the role of international actors in transferring policy and diffusing knowledge, a dynamic
for the transnationalization of policy comes into analytical sight. In particular, ‘soft’ forms of
transfer – such as the spread of norms and expertise in which non-state actors play a more
prominent role – complements the hard transfer of policy tools, structures and practices
pursued by government agencies and international organizations.

As a philanthropic entity, OSI has multiple identities as a global, regional, national and local
policy actor. To be sure, its capacities as a policy actor at all these levels are quite limited
and constrained. However, given OSI’s resources, its innovative organizational form, and
overt policy (transfer) ambitions, analysis of this network’s regional and global impact is long
overdue (Stubbs, 2005: 79).

The paper is structured into two parts. The first section outlines the concept of policy
transfer, knowledge networks and their connection to global public policy via transnational
philanthropy. The second section applies some of the ideas of the preceding two sections to
the Open Society Institute which is portrayed as having the multiple character of different
kinds of network. Sometimes it conforms to the concept of a transnational advocacy network

3
(TAN). For instance, the OSI’s East-East Program is an exchange program to “develop
advocacy networks for the transnational promotion of open society” (OSI, 2004: 156).
Occasionally, parts of OSI display features of a knowledge network or epistemic community
(Haas and Haas, 1995) while at other times the OSI is a stakeholder alongside other
international partners in ‘global public policy networks’.

The legitimacy and credibility of OSI’s expertise is drawn through a circular process between
the knowledge it produces and the audiences that help legitimize and institutionally
consolidate that knowledge. It becomes a mutual validation process, but one that helps give
intellectual credibility to OSI norm advocacy and policy transfer. This credibility
construction so as to better inform policy deeply implicates OSI in global governance. Its
guise as an independent philanthropic network, sponsoring autonomous foundations in the
separate domain of (global) civil society becomes questionable.

1. Policy Transfer, Networks and Philanthropy

Policy Transfer
Policy transfer is a transnational policy process whereby knowledge about policies,
administrative arrangements or institutions in one place is used in the development of policy
elsewhere. The objects of transfer can include (i) policies, (ii) institutions, (iii) ideologies or
justifications, (iv) attitudes and ideas, and (v) negative lessons (Dolowitz, 1997).
Additionally, there are different degrees of transfer in that actors engage in straight-forward
copying of policy, legislation or techniques as well as various forms of emulation, synthesis
and hybridization, and inspiration (Dolowitz & Marsh, 1996: 351).

Policy and normative transfers can be either voluntary or coercive or combinations thereof.
Terms such as ‘lesson-drawing’ portray transfer as a voluntary and somewhat rational activity
(Rose, 1993). Other terms emphasize compulsory conformity; that is: ‘penetration’ by
international policy actors (Bennett, 1991). By contrast, the more atmospheric term of
‘diffusion’ has been used in World Bank circles (Stiglitz, 2000). For this institution, the word
has more neutral overtones of a natural, gradual and apolitical process. In some of the
international relations literature, the word ‘diffusion’ been used in more precise manner to
recognize the roles played by agents and the prospects for individual and organizational

4
learning where interdependent decision making by national governments is promoting policy
transfer. But this literature leans towards methodological nationalism when it asserts that:
“International policy diffusion occurs when government policy decisions in a given country
are systematically conditioned by prior policy choices made in other countries” (Simmons,
Dobbin, & Garrett, 2006). Assuredly, policy diffusion occurs between countries. However,
this paper goes beyond this hypothesis of interdependence to suggest policy transfer helps
create transnational policy spaces as well.

The mechanisms of transfer are multiple. One mechanism is coercion such as exercised
directly or indirectly by powerful nations or international organizations. The Bretton Woods
institutions have long been accused of dispensing ‘one-size-fits-all’ policies coercively
imposed through loan conditionality (although institutions such as the World Bank now use
the discourse of “putting countries in the driving seat” – Stiglitz, 2000). Coercion is not at
the disposal of a non-state actor like OSI. Yet, ‘soft’ ideational modes of coercion – such as
reflected in accusations from politicians that OSI actors played a catalytic behind-the-scenes
role in the Rose and Orange Revolutions of Georgia and Ukraine – suggest that norms and
policy activity promoting them can have (counter-) hegemonic impact. Russian President
Vladimir Putin accused George Soros of orchestrating the ‘color revolutions’. The contagion
potential of “powerful pro-democracy groups in neighboring countries apparently represent a
very effective power resource for would-be democratizers” (Gray, 2006).2

Similarly, competition as a mechanism of diffusion has been seen primarily as consequence

of economic competition among countries. States compete with each other to attract foreign
capital. “Simplifying regulatory requirements, ameliorating investment risks, and reducing
tax burdens” are often viewed as efficient policy choices but there are also “those who see
competition producing a “race to the bottom” as countries slash taxes and social spending,
environmental and labor regulations, to win over investors and export markets” (Simmons,
Dobbin, & Garrett, 2006). Rather than competition, the literature on Europeanization has
stressed harmonization as a policy dynamic of political and economic convergence.
Convergence also occurs among non-state actors who are increasingly subject to common

2
A number of interviewees distanced the OSI from such interpretations, considering such journalistic
portrayals to be over-simplification of the role that OSI played and a gross over statement of its
influence. “We are not the driving force behind oppositions in Ukraine or Georgia … they would be
there anyway” (interview 3).

5
standards imposed by their donors on project design, budgeting and reporting when
competing for grants. Even so, the competition in ideas and spread of counter-hegemonic
projects also happens when bodies like OSI engage in ‘soft’ policy transfer of opposition
ideas, alternative experts and ‘open society’ norms.

Transfers of ideas or programs are sometimes underpinned by a deeper process of learning.

Here, the emphasis is on cognition and the redefinition of interests on the basis of new
knowledge that affects the fundamental beliefs and ideas behind policy approaches (Hall,
1993). The concept of learning has been subject to numerous interpretations (Bennett &
Howlett, 1992: 277; Checkel, 1997) but can be contrasted to simple copying or emulation.
Policy scholars suggest that cross national lesson-drawing occurs via transnational ‘epistemic
communities’ (Rose, 1993) or that policy oriented learning occurs within advocacy coalitions
(Sabatier, 1991). It is also a concept that resonates with George Soros’ idea of reflexivity.3

Learning can lead to the development of ‘consensual knowledge’ by specialists and epistemic
communities about the functioning of state and society. When consensual knowledge is
developed at a transnational level, the potential exists for the exchange of ideas providing
impetus for policy transfer. Learning in transnational networks helps promote an
‘international policy culture’ but it is not automatically the case that learning will
institutionalize in national government policy or be put into practice at local levels. Learning
is uneven and imperfect across different actors within a policy network, as well as highly
differential in implementation. Political and bureaucratic interests are constrained by electoral
considerations, issues of feasibility, funding shortfalls, war or famine; that is, factors that
prevent ‘harder’ institutional forms of transfer. Certain actors may have a greater capacity for
learning whereas others may adopt lessons for symbolic purposes or as a strategic device to
secure political support or development assistance rather than as a result of reformed policy
understanding. In short, there may be transfer of policy knowledge but not a transfer of policy
practice. With regard to OSI, it would require extensive fieldwork with national foundations
and interviews with staff to assess how open society values are imbibed and translated into
practice, if at all, and the degree to which the OSI normative frame is, or is not, accepted

3
John Gray (2006) paraphrased Soros’ position as follows: “social objects are partly created by
human perceptions and beliefs, and when these perceptions and beliefs change, social objects change
with them. This introduces an element of uncertainty into our view of the world: … we can never
have objective knowledge of society”.

