ASSEMBLAGES, TRANSLATION, AND INTERMEDIARIES IN SOUTH EAST EUROPE: Rethinking transnationalism and social policy

Social policy is no longer a matter for domestic actors alone. Whilst there are significant theoretical disagreements about how to study the transnational dimensions of social policy, and empirical disagreements about the relative weight to assign to international and domestic actors in terms of specific policy domains, no study of the making of social policy is complete without, at least, some recognition of the role of external actors. Questions remain, however, about how far the increased salience of international actors, trans-national processes, various kinds of network-based organisations, and new intermediaries has effectively depoliticised social policy, or how far it has led to a more intense internationalisation of political struggles and mobilisations.

This paper addresses these issues theoretically and empirically, through a review of recent work on trans-national processes and flows, and an overview of the making of social policy in South East Europe.¹ Theoretically, the paper argues that notions of ‘social policy’ and of the ‘external’ and ‘internal’ cannot be held as constants in these debates but must, also, be questioned and seen as socially constructed. Empirically, South East Europe is of considerable interest as an emergent regional space, largely ascribed from outside, in which political and institutional arrangements have been profoundly destabilised, and sub-national, national and regional spaces and their institutionalisations, inter-relationships and re-scalings are heavily contested, not least in terms of the delineation of sovereignty. There is a continuing ‘clash’, therefore, between an uneven and unclear European trajectory and a fragmented, projectised, development and reconstruction agenda. Politically, since, in this region, ‘governance and the subjects and objects of governing are in process of simultaneous and mutual invention or constitution’ (Clarke 2008), the construction, deconstruction and reconstruction of diverse kinds of solidarities is an ongoing process, itself affecting and affected by, social policy processes and flows, rendering stable notions of ‘citizenship’ and, indeed, of social policy itself, deeply problematic.

The next section introduces the idea of social policies as fluid, complex, multi-actor assemblages rather than as path-dependent regimes and draws on notions of trans-national policy as translation rather than transfer. Section 3 reframes social policy in South East Europe based, in part at least, on insights derived from a series of linked case studies. It explores, in particular, the limits of a construction of the region as a site of neo-liberal governmentalities, based on partnerships between reconfigured states, new ‘service-oriented’ NGOs, technocratic professionals, and powerful international actors. In conclusion, section 4 explores the same themes from a normative perspective in terms of the ways in which depoliticised technocracy may erode a public sphere of open debate and choices, whilst containing the possibilities for a re-politicisation as new spaces are opened up and new coalitions and discourses emerge which resist and challenge these processes.

Clearly, international and trans-national processes pose distinctive challenges to the traditional boundaries and conceptual frameworks of the discipline of social policy. International actors can no longer simply be ‘added on’ to traditional national studies. Rather, a new vocabulary and epistemology is needed to capture the complexity and liminality of the encounters between actors, sites, discourses, scales and contexts. A wholesale deconstruction of the taken-for-granted conceptual apparatus around ‘social policy’ and ‘welfare regimes’ developed in the Global North is needed to allow for ‘the emergence of a multifaceted perspective that recognizes hybridity, incorporates diverse insights and promotes a truly global understanding of social welfare’ (Midgley 2004: 217). In thinking ‘difficult thoughts’ about welfare (Clarke 2004: 2), there is, of course, a danger that we over-emphasise the ‘newness’ of the approach being suggested, rather than tracing its much longer lineage in social science and its significant, although usually peripheral or marginal, role in social policy studies. The approach seeks to keep alive a conversation about welfare arrangements as constructed, contested and contradictory (Clarke 2004: 5), borrowing from traditions which are in tension with each other and, crucially, rarely impact on the core of analyses of social policy. The dynamics of a region in flux, in the context of the rather complex role of transnational agencies, actors, resources and ideas make the approach we adopt, in our view, particularly useful for extending these conversations and shedding new analytical light on processes, structures and spaces of welfare. Here, we can do no more than sketch some of the elements of a new conceptual architecture in terms of the importance of a number of concepts: ‘assemblages’, ‘translation’, ‘intermediaries’, and ‘space and scale’.

Assemblage as a concept, emerging from the work of Deleuze and Guattari (2004), seeks to grasp both the structural and fluid aspects of social phenomena pointing ‘to complex becoming and multiple determinations’, being ‘sensitive to time and temporality in the emergence and mutation’ of phenomena (Venn 2006: 107). It, thus, stands in stark contrast to the rigidifying reification of final or stable states (Marcus and Saka 2006: 106) which haunts the work of classic social theory and mainstream social policy studies. It rejects the Western imperialism of a modernist epistemology in which emerging welfare systems are squeezed into one of a number of ideal typical welfare regimes, whether liberal, conservative corporatist, or social democratic (Esping-Anderson 1990), and resists the temptation to name new ideal types whether Southern European (Ferrera 1996) or post-communist (Deacon et al. 1997).

