Multiple functions in a connected transnational space

Some research on EU politics and policy-making has addressed historical dimensions, if sometimes implicitly. Thus, starting with Haas (Reference Haas1958), neo-functionalist approaches have concentrated on analysing incremental change between treaty negotiations and revisions. Historical institutionalists such as Pierson (Reference Pierson2004) have emphasized the crucial importance of early decisions for creating path dependencies. More recently, several scholars of networks have also raised hitherto heavily neglected issues with strong historical connotations. Thus, Sørensen and Torfing (Reference Sørensen and Torfing2008: 303) have asked whether and in what ways the formation of governance networks is path dependent or how their formation and evolution over time has created such path-dependencies for policy solutions.

From a different theoretical position, Bevir and Rhodes (Reference Bevir and Rhodes2006; Reference Bevir and Rhodes2003) have argued for a ‘de-centred analysis’ of what Hertting (Reference Hertting, Sørensen and Torfing2008: 44) has called ‘the micro-mechanisms’ of the formation, nature and activities of policy networks. Drawing upon Geertz (Reference Geertz1973), Rhodes has proposed ethnological ‘thick descriptions’ of ‘people's beliefs and practices’ (2007: 1251) within networks which in this view are informed more by what these authors call ‘traditions’ and ‘story-telling’. If, however, actors' past individual and collective experience largely guides their network behaviour and decision-making, it would clearly be crucial to know more about how relevant ‘traditions’ have grown over time and under what conditions they have changed or are likely to change in the future. While Bevir and Rhodes (Reference Bevir, Rhodes, Sørensen and Torfing2008: 82) propose that change in networks occurs when actors face ‘dilemmas’ that unsettle ‘traditions’, the contingent response by policy network actors to any such past dilemmas has not been studied systematically either in a national or in the EU context.

The new interest in political science in the role of ideas as ‘broader outlines of policy’ (Daugbjerg and Pedersen Reference Daugbjerg and Pedersen2004) in the shaping of politics and policy-making also suggests taking historical dimensions more seriously. Discussing policy framing in the EU, Daviter (Reference Daviter2007: 655) has argued, for example, that ‘issue frames have been found to influence ensuing policy dynamics over the long run [my emphasis] to the extent that the specific representation and delineation of policy issues shapes the formation of substantive interests and at times restructures political constituencies’. This framing of issues, not just participation in problem-solving, is clearly one important function of networks. Linked to this, network actors can also play a crucial role in setting policy agendas (Princen, Reference Princen2007: 35). Even sceptics of treating policy networks as a new form of governance rather than one important dimension of it, readily admit that networks can play a crucial role in the ‘soft stage’ of the policy cycle (Börzel Reference Börzel and Tömmel2007; Reference Börzel and Schuppert2005).

Finally, drawing upon fundamental structural changes such as the alleged ‘hollowing-out’ of the state, scholars of policy networks have usually claimed that both networks and governance are phenomena that have grown, in Hirst's words (2000: 19), ‘on the ruins of the 1970s’. In line with the recent conceptualisation by Kohler-Koch and Finke (Reference Kohler-Koch and Finke2007: 209) of three ‘generations’ of ‘EU-society relations’, the origins of EU governance tend to be traced back to the 1980s, somewhat later than in the national context. However, as historians have pointed out and some political scientists have recognized (Gamble Reference Gamble and Pierre2000: 13), even in the hey-day of the nation-state and nationalism around 1900 policy areas like transport were in fact characterized – nationally and transnationally – by forms of informal cooperation and policy coordination between state and non-state actors akin to modern ‘governance’. As for the present day EU, moreover, it was never a centralised state nor did it have hierarchical nation-state ‘government’. It had different layers of governance and involved non-state actors in policy deliberation from the beginning. As a result, it would seem crucial to ascertain the precise nature of any change in EU governance from early integration through to the present day.