6
within different communities of a country. Even so, learning is not restricted to the nation-
state level. OSI itself is a ‘learning network’. Learning can also occur through its partnerships
with other international organizations or non-state actors. Again, whether the OSI normative
frame is accepted throughout the Network, or among staff of OSI offices, can be a case of
tactical or instrumental learning (to secure a job or project funding), rather than a deep
philosophical commitment.4 Indeed, one interviewee (#3) stated that the term ‘open society’
was “a label used as a reflex without thinking about it”.

‘Open society’ principles as conveyed by the OSI have had mixed reception within the
organization in the development of its organizational culture, as well as within target
countries and communities. National foundations have faced real difficulties in Belarus,
Uzbekistan and Tajikistan due to the oppressiveness of the incumbent regimes towards OSI
as well as other non-governmental actors. Norm diffusion and policy transfer can meet
substantive opposition.

Alongside government agencies and officials at the state level, key actors in the mechanics of
policy transfer are international organizations and non-state actors such as NGOS, think
tanks, consultant firms, law firms and banks. These non-state actors can have considerable
agenda setting influence when they function as part of ‘transnational advocacy networks’.
Recognition of non-state and international organization roles complicates understanding of
policy transfer processes beyond that of simple bilateral relationships between importing and
exporting jurisdictions to a more complex multilateral environment and the
transnationalization of policy.

This paper is concerned primarily with non-state policy transfer, albeit recognizing that there
is dependency of this third form on the two other modes of transfer. Non-state actors in
transnational advocacy networks may be better at the ‘soft transfer’ of broad policy ideas
(Keck & Sikkink, 1998). That is, influencing public opinion and policy agendas as ‘norm
brokers’ (Acharya, 2004; Riggirozzi, 2006). Certainly OSI is involved in the spread of best

4
Among responses from interviewees during discussion about the extent of staff commitment to
‘open society’ values (broadly defined), it twice occasioned the view that OSI was not about
“indoctrination”. The only ‘regularized’ expression of OSI mission and values was in induction
sessions provided for new Board members of national foundations introducing them to the history,
structure and mission of the organization.

7
practices on transparency themes at a country and region wide level (interview 6).
Philanthropic capacity building, expertise based organizations like the Open Society Institute
transfer knowledge, practice and people. In theory, bodies like the OSI have the institutional
capacity to scan the international environment and undertake detailed evaluations of policy
that will help prevent the simplistic, ad hoc copying of policy that leads to inappropriate
transfer and policy failure. The strong local ethos of OSI and the autonomy bestowed on
national foundations is designed to promote the most favorable environment for learning and
‘local ownership’ of policy ideas (Carothers, 1996).

The ‘soft’ transfer of ideas and information is relatively easy. It is a more difficult enterprise
first to see ideas or knowledge about ‘best practice’ structure official thinking or public
discourse and secondly, to ensure that those ideas become institutionalized. Notwithstanding
evidence of considerable degree of information sharing, policy research and expert advice
incorporated in OSI head offices and the various national foundations, the causal nexus
between transferred policy ideas and their adoption is muddied by many intervening
variables. While some ideas may capture the political imagination, many more fall by
wayside. Non-governmental modes of knowledge transfer and normative diffusion are more
extensive than harder transfers of policy tools and practice. The non-governmental status of a
philanthropic body is a major structural constraint to policy transfer. OSI cannot bring about
policy transfer alone but is dependent on governments, international organizations and local
communities to see policy ideas accepted and instituted. Ideas have the power of persuasion,
but they need institutions. Accordingly, non-state actors are often to be found in partnership
or coalition on either an ad hoc or more permanent basis with government departments and
agencies, international organizations or with other NGOs.

Networks
Principal in its armory to inform policy debate and educate opinion has been OSI’s structural
properties as an international network. There are at least three types of network concepts
relevant to analyzing OSI operations:
• Global Public Policy Networks are tri-sectoral and share interests. That is, they are
alliances of government agencies and international organizations with business (usually

8
 corporations) and civil society Reinicke & Deng, 2000).5 Actors invest in these GPPN
networks to pursue material interest but have in common a shared problem. Their
interactions are shaped by resource dependencies and bargaining. They tend to cohere
around international organizations and governments that have publicly entered into a
policy partnership for the delivery of public policy. Examples include the Campaign to
Ban Landmines, the Global Water Partnership and the Intergovernmental Panel on
Forests. There are, however, many more networks. Virtually, all draw in experts and
advisers along with various NGOs, community groups and business interests specific to
the policy focus of the network. Over time the network may become institutionalized with
the creation of formal arrangements such as advisory committees, consultation procedures
and recognition by state and multilateral agencies in the implementation of policies to
gradually become governance structures.

• Transnational advocacy networks encompass a range of non-governmental organizations

and activists with shared beliefs. These networks seek to shape the climate of public
debate and influence global policy agendas and are much less integrated into policy-
making than GPPNs. They are bound together by shared values, dense exchanges of
information and services, and a shared discourse. They are called advocacy networks
because ‘advocates plead the causes of others or defend a cause or proposition’. TANs
give ‘normative resonance’ to cause groups by pulling together the symbols, language
and ‘cognitive frames’ that portray ‘morally compelling’ issues in a concrete manner to
which the public can respond. Participants in advocacy networks can sometimes lack the
status of recognized professional judgment of ‘experts’. However, these networks have
been prominent in ‘value-laden debates over human rights, the environment, women,
infant health and indigenous peoples, where large numbers of differently situated
individuals have become acquainted over a considerable period and have developed
similar world views’ (Keck & Sikkink, 1998: 8-9). These networks cohere around
‘principled beliefs’ – normative ideas which provide criteria to distinguish right from
wrong – unlike epistemic communities (below) which form around ‘causal beliefs’ or
professional understandings of cause and effect relationships. Consequently, transnational

5
The Global Public Policy Initiative – GPPi – has produced numerous empirical studies of these
networks: http://www.gppi.net/about/

9
 advocacy networks (TANs) are more effective in valuing ‘grass roots’, traditional and
non-scientific knowledge.

• Characterized by shared scientific understandings, an ‘international knowledge network’

is “a system of coordinated research, study (and often graduate-level teaching), results
dissemination and publication, intellectual exchange, and financing across national
boundaries” (Parmar, 2002: 13). 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 to practice. The expertise, scientific knowledge, data and
method, analysis and evaluations that help constitute knowledge networks provides the
experts within them with some authority to inform policy (Stone & Maxwell, 2004). The
knowledge credentials and expertise of network actors (PhDs; career profile in a think
tank, university or government research agency; service on blue ribbon commissions or
expert advisory groups, etc) bestows some credibility and status in policy debates that
gives weight to their recommendations. It is an elitist view of experts informing policy.
Although from quite different theoretical standpoints, the category includes the concepts
of epistemic communities (Haas & Haas, 1995), “transnational expertised institutions” (St
Clair, 2006) and neo-Gramscian ‘embedded knowledge networks’ (Sinclair, 2000).

Due to the diversity of its operations, OSI can be seen at various junctures to be exhibiting
features of all these three types of network. This does not mean that OSI is a schizophrenic
or incoherent organization. Instead, due to its decentralized structure and diversity of
projects, some initiatives are more normative in character and typified by advocacy, whereas
in other instances, OSI initiatives are more technical and social scientific.