The concept has been deployed in John Clarke's brilliant critique of the objectivism and structuralism of mainstream welfare studies, played out in a ‘big picture’ obsession with the ‘usual suspects’ such as ‘globalization, neo-liberalism, … post-Fordism, or the needs/interests of capital’ (Clarke 2004: 3) which serves to obliterate questions about the contradictory, complex, and uneven processes at the heart of contemporary transformations of the social, and tends to leave out all which does not fit into a linear, unidirectional narrative. Theorising in terms of ‘welfare assemblages’ allows us to consider ‘how a specific moment is shaped by multiple and potentially contradictory forces, pressures and tendencies’ (Clarke 2004: 25), with outcomes, in terms of new ‘welfare settlements’, much more shaky, unstable, unfinished, and complex than can ever be captured by ‘welfare regime’ studies. In Clarke's terms, context and conjunctures matter because ‘they embody (contested) imaginaries’ (Clarke 2004: 47). In South East Europe, we would argue, multiple and competing meanings of ‘social policy’, ‘transition’, ‘reconstruction’, and ‘development’ cannot be squeezed into a simplified notion of an emergent welfare regime but, rather, need to be addressed as complex, contradictory and contested, constructed in and through encounters within and between various actors, local, national, trans-national and inter-stitial.

Latour and Callon's work on ‘translation’ takes the concept way beyond linguistics, in an attempt to emphasise the fluid and dynamic nature of the social world, where meanings are constantly transformed, translated, distorted and modified (Latour 2005). The notion of translation problematises social policy as ‘policy’, seeing this as a continuous process of ‘displacement’, ‘dislocation’, ‘transformation’ and ‘negotiation’ (Callon 1986). ‘Translation’ occurs in a complex web of social actors and non-social actants, in actor networks, with everybody and everything enrolled in the network being active members and mediators shaping and transforming claims, artefacts, discourses, and interpretations according to their different projects (Latour 1987). Hence, while the concept is inseparable from ‘the building, transforming or disrupting of power relations’ (Sakai 2006: 71–2), translation is ‘a continuous process through which individuals transform the knowledge, truths and effects of power each time they encounter them’ (Herbert-Cheshire 2003: 456). Importantly, the notion of translation works with a poststructuralist understanding of power, where power is understood as a temporal, fluid and interactive process, working through distortions, displacement and transformation resulting in a constant inscription and re-inscription of power relations.

Above all, an emphasis on policy as translation questions the realist ontology of an orthodox, and influential, policy transfer literature in which ‘policy’ exists as a kind of ‘package’ ready and able to be transplanted or transferred from one setting to another. The policy transfer literature, notwithstanding more nuanced contributions such as de la Porte et al. (2001) on the EU's social policy as a new form of governance, tends to work through binary oppositions: either policy is institutionalised in another place or resisted; it either ‘fits’ or it does not fit; it is picked up by institutions and actors or it is blocked by veto players and/or at institutional veto points. Crucially, however nuanced are understandings of power games and social learning, the literature is still dominated by a rather linear, institutionalist, perspective. By reconsidering our understanding of the policy transfer process from the point of view of translation theory we would argue, instead, that the policy transfer process should be seen as one of continuous transformation, negotiation, and enactment and as a politically infused process of dislocation and displacement. Emphasising processes of formation, transformation and contestation implies that policy transfer is never an automatic or unproblematic, taken-for-granted, process. Rather, it suggests the need to pay attention to the ways in which policies and their schemes, content, technologies and instruments are constantly changing according to sites, meanings and agencies (Lendvai and Stubbs 2007b).

The trans-nationalism of policy as translation focuses on the attempt to render certain specific policies as universal and to ‘re-transcribe’ (Venn 2006: 82) existing socio-economic, administrative and cultural practices within its idiom. There is nothing new in the movement of ideas, institutional blueprints, discourses and knowledge claims between and across sites, scales and actors. However, in the last 30 years, and particularly in so-called ‘countries in transition’ over the last decade and a half, these processes have intensified. Hence, acts of translation are, themselves, crucial in the construction of new welfare assemblages. South East Europe, at the confluence of ‘Europeanisation’, ‘anti-poverty developmentalism’ and ‘post-war reconstructionism’ is being multiply signified so that ‘national social policy’ frameworks become deconstructed, reconfigured, reframed and re-coupled (Lendvai 2007). Social policy becomes multiply framed as ‘combating poverty’, ‘meeting the needs of the vulnerable’, and as ‘social inclusion’, with the concept of ‘social protection’ meaning many different things to different actors in different frames. All manner of diverse and different ‘scripts for social policy’ necessarily, therefore, stretch domestic understandings which, in fact, are never as purely domestic as they may first appear.