For a long time diplomatic and economic historians conceived of western European integration after 1945 as not much more than a process of multi-lateral bargaining of ‘national interests’ (Kaiser Reference Kaiser, Cini and Bourne2006; Gilbert Reference Gilbert2008). More recently, however, a new generation of contemporary historians have re-conceptualised early European integration as the slow formation of an incipient ‘trans- and supranational polity’ (Kaiser, Leucht and Rasmussen Reference Kaiser, Leucht and Rasmussen2009) and some have explicitly utilized the network approach as a heuristic device (Kaiser and Leucht Reference Kaiser and Leucht2008), for example for reconstructing the crucial role of Christian democratic party networks in ‘core Europe’ integration (Kaiser Reference Kaiser2007) or the impact of transatlantic expert networks on the anti-trust and institutional provisions of the treaty that created the European Coal and Steel Community (ECSC) in 1951–52 (Leucht Reference Leucht2008; Reference Leucht2006). Other scholars have analysed important dimensions of supranational politics and policy-making which are highly relevant to understanding the role of network-type relations for ECSC/EEC politics, such as the role of supranational institutions like the Commission (Ludlow Reference Ludlow2006) or of interest groups (Knudsen Reference Knudsen2009a) in shaping new supranational policies.

Historical research on networks in the EU emphasizes that their formation, governance and policy impact is highly contingent. It stresses multiple causes for explaining networks in EU governance, rephrasing (and overcoming) classic ontological dichotomies in the social sciences such as agent versus structure or ideational versus material factors and their role for motivating network actors as empirical questions to which only concrete historical answers can usefully be provided. At the same time, historical research on networks has highlighted their multiple functions. The greater interest in their various internal functions, or what Dakowska (this issue) calls the ‘dynamics of political configurations, resources and opportunities’, not only direct policy impact, has significant implications for research.

Alongside impulses from political science, historical research on the EU is also informed by the new transnational disciplinary agenda that conceptualises the history of modern Europe as a ‘connected’ space (Werner and Zimmermann Reference Werner and Zimmermann2006; Misa and Schot Reference Misa and Schot2005) with dense cross-border linkages. In an attempt to correct the excessive nation-state focus of European historiography, research in this vein is interested more generally in ‘circulatory regimes’, emphasizing the importance of ‘intertextual (reading, translation, quotation) and interactional (visits, correspondence, formal and informal organisations) communities’ (Saunier Reference Saunier2008: 174) and their role in processes of cultural and political transfer.

Historians also assume that most actors have multiple identities. Thus, referring to the Catholic Church and the Vatican as a particularly pertinent example, Patel (Reference Patel and Osterhammel2008: 72) has argued that a clear distinction between state and non-state actors depending on their formal institutional status often turns out to be ‘misleading’ in historical perspective. Policy network research in political science is predicated on this distinction, however: no non-state actors, no policy networks in governance. In fact, the alternative differentiation between public and private actors (Börzel and Heard-Lauréote, this issue) is even more rigid as it treats political parties as public, not non-state actors, so that only business and ‘civil society’ actors qualify as private actors. In the section on networks in EU governance it will become clear, however, that these multiple actor identities have larger implications for how we conceptualise the nature and impact of what Scharpf (Reference Scharpf1994: 38–40) first called the ‘shadow of hierarchy’ cast by state actors (Héritier and Lehmkuhl Reference Héritier and Lehmkuhl2008), and thus, for our understanding of how networks matter in EU politics.

Path-dependencies, brokers and multiple identities

Regarding the governance of networks, contemporary historical research has brought out four main points. Firstly, it demonstrates very clearly the crucial importance of early decisions on the structures and governance of networks for their long-term evolution. Long-term path dependencies concern in particular, the original criteria for inclusion and exclusion in networks, actors' core beliefs which help sustain them in the long run, and their main policy objectives. Thus, the largely informal political network of Christian democratic leaders in post-war western Europe carefully controlled access to the newly formed network, which was made dependent on strong support for Franco-German rapprochement, some form of supranational integration and the initial (self-) exclusion of Britain from ‘core Europe’ (Kaiser Reference Kaiser2007). Politicians sceptical about or antagonistic towards this concept like some French Left-Catholics and German conservative-protestant and ordo-liberal Christian democrats, were strategically excluded from these transnational contacts. The network took great care to embed its broadly federalist agenda in national party programmes and government policies as well as in transnational political programmes and access conditions for participation in formalized party cooperation, in the predecessors of the present-day European People's Party (EPP) formed in 1976. In fact, this political network's path dependency has been so strong that the EPP has been very resilient in protecting its federalist programmatic core across several EU and EPP enlargements to include more traditionally conservative parties from countries such as Spain.

