Gaustadalléen 30 (map)
ARENA Working Papers
Reforming European Institutions of Governance
Johan P. Olsen
Comprehensive institutional reform
and the search of political order
Institutional reform has -again- become a salient issue on the political agenda of the European Union. The allegation is that the EU institutions of governance have major weaknesses, that they lack effectiveness and legitimacy, and that comprehensive reform is needed. The claim includes single institutions, as well as the relationships and balance among institutions. Therefore, the basic principles and rules for constituting, distributing, controlling and legitimizing power are involved.
My focus is on comprehensive reform of the European political order and the prospects for a deliberate rearrangement of the relations among key institutions. What can students of political institutions contribute to a better understanding of institutional reform in a complex, dynamic, multi-cultural and pluralistic setting like the EU? How useful are existing theoretical ideas about institutional dynamics, that is, how political orders are established, maintained, changed and abandoned? What do these ideas tell us about the processes through which a new European order is being established, to what degree institutional engineering is feasible, and what can be achieved through deliberate reform?
There is no generally agreed upon theory of institutional dynamics that explains how and when institutions of governance change, and with what implications. Neither is there agreement on the significance of deliberate intervention in processes of change. There are different ways of understanding reform, based on competing conceptions of political institutions, actors and processes of change.
A reform perspective emphasizes the role of political agency and the attributes of identifiable and autonomous actors - such as their will, understanding and power. The working assumption of this paper, however, is that an adequate understanding of comprehensive reform in the EU presupposes that actors and their will, understanding and power may change during reform processes. The institutional perspective used also assumes that such changes are influenced by the institutional contexts within which reform takes place. Making actors, their normative and cognitive beliefs and power endogenous, rather than treating them as predetermined and exogenous, complicates our models of change. However, this starting point may also deepen our understanding of how formally organized political institutions change and how they may be deliberately improved.
I start with the demands for comprehensive reform in the EU. Next, a simple model of institutional engineering is sketched. Then, the complications of seeing actors, political will, understanding and control as endogenous to reform processes are introduced. Finally, I return to the prospects of institutional engineering in complex and dynamic settings like the EU.
The demand for comprehensive
institutional reform in the EU
The assertion that European institutions are not working properly -- that they are unable to cope with major social and economic problems, to do what they are intended to do, and to adapt to a rapidly changing world -- is hardly new. Neither is the demand for comprehensive reform completely without precedence. Still, at the end of the 1990s the contested legitimacy of European institutions are by some portrayed as part of a constitutional crisis (Cowles and Smith 2000: 3). It is argued that the Union has to rethink and reshape the way institutions are organized, governed and changed. The EU needs to reform its institutions as well as the ways in which institutions are being revised.
Demands for reform have a threefold basis. First, policy reforms, and an expanding social and economic agenda leaving few spheres of life unaffected by EU-policy, have already created a need for parallel institutional reforms. Secondly, the coming enlargement will increase the size and heterogeneity of the Union and will have a significant impact on the functioning of institutions. Finally, reform needs are reinforced by changes in the world economy and geo-politics, and by the EU desire to strengthen Europe's role in the world and harness globalization.
The demand for comprehensive institutional reform has been aired from Maastrict through Nice. Reform will also be high on the agenda of the 2004 IGC and have a prominent place in the announced Commission White Paper on new forms of governance in Europe. According to the President of the Commission, Romano Prodi, the issue is nothing less than "the grand project of creating a European Union" and the fundamental nature of the Union in the future. Comprehensive reform involves building a political union and not only a trading bloc, improving the Union's external power and internal solidarity, and protecting and affirming the values of democracy, social cohesion and justice (Prodi 2001). 
Reform aspirations are reflected in the instrumental language used. Reformers want to "build Europe" by changing "the European institutional architecture" and the "institutional machinery". Institutions are portrayed as tools for implementing policies and visions of a better society. According to the Treaties, the Union is supposed to endow itself with the appropriate institutional instruments and other means necessary to attain its objectives and carry through its policies. "The Union shall be served by a single institutional framework which shall ensure the consistency and the continuity of the activities carried out in order to attain its obectives". 