Translation as a process needs translators, including a growing cadre of transnational and local policy consultants and experts, who do more than merely provide ‘technical’ knowledge and ‘disinterested analysis’, but are interested actors in their own right. The importance of these new intermediaries whether termed ‘strategic brokers’ (Larner and Craig 2005), ‘boundary spanners’ (Williams 2002); ‘interlocutors’ (Belier and Wilson 2000), ‘border crossers’ (Ferguson and Gupta 2002), ‘transactors’ (Wedel 2004), or ‘cultural brokers’ (Trevillion 1991) is increasingly recognised, but rarely addressed in the social policy literature.

Intermediaries come in different guises; often as policy experts, advisors, and consultants of various kinds but, increasingly, as non-specialist ‘fixers’ who, whilst not claiming any sectoral competence, claim to be able to connect, facilitate and smooth the passage between diverse actors in a chain of funding, ideas, and/or projects. Typically, intermediaries are located in ‘hybrid’ or ‘interstitial’ spaces: in-between scales, organisations, discursive practices, knowledge systems, and geographies. This liminal space is best characterised in terms of a ‘blurring and merging of distinctions’ (Czarniawaska and Mazza 2003). It is produced by, and itself productive of, a fundamental shift in terms of the relationship between individual actors and formal organisations, with organisations now much looser, with many staff on temporary, short-term, and consultancy contracts measured in days; an increasing ‘revolving door’ in terms of exerts working for different organisations in a short time period; and, above all, in a separation of knowledge-based and bureaucratic-based imperatives.

One strand of this, pointed to in Janine Wedel's studies of Western aid to Eastern Europe (cf. Wedel 2000,2004), is the importance of multiplex networks (Wedel 2004: 165), where players know each other, and interact, in a variety of capacities, with multiple identities (which she terms ‘transidentities’), and in a variety of roles. Brokers, working in innovative spaces, creating new alliances, new meanings and forging new potentialities, are crucial to these networks. Their ‘enablement skills’ moves them beyond the formalistic and bureaucratic requirements into a space where ‘things get done’ because ‘they have the vision, the networks and practical implementation skills to take things a whole step further’ (Larner and Craig 2005: 416). In creatively engineered spaces, brokers are social entrepreneurs who empower, mentor and facilitate new communication channels, new associations, and generate institutional and situational reflexivities. At the very least, they have skills presenting ideas in the most graphically appealing form. At the same time, they operate in a political context, which unsettles and complicates their seemingly technical skills. Indeed, as Larner and Craig suggest, brokers often become professionalised and skilled in multiple intersected and intertwined institutional and organisational sites, and can become politically embedded and ‘governmentalised’ in far less progressive, transformatory, counter-hegemonic ways as:

Considering the key role played by consultants in projects of welfare reform in SEE, the various ‘technologies’, inscriptions, and non-human actants constitute a crucial element in the way consultants produce and re-produce knowledge, put forward knowledge claims, allow certain agendas to emerge, while actively silencing others and constructing (both discursively and practically) the subject positions of policy makers, politicians, professionals, ‘welfare subjects’ and communities. In this sense, ‘welfare reforms’ are neither simply rationally engineered processes, nor are they reducible to discursive structures. Rather, they are complex, multiple and fluid processes in which various kinds of intermediaries play increasingly important roles.

In contrast to realist studies of international actors and social policy, which see scales or, more usually, levels as unambiguous, there is a need to render space and scale contingent, complex and socially and politically constructed (Stubbs 2005), as an ‘outcome of the tensions that exist between structural forces and the practices of human agents’ (Marston 2000: 220). Rather than being taken-for-granted, a ‘politics of scale’ refers, in Brenner's terms, to ‘the production, reconfiguration or contestation of particular differentiations, orderings and hierarchies among geographical scales’ (Brenner 2001: 600). In this sense, trans-national policy is no longer about linear transfers from one place to another but, rather, concerned with ‘multi-scalar networks’ which ‘link local and trans-local processes, producing and consolidating social constructions of place’ (Jones et al.2004: 104).