Civil Society and Transnational Philanthropy

Civil society feeds on, and reacts to, globalization. There is a complex relationship between
firstly, economic globalization, and secondly, the thickening in international rule of law and
new forms of political authority as another driver of globalization, then, thirdly global civil
society as the other ‘driver’ of globalization.6 These drivers are creating new transnational
processes of policy making. Rather than network density and diversity disrupting hierarchies,

6
For a discussion of the debates on conceptualising global civil society see the semi-annual Global
Civil Society produced at the London School of Economics.

10
opening participation and dispersing power, networks can also represent new constellations
of privatized power. Instead of being civil society manifestations of bottom-up, non-statist
globalization, networks are viewed here as ‘mutually implicated’ in the affairs of states and
international organization (Baker, 2002: 936; also Sending and Neumann, 2006). As will be
discussed, there is considerable blurring at global and regional domains of civil society where
authority is not only embedded in the state.

One neglected aspect of global civil society is the ‘elite’ forms of associational life. This
includes a variety of groups with different modes of membership, networking and
organization. This diversity can only be itemized here with a few examples:
1. Foundations such as the Bill and Melinda Gates Foundation; the Ford Foundation
or the Aga Khan Foundation that provide funding and resources to other civil
society organizations or GPPNs;
2. Policy Dialogue Groups such as the World Economic Forum (WEF – Pigman,
2002) in Davos which acts as a transnational convener of opinion leaders in
government, business, academe and NGOs;
3. Promotional groups such as Freedom House that are activist engaged in the
advocacy of certain values and ideals.
4. Business Associations such as the Trans Atlantic Business Dialogue or the Global
Business Coalition on HIV/AIDS;
5. Scientific associations and research groups; for instance, the Global Research
Alliance or the Global Development Network.
These are professional bodies with substantial financial resources or patronage as well as a
high international profile. They are aimed at influencing policy. This sphere of global civil
society is not only hierarchical, but accessible primarily to those with resources; that is
wealthy, westernized professionals and their agencies. It is an uneven playing field
characterized by both intense competition for funds, resources, donor patronage or political
recognition as well as collaboration and gate-keeping.
In this rough schema, OSI overlaps with all types. First and foremost, OSI is a philanthropic
body and is often involved in funding or patronizing the other types of groups listed above.
For instance, Soros has been a speaker at Davos for the World Economic Forum. Secondly,
some of its constituent elements operate as policy dialogue venues. East-East supports
exchanges among actors from civil society in order, inter alia, to share best practices and

11
lessons learned in social transformation.7 Third, OSI advocates normative projects. For
instance, EUmap is a monitoring and advocacy program that works with national experts and
nongovernmental organizations to encourage broader participation in the process of
articulating the EU’s common democratic values as well as in ongoing monitoring of
compliance with human rights standards throughout the union.8

Fourth, and although OSI is not a business association, it has a number of programs aimed at
promoting entrepreneurship as well as ‘economic and business development’.9 Finally, OSI
supports various research groups and think tanks inside and outside its organizational
domain. Although it is legally separate from OSI, parts of the Central European University
(CEU) are connected with OSI activities.

OSI can be contrasted to (but is not unconnected with) the anti-globalization movement, often
portrayed as ‘bottom up’ globalization. It is a diverse and quite fragmented aggregation of
groups and interests, sometimes convening at public meetings like the Global Social Forum,
an example of so-called “contemporary progressive civic transnational advocacy” (Khagram,
2006: xx). Groups in the anti/alter-globalization movement frequently decry limited access
into and lack of accountability of international organizations. Their role in the deliberative
processes of multilateral forums has not been normalized in the manner accorded to the
associations of corporate actors or other groups at the elite end of global civil society. As a
consequence, the decision-making processes of the international institutions remain contested
domains of legitimacy.

Access is partly conditioned by official recognition and public perceptions of legitimacy to

participate. The authority and legitimacy for non-governmental public action in global
affairs is not naturally given but cultivated through various management practices and
intellectual activities. The private authority of WEF or Freedom House or OSI rests in large
degree with their establishment as non-profit or charitable organizations. Their executives
can argue on the one hand, they are not compromised by the need to generate profits in

7
http://www.soros.org/initiatives/east/about
8
http://www.soros.org/initiatives/eu
9
http://www.soros.org/initiatives/business

12
tailoring policy analysis to the needs of clients, and on the other hand, that they have
independence or autonomy from bureaucracies and political leaders. Indeed, the Annual
Reports of the Soros Foundations Network are littered with references to the ‘independence’
and ‘autonomy’ of the national foundations (see also Carothers, 1999: 273).

Another strategy to enhance legitimacy is rhetorical resort to the professional and scientific
norms of scholarly discovery and intellectual investigation. Universities, including CEU,
have long held this status. Think tanks set themselves apart from other non-state actors as
independent knowledge organizations, and often cultivate a reified image as public-minded
civil society organizations untainted by connection to vested interest or political power.
Mixed sources of funding from private or independent philanthropy reinforce this discourse
of dispassionate expertise and critical distance.

A related rhetorical tactic is when non-state actors adopt the mantle of protectors of the
principles and philosophies underlying democratic societies. Numerous organizations lay
claim to participation in public debate by ‘representing’ the interests of minorities or the
human rights of oppressed communities and future generations.

The ends of both the donor and the grantee organization are served. Such discourses of
authority and legitimacy are a necessary component in effectively diffusing ideas and
propelling them into official domains. Via these three discourses of conduct – non-profit
legal and financial independence; dispassionate scientific endeavour; and democratic
representation – credibility is manufactured for non state actors. But in creating their
credibility, they become ‘harnessed to the task of governing’ (Sending & Neumann, 2006:
656).

Civil society dialogues with governments and international organizations have become more
frequent where such groups are treated as ‘partners’ and ‘stakeholders’ in international
development and global governance. It is within this context that the Open Society Institute
can be found. Like most private philanthropic enterprises, the OSI is a legally independent
organization. However, various units of the OSI are to be found in partnership with UNDP,
the World Bank or parts of the European Commission. As OSI expands its “global agenda,
partnerships with other donors are becoming ever more significant” (OSI, 2006: 174). Many
more organizations that are recipients of OSI grants are likewise enmeshed in regional policy

13
dialogues, international alliances or multilateral initiatives. Instead of global civil society
being a flat open sphere for apolitical ‘associational life’, ‘social engagement’ and ‘non-
governmental public action’, it is also a politicised domain traversed by rich and powerful
groups and networks seeking influence through policy analysis and quiet diplomacy or
lobbying. With their substantial financial resources, foundations are in a prime position for
promoting norms and setting agendas for policy debate.

For some, private philanthropy is a privileged strategy for generating new forms of “policy
knowledge” convergent with the interests of their promoters (Guilhot, 2008; Parmar, 2002).
This is apposite when assessing the purposes of the OSI and its sister institution CEU
regarding the political and ideological functions of philanthropic initiatives aimed at higher
education.
… it gives us indications regarding the strategic value of these fields as
laboratories of social reform – both as the training ground of new elites and as
generators of policy knowledge. Investing in higher education does not only earn
philanthropists some social prestige: it allows them to promote “scientific” ideas
about social reform and to define the legitimate entitlements to exercise power
by reorganizing traditional curricula and disciplines. Educational philanthropy
allows specific social groups, using their economic and social capital, to shape
the policy arena not so much by imposing specific policies as by crafting and
imposing the tools of policy-making (Guilhot, 2008: forthcoming).