The rhetoric indicates a belief in the possibility and importance of governing through deliberate, comprehensive institutional reform. Institutional engineering is seen as a possible strategy to achieve an intended social development.  Still, there have been disappointments. The Union does not seem to be capable to reform through a one-time shot of a constitutive assembly writing a constitution and specifying the basic principles of political organization. For example, for reform enthusiasts the achievements reflected in the Amsterdam Treaty and the Treaty of Nice were seen as inadequate.
While comprehensive reform has been difficult to achieve, the institutional balance of the Union is dynamic, not static. In the words of Romano Prodi: "Since 1981, first the Community and then the Union have been going through a continual process of adjustment, restructuring, enlargement and adaptation". There has been a constant seismic shock shaking the institutions and making the general public less sure "what Europe is all about or whether it is headed in the right direction" (Prodi 2001:5).
Still, over the years there has been a consistent development towards institutionalized cooperation, more so than in any other region of the world. Member states' authorities have become an integral part of European governance (Commission 2000: 13). The EU has been transformed from a bargained agreement among nation-states, to a quasi-federal polity (Stone Sweet and Sandholtz 1998:1).
A simple model of institutional engineering
Institutional design and reform involve how institutions might be, and ought to be, adapted to human purposes in order to function well and create improvement (Simon 1970). Then institutions are seen as organized arrangements created and run by purposeful people (Egeberg 1987, Stinchcombe 1996, 1997). Institutions are supposed to govern systems of activities, raising the question, how and to what extent they are capable of coping with the problems they are supposed to solve and produce substantive results (Underdal 1995).
Institutional engineering suggests an understanding that focusses on changes in the will (intention), knowledge (intelligence) and power (resources and control) of identifiable political actors. An institution represents a solution to a shared problem in a consensus system, or an imposed, coerced solution from a winning coalition or a conqueror. Institutions arise and are maintained or transformed as a function of the degree to which they serve the purposes of the relevant actors (March and Olsen 1995, 1998).
A simple model of institutional engineering suggests that institutions can be deliberately designed, chosen and reformed by actors:
(1) Who know what they want. Reformers have clear, consistent and stable objectives or normative criteria over the time period studied. These criteria define tasks, goals, performance failure, improvement and progress.
(2) Who know what it takes to achieve their objectives. Organizational form is a significant determinant of performance and the actors understand how alternative institutional forms affect performance. Claims for comprehensive reform follow from a perceived major performance failure or unexploited potential for improvement.
(3) Who have the power to do what is needed to achieve a desired result. Choices made by political actors are the most important determinants of institutional form.
In sum, in the simple version of institutional engineering rational leaders with clear objectives develop policies and then design or reform institutions as rational organizational instruments for implementing their policies. Political institutions are conceptualized as malleable organizational instruments. Legitimacy is functional-instrumental, that is, understood in terms of cost-effective technical performance and improved substantive results.
This view is basically consistent with the democratic idea that citizens, collectively and as equals, should determine how they may live together and organize politically (Dahl 1989). The winners of elections should have the authority and power to shape and reshape political and administrative institutions. The view is also consistent with parts of organization theory seeing formal organizations as "consciously planned, deliberately constructed and restructured" (Etzioni 1964:3). Furthermore, it is consistent with mainstream international relations theory. Neo-liberal institutionalists see international institutions and cooperation as rational attempts to counteract the inadequacy of one's own resources and discover Pareto improvements (Keohane 1982, 1984). Realists also portray international institutions as tools. Structures reflect the interests of victorious states and are the outcome of strategic interaction, alliances, coercion, competing interests and relative power. Changes in order results from changing powers and material capabilities (Strange 1983, Grieco 1988, Mearsheimer 1994).