Seeking to study the relationship between states, space and scale, Ferguson and Gupta suggest that both ‘supranational’ and ‘grassroots’ actors are engaged in similar practices of verticality and encompassment. Hence, they are critical of a perspective which treats the ‘global’ ‘as if it were simply a superordinate scalar level that encompasses nation-states just as nation states were conceptualized to encompass regions, towns, and villages’ (Ferguson and Gupta: 990). Hence, to quote Jeremy Gould:

What is important here is that social policy making is not taking place at different taken-for-granted levels of governance but, rather, that key policy players are transcending each level at any one moment. The ‘global is in the local’ and the ‘local in the global’ captures some of this, providing we re-assert a notion of power in terms of ‘uneven reach’, ‘differential intensity’ of places and spaces, and the differential ability to ‘jump scale’ (Moulaert and Jessop 2006). Essentially, we would assert that certain policy spaces open up, and others close down, in the encounters between international organisations and actors and national governments and that those who are better able to travel between these scales: consultants, INGO experts, think-tankers, policy entrepreneurs, and so on, are, often, better placed to influence policy. The precise forms this takes will, however, vary from context to context.

Hence, we are arguing here for an understanding of welfare and its transitions and transformations as complex, multiple and fluid processes of knowledge production, meaning-making, and claims-making. Instead of seeing cases as conforming and moving towards different (yet universal) types of ‘welfare regimes’, cast by separable ‘state’ and ‘non-state’ actors of ‘institutionalised politics’, the alternative perspective presented here privileges a view of welfare reforms as a set of interactive, intensive, and liminal processes. Concepts such as ‘assemblages’, ‘translation’, ‘intermediaries’ and ‘space and scale’ offer a critical entreé into the transformations of ‘the social’, of ‘policy’, and, above all, of ‘social policy’ forged in the encounter between South East Europe and diverse trans-national forces and actors. What follows is an, all too brief, attempt to utilise these concepts in the specifics of social policy reform in South East Europe, taking on board the ‘three challenges’ for ‘global social policy studies’: to see policy as ‘political’ less in terms of ‘grand narratives’ and more in terms of processes; to embrace ‘methodological reflexivity’ in ways which are sensitive to positionalities and standpoints; and to recognise power relations in terms of the voices which are enabled and the silences which are produced in the reform processes (Lendvai and Stubbs 2007a).

Just as, in the section above, taken for granted notions of ‘social policy’ and, indeed, of ‘international actors’ have been questioned, what is meant by South East Europe is by no means self-evident either. Regions are political constructs which create ‘imagined communities’ at levels beyond that of the nation state (cf. Anderson 1991). This notion of regions as ‘relatively malleable entities contingent on various social practices’ (Benchev 2006: 5) is important, not least in terms of the complex dynamic between notions of identity, nationhood and above all, the spectre of (Western) Europe and its ‘Other’. At its most acute in terms of the frozen notion of ‘the Balkans’ (Todorova 1997) and only slightly more nuanced in terms of the EU's construction of ‘the Western Balkans’, the region tends to be defined in negative terms and, in the 1990s, as essentially conflict-prone and underpinned by deep-rooted historical animosities. The difficulty of constructing an antithesis from within, in terms of ‘Balkan is beautiful’ (Rasza and Lindstrom 2004) or, more mildly, South East Europe for itself, is reinforced by the real and imagined uneven geo-politics of accession to the European Union, itself constructed in terms of modernity, as states seek to ‘join or rejoin Europe’.

The impacts of wars and transition on social policies in South East Europe have been documented elsewhere (cf. Deacon et al.1997: Chapter 7), and cannot be addressed in depth here. Nevertheless, the wars of the Yugoslav succession, beginning with the short conflict in Slovenia in 1991, spreading to Croatia from 1991 to 1995 and Bosnia-Herzegovina from 1992 to 1995 as well as the conflict in Kosovo and Serbia in 1999, and the instability in Macedonia in 2001, saw a complex and changing mix of international humanitarian and security interventions which directly affected processes of social and political change. The concept of ‘state-building’ is in danger of technicising the complex social and political engineering which is being attempted in parts of the region. The unfinished nature of this, not least in terms of processes within Bosnia-Herzegovina and the status of Kosovo, are also important factors which need to be addressed within a frame which links social policy with other discourses, notably security, refugee return, and democratisation. In a sense, elements of the de-territorialisation and de-institutionalisation of welfare, in terms of diasporas, remittances, cross-border claims and entitlements, ‘enclave welfare’, forced migration and return, the grey economy, informal welfare, the role of inter-household transfers, and ‘parallel power networks’ (Solioz 2005: 80) are all, albeit unevenly, relevant to, but often largely absent from, the study of social policy reform in the region.