Indeed, the inspiration for the organization originates with George Soros (and advisors
around him) and in particular, his fascination with the work of Karl Popper. In The Open
Society and its Enemies (1945), Popper argued that totalitarian ideologies have a common
element: a claim to be in possession of the ultimate truth. Ideologies such as communism and
Nazism resort to oppression to impose their version of truth on society. The ‘open society’ is
presented as an alternative that holds there is no monopoly on truth and that there is a need
for institutions to protect human rights, freedom of speech and freedom of choice. After
making billions from hedge fund speculation, Soros established the Open Society Fund with
the objective of “opening up closed societies; making open societies more viable; and
promoting a more critical mode of thinking” (Soros, 1997).

14
However, the interpretation and implementation of the ‘open society vision’ as held by Soros,
by the Board members and by directors of major OSI programs, is dependent on its staff and
their values and interests. Moreover, the recipients of fellowships, scholarships and higher
education support grants are motivated by their own interests. The ‘open society’ is “an
abstract idea, a universal concept” (Soros, 1997: 7). It will be put into practice in a
multiplicity of imperfect, unintended but creative ways. This is all the more the case given
that OSI operates as a decentralized network. In the words of interviewee 1:
“OSI is not one institution… it has a flat structure” “Everything is free
standing”. Sometimes, there is a “failure to capitalize and develop the
multiplying effects” of OSI initiatives. This is because “you have the freedom to
do what you want”...
Rather than this being seen as poor control and problematic coordination, this diversity is
welcomed. OSI is a more complicated organization than a mere cypher and disseminator of
Soros’ philosophy. OSI functions as a vehicle for public action as well as private interest, and
not only the private interests of its founder. Organizational coherence and unity of purpose is
further confounded by its fragmented character. As interviewee 5 put it: OSI is “an octopus of
an organization”… “one part does not know what other parts are doing” ... and there is “no
system wide strategic planning”.

2. The Open Society Institute, Transnational Advocacy and Crafting Policy Expertise10

Most information about the Open Society Institute is found on its web-site. There is very
little independent scholarly material available to access and evaluate the OSI. By contrast,
there is a considerable amount of journalistic material that focuses on George Soros as the
billionaire hedge fund speculator and philanthropist. The focus is on the individual rather
than the organization. Similarly, more attention is devoted to the Bill Gates the individual,
rather than to the operations of the Bill and Melinda Gates Foundation (see for example, The
Economist, 2006).

10
Quotations are not referenced in this section but are based on interviews from October 2006. They
are supported by participant observation and personal conversations with OSI related staff in
Budapest over the period January 2004 through March 2006.

15
This methodological individualism is detrimental. It deflects social science attention away
from organizational manifestations of non-governmental public action. By focusing on the
individual donor as the source of funding, a tendency comes into play where the motivations
of the philanthropist are assumed to guide and direct the operations of the foundation.
Analysis is foreshortened without delving into the organizational ‘black box’. Removed from
sight is the internal politicking, the deviations from principles, the poor implementation
records, the experiments, mistakes and financial misappropriations that may or may not
occur. Importantly, other stakeholders of an organization also shape its vision and strategies.

On its web-site, the Open Society Institute is introduced as:

… a private operating and grantmaking foundation, (that) aims to shape
public policy to promote democratic governance, human rights, and
economic, legal, and social reform. On a local level, OSI implements a range
of initiatives to support the rule of law, education, public health, and
independent media. At the same time, OSI works to build alliances across
borders and continents on issues such as combating corruption and rights
abuses.

OSI was created in 1993 by investor and philanthropist George Soros to

support his foundations in Central and Eastern Europe and the former Soviet
Union. Those foundations were established, starting in 1984, to help countries
make the transition from communism. OSI has expanded the activities of the
Soros foundations network to other areas of the world where the transition to
democracy is of particular concern.11

OSI has been built as an international network but it overlays and funds a series of national
foundations. The foundations network consists of national foundations in 29 countries,
foundations in Kosovo and Montenegro, and two regional foundations, the Open Society
Initiative for Southern Africa (OSISA) and the Open Society Initiative for West Africa
(OSIWA) while a third initiative for East Africa is in gestation. OSISA and OSIWA, which

11
http://www.soros.org/about/overview, accessed 17th March 2006

16
are governed by their own boards of directors and staffs from the region, make grants in a
total of 27 African countries.

The two main central offices are located in New York and in Budapest with additional
independent offices in Brussels, London and Paris. These two offices provide administrative,
financial, and technical support to the Soros foundations. OSI-New York operates initiatives,
which address specific issues on a regional or network-wide basis internationally, and other
independent programs. OSI-New York is also the home of a series of programs that focus
principally on the United States.

The Soros foundations operate as autonomous organizations with a local board of directors
and considerable independence in determining how to implement the ideals of the open
society. Due to the strong ethos of localism and of budgetary control of national boards, “the
Soros national foundations are often perceived in their host countries as being organizations
of those countries” rather than subordinates of OSI-NY or subject to the personal whims of
Soros (Carothers, 1999: 273). In addition, there have been gradual pressures on the national
foundations to become more self sufficient and less reliant on OSI funds. To varying degrees,
these foundations participate in Network wide activities coordinated from New York and
Budapest. Given the concern with transnational advocacy and global public policy, this
paper is primarily focused on the network wide activities as these initiatives are those that are
most transnational in design.

For instance, an early program developed out of New York, and now managed from London,
is the ‘East-East: Partnership Beyond Borders’ program. After 1989 with the onslaught of
East-West exchanges, Soros wanted to provide opportunities for an ‘East-East’
“communications space”. A remarkably low-cost program, it was designed to educate people
into the idea that there could be more to learn from each other rather than going to the West
(Paris, London or New York) where inappropriate models, different historical and economic
circumstances and mismatch of experience could occur. East-East has been “de-mystifying”
who you can learn from. Rather than displacing East-West exchanges, the East-East program
is “additional or supplemental”. East-East is rooted in the national foundations. The program
works because it is “bottom-up” and where the London-based director relies upon people in
the national foundations “who have the pulse” and can deliver ideas and plans on what kinds
of exchanges are needed or wanted. “The national foundations fill in the content”. East-East

17
also serves as a mechanism to link the national foundations to prevent them working in
isolation (interview 1.). In once clear example of policy transfer, one initiative was “to help
European Union actual and prospective candidate countries learn from the experiences of
Central European countries that succeeded in acquiring EU membership” (OSI, 2006: 138).
But whether such exchanges promote policy learning, or merely the dissemination of
information and techniques, remains a moot point.

The multiple snowballing of programs and initiatives has meant that diversity, plurality,
innovation and creative responsiveness to very different local political, economic and social
circumstances has been the order of the day. This ethos is dominant vis-à-vis a bureaucratic
emphasis on audit and financial accountability. There is a sense of creative chaos among a
“concentration of very talented people” (interview 1). Soros “is a man who believes that
OSI is driven by its philosophy”.12 He is also averse to bureaucracy. Consequently, there is
less attention to monitoring and evaluation of organizational methods and impact. Yet, for an
organization that advocates ‘best practice’ on transparency and local government budgeting
through projects such as Revenue Watch, its own internal practices are a step behind that of
the general thrust of the Network’s public advocacy (also Carothers, 1996: 22).

