The lack of a European demos mentioned above should have made clear that full-fledged input legitimacy is unrealistic at the European level. But whilst the EU will increase input problems, at the same time it will decrease problems of output. Effectiveness will have to come at a democratic cost. This is very much in line with the democratic dilemma pointed out by Dahl - system effectiveness beyond the nation-sate runs counter to citizen participation.
But Dahl also argues that we are witnessing a third transformation of democracy from territorial states to multi-level systems. The implication is that just as democracy on the scale of national states required new patterns of political institutions radically different from those of the small scale city-states, democracy on a transnational scale will similarly require new sets of institutions sometimes radically different from those of the familiar political institutions in modern representative democracies.
(Dahl, 1994, 27) Dahl's argument appears inherently convincing and could help explain why filling the democratic gap with elements found at the national level is a wrong approach. Since the European Union is not a state, its institutions ought not to be assessed on the basis of those found in the member states. Merely transferring those institutions from the national to the European level would not make the EU more democratic or legitimate. One should rather look if a new form of democracy could come to exist and how this form of democracy could operate differently from that within the nation-states.
As the notion of input legitimacy is under direct constrain from external economic and institutional factors, this new form of democracy would inevitably have to depend on output legitimacy. Policy choices that would be domestically popular must be avoided due to European law or because they could affect negatively the international competitiveness of the nation. As a result, the role of popular preferences and demands will be rendered more insignificant and expertise and specialized knowledge will increase, especially if solutions are achieved though interstate negotiations.
In Scharpf's words, it means that "popular approval and popular demand are becoming less and less sufficient for assuring, or even for justifying, corresponding policy choices. As a consequence, input-oriented legitimating arguments will become less plausible, and government at the national level must increasingly depend on output-oriented legitimation arguments alone. " (Scharpf, 1998) Policies guided solely towards input legitimacy in a Europe without common identity could actually signify de-legitimization of its policy.
Unlike input legitimacy, output legitimacy only requires the perception of common interests but no common identity. Choices will still be legitimate because they effectively promote the common welfare of the constituency by giving solutions to problems that can only be achieved through collectively binding decisions. Furthermore, there exist several mechanisms with which legitimacy in the Union can be sustained. For instance, Majone argues that pareto-improving policies should be left the regulatory agencies in Europe, and only redistributive policies would need democratic legitimation.
Although these independent agencies seem to violate democratic principle since they are not subject to majoritarian control, the American model showed that agency independence and public accountability should be viewed as complementary and mutually reinforcing rather than antithetical values. These agencies will be legitimized by the results that they achieve. Factors such as professionalism, statutory goals, procedural requirements, and judicial review are all elements of a complex system of immanent control more important than majoritarian rule that ensure legitimacy and accountability.
(Majone, 1996) All these arguments converge at one point: European governance will have to rely more on efficiency due to the same effects that partially caused European integration, namely globalization. The democratic quality of the EU will now have to be assessed more in terms of the outcomes that the EU produces instead of its democratic accountability. Input Legitimacy Revisited The implications of the previous section should under no means imply that input legitimacy should or could be totally neglected.
There is still a need to provide citizens with a sense of control, voice, and participation in specific policy contexts. Kohler-Koch has pointed out that "legitimacy is a matter of performance. The focus of governments and the Commission is clearly on output legitimacy: efficiency comes first. But the lessons of Maastricht (governments facing an obstinate public, reluctant to follow or even turning down what had been agreed upon at the European level) have been learned too.
" (Kohler-Koch, 1999) Democracy will be needed in order to make justifiable decisions when collective choices will interfere with or interrupt established interests and identity-based concerns. According to Jachtenfuchs, there is a normative and a pragmatic reason for democracy. The argument is that by taking collectively binding decisions which intervene into the lives of people, these people must have some voice in the system and be able to change the course of events.
At the same time, for the system itself it would be impossible to govern against strong opposition from the part of the citizens, so democracy would increase system stability and problem-solving efficiency. (Jachtenfuchs, 1997, 6) In either case, therefore, there will still have to be a combination of output and input legitimacy. The aim should hence be to determine the optimal level of democracy, meaning how much and when democratic means are inherently required for decision-making. In light of what has been mentioned above, plausible solutions will have to address two key issues.
First, it is imperative that the European people come to understand and accept what has been the argument of this paper so far, namely that output legitimacy is the only mode to deal with globalization in the absence of a demos. Second, the European people must become more involved in the decision-making process since their role at the moment is rather insignificant. Both issues could easily be solved by the same mechanism - public discourse. These public discussions could perform a guiding and legitimating role if they communicate the wider framework of current controversies and the associated political implications.
They would similarly enhance issue-awareness necessary for approving or rejecting certain policy choices, under which "political support or opposition can indeed claim, and bestow, a maximum of democratic legitimacy. " (Scharpf, 1998) Public discourses could also ensure that citizens have a say on the European agenda, implying what important problems would be discussed next. Yet, two conditions seem necessary for the establishment of such public discourses. First: a systematic reform in the process of policy-formation to increase opportunities for public debate and increase the representative character of the democratic structures.
Second: an improvement of the information reaching the public through better use of available opportunities such the media and information technologies, making European policy more transparent and inevitably more relevant to the citizens. But as long as these "issues remain untouched, civil society will continue to feel disengaged - indeed disenfranchised - from European action. " (Lebessis and Paterson, 1999, 17) To sum up, a system where people have a dialogue and a sense of involvement is the most plausible way of reducing the democratic deficit without substantially reducing system effectiveness.
This proposal is reflected in the deliberative notion of democracy, which may be defined as collective decision-making by those who will be affected by the decision or their representatives "by means of arguments offered by and to participants who are committed to the values of rationality and impartiality. " (Goldmann, 2001, 151) The essential idea of deliberative democracy is that policy is made by means of consensus-seeking on the basis of free and rational debate and that the public sphere is not only open to powerful organizations such as political parties.
The quality of decisions taken will then depend on the quality of the political deliberation, implying rational discourse using well-developed arguments. Many would question how such a deliberative system could function since a European public sphere is partly missing. However, several authors have commented that creation of a public sphere is not an intangible idea, for instance, due to the emergence of "new European audio-visual spaces - newspapers, television, internet, and English maybe as a bound to be first language - and the new social movements, NGOs and identity politics across borders.
"(Eriksen, 2000, 55) Especially the Internet provides several possibilities since network media might facilitate community-building and mobilization for civic engagement, and electronic discussion fora could help develop and clarify arguments. Conclusion The democratic deficit in the European Union is not a mere postulation, it is indeed present. However, I have argued that since Europe is characterized by the absence of a demos, "filling" this gap completely would be impossible and also undesirable.
But the democratic deficit should not be seen as the end of the legitimacy of European governance. A new form of democracy must come to be understood, one deriving its legitimacy primarily from efficiency rather than from democratic participation. This will be intrinsically significant for dealing with the several challenges posed by globalization. Nonetheless, if the European Union does not want to completely alienate itself from the 360 million citizens it governs, some degree of public participation must be ensured. I have argued that enhancing public discourse is the most plausible solution.
This proposal would not intend to completely reform the European institutions, but make European policy more transparent and relevant to the citizenry, and incorporate it in the decision-making process. If this can be achieved, European governance could then for the first time truly be considered effective and legitimate.
Bellamy and Castiglione (2000) 'The uses of democracy: reflections on the European democratic deficit,' in Eriksen and Fossum (eds), Democracy in the European Union. London/New York: Routledge.