There is a bewildering array of international actors and their representatives, some of whom wear more than one face, all directly or indirectly shaping the social policy of the region and, to an extent, competing with each other in terms of influence through projects, programmes and policy reform agendas. Indeed, in this context, new intermediaries and brokers emerge with major implications for transparency and ownership. It is still true that the World Bank and the EU are the major international actors influencing welfare reforms. However, in addition to the presence of the World Bank, the EU, and the UN agencies including the UNDP, ILO and UNICEF, the region is marked by a proliferation of actors, some of which are completely new and largely incomparable with any other bodies elsewhere, and all of which contribute, explicitly or implicitly, to a crowded arena of policy advice, project implementation, and strategic alliance-building in social policy. These include the Stability Pact for South East Europe, in the process of transforming into a Regional Co-operation Council, a vast array of International Non Governmental Organisations including those with a history of work in developing and post-conflict societies, and those which are more based on trans-European kinds of solidarities.

The competition-cooperation binary between the EU and the World Bank takes many shapes and forms. On the face of it, the two agencies offer very different ‘technologies of involvement’ and ‘technologies of enumeration’. From the point of view of social policy, the World Bank has a strong and often pervasive ‘structural adjustment’ framework, which addresses key social policy areas, such as pensions, social protection (seen as social assistance and social services), and health. On similar ‘core’ issues, the EU largely remains silent or plays a coordinating, but not a decisive role in those countries for whom membership remains a long-term goal. Hence, while the EU seems to be using ‘soft’ technologies such as supporting and monitoring the adoption of the regulatory framework of the acquis, the World Bank is relying on ‘hard conditionalities’ reinforced by loans. Thirdly, of course, the World Bank's insistence on the importance of absolute poverty lines stands in contrast to the emphasis on relative poverty by the EU, with the World Bank often co-opting local experts to produce ‘definitive’, and selectively editorialised, studies demonstrating that ‘the legitimacy and credibility of the Bank's expertise is drawn through a circular process between the knowledge it produces and the audiences that legitimise that knowledge’ (St. Clair 2006: 77).

Beyond the surface, there are important similarities between the two agencies. Firstly, often key international actors such as the World Bank and the EU pool from the same social policy ‘experts’ and rely on the same consultants. Secondly, we see similar practices of enframing in terms of ‘data’ and 'knowledge’ production throughout South Eastern Europe, whereby important studies, reports and databases are developed to be acted upon. In this regard, a careful analysis of key documents produced by the World Bank and the EU show textual similarities. In this sense World Bank discourse is moving from traditional structural adjustment towards more transitional programmes in order to support these countries in their integration to the European Union, at least rhetorically.

We now explore three cases of social policy reform in South East Europe, where, in each case, the role of policy translators and intermediaries operating in the new breed of flexible, hybrid, fluid, and less predictable organisations, networks, temporary coalitions, and informal policy networks has been important, albeit in different ways at different times. In each case, that of Bosnia-Herzegovina, Kosovo and Croatia, we sketch some of the actors, agencies and institutions framing social welfare reform, address the unexpected discursive chains or connections which have been made, and explore some of the displacements, silencings and contestations which are associated with the fragmentation and sub-contracting of welfare reform strategies.

Tracing social policy making in Bosnia-Herzegovina (BiH) is an immensely difficult process. The state is highly fragmented, essentially composed of two entities and a district, with one entity further divided into cantons or regions. It is a kind of semi-protectorate marked by large-scale, but also largely un-coordinated, international agency interventions. Overlapping ‘contact zones’ exist in which a range of international and domestic actors interact, with a wide range of flexible and parallel power networks operating. In a sense, donor effort on strengthening a kind of virtual central state including the creation of new, flexible, agencies at the central state level, offers a new platform or playground for these networks, and creates a privileged site for ‘technocratic’ depoliticised interactions between key politicians, civil servants and international actors (cf. Maglajlić Holiček and Rašidagić 2007; Stubbs 2007b).

Whilst a number of authors trace policy developments in Bosnia-Herzegovina in terms of the playing out of ‘liberal peace’ (Pugh 2001), ‘therapeutic governance’ (Pupavac 2001) and the gradual movement from the colonial power of the High Representative to the disciplinary power of the European Union (Chandler 2002,2003), these are examples of where ‘theory determines the evidence’, privileging linear essentialism over a nuanced understanding of the multiple, fractured and chaotic nature of policy making. The argument that the World Bank continues to dominate social policy making in BiH represents no more than a kind of literal truth insofar as Bank staff continue to press for ‘urgent’ reforms (cf. Maglajlić-Holiček and Rašidagić 2007). The space left by the absence of macro-level reform has been occupied by micro-level ‘pilots’, involving partnerships between a range of donors, international and local NGOs, local professionals, consultancy companies, academics, and politicians which operate on the border between the formal and the informal, the public and the private, and which serve to create uneven, contradictory and, above all, unsustainable localised practices and policy inputs. In the end, project resources tend to produce ‘false positives’, inducing a kind of ‘instrumental conformism’, through which Deputy Ministers, heads of public agencies, academics, and others, are paid honoraria to reproduce a new consensus on social policy. To quote from a recent study:

Ultimately, then, ‘success’ was judged less in terms of the implementation of this consensus and more in terms of the ability to organise conferences, produce weighty publications and, above all, run well-managed projects. Recently, a move from ‘projects’ to ‘strategies’ has occurred with donor support, particularly from the UK Government's Department for International Development (DFID), for the transformation of an initial World Bank driven ‘Poverty Reduction Strategy’ to resemble, more, an EU development strategy. Again, however, this kind of linear narrative, framed in terms of formal organisations, is deceptive, without reference to the creation of a central governmental unit, attached to the Council of Ministers, whose formal status is now clarified but whose informal power is considerably reduced following the departure of a key local policy entrepreneur. In the end, ‘the services one receives still largely depend on where one lives’ (Maglajlić-Holiček and Rašidagić 2007: 163) and welfare users are still, largely, silenced or rendered as objects rather than subjects (cf. Stubbs 2002: 324).

Few sites can have experienced more turbulence and changes in terms of ‘who governs’ and ‘how’ than Kosovo in the last 20 years. Following the curtailment of the province's autonomy by Serbia in 1989, the majority Albanian population began building a ‘state within a state’ based on the creation of parallel structures (Clark 2000). Social welfare for the Albanian community was, at the time, organised by a local non-governmental actor, the Mother Teresa Society. By 1998, in the face of continued pressure, a new military grouping, the KLA, emerged with considerable support from the diaspora community. Following the NATO campaign and the withdrawal of Serbian forces, UN Security Council Resolution 1244 created a protectorate under a UN Interim Administration Mission in Kosovo (UNMIK) and a NATO-led Kosovo Force (KFOR). Kosovo declared independence on 17 February 2008, recognised as of late June 2008 by 43 out of 192 UN member states and by 20 out of 27 EU member states. The new nation will allow NATO forces to remain, together with a new EU ‘Rule of Law’ mission.

Initial post-conflict social welfare programmes in Kosovo grew out of an international mandate for emergency humanitarian relief, an effort organised by international relief agencies such as Catholic Relief Services, Mercy Corps, and CARE International ‘whose strategic priority was initially emergency response and, subsequently, disengagement’ (Cocozzelli 2007: 203). From late 2000 onwards, an internationally driven social assistance programme and a World Bank-DFID social protection reform programme gave free rein to a wide range of international consultancy organisations and individual policy entrepreneurs, marginalising local actors including politicians, trades unions, local NGOs, and so on. Later, a new partnership between international donors, consultants and a new group of Kosovan public officials has sought to establish more sustainable social policy systems.

Despite elections in November 2001, and the establishment of Provisional Institutions of Self-Government in February 2002, local political processes still had little direct influence over the content of social policy. This was most clearly demonstrated in terms of a new pensions system drafted by a USAID funded consultant, the regulation governing which was rushed through before the formation of the PISG ‘apparently in order to ensure a workable scheme free of local political involvement and bargaining’ (Cocozzelli 2007: 213), with only details such as the date of introduction and the benefit rate left for the new Kosovo Assembly to decide.

A major strategic intervention was the Kosovo Social Protection Project, jointly financed by the World Bank and the United Kingdom's Department for International Development (DFID), which ran from July 2001 beyond its initial term, ending in August 2006. Each component and sub-component was subject to competitive tender leading to a project consisting of poorly co-ordinated, poorly sequenced, and differently conceived interventions, an over-emphasis on the one hand on largely irrelevant or highly academic training and study visits or, on the other hand, copies of European laws and regulations with little connection with on-the-ground realities and even less attention to the dynamics of implementation. International oversight focused, narrowly, on adherence to sub-contracting rules whilst mixed messages were given to an emerging cadre of Kosovan administrators and politicians, in terms of being asked to ‘steer’ the scheme but, also, being sanctioned for any behaviour which appeared to be over-politicised or self-interested.

Currently, the World Bank, since Kosovo is not a country and does not qualify for loans, is not involved in follow-up work. DFID, through a consultancy company previously involved in BiH, is focused on decentralisation, further technicising reform efforts. Whilst this certainly demonstrates the absence of ‘policy-making institutions that are representative of the Kosovo public’ (Cocozzelli 2007: 219), it is less that ‘the province is left to be buffeted by powerful global trends as articulated by international actors such as the World Bank, or donor governments such as the US or the UK’ (Cocozzelli 2007: 219), than the fact that all manner of diverse international-local interactions, divorced from a political sphere, continue to produce a chaotically engineered, yet fundamentally hybrid, social policy.