A matter for discussion in interviews concerned tendencies to “universalization” within the

OSI. That is, the assumption that there is commonness of the Eastern European experience (a
homogenization of the different country and local experiences) and a policy belief in the
replication of the experience in other venues. That is, what was done in Eastern Europe can
be transmitted into Central Asia. Such a development in thinking loses what was innovative
and special in the encouragement of “local knowledge” and “local ability” via the national
foundations during the 1990s. The bureaucratization of OSI as the organization consolidates
has prompted a universalizing dynamic that it is in tension with the ‘reflexive’ spirit favored
by Soros. It presents a challenge for OSI given limited resources and insufficient expertise
that prompts a default to “second world lessons for third world contexts … we need to avoid
this” (interview 7).

As the national foundations are encouraged to become more autonomous and independent of
OSI, a subtle transformation is occurring. OSI gradually moves away from a ‘bottom-up’

12
Personal conversation with OSI Board member, Budapest 21st March 2006

18
strong contextual approach to capacity building of local and national communities. A
dynamic is growing of a ‘top down’ professionalized mode of policy interaction with
decision makers. That is, “more like a traditional foundation with program officers in New
York” (interview 5). In short, OSI is moving away from public action that is focused on
capacity building at local and national levels (built in the historical context of post-
communist transitions) to public actions also aimed at transnational levels and at higher level
policy processes. “The losers are the traditional national foundations” (interview 7). This
centrifugal dynamic has created a more vertical set of relationships within OSI constituent
parts, and with its grantee organizations.

Open Society Policy Transfer and Norm Brokerage

OSI is a transnational transfer network. It is engaged in explicit form of normative transfers.
Thus in many respects its can be described as a TAN. Indeed, the OSI motto – “Building a
Global Alliance for Open Society” – is indicative of the organization function of brokering
norms.

A ‘norm broker’ is an agent that generates, disseminates and institutes norms regarding a
political-economic model. The term ‘norm’ includes standardized knowledge and ideas,
principles and practices that are usually framed into paradigms or policy proposals
(Riggirozzi, 2006). However, the extent to which norms find acceptance or
institutionalization is dependent to some large degree on ‘norm takers’; that is, how “local
agents reconstruct foreign norms to ensure that norms fit with the agents cognitive priors and
identities” in a dynamic process of ‘localization’. Norm takers “build congruence between
transnational norms … and local beliefs and practices” (Acharya, 2004: 239-241).

In other words, policy change does not entail a one-way exercise of power and imposition of
paradigms, but rather the capacity of the amalgamation and compromise between the OSI and
local knowledge(s) found within community groups, local and national governments. The
brokerage role articulating between global ideas and local expertise can make the difference
between achieving consensual long-term reforms, or failure from the lack of support and
legitimacy for the reform of institutions (Riggirozzi, 2006).

The OSI operates as a norm broker for ‘open society’ values or paradigm. That is, “rule of
law; respect for human rights, minorities, and minority opinion; democratically elected

19
governments; market economies in which business and government are separate; and thriving
civil societies” (OSI, 2003: 187). At a more specific level, OSI engages in policy transfer
primarily as a generator and disseminator via network wide initiatives, and less so as an
implementer. The national foundations have been more closely involved in implementation.
And these foundations can be seen as both norm brokers and exercising choice as norm
takers.

OSI has played a prominent role in the region “promoting policy research, evaluating policy
options, initiating and disseminating best practices, and monitoring policies…”. As the
“Communist menace” receded in the early 1990s, it “pursued individual grant making for
scholarly research, academic advancement of the local expert communities, and enhancing
diversified civil societies and independent media” (Krizsán and Zentai, 2004: 169-70).13 In
particular, OSI funds a number of think tank initiatives including the new European Council
on Foreign Relations (ECFR) which will “promote a more integrated European foreign policy
with open society values at its core”.14 It is partially emulating the US Council on Foreign
Relations.

The Local Government and Public Sector Initiative (LGI) and PASOS are good examples of
the OSI as a generator and disseminator of policy ideas. PASOS is the Policy Association for
an Open Society – a network of policy institutes from 23 Central and Eastern Europe
countries and the Newly Independent States.15 It provides institutional infrastructure for
pooling and exchanging policy-related knowledge. LGI is an older OSI initiative to promote
democratic and effective governance in the countries of the Soros Foundations Network. It
has specialized in financial management reforms. To that extent, it operates more as a
disseminator and grant-maker than a generator of governance knowledge. LGI provided start-
up funds for PASOS.

13
The Director of CPS, Viola Zentai was initially employed by the OSI when the Center for Policy
Studies was established, but is now full time employed by CEU.
14
Email From: Mabel van Oranje-Wisse Smit Sent: Friday, June 08, 2007 12:24 PM
15
Aside from PASOS and ECFR, other think tank projects supported by OSI include a network of
human rights related think tanks co-ordinated from Budapest
http://www.soros.org/initiatives/hrggp/focus_areas/think .

20
An important component of PASOS activity is to improve the capacity of the participating
centers through exchange and sharing of best practices in a collaborative manner. LGI has
targeted both managerial capacity of the centers and their capacity to prepare better policy
documents and advocacy (through training workshops and mentorship). Considerable
attention is dedicated to twinning centers and sharing of good practices. One area where
policy transfer has already been facilitated by LGI as a ‘knowledge broker’ concerns the
spread and adoption of ‘quarterly economic indicators’ in Ukraine, Moldova and Kazakhstan
in national accounting systems (Ionescu et al, 2006).

One PASOS member is the Center for Policy Studies (CPS), an academic unit based at CEU.
CPS is an important link between the OSI and the CEU. CPS conducts “research and
advocacy” on public integrity and anti-corruption; social diversity and equal opportunities;
rural development and equal opportunities. Project areas where there is an impetus for policy
transfer include research work on “European Integration and Policy Making’ and the
European Commission funded program on ‘economic cultures’. The latter aims at
identifying the types of cultural encounters in the European economy during and after the
enlargement, mapping the major cultural gaps and strategies to bridge them, and enable the
EU to draw lessons for the next rounds of accessions.

CPS incubated a Masters of Public Policy that eventually spawned an independent

Department of Public Policy at CEU. In tune with the wider objectives of the University, it is
“aimed at training professionals to become sensitive to transnational policy/global
governance” (Matei, 2008). CPS also manages the International Policy Fellowship initiative
(Pop, 2006). The motivation for this program is largely to counter ‘brain drain’ by giving in-
country fellowships to researchers and activists who have potential as open society leaders.
Policy transfer occurs through the mentoring process of fellows who gain professional advice
on how to write policy documents, spark public discourse in transition countries, and propel
their ideas into official domains (OSI, 2003: 159).

The transfer undertaken is of western standards of policy professionalism. Indeed, one of the
most popular and widely utilized publications of LGI has been Writing Effective Policy
Papers (Quinn & Young, 2002). The book adapts Anglo-American ideas of policy writing for
post-transition regional audiences adapting to new policy environments. Its authors are
regular participants in various capacity building workshops organized by OSI and other

21
multilateral donors in the region. They transfer professional experience and provide ‘hands-
on’ technical advice about public policy processes.