Developing political will,
understanding and control
Sometimes reformers have precise preferences, perfect knowledge and full control. More often, however, they have a host of motives and concerns. The criteria for what makes one set of institutions more desirable than others are ambiguous, inconsistent or changing over the period studied. Likewise, the understanding and control of institutional dynamics may be more or less perfect. Reformers are not omnisciently rational. They do not necessarily know all possible institutional alternatives and their implications for the relevant objectives and principles. Neither are reformers omnipotent, controlling perfectly all factors having an impact on institutional change.
An implication is that political will, understanding and power may be seen as endogenous, not exogenous, to processes of comprehensive institutional reform. We need to understand goal formation processes; how information is obtained, interpreted and used; as well as how power, resources and capabilities are influenced by reform processes. Both the perceived need for comprehensive reform and the ability to implement reforms, are influenced by the institutional context and the processes of interaction within which reform takes place. What, then, are the implications for institutional engineering in the EU?
Processes of will-formation. What do EU reformers want? What do they expect from their institutions? What criteria (values, principles, norms, preferences, interests) do they use when they try to improve institutions?
Consider, first, that institutional reform reflects the predetermined and stable policy preferences of the reformers. Actors have preferences over outcomes, but not over social relations. There are no intrinsically valuable forms of association and cohesion (Lake 1996: 12,13).
Consider, secondly, that reformers have predetermined and stable preferences over institutional arrangements. They may adhere to durable general principles of political organization, only loosely connected to specific policy choices. Practices and procedures may become valued beyond their technical-functional properties (Selznick 1957). Then, institutions are not solely organizational tools for the policy achievements of one group of actors. They are part of a political order making it possible to live peacefully together with enduring tensions and conflicts. Legitimacy depends on the degree to which structures, procedures and rules conform with societal beliefs about appropriate institutions (Meyer and Rowan 1977). For instance, representative democratic government is not defined by particular actions at a particular moment, but by long-term systemic arrangements and how they function (Pitkin 1972: 234). A complication is that institutional legitimacy and policy efficiency do not necessarily coincide. There are illegitimate but technically efficient means, as well as legitimate but inefficient means (Merton 1938, Weber 1978).
Consider, thirdly, that objectives, preferences, interests, norms, principles and values may change over the period studied. This may be so both in terms of the attention they are given and how their content, consistency and significance are interpreted.
The motivations of EU reformers are complex and shifting. They want many, different and not necessarily consistent things. Like other reformers (Nystrom and Starbuck 1981:xiii, Goodin 1995:39), political leaders in the EU try to make institutions more rational and efficient, more humane, representative, responsive, transparent and accountable. They also try to make them more useful to societies, more profitable for owners, more submissive to top managers, more stable, predictable and robust, or more flexible, adaptive and able to learn.
Purposes and objectives include substantive policy criteria as well as maintaining and developing democratic institutions and processes. As the social and economic agenda has expanded, so has the number of relevant policy goals. In addition, in its self-presentation the EU adheres to several fundamental principles of governance, independent of the single policy issue at hand. Examples are, democracy, transparency, subsidiarity, respect for human rights and fundamental freedoms, the rule of law, and a competitive market economy.
The complexity of the Union's normative crieria makes it less likely that reformers can act on the basis of a single normative principle, or maximize a single objective or a consistent and stable utility function. They are likely to attend to a variety of partly competing and changing normative criteria. Consequently, students of institutional dynamics need to understand which criteria are attended to when, and how tensions among them are resolved. Conflicts between states and adherents of competing visions of European cooperation are well known. Here, attention is called to two less discussed complications.
First, sometimes reform focussed on adapting institutions to desired policy outcomes and improved effectiveness, and reform focussed on system improvement according to general organizational principles, create problems for each other. Policy-driven reform typically adapts institutions to a specific task- or policy environment. As each part of a system of governance adapts to its immediate task environment in a more or less myopic manner (Levinthal and March 1993), the system as a whole may become incoherent and ineffective. There is an increasing demand for coordination across functional sectors. Yet, functional differentiation and integration, as well as institutional "fusion" between levels of governance, make system-wide coordination difficult (Rometsch and Wessel 1996, Brunsson and Olsen 1998).