Since Croatian independence, the ground upon which social policy operates and has effects has become much more fragmented and contradictory. Successive waves of reforms, and the presence and influence of different kinds of international actors always, of course, working in implicit or explicit alliance with national forces, have produced a kind of uneven welfare patchwork. Rapidly changing modalities of international assistance have rubbed against the rapidly changing internal political and social landscape of Croatia, as the war and authoritarian nationalism were gradually replaced by democratisation and pursuit of EU membership. At first, a kind of ‘welfare parallelism’ emerged (Stubbs and Zrinščak 2007) with ‘implicit social policy’ dominating over any longer-term planning. Interestingly, the case of pensions reform (Stubbs and Zrinščak 2006) demonstrates how reforms could be achieved based on the establishment of policy coalitions of internal and external actors, making a supposedly ‘technical’ case at the height of political uncertainty, with a leading role played by reformers from Chile and World Bank staff seconded into the responsible Ministry.

Following the election of a Social Democratic Party-led coalition government in January 2000, the World Bank, together with the UK Government's DFID and the Government of Japan, worked with the Government on the negotiation of a Technical Assistance Credit, to be followed by a significant World Bank loan, to support the Government in developing a coherent and complete Social Welfare Reform Strategy. Teams of consultants were recruited and began work in April 2002. As perhaps one of the more dramatic examples of the problems of sub-contracting and the role of consultancy companies (cf. de la Porte and Deacon 2002; Stubbs 2003), no less than eight consultancy teams or companies were contracted to work on the reforms, all but one based on competitive tendering, covering social assistance; social services; labour and employment; fiscal issues and decentralisation; administrative strengthening, IT and database issues; poverty monitoring; as well as an overall team leader and a local resources team. Over the year of the work of the teams, levels of shared understanding of the task, and, indeed, trust between the teams, between international and local consultants, and between the teams and government, became quite low. In the end, the synthesis report was, in fact, written by the Fiscal and decentralisation team which was more experienced in working on USAID fiscal programmes questions in Central and Eastern Europe than on social welfare reform. The pinning of hopes for reform on an individual ‘change agent’, in this case an Assistant Minister plucked from academia, can be seen to have had unintended consequences, and may even have sharpened some resistance to change.

The HDZ-led government, elected in late November 2003, reorganised the structure of government so that social welfare now became part of the health-dominated new Ministry of Health and Social Welfare. The former Assistant Minister was promoted in the new Government to the position of State Secretary although, perversely, his influence weakened, in part because of the political nature of the position, filled according to the interests of a minor coalition partner. Importantly, in the aftermath of deaths in an institution for people with learning difficulties, the new Minister made it very clear that, if there were to be any World Bank loan in this sector, the largest part should be spent on repairs to essential infrastructure in institutional care facilities. In addition, the role of co-ordinating the reforms was given not to the State Secretary but to a leading HDZ MP, a medical doctor, who had established her own NGO providing institutional care for people with disability.

Under some internal criticism for the time spent from initial credit to loan signature, the World Bank put in place a number of new consultants, and supported a number of consultative seminars before, finally, a loan agreement was signed with support from the Swedish Government. The revamped project represents some continuity although, interestingly, it re-introduces the notion of pilots in three counties, all HDZ controlled, notwithstanding the fact that counties have, until now, played a very limited role in social welfare in Croatia. There is some attention to transformation of institutions, to innovation and to the establishment of new reference or referral centres although all remain somewhat vague and subject to diverse interpretations. A third wave of consultants, far removed from the original concepts of the reform, are now engaged to make the project meaningful.

At the same time, the preparation of Croatia's Joint Inclusion Memorandum prior to EU membership, which began in autumn 2005, signalled a more active interest and involvement in social policy by the European Commission. The influence of Europeanisation both in terms of harmonisation of social statistics and in consultation with stakeholders has already become visible, representing changes in ‘technologies of involvement’ (Haahr 2004) and in ‘technologies of enumeration’. The EU accession process creates new opportunities for intermediaries and consultants, with those who previously worked with the World Bank and/or UNDP now re-positioning themselves in relation to a new discourse.