Notwithstanding historical and individual ties between OSI and various parts of the
University, the relationship between the two is increasingly marked by different trajectories.
Like many universities, disciplinary boundaries are hardening in that scholars stick to their
departments and research interests. Employment contracts no longer require CEU staff to
devote 20 days work to OSI. The pursuit of academic norms has implications in providing
little incentive for faculty to engage with OSI. The University ploughs its own furrow in its
pursuit not only to “help to create regional identity” but also to cultivate “Awareness among
actors in the region about the need to work for regional and global solutions” (Matei, 2008).

The ‘Global Turn’ and the ‘Policy Awakening’

From the end of the 1990s, the Network engaged itself in various debates regarding global
transformations and as a consequence, ventured to reach out to new regions of the world
(Palley, 2002). The programs of the Network sought to critically examine issues of emerging
democracies not only in a post-socialist but in a global context (Krizsán and Zentai, 2004).
As Soros become interested in globalization, he was less inclined to close down OSI as
originally intended (interview 2).

The ‘global’ strategy was also pushed by the accession of the new 10 member states to the
EU. This ‘global turn’ is perceived by interviewees 1. and 3. as a horizontal stretch of OSI
activity across geographies than a vertical conceptual expansion of influencing global debates
and international organization. This view can be contrasted to the public opinion of George
Soros who stated: “Our global open society lacks the institutions and mechanisms necessary
for its preservation, but there is no political will to bring them into existence” (Soros, 1997:
7). One event to mark the newly re-fashioned goals of the OSI was a conference in 2001 at
the Central European University for its 10th Anniversary conference. For CEU and its sister
institution OSI, the conference was a venue to publicize their joint commitment to the ‘global
open society’ and the objective “to articulate critical and policy views in the global public
sphere” (Krizsán and Zentai, 2003: 37, 35). An objective has been to get the “network
programs to think global, to spread expertise” (interview 7)

22
Through the mechanism of the Chairman’s and Presidential Grants, there are further signs of
the global agenda. Grant giving to transnational advocacy programs (interview 7) is
especially apparent in the fields of human rights and anti-corruption. In 2003, funding went
to bodies such as Global Witness16, the Data Foundation (for educating the US public about
debt relief, aid and trade), the Altus Global Alliance17, TIRI,18 and long standing OSI partners
such as Human Rights Watch and the International Crisis Group (OSI, 2003: 190).

Another indicator of the ‘global turn’ is the degree of interaction and partnership between
OSI and international organizations such as the World Bank, the European Union, UNDP and
the WHO as well as a range of other non-state international actors (OSI, 2006: 174). One
important example has been the long standing record of work of OSI regarding Roma
communities, support for the establishment of the European Roma Rights Center and the
regional Roma Participation Program amongst other initiatives. Much work involved surveys
and data gathering simply to understand the dimensions of the situation faced by Roma. In
mid 2003, the OSI in conjunction with the World Bank initiated the ‘Decade of Roma
Inclusion: 2005—2015’. The two institutions have subsequently brought to the partnership
most regional governments as well as that of Finland and Sweden, the European
Commission, UNDP and the Council of Europe (Krizsán and Zentai, 2004: 175-80). Others
in OSI also see the Roma Decade as a success albeit more as a “rhetorical device” (interview
3) and an “empty frame” to fill (interview 5). Even though Roma concerns are advocated in
regional or global institutions, bring about a change in attitudes and practice in local
communities and national administrations is a slower process.

As such, the ‘global turn’ has many dimensions. It includes the internationalization of civil
society at national and sub-national levels through capacity building initiatives to educate
local communities and policy actors into the impact of globalization and regionalization.
Additionally, OSI partners with international organizations and governments, sometimes in

16
http://www.globalwitness.org/
17
http://www.vera.org/project/project1_1.asp?section_id=9&project_id=69 Established by the Vera
Institute of Justice
18
http://www.tiri.org/ TIRI was originally an acronym for Transparency International Research
Institute (spun-off from Transparency International after a leadership dispute) but the organization
prefers to be known as TIRI. It describes itself as one of “a new generation of global policy network”.

23
arrangements that are similar to ‘global public policy networks’ (OSI, 2006: 174). OSI was a
late donor to the Campaign to Ban Landmines. It has also partnered with the Consultative
Group to Assist the Poorest, and the Global Fund to Fight AIDS, Tuberculosis and Malaria,
all of which are considered GPPNs (Reinicke & Deng, 2000). Indeed, the Roma Decade
partnership can be thought of as such a network. Elsewhere OSI has been described more
generally as a ‘global democracy promotion public policy network’ (Sisk, n.d).

The global turn is also apparent in the re-articulation of the Network’s driving principles for a
‘global open society’. On this later score, the OSI displays many features of a TAN: It is a
human rights advocacy network. But there are also some important differences from the TAN
concept. The OSI network “is a more formal and institutionalized network with an
established bureaucracy, relatively secure funding and gradually centralising structure”
(Krizsán and Zentai, 2004: 171).

Rather than resembling an ‘epistemic community’ in its strict definitional sense, OSI
nevertheless operates as, and with, knowledge networks. For instance, OSI supports research
in the public health field such as the international effort on extremely drug-resistant
tuberculosis (XDR-TB).19 Additionally, the Public Integrity Education Network (jointly run
by CPS at CEU and TIRI, and part funded by OSI) is a network of universities and civil
service colleges in 60 countries engaged in the development of new courses and resources in
the field of governance, integrity and administrative reform.

Notwithstanding the different orientation or composition of these networks, they act as

vehicles for the international diffusion of norms, ideas and practice. However, the global turn
is not an evenly spread dynamic throughout OSI. Few of the national foundations work on
regional issues, or on international organizations, or even on other countries. “Most work on
public policy”; that is, work at the national level (interview 4).

Related to the ‘global turn’ is its ‘policy awakening’. As OSI has matured, it has advanced
from a focus on capacity building to using built capacity to influence policy. For instance, as
stated by the former LGI Research director:

19
Accessed 30 April 2007 at:
http://www.soros.org/initiatives/health/focus/phw/news/drugresistant_20070314

24
 We… have started to gradually move towards new forms of international
development. Beyond traditional action-oriented, grant-giving and capacity-
building activities, we are actively involved in policy design and policy-making
(Gabor Peteri, preface to Quinn & Young, 2002).

Institution building and open society advocacy has not been supplanted. However,
recognition that “the collapse of a repressive regime does not automatically lead to the
establishment of an open society” prompted more nuanced, targeted and policy focused
approach in the Network (Soros, 1997: 10-11). Indicative of the policy awakening is the
support given to PASOS and other think tanks as well as the International Policy
Fellowships.

This ‘policy turn’ has been by design, but was also an economic necessity:
Grant making in 1990s style is no longer feasible. The US dollar was stronger and
went a long way. We have expanded our reach, so foundations have become
smaller in real dollar terms (interview 3.);

The global and policy turn happened simultaneously. We had focused on

generalized civil society development but resources were starting to decrease. We
concentrate on policy change now rather than a thousand flowers blooming
(interview 7.)

The focus on policy work and policy development is succeeding civil society development,
and represents a more self conscious approach in OSI. At the same time, there is an “on-
going internal discussion” about the crisis of sustainability for many civil society
organizations in Central and Eastern Europe (interview 5, also, Koncz, 2006).