Comprehensive institutional reform, in particular at the "constitutional moments" in the history of a polity (Ackerman 1991), has a different focus. Here the main concern is to develop a coherent order according to general principles of political organization and governance. Such processes, constitutional in nature, tend to create problems for specific policy sectors because they rarely attend in any detail to policy- or sector specific characteristics that affect performance.
Secondly, reformers may face a dilemma in reconciling short term effectiveness and long term adaptation and survival. A major institutional problem in sustaining adaptive capability is how to balance processes of exploitation of existing standard operating procedures which have proven to be effective in the past, with the exploration of new institutions and procedures that may improve long term survival (March 1991, 1994).
For instance, Peterson and Bomberg (2000: 39) argue that risk-aversive behavior is a prime characteristics of EU decision making. If so, political leaders are likely to give priority to improving effectiveness in the short run at the cost of sustaining long run adapatation and survival. They will be exploiting the status quo of established arrangements, rather than exploring the potentials of experimenting with new forms, both in terms of empowering European supranational institutions and decentralizing power already transfered to the Union level.
Since reformers face a balance problem, there is no neat solution. Achieving balance, however, depends on the ability to simultaneously ensure that experiments in new possibilities, as well as persistence in them, are stimulated, and that the lessons of experiments are diffused across time and space (March 1991, 1994). The "new open coordination method" in the EU, involving score cards, benchmarking and dissemination of best practice across member states, may make diffusion processes more efficient. Simulteneously this method may produce more homogenization across domestic systems and therefore less experimentation and variation, which in the long run may inhibit learning and adaptation.
In a Europe of complex and dynamic preferences, sustaining institutional adaptive capability, rather than implementing predetermined objectives and principles, may be the principal challenge. For the EU, then, the challenge is not necessarily to make comprehensive reforms possible, but to foster the continuous learning and adaptation that makes great leaps unnecessary (Olsen 1997a). This approch is illustrated by the Finnish Prime Minister, Paavo Lipponen, who has argued that if the EU were first to define what is wanted, it might be easier to establish how it can be achieve. However, in a changing world this may be an impossible task. Reformers have to develop a Union that is responsive to change and challenges as a continuous process (Lipponen 2000).
My argument is that a simple model of institutional engineering, assuming a predetermined, coherent and stable political will, is unlikely to provide an adequate understanding of comprehensive reform in the EU. Institutional engineers can not assume a shared vision of the future Europe, providing clear normative criteria of institutional improvement. Instead, a variety of competing visions and normative criteria are seen as relevant. Reform goes beyond designing or reforming institutions fit for implementing exogenous objectives or normative principles. To a great extent, the reform debate in the EU involves what type of Europe and European governance is desirable and what democracy may imply in the EU context (Eriksen and Fossum 2000).
If it cannot be assumed that reformers take all relevant normative concerns into account; that they through exchanges, deliberate trade-offs and side payments organize their preferences into a consistent and stable utility function; and that they act on the basis of these preferences; an alternative understanding of the role of human purpose and political will in institutional change is needed.
One possible approach is suggested by Cyert and March (1963). They argue that the multitude of relevant preferences, purposes and principles can be understood as independent, aspiration level constraints on reform processes. That is, each constrain defines a standard that an acceptable institutional solution must meet. Understanding reform, then, requires detailed studies of a variety of processes and mechanisms:
-- institutional identification, where decision makers are socialized into giving priority to specific normative criteria,
-- "local rationality", where different institutions attend to, and protect different goals,
-- selective and sequential attention to goals, where shifting attention over time may change the reference group of social comparison and the normative criteria of good performance,
-- the significance of slack resources in buffering competing normative claims, suggesting that reduced slack may produce demands for more coordination and institutional reforms.
The search for knowledge. Cognitive beliefs about instititutional change are not always, anymore than normative beliefs, predetermined, reliable and stable. A comparative institutional analysis specifying the short- and long-term consequences of reforming an institutional order, is no trivial task. It is often difficult to form rational expectations about the future. Neither is it always easy to interpret collective and individual experience and the lessons of the past (March and Olsen 1975).