Similar path dependencies appear to have governed policy networks and their role in the formation of early Community policies such as competition policy. Thus, many core features of EEC competition policy originated in the anti-trust provisions of the ECSC treaty which were shaped to a large degree by transatlantic expert networks (Leucht Reference Leucht2008; Leucht and Seidel Reference Leucht and Seidel2008). Similarly, as Ramirez (Reference Ramírez Pérez2009) has shown, the network that eventually created the European Round Table of Industrialists (ERT) in 1983, actually formed in the early 1970s as a result of contacts forged between European car manufacturers and some Commissioners and leading officials like Robert Toulemon, to prevent the adoption of costly US car safety regulations and to promote a Community structural policy. Thus, the European producers excluded the European arms of Ford and General Motors from the network. This in turn provided a crucial impetus for the eventual formation of the ERT as an organisation only of CEOs from multi-national companies with headquarters in Europe, contributing to the Europeanisation of transnational business networks. Although the network failed concerning its initial objectives, its structures and governance were in place long before it developed any significant policy impact.

Secondly, competition policy highlights the great influence of the Commission on the formation and maintenance of policy networks, a specificity of EU governance. The initial choice of consultants by the German Commissioner Hans von der Groeben was crucial for the ideological orientation of the competition policy network (Seidel, Reference Seidel2008; Hambloch, Reference Hambloch2007). Similarly, the Commission brokered the formation of a European-level policy network in agriculture (Knudsen Reference Knudsen, Gehler, Kaiser and Leucht2009b), first encouraging the setting-up of the supranational farmers' organisation COPA in 1958 and then inducing in some cases reticent national farmers' organisations to engage more pro-actively in EEC level policy-making instead of relying heavily or even exclusively on the national route of lobbying. Especially through its committee structures, the Commission was able from the beginning to offer relevant resources to non-state actors to engage in supranational policy-making. Historical research on the formation of the Common Agricultural Policy (CAP) has shown (Knudsen Reference Knudsen2009a; Ludlow Reference Ludlow2006) that the emerging transnational agricultural policy network was a crucial asset for the Commission in drafting legislation in line with its own institutional preferences and geared towards striking workable compromises in the Council of Ministers, where most of its proposals were in fact adopted without much controversy.

Networks including non-state actors have enhanced the Commission's technical competence, the efficiency of its policy-making and output legitimacy (Kohler Koch and Finke Reference Kohler-Koch and Finke2007). Alongside functional advantages, fostering policy networks in the early stages of the EEC was also in the Commission's institutional self-interest of maximising its influence, as networks provided it greater legitimacy vis-à-vis member-state governments. It appears that the Commission's brokerage role in network formation and governance has been particularly strong in the process of transfer of policy areas from the national to the EU level, especially where few national models existed.

Thirdly, historical research on networks in EU governance also suggests the need to ‘bring people back’ into the study of networks (Kaiser Reference Kaiser, Phinnemore and Warleigh2009, see also Rhodes Reference Rhodes2002). While older personalized Whig histories of European integration, which ascribed Jean Monnet a decisive role as ‘founding father’ of the ECSC, have been refuted, the leading French official and first president of the ECSC High Authority did play a crucial role in networks of officials, businessmen and legal experts, and in bringing these networks together at the Schuman Plan conference to guarantee a successful outcome of the inter-state negotiations (Leucht Reference Leucht2008). Similarly, Commissioners Mansholt and von der Groeben invested substantial personal resources in building and sustaining transnational networks in the policy fields for which they were responsible from 1958.

Interestingly, second-order politicians and private citizens with excellent personal networks sometimes played a crucial role in creating and maintaining networks. Thus, the informal network of Christian democratic leaders, who met secretly in Geneva from 1947 onwards, was managed by two private citizens, Victor Koutzine and Johann Jakob Kindt-Kiefer, who acted as interpreter, provided cover to disguise the secret meetings and also funding to hold them (Kaiser Reference Kaiser2007; Gehler and Kaiser Reference Gehler and Kaiser2004). Koutzine was close to the last leader of the French Resistance and French foreign minister until 1948, Georges Bidault, and Kindt-Kiefer had close contacts with Konrad Adenauer, the leader of the German Christian Democratic Union (CDU) and first chancellor of the Federal Republic from 1949 to 1963. Another example is Alfred Mozer, who was international secretary of the Dutch Labour Party before he joined the personal cabinet of Mansholt in the Commission (Knudsen Reference Knudsen, Gehler, Kaiser and Leucht2009b). Mozer, who had fled from national-socialist Germany in 1933, was no agricultural expert, but he played a crucial role in maintaining informal transnational party networks to guarantee sufficient support for the emerging CAP in national governments, parliaments, and the European Parliament (EP).