OSI combines activities with normative aspirations and advocacy as a TAN alongside
scientific analysis that might be associated with a knowledge network. The Network has
“functioned as a mechanism of bridging knowledge production and policy…” (Krizsán and
Zentai, 2004: 169). The organization is sufficiently broad and flexible to encompass a variety
of differently motivated actions. While there are contradictions or tensions inherent in such
combinations, there are also potential benefits in consolidating the mission of the OSI. The
network structure potentially facilitates the incorporation of local expertise into more

25
traditional and elite research approaches. Parts of the Network can be engaged in policy from
conceptualization through to policy advocacy, concrete action and monitoring, although this
does not happen as often as hoped (Krizsán and Zentai, 2004: 174, 182).

Its image as a relatively rigorous knowledge actor is a significant source of authority for OSI.
Many of its operatives have social status as experts and reputable policy analysts. Many OSI
operations are think tanks or other types of research and analysis organization. Attributed as
public-spirited and with a steadfast commitment to independence, objectivity and scholarly
enterprise bestows authority on OSI in a dynamic that also boosts the reputations of the
individuals associated with it. These groups (and sometimes the media in its quest for expert
commentary) legitimate OSI staff as ‘serious’ and ‘expert’ persons. To maintain their
organizational reputation and repudiate accusations of politicization, advocacy and lobbying,
or ideological polemic, OSI executives have encouraged engagement with academic
communities. In this regard, the strong relationship and physical proximity of OSI-Budapest
to its sister institution, the Central European University is important.

In sum, the OSI has been in constant renegotiation and reconstruction of its identity and in
pushing out its socio-political boundaries as a transnational actor. As a transnational network
it does not operate in separate domains at global/regional levels that are apart from the
national/local. Instead, it is engaged in making critical cross cutting links. In re-inventing
itself from a norm broker in opposition to communism and advocating open society values to
a body with stronger research capacity, it has also sought to bridge social science and praxis.
Finally, although it is a non-state actor, its partnership activities and policy aspirations
substantially blur the distinctions of OSI as an independent civil society organization.

3. Global Policy and Non Governmental Public Action

The source of OSI power and influence does not lie in numbers; it does not have or seek
electoral support, and it is not a social movement. Nor does it have the power and authority of
public office; it is outside the international civil service of intergovernmental organizations
and state bureaucracies. In terms of material power, OSI is puny compared to that of
corporations and the economic clout of business, notwithstanding the hundreds of millions
ploughed in by its founder. Instead, the sources of its power in policy lie in the appeal of its
norms, knowledge and networks. That is, the norms of the open society and human rights

26
bolstered by knowledge creation through think tanks, university and policy fellowships that is
jointly disseminated and advocated through collective action of alliances, partnerships and
networking.

OSI seeks to provide the conceptual language, the normative paradigms, the empirical
examples that then become the accepted assumptions for those making policy. OSI does not
act alone in such intellectual action, but more usually in coalition with like-minded thinkers
and activists in journalism, the professions, universities and so forth. Through its networks,
OSI has ‘boundary transcending’ qualities. It draws together intellectual resources allowing
the OSI to do the work of articulation between the national, regional and global levels of
governance. The very concept of ‘open society’ has also been represented by Soros as one
that has boundary transcending qualities:
… the open society as a universal concept transcends all boundaries. Societies
derive their cohesion from shared values. These values are rooted in culture,
religion, history and tradition. When a society does not have boundaries, where
are the shared values to be found? I believe there is only one possible source:
the concept of the open society itself (Soros, 1997: 7).

However, rather than just transcending boundaries, the OSI – and numerous other
transnational actors – are carving out new transnational spaces for public action. The
Habermasian notion of a public sphere goes some way to accommodating this idea of a realm
for the evolution of public opinion (Hodess, 2001: 130). However, this notion is based
predominantly on debate and dialogue, neglecting in considerable degree the variety of
institutional developments that populate this space and the global policy processes that
networks form. Instead of a simple co-option into governance, OSI is proactive in the
creation transnational policy transfer processes through its own international network
infrastructure and identity, as well as through its multiplicity of policy partnerships.

Within this sphere of global policy debate and networking, OSI has been consolidating its
own credibility and authority in part by creating its own audiences and reference points.
Funding intellectuals, NGOs, the CEU and other academic centers helps build clientele
relationships between these grantees and OSI, as well as with other foreign donors. By no
means is this exceptional to OSI, but a common feature of philanthropic foundations
(Roelofs, 2003: 188). The point, however, is that OSI subsidizes various experts and

27
intellectuals to inform civil society and professional or bureaucratic audiences. Instead of the
linear transmission of knowledge with OSI as a conveyer belt of policy ideas, a circular
process is in operation whereby the constituencies of OSI are sources of legitimation of OSI
as a “transnational expert institution” (St Clair, 2005). Recognition of OSI supported think
tanks as centers for expert, scientific and authoritative advice occurs because of the scholarly
credentials and output of these organizations. It also happens because of the relationship with
public institutions and donor groups that have a vested interest in the general belief that
policy institutes and are rational social tools for policy planning. Commissioning and funding
studies, these interests want independent and rigorous analysis. On the other side of the coin,
these international agencies can then legitimize their policy position by arguing that they are
interacting with and consulting independent civil society organizations. Moreover, the
various policy networks of the national foundations further embed OSI in a range of official
actions and public policies. Clear distinctions between state and non-state, public and
private, actors become blurred.

OSI becomes a ‘meta-NGO’ where its primary purpose is to provide support to other NGOs
and groups but can “end up ‘governing’ other NGOS” (Stubbs, 2005: 81). And in taking a
‘global turn’ in its ‘policy awakening’ OSI-NY and Budapest become more distant from local
associations and closer to international organizations and other ‘transnational expert
institutions’.

To conclude, this paper avoids assumptions that civil society is a separate and distinguishable
domain from states and emergent forms of transnational authority. Focusing on the ‘soft’
ideational and normative policy transfer undermines notions of clear cut boundaries between
an independent philanthropic body in civil society and highlights the intermeshing and
mutual engagement that comes with networks, coalitions, joint funding, partnerships and
common policy dialogues. While OSI has origins as a grass roots civil society actor in the
post communist countries, it has become an elite global policy organization.

28
References
Acharya, Amatav. (2004) ‘How Ideas Spread: Whose Norms Matter? Norm Localization and
Institutional Change in Asian Regionalism’, International Organization, 58 (Spring):
239-275.

Baker, Gideon. (2002) ‘Problems in the Theorisation of Global Civil Society’, Political
Studies, 50(5): 936.

Bennett, Colin. J. (1991) ‘Review Article: What is Policy Convergence and What Causes It?’
British Journal of Political Science, 21: 215-233

Carothers, Thomas (1999) Aiding Democracy Abroad: The Learning Curve, Washington
DC., Carnegie Endowment for International Peace.

Carothers, Thomas (1996) ‘Aiding Post-communist Societies: A Better Way?’, Problems of

Post-Communism, September/October: 15-24.

Checkel, Jeffrey. (1997) ‘International Norms and Domestic Politics: Bridging the
Rationalist-Constructivist Divide’, European Journal of International Relations, 3(4):
473-95.

Dolowitz, David. (1997) ‘British Employment Policy in the 1980s: Learning from the
American Experience’, Governance, 10 (1):

Dolowitz, David. & Marsh, David. (1996) ‘Who Learns from Whom: A Review of the Policy
Transfer Literature’, Political Studies, 44(2): 343-57.
Economist (2006) ‘A Survey of Wealth and Philanthropy’, The Economist,

Gray, John. (2006) ‘Moving Target’, New York Review of Books, 53(15) October 5.

