A language that takes a traditional concept of a nation-state as a legitimate and hierarchically organised authority over a people and a given territory with the monopoly over the use of force for granted, misses much of the current reality of modern European nation-states and of the European Union (EU) itself. At least, we need to distinguish between formal and material sovereignty, the latter being defined in degrees of the capacity for autonomous action. As to the former, sovereignty is already divided, as well as shared, to a large extent between EU authorities and the Member States. As to material sovereignty, neither the EU nor the modern welfare states enjoy the capacity for autonomous action of a 19th Century nation-state.
If we use state-centric concepts to describe the realities of modern welfare states, we encounter conceptual problems because such language implicitly maintains two distinctions which are increasingly problematic: first, the distinction between `state' and `society,' and, second, between the `domestic' and the `international' orders. As to the first distinction, it is increasingly difficult to describe the modern nation-state as one in which governing functions remain the exclusive property of state officials or governments. In comparative policy analysis, scholars increasingly talk about the `co-operative state,' the `negotiating state,' the `co-operative administration,' or `policy networks'.6 These and other terms imply the fact that governing functions are increasingly taken over by negotiating networks encompassing governments (national, sub-national, and local) as well as private actors (firms, interest groups, etc.) and representatives of civil society (such as non-governmental organisations [NGOs]). Modern welfare states look increasingly less like hierarchical structures of legitimate authority, and more like multi-level bargaining and negotiating networks in which public actors are not obsolete, but can only fulfil their functions by co-operating with private actors and/or groups. This is even true for the quintessential European nation-state, France. The authority of the French centralised state is balanced by dense formal and informal networks linking local, regional, and central-state authorities with private actors at the various levels of governance.
As to the second distinction between the `domestic' and the `international' realms, international relations scholars tend to emphasise that the traditional division between domestic and foreign affairs is obsolete in an age of globalisation and Europeanisation. While Joschka Fischer remains the German foreign minister, his colleagues in the finance and economics ministries are as much involved in external relations as he himself is. The distinction between the `domestic' realm characterised by hierarchy and legitimate order, on the one hand, and the `international' realm of anarchy in the absence of a world government, on the other, is less and less useful as an instrument to describe, let alone explain, the current international-domestic order. Moreover, authority in such `intermestic' systems is increasingly organised along functional lines. The World Trade Organisation, for example, represents such a functional organisation regulating the international economic order.
As a result, international relations and comparative politics scholars tend to use the term `governance' to describe the current reality of political life both inside and beyond the nation-states. It accounts for the finding that national states (governments) have lost their exclusive authority in the policy-making process, sharing it with international and supranational institutions on the one hand, and with private actors, such as multinational firms and representatives of civil society, on the other. The regulation authority of international institutions and accompanying networks may vary across issue-areas. Yet, material sovereignty no longer resides in the nation-state, but is divided and shared between multiple levels of governance.
The EU can then be described as a peculiar multi-level system of governance7 which, first, not only encompasses national governments and supranational institutions such as the Commission, the European Parliament, the European Court of Justice, and the European Central Bank, but also transnational interest groups and other private actors in governance networks of varying density and scope. Second, the EU encompasses a variety of functional regimes with different scopes and depths as far as the nature of the regulations are concerned-from EMU to the Common Agricultural Policy to environmental regulations, and the Social Protocol. Third, the EU is multi-layered in the sense that supranational, national, and sub-national authorities interact regularly in these networks. Clearly, the various treaties constituting the European Union regulate the ultimate decision authority of the various layers. But what does it mean that the Commission has the right of initiative, if not one single initiative is taken behind closed doors in Brussels, but usually only after extensive consultations with state officials in the various Member States and with interest group representatives both in Brussels and in the national capitals? What does it mean that the European Council representing the Member States negotiates changes in the EU treaties when these changes have been prepared and talked about in intergovernmental conferences, which again include national and supranational actors and consult regularly with private actors and interest groups? In other words, how much do we analytically capture the EU's daily life if we focus exclusively on formal powers and legislative structures?
The conceptualisation of the EU as a multi-level structure of governance is widely accepted. The term has recently even entered the EU vocabulary through Commission President Romano Prodi. But how much mileage do we gain when talking about the future European political order? Does this language help us when moving from description and analysis to prescription and visions about the `finalité politique' of the European order? First, we avoid the language of statehood when talking about the future of Europe and, thus, are no longer wedded to the false alternative of either a `Europe of nation-states' or a European federal state. To the extent that Joschka Fischer also struggles to avoid this alternative, political scientists might then offer him different concepts to express his vision. Second, if we conceptualise both the EU and its Member States as multi-level structures of governance, our understanding of what constitutes a `state' changes. It is no longer wedded to hierarchical structure of legitimate authority, but we can now speak more easily of `divided sovereignty' as a central concept in Fischer's speech. Third, however, the notion of `multi-level structure of governance' which focuses on the structure of material sovereignty or action capacity, does not easily translate into a constitutional language, which should delineate who is in charge of what and when, and thus, should define structures of formal sovereignty.
The constitutional language of federalism and of federalist orders allows the dividing and sharing of sovereignty in a multi-level system of governance to be discussed. Clearly, we are not referring to the notion of federalism as described in the `Federalist Papers' and which is usually associated with the creation of a European federal state, where sovereignty would be fully transferred to the European Union and the Member States would lose their state quality. Nor do we mean a federal union or a confederacy, in which sovereignty exclusively resides in the Member States and is only pooled at European level. Most versions of federalism are about divided and shared sovereignty. We will now discuss various federalist orders and how they can be used to construct a European federation.
6 See, for example, Benz et al (1992); Rüdiger (1995); Czada et al (1993); Börzel (1998).
7 See, for example, Marks et al (1996); Jachtenfuchs & Kohler-Koch (1996); Kohler-Koch & Eising (1999).