For individual network brokers, their transnational connections and inter-cultural competence were crucial assets for networked politics. The structural conditions for their entrepreneurial leadership in political and policy networks were probably better up to the first enlargement of 1973 due to the much smaller size and greater political and cultural cohesion of the ECSC/EEC compared to the EU of 27 member-states.

Fourthly, individuals have had multiple and partly overlapping identities. To conceive of them as institutional state/non-state or public/private actors may be of some use as a category descriptor, but not as an analysis filter, as these categories cannot adequately capture the social complexities of EU governance. Thus, as CDU party leader, Adenauer played a leading role in the informal Geneva Circle and, as chancellor, in intergovernmental bargaining. Both forums – the party network and inter-state diplomacy – were characterized by different logics, languages and behaviour. Crucially, such multiple actor identities were, and still are, by no means limited to politicians. As Vauchez (Reference Vauchez and Petersen2008a) and Cohen (Reference Cohen2007) have shown for the example of the European law network, which has been so crucial for supranational integration over the decades, the vast majority of individual actors switched and accumulated roles. The professional mobility of these network actors was very high. At various stages they were lawyers, judges, professors at universities, consultants or, in some cases judges in the European Court of Justice (ECJ). These European lawyers were individuals whose changing beliefs, ideas and moods potentially affected the governance of networks and their impact. However, they were never willing or able to free themselves from the constraints of the larger network, or what Vauchez (Reference Vauchez and Petersen2008a,Reference Vauchezb) calls (with reference to Bourdieu) the European legal field.

Differentiated polity and shadow of hierarchy

Concerning the impact of networks on EU governance in long-term historical perspective networks firstly have played a crucial role in structuring the political integration space of western Europe. Crucially, not all networks have been geared towards fostering supranational integration and policy-making. In their study of network relations in the Council of European Industrial Federations (CEIF) created in 1949, Rollings and Kipping (Reference Rollings and Kipping2008) have confirmed that business interests (especially the steel producers) were divided and ineffective in influencing the ECSC treaty. These authors have also demonstrated, however, that the CEIF was an important platform for intensified network links especially between British, Nordic and Swiss business interests. In the Spring of 1958, the business initiative to create a small free trade area in the absence of a larger solution also covering the EEC countries originated in the CEIF. Subsequently, this transnational business network played a crucial role in inducing partly reluctant governments, especially the British, to proceed with the creation of the European Free Trade Association in 1959–60.

Other transnational networks with traditions reaching back to the nineteenth century played a key role in preventing the early supranationalisation of policy-making in a geographically confined ‘core Europe’. Thus, as Schot (Reference Schot, Badenoch and Fickers2009) has demonstrated for the transport sector, the existing expert network strongly preferred all-European and international networking and regulation. Despite the early proposal for sectoral ‘core Europe’ integration to be expanded to transport and its later inclusion in the EEC treaty, this sectoral expert network blocked any move towards a meaningful supranational EEC transport policy.

Secondly, historical research also suggests reconnecting more narrowly focussed public policy network research with processes of constitutional and institutional reform and EU enlargement. For example, Christian democratic party networks decisively influenced the formation of ‘core Europe’ without Britain which in turn was a precondition for the drafting of the supranational CAP in the 1960s. Social democratic party networks played an important role in the transition in Spain and Portugal in the 1970s (Ortuño Anaya Reference Ortuño Anaya2002). This in turn facilitated these countries' integration in the EC, enhanced their Europeanization beyond the formal adoption of the acquis communautaire, and impacted on the EC's structural and regional policies. Thus, networks comprising non-state actors have reinforced polity-building efforts of supranational institutions such as the Commission, the EP and the ECJ, with significant repercussions for policy-making.

Thirdly, networks have important policy-framing and agenda-setting functions that should also be taken seriously, as Berghahn (Reference Berghahn, Gehler, Kaiser and Leucht2009) has argued with respect to the role of (transatlantic) networks of intellectuals in the early Cold War confrontation and debate about the desirable forms and functions of European integration. Framing discourses in decisive ways can define what is seen as politically legitimate, constrain state actors and thus, indirectly influence policy outcomes. The rhetorical framing of the objectives of the CAP, which largely adopted well established national legitimising concepts like the desirability of self-sufficiency and of the retention of ‘healthy’ rural social structures and lifestyles, helped support a wasteful policy at the expense of weakly organized consumer interests. The agricultural policy network acted as a ‘strategic frame manipulator’ (Daviter Reference Daviter2007: 656). Such networks have been especially important for agenda-setting at EU level because they are very well equipped for developing policy ideas.