In conclusion, one more binary opposition needs to be re-thought in order to make sense of the empirical data above. This is the division, so far rendered unproblematic in this text, between political and technocratic processes. Seeing both concepts more as metaphors than as accurate descriptors, also profoundly shaped by the confluence of national and trans-national processes, offers a new way of understanding sites of struggle, the nature of resistance, and the possibility of new alliances for change. We share John Clarke's concern that, too often, studies narrating a ‘compelling story of a totalizing system, strategy or force’ contain ‘a tagged-on last paragraph about resistance’ (Clarke 2004: 158). His concluding questions, albeit derived much more from a UK–USA–Western European trajectory, seem to us to be highly relevant in terms of beginning the urgent work of locating ‘the possibilities of refusals, resistances and alternative possibilities within the analysis’ (Clarke 2004: 158).

In all of the countries and territories discussed above, there is a healthy scepticism amongst professionals, policy makers and the informed public about ‘consultants’, renamed ‘insultants’, whose reports gather dust in drawers, and who have introduced a pseudo-language of ‘ubleha’ (Šaviha-Valha and Milanović-Blank, no date). In part, the task at hand is to raise this scepticism to the level of critical scholarship and action. Whilst this may appear to be too close to the ‘audit culture’ of technocratic new public management, a call for a register of consultants in which ratings of performance are systematically made (cf. Stubbs 2003) might offer one way of rendering consultancy more accountable if only, in a sense, to ‘weed the apples from the pears’ (Gould 2003). Beyond this, making explicit the difference between political and technocratic approaches would, also, be of use. In addition, the emphasis on ‘technologies of participation’ at the centre of trans-national processes offers an opportunity for other voices to be heard.

This paper has emphasised many of the inherent contradictions in reform processes: the need for explicit and never complete translations; the possibility of allowing divergent voices to be heard; the ‘accidental’ nature of many of the processes at hand and, above all, the possibility, always, that depoliticised processes will become repoliticised. This can occur through a re-insertion into formal domestic politics; through a new internationalism as, for example, European-wide social movements become involved as active subjects in debates; and through myriad small-scale acts of resistance or sheer recalcitrance. One manifestation of this is the explicit challenging of material inequalities which can be dramatic in trans-national contexts as, for example, a senior career grade civil servant in Kosovo who points out that his salary is 400 Euros per month whilst he signs big contracts with consultancy companies whose consultants are paid up to three times as much per day. Again, as outlined above, part of the task of critical scholarship is to discuss these imbalances in remuneration and to promote both local actors and, crucially, ‘localised’ knowledge and skills.

Elsewhere, one of us has argued that it is welfare users who are most marginalized in these processes (Stubbs 2002). Nevertheless, ‘technologies of participation’ do provide a space to ensure users’ voices are heard, not in a technocratised, projectised or fetishised way but, rather, through new processes of democratic accountability. Trans-national user and advocacy groups can play an important role here. In addition, networks of political party support (such as the German-based Friedrich Ebert Foundation) can help to build the capacity of Parliamentarians to scrutinise reform processes and to put forward social democratic alternatives.

This paper has not discussed the use of nationalistic concepts of ethnicity at the centre of emerging welfare settlements in each of the case study countries and territories. Of course, a trans-national ‘human rights’ discourse and practice is not always best placed to trace and undermine these. However, there are, in each country and territory, forms of localised everyday cosmopolitanism which reject ethnicised labels and which articulate an alternative common-sense. Throughout the region, memories of self-management socialism, whilst mixed, also provide a basis for an alternative conceptualisation of solidarity and for an alternative sense of the political.

Ultimately, acting locally, nationally, regionally and trans-nationally in the ‘spaces between … contending and overlapping (partial) authorities’ (Deacon 2007: 178) may, indeed, be where a ‘social reformist’ politics meets theories of complex, contested, welfare assemblages. The struggle for an alternative politics of welfare in South East Europe is difficult to envisage outside of a critical challenge to Northern and Western-based theoretical structures, political practices and policy prescriptions. Re-scaling of social policy merely reconfigures the terrain on which complex relationships between the political and the technical are played out. New intermediaries and translators are both a product of this shift and, potentially, a site of re-politicisation. Emerging regional spaces and networks both transnationalise local spaces and reshape the possibilities, in both form and content, for different kinds of ‘vocality’ (cf. Ferrera 2005: 179), expanding the possibilities for some actors and limiting the spaces for others. The nature of these new spaces of opportunity are very much a subject for further research. The need for action research emerging from the encounter between critical ethnography and critical pedagogy, whilst risking being seen as idealistic and/or relativistic, is the kind of praxis which this essay perhaps prefigures.

Noémi Lendvai is a lecturer in Comparative Social Policy at the University of Bristol, UK. Her main research interests are EU accession and enlargement, post-Communist social policy and the sociology of translation.

Paul Stubbs is a Senior research Fellow in the Institute of Economics, Zagreb, Croatia. His Main research interests are : international actors and social policy; new regionalism in South East Europe; policy translation; and computer-mediated activisms.