The task is also complicated by the many possible functions institutions may have - as instruments of command and coercion; arrangements for regulating and facilitating exchange; tools for collective problem-solving; vehicles for the redistribution of resources and power; and as frameworks for constructing meaning, defining appropriate behavior, constructing individual and collective identities and building a political culture. Impacts, then, include how institutions may affect policy substans, as well as the sense of political community, identity and shared purposes, concepts of right and truth, a shared political vocabulary, and the distribution of institutional resources and capabilities (March and Olsen 1995).
The idea of bounded rationality suggests that reformers act on a simple model of a complex reality. A theory of choice is to a great extent a theory of search. Information has to be obtained and exposure to and search for information is selective (Cyert and March 1963). Instead of assuming a choice between available institutional arrangements with known consequences, it is necessary to understand how reformers get information about institutional alternatives, their implications, and what can be achieved through reform.
Reformers may, or may not, share beliefs about institutional impacts. Sometimes there are "reality checks", like a performance crisis or obvious performance differentials convincing everyone that rethinking and reshaping institutions is called for. For instance, Pérez-Díaz portrays Spain's exposure to western Europe as a meeting with institutions and a culture which were far more efficient in achieving some of the traditional Spanish objectives, as well as other objectives which Spaniards were rapidly learning to appreciate. They learned from, imitated and came to identifying with the peoples of western Europe, their institutions and their way of life (Pérez-Díaz 1993: 13). Scharpf also argues that Europeans through a reality check may learn about international interdependencies and the loss of national "fate control", so that they adapt aspiration levels, internalize dependencies and the interests of others, avoid wishful thinking and concentrate on alternatives effective under current international interdependencies (Scharpf 1999: 283-6).
Often, however, rational expectations of the future are uncertain and interpretations of history are not inherent in the events themselves. The meaning of principles and interests in concrete situation has to be interpreted. The indeterminacy of facts, norms, interests and situations makes it difficult to conclude what they imply, how they are going to be applied, and whether they correspond to a specific organizational form (March and Olsen 1995:44, Habermas 1996, Kohler-Koch 2000).
In the EU the struggle over future forms of governance and organization is to a great extent a struggle over competing cognitive frames (Kohler-Koch 2000). Increased interdependence and extended interaction and communication are crucial elements of the shared experience of Member States. Shifts in cognitive frames may then become a main vehicle for system transformation (Finnemore and Sikkink 1999: 254). Yet, such shifts may involve shifting fashions, as much as new knowledge. Interpretations are influenced by both the institutional context and the normative environment in which they take place. Therefore, an adequate comprehension requires studies of the mechanisms by which external influences combine with endogenous processes to produce change (Pérez-Díaz 1993: 13). One aspect is the role experiential learning may have for the emergence of common understandings and identities.
All institutions develop experience-based standard responses for dealing with changing circumstances. As a result, institutions may be transformed through mundane processes of learning and incremental adaptation (March 1981). Institutions adapt smoothly as they codify their changing experience, wisdom and morality. Yet, in practice, there are many impediments to learning and adaptation. In most organized settings, experiential learning is complicated by limited experience, limited capability to make inferences from experience, limited ability to act on the basis of new knowledge, and limited memory and capability of retaining information (March and Olsen 1995: 206-223). In addition, a complication is that often several actors are learning simultaneously (March 1991, 1994).
An implication is that students of institutional dynamics need to understand the main sources of inefficiency in routine institutional processes of learning and adaptation.  On the one hand, a well-functioning democracy -- based on representative and accountable government and a well-developed civil society with strong voluntary associations, social movements and free public debate -- is supposed to facilitate continuous learning and adaptation. The expectation is that experience will improve the intelligence, effectiveness and adaptability of governance. Governments are supposed to detect and counteract failures, and to improve their performance as well as the polity's fitness for the future. Likewise, citizens are supposed to adapt their aspirations. Therefore, in an ideal democracy, equilibrium institutional solutions are assumed to be common. That is, no actors are likely to act so as to challenge radically existing institutions (March and Olsen 1995, Olsen 1997a).