Historical research suggests fourthly that the network impact on politics and policy-making cannot easily be categorized or measured. Börzel and Heard-Lauréote (this issue) argue that we should only speak of ‘network governance’, i.e. networks as a new form of governance, if policy networks do not just ‘prepare’, but ‘determine’ policy. To differentiate between ‘preparing’ and ‘determining’ policies does not always capture the political realities of decision-making in EU governance, however, especially concerning the bridging role of political networks. Thus, the network of Christian democratic leaders clearly ‘determined’ the shape of the ECSC by ‘preparing’ the Schuman Plan, which combined a supranational design and sector integration, and provoked the anticipated and desired self-exclusion of Britain by its Labour government.

Moreover, the multiple actor identities characterized above as especially typical of EU governance are key to understanding where state actors cast what kind of ‘shadow’ delineating the scope and influence of networks. Interestingly, historical research suggests that networks in the EU often succeed in acting ‘through’ supranational public actors, so that they themselves influence where the ‘shadow’ falls and how strong or light it is. In one particularly pertinent example (Rasmussen, Reference Rasmussen, Poiares Maduro and Azoulai2009), the first path-breaking integrationist ECJ decision on the direct applicability of Community law in 1963 was actually taken by the narrowest of margins (4:3) with the casting vote coming from Robert Lecourt, who had been appointed in the previous year by the radically ‘intergovernmentalist’ French President Charles de Gaulle. His appointment probably resulted from a domestic political deal with the Christian democratic Mouvement Républicain Populaire (MRP), a coalition government partner with the Gaullists for a short time in April 1962. The former parliamentary party leader of the strongly federalist MRP, Lecourt was active in the transnational Christian democratic party network and as a committed federalist and constitutional lawyer, in the European law network. In this perspective, the ECJ's 1963 judgement looks more like a decisive victory, for broadly federalist political party and European law networks in a highly politicised battle over the future of integration than the ‘casting’ of a ‘shadow’ by a public supranational actor.

Change over time: from hierarchy to networks?

Concerning change over time in forms of governance in Europe, just how hierarchical was the national welfare state as an actor in the politics of integration after 1945? Some political scientists have called into question the assumption that a radical change from hierarchy to networks must have occurred at some point after the 1970s. Thus, Michael Keating (Reference Keating2008: 75–6) has provocatively argued in relation to the British policy network literature that the ‘network governance’ thesis depends on a ‘stylised account’ of ‘a fictional world before governance in which there was a unitary, centralised state autonomous of social interests’. For a federal EU member-state such as Germany such a narrative would make no sense. However, even previously highly centralised nation-states prove to have been highly fragmented actors in ECSC/EEC politics. Political parties, coalition governments, national bureaucracies and interest groups were deeply split on major strategic questions of integration, and they searched for allies for their preferences not only among other national actors, but also in transnational networks and alliances and in the new supranational institutions.

Within the emerging increasingly pluralistic EEC polity even the governments of Fifth Republic France had great difficulties to define clear preferences and to project influence, as the close cooperation of liberal-minded officials from the economic ministries in Paris with national ministries in other member-states and the Commission shows (Warlouzet Reference Warlouzet2007). From the beginning, supranational institutions gave experts from universities, societal organisations and interest groups access to deliberative processes. The manifold examples of network-type informal coordination including non-state actors from before the first enlargement of 1973 demonstrate that the EU never saw an abrupt change from hierarchical government to governance in networks. It would also be inappropriate to imagine any change over time as the linear growth of transnational contacts and cooperation in networks and their ever greater influence on decision-making ultimately resulting in the emergence of a new form of governance.

How then could we conceptualise change over time in networks in the EU? Although not targeted at EU governance, Bevir and Rhodes (Reference Bevir, Rhodes, Sørensen and Torfing2008) have argued that political ‘dilemmas’ have been catalysts with the potential to initiate such change in the constitution and dynamics of networks without necessarily impacting directly on their influence on policy-making. Dilemmas call into question established forms and activities of networks, governing traditions and solutions for particular policy sectors. We hypothesize that four types of dilemmas are of critical importance, to instigate the transformation of existing and the creation of new networks and phases of intensive ‘networked politics’. The first concerns constitutional-institutional change. Thus, the setting-up of the ECSC and of the EEC created an entirely novel supranational order in which the new institutions and transnational societal actors had to carve out a role for themselves. In itself, this was a great incentive to activate existing networks or to create others with a view to fulfilling the new legislative tasks. In fact, for some crucial policy sectors like agriculture and competition, these tasks were only very vaguely defined in the treaties. Subsequently, the greatest constitutional reform took place in the Maastricht Treaty of 1992. From a network perspective, this reform contributed greatly to upgrading the role of the EP as an actor in networks and a locus of processes of deliberation and negotiation with civil society groups.