Guilhot, Nicolas. (2008) ‘Reforming the World: George Soros, ‘Global capitalism and the
philanthropic management of the social sciences’, Critical Sociology, forthcoming.

Haas, Peter. M. & Haas, Ernst. B. (1995) ‘Learning to Learn: Improving International
Governance’, Global Governance, 1 (3): 255-285

Hall. Peter (1993) ‘Policy Paradigms, Social Learning and the State: The Case of Economic
Policy making in Britain’, Comparative Politics, 25: 275-97.

Hodess, Robin. (2001) ‘The Contested Competence of NGOs and business in public life’, in
Daniel Drache (ed) The Market or the Public Domain: Global Governance and the
Asymmetry of Power, London, Routledge.

Ionescu, Adrian, Prohnitchi, Valeriu., Mukhambetova, Gulnar., and Stone, Diane. (2006)
“Transferring Expertise and Methodologies: A Case Study of the Development of
‘Quarterly Economic Indicators’ in Moldova and Kazakhstan”, in Robert Klitgaard
(editor) Linking Think Tanks, Santa Monica, California, RAND (forthcoming).

Keck, Margaret, and Sikkink, Kathryn (1998) Activists Beyond borders: Advocacy Networks
in International Politics, Ithaca NY., Cornell University Press

29
Khagram, Sanjeev. (2006) “Possible Future Architectures of Global Governance: A
Transnational Perspective/Prospective”, Global Governance, 12(1): 97-117

Koncz, Katalin. Ed. (2006) NGO Sustainability In Central Europe: Helping Civil Society
Survive, Budapest, Local Government and Public Service Reform Initiative, Open
Society Institute.

Krizsán, Andrea. and Zentai, Violetta. (2004) “From Civil Society to Policy Research: The
Case of the Soros Network and Its Roma Policies” in Diane Stone and Simon Maxwell
(eds) The Challenge of Transnational Knowledge Networks: Bridging Research and
Policy in A Globalising World, London, Routledge.

Krizsán, Andrea. and Zentai, Viola (2003). ‘Introduction’ in Andrea Krizsán and Viola Zentai
(eds) Reshaping Globalization: Multilateral Dialogues and New Policy Initiatives,
Budapest, Central European University Press.

Ladi, Stella. (2005) Globalization, Policy Transfer and Policy Research Institutes,
Cheltenham, Edward Elgar

Levi-Faur, David. (2005) ‘The Global Diffusion of Regulatory Capitalism’, in David Levi-
Faur and Jacint Jordana, (eds) ‘The Rise of Regulatory Capitalism; The Global
Diffusion of a New Regulatory Order’, Annals of the American Academy of Political
and Social Science, vol 598, March: 27.

Matei, Liviu. (2008) ‘How a Small University Can Contribute to Global Governance’, Global
Governance, forthcoming.

Open Society Institute (2006) Building Open Societies: Soros Foundations Network 2005
Report, New York, Open Society Institute

Open Society Institute (2004) Building Open Societies: Soros Foundations Network 2003
Report, New York, Open Society Institute.

Palley, Thomas. (2003) ‘The Open Institute and Global Social Policy’, Global Social Policy,
3(1): 17-18.

Parmar, Inderjeet. (2002) ‘American Foundations and the Development of International

Knowledge Networks’, Global Networks, 2(1): 13-30.

Pigman, Geoffrey Allen. (2002) ‘A Multifunctional Case Study for Teaching International
Political Economy: The World Economic Forum as Shar-pei to Wolf in Sheep’s
Clothing? International Studies Perspective, 3: 291-303

Pop, Daniel. (2006) ‘Policy Research Fellowships and international networking as bridging
devices’, Bridging Brief Series, Global Development Network’s Bridging Research and
Policy Project: http://www.eerc.ru/details/download.aspx?file_id=10015

Quinn, Lisa and Young, Eóin. (2002) Writing Effective Policy Papers: A Guide for Policy
Advisors in Central and Eastern Europe, Budapest, Local Government and Public
Service Reform Initiative, Open Society Institute.

30
Reinicke, Wolfgang. & Deng, Francis (2000) Critical Choices: The United Nations,
Networks and the Future of Global Governance, executive summary, available at:
www.globalpublicpolicy.net/

Riggirozzi, Pia. (2006) “Norm brokers” in Diane Stone and Chris Wright (eds.) The World
Bank’s Decade of Reform and Reaction, London Routledge.

Roelofs, Joan. (2003) Foundations and Public Policy: The Mask of Pluralism, Albany,
SUNY.

Rose, Richard. (1993) Lesson Drawing in Public Policy: A Guide to Learning Across Time
and Space, Chatham, N.J., Chatham House

Sabatier, Paul. (1991) ‘Political Science and Public Policy’, PS: Political Science and
Politics, 24(2): 144-156.

Sending, Ole Jacob. And Neumann, Iver. B. (2006) ‘Governance to Governmentality:

Analyzing NGOs, States and Power’, International Studies Quarterly, 50:651-672.

Simmons, Beth., Dobbin, Frank., and Garrett, Geoffrey. (2006) ‘Introduction: The
International Diffusion of Liberalism’, International Organization, 60(4) 2006: 781-
810

Sinclair, Timothy. J. (2000) ‘Reinventing Authority; Embedded Knowledge Networks and

the New Global Finance’, Environment and Planning C: Government and Policy, 18:
487—502.

Sisk. Timothy. (n.d.) “Global Networks for Democracy Promotion: Enhancing Local
Governance’, Case Study for the UN Vision Project on Global Public Policy Networks.

Soros, George (2001) On Globalization, New York, Public Affairs.

Soros, George. (1997) “The Capitalist Threat”, Atlantic Monthly, 279(2): 45-58. accessed 11th
April 2006 at:
http://people.hofstra.edu/faculty/Robert_Bohl/News%20Clippings/Political%20Econo
my/The%20Capitalist%20Threat%20-%20Soros.pdf

Stiglitz, Joseph. (2000) ‘Scan Globally, Reinvent Locally: Knowledge Infrastructure and the
Localization of Knowledge”, in Diane Stone (ed.) Banking on Knowledge: The Genesis
of the Global Development Network. London: Routledge

St Clair, Asunción. (2006) ‘The World Bank as a Transnational Expertised Institution’,

Global Governance, 12(1): 77-95.

Stone D. & Maxwell S. eds (2004) Global Knowledge Networks and International
Development: Bridges Across Boundaries, London, Routeledge.

Stubbs, Paul. (2005) ‘Stretching Concepts Too Far? Multi-level Governance, Policy Transfer
and the Politics of Scale in South East Europe’, Southeast European Politics, 6(2): 66-
87.

31
Zentai, Violetta. (2000), ‘Messianic Fantasies’, chapter 5, doctoral dissertation, Rutgers
University.

List of Interviews:

Tom Carothers, Carnegie Endowment for International Peace, Washington DC., 8th March
2007

Allison Giffen, Policy Analyst, Washington DC. 6th March 2007

Robert Kushen, Director of International Operations , OSI New York, 26th February 2007

Mary Frances Lindstrom, Cambridge House, Open Society Foundation office, Hammersmith,
London, 7th December

Bronwen Manby, Cambridge House, Open Society Foundation office, Hammersmith,

London, 7th December

Jonas Rolett, Regional Director for Southern Central and Eastern Europe, Washington DC.
6th March 2007

George Vickers, Regional Director for Latin America, Washington DC. 6th March 2007.

32