On the other hand, inefficiencies in institutional learning and adaptation are integral and planned parts of governance in constitutional democracies. Constitutional rules both institutionalize orderly change and constrain change. In practice, it is also difficult for democracies to live up to the norms, expectations, and claims of experiential learning. Causal and normative beliefs, behavioral patterns and institutional designs do not easily change in light of experience (March and Olsen 1989,1995; Rose 1993, Olsen and Peters 1996).
Disenchantment with public institutions and their ability to learn and adapt has made it commonplace to suggest market mechanisms and competitive pressure as a panacea. Institutions incapable of rapid adjustment to changing circumstances are assumed to be eliminated by competition. However, the vision of an ideal self-regulating market society has to be held together with a practice of more or less perfect markets and with public and private organizations more or less able and willing to adapt to their environments (Nystrom and Starbuck 1981). There is no guarantee that competitition will drive out inefficient institutions or firms (North 1990). Sometimes institutions are rigid in spite of changing environments and deliberate reform attempts. They outlive their functional efficiency as well as their normative support. They are outdated, promote superstition and allow exploitation. Then, transformations may be radical or revolutionary and characterized by conflict, crisis, and institutional break-downs. Change in a system can fail so badly that it generates change of the system itself (Kochanek 1971:319).
Understanding change, then, requires comprehension of how the need for deliberate comprehensive reform may depend on how routine processes of institutional learning and adaptation are working (Olsen 1997a). Improving transparency, participation and other democratic mechanism in the EU may, for example, improve processes of learning and adaptation and make comprehensive reforms less important.
For behavioral students a challenge is that the dynamics of the EU institutions are usually discussed in formal-legal terms, that is, in terms of the powers formalised in treaties and law. The Union is a project of legal integration, representing a renewed trust in governance by law and the legal integration of society. The EU is inspired by a continental European legal culture and mentality emphasizing the importance of abstract constitutive principles, a political-legal order and generalized codes of conduct. Citizens are assumed to accept the obligations and rights following from membership in a political community, decided through positive law-making and legitimized by the principles on which the form of government is founded (Friedrich 1951: 34-5).
Understanding European institutional dynamics, however, requires a better grasp of how the formal-legal institutions set up by the Treaties, are exercised and translated into practices and behavior. Political scientists have been sceptical to the importance of constitutional and institutional design and reform (Dahl 1998: 127-8, 139). Likewise, lawyers have warned against "excessive legalism" and asked for a better understanding of how the formal-legal constitutionalism corresponds with the ways in which power is actually exercised. The assertion is that there is little correspondence between how institutions are supposed to function according to the Treaties and how they operate in practice (Búrca 1999: 61-66).
An implication is that both reformers and scholars have to go beyond the judicial realm, the Treaties and the formal-legal institutions. Empirical studies are called for. There is a need for knowledge about the relationships between, on the one hand, formal-legal institutions, legally binding decisions, and authorized texts, and, on the other hand, rule-implementation, "living institutions" and actual political conduct and outcomes.
Developing support and capabilities. In Discourses, Machiavelli (1950:138) concludes that, "to found a new Republic, or to reform entirely the old institutions of an existing one, must be the work of one man only." Machiavelli was right in observing that comprehensive reform requires strong organizational capabilities to stabilize attention, provide adequate analyses, mobilize resources and cope with resistance.
In contrast, democratic reformers under normal circumstances have a limited capability for comprehensive design and reform. In practice, political institutions are usually set up in a piecemeal way and they cannot be changed into any arbitrary form (March and Olsen 1983). Institutional developments are seldom driven by over-arching conceptions of the preferred design of the polity as such. There is typically no single design or designer, but rather a process involving several competing designers and localized attempts at partial design (Goodin 1995:28). As a concequence, reformers and students of institutional dynamics have to attend to the power and conflict dimensions of reform and how resources and capabilities may develop within the process itself.
"Politics is eternally concerned with the achievement of unity from diversity" (Wheeler 1975: 4) and the conflict-consensus mix impacts the legitimacy of comprehensive reform. Heterogeneous societies, in particular, demand strongly qualified majorities to change the power of different branches and levels of government or the relative power of public authorities and citizens (Weaver and Rockman 1993:464). The integrative power of democratic politics depends on the ability to rally broad public support for institutions and policies. Experiences from how conflicts of comprehensive reforms are resolved, therefore, are likely to impact future reforms. The ability to reconcile adaptiveness and system identity and integrity is likely to make future comprehensive reform more acceptable.
In the European Union governance takes place in polycentric, multilevel policy networks of public and private actors (Jachtenfuchs and Kohler-Koch 1996, Kohler-Koch and Eising 1999, Hooghe and Marks 2001). Reformers are not omnipotent. There is no single sovereign center with authority and power to change fundamentally the political order and many other factors than reformers' choices influence change. Furthermore, reform capabilities often have to be developed as an inherent part of the reform process, a key issue in many applicant countries (Nakrosis 2000).
Comprehensive reforms tend to be highly divisive and European reformers face enduring differences that cannot be hidden behind an apolitical rhetoric. The challenge is to make binding collective decisions while preserving the social fabric of the Union. Support has to be mobilized or created. Reformers have to convince political leaders and organized interests that reform is possible and desirable. Furthermore, harnessing public support has become a key issue in the EU (Commission 2001:4).
Over the last few years, the EU has invested more attention and resources in institutionalizing institutional reform. This makes such work more continuous rather than ad hoc and episodic. It becomes somewhat easier to buffer short-term fluctuations in attention. It also becomes easier to divide large-scale reform into deliberate, incremental and consistent reforms, with which the political system of the Union can cope.
Still, the Union is in search of an institutionalized process for changing its political order. The Union has become involved in some kind of semi-permanent IGCs focussed on institutional reform. Ad hoc-committees are frequently used (e.g. Dehaene, von Weizsäcker and Simon 1999). As illustrated by the responses to the German foreign minister Joschka Fischer's Humboldt lecture on the future organization and governance of Europe, the EU also lacks a shared vocabulary and shared concepts and cognitive frameworks for dealing with comprehensive reform (Joerges, Mény and Weiler 2000, Kohler-Koch 2000).
This openness may be seen as a liability. However, the future organization of Europe involves a struggle over peoples' minds, their identities and normative and causal beliefs, and the future distribution of power depends on the outcome of the struggle. As observed by Aristotle (1980: 299), "one needs to learn to be a citizen as much as a craftsman needs to be trained", and debates over the nature of political community are rhetorical events with a potential for educating and socializing Europeans. In periods of transformation it is argued over appropriate institutional arrangements. Standards of evaluation are made more visible and concepts such as democracy, equity and justice may change (Yack 1985). So may trust in contestants for authority and power and consequently the actual distribution of resources and capabilities.
A challenge for European institutional engineers, then, is to establish processes of change which nurture and develop good settings for reflective processes where participants can critically examine their own normative and causal beliefs and identities (Offe 1997, March and Olsen 1995). Another challenge is to regulate the access of issues and participants to such processes. Consider, for example, a development from the rather closed participation and intransparency of many earlier IGCs, to a Convention based on extensive participation and public debate, like the one used in the development of the Charter of Fundamental Rights. A new IGC-model, bringing together representatives of governments, national parliaments of both member states and candidate states, Union institutions, and civil society (e.g. Lipponen 2000), is likely to make the reform process more unwieldy. Empovering new participants may possibly lead to a stalemate. Yet, strengthening participatory democracy may possibly provide legitimacy and help reframing issues so that normative and causal beliefs converge.
Reformers may also be helped by the ongoing dynamics of imperfectly integrated political orders. A major historic development in Europe is the emergence of differentiated and partly autonomous institutional spheres with distinct logics of action, meaning and resources. Each sphere legitimates