Secondly, EU enlargements also create major dilemmas, especially when they appear to threaten either the essentially federalist political objectives of ‘core Europe’ or the EU's internal political and economic balance as a result of a new weighting of votes in the Council or significant budgetary transfers and reforms. Thus, the accession of Britain provoked a strengthening of ‘core Europe’ policy networks to prevent, so to speak, the meltdown of the integration ‘core’. Similarly, Henning's analysis in this issue of changes in agricultural interest group networks before and after the eastern enlargements of 2004–07 suggests that supranational deepening of networks occurred in response to fears over the dilution and reform of the CAP.

Thirdly, socio-economic crises create dilemmas which lead to increased expectations concerning the EU's problem-solving capacity, requiring greater member-state coordination and instigating the Europeanisation of policy fields. This applies especially to the time following the first oil crisis of 1973–4, which was characterized by low growth, increasing unemployment and spiralling budget and state deficits. Until then, most business interest groups and companies had been broadly supportive of market integration. They had not invested many resources in supranational political networking or lobbying, not least in light of the fact that during the 1960s, the Commission was loath to be seen as dependent on business interests. The economic and energy crisis, the increasing distribution conflicts and growing concerns about the future competitiveness of European industry in the light of US technological superiority and Japanese productivity gains induced companies to extend their networking activities. They also increasingly engaged in public debates over EC reform, culminating in the formal creation of the ERT and its Single Market initiative in the early 1980s.

The fourth dilemma, which is especially relevant for the constitution and transformation of networks, results from legitimacy crises in integration. We hypothesize that such crises induce the supranational institutions, especially the Commission, to invest greater resources into activating and including societal actors in informal consultation mechanisms and transnational networks, not least in the hope of enhancing input legitimacy (see also Börzel and Heard-Lauréote, this issue). The 1970s were still characterised by a fairly linear growth of the ‘permissive consensus’ (Lindberg and Scheingold Reference Lindberg and Scheingold1970: 249), i.e. increasing general support among the citizens of the new member-states for integration, combined with lack of interest in concrete European policy issues. However, as Down and Wilson (Reference Down and Wilson2008) have rightly pointed out, the 1970s more than the 1990s, was also a period of extreme polarisation in integration with an aggressively articulated minority opposition to predominately economic integration. In those days the student movement and left-wing protest movements saw the EC as the supranational embodiment of capitalist ‘exploitation’. The new social movements called into question established forms of politics not only at the national, but also at the European and global levels. The Commission tried to contain such discontent by assisting new social movements to develop and formalise transnational contacts and by including them in the Europeanisation of new policy fields like environmental policy after the threat of a collapse of the global eco-system was so poignantly exaggerated in the report by the Club of Rome in 1972.

Thus, it appears that the EU underwent an intensification of network contacts and networked politics when a combination of constitutional-institutional reforms, critical enlargements and socio-economic crises radically called into question the existing formal and informal institutional arrangements. This was especially the case in the early formative phase of EEC integration between 1958 and approximately 1963, again between the first enlargement and the oil crisis in 1973–4 and the early 1980s and once more after 1992. Moreover, legitimacy crises especially after 1973 and again from the mid-1990s led to a greater plurality of actors in networks, especially the activation on a large scale of business actors in the 1970s and ‘civil society’ actors in the 1990s. In contrast, the rapid quantitative growth of interest group representation in Brussels in the context of the Single Market programme in the late 1980s did not mark a new phase of networks in governance, but amounted to a transformation in lobbying practices (Greenwood Reference Greenwood2007) that became more geared towards the supranational level. From this perspective, the 1970s, which have traditionally been portrayed as the ‘dark ages’ of European integration (Keohane and Hoffmann Reference Keohane and Hoffmann1991: 8; Bache and George Reference Bache and George2006: 138), were actually a crucial founding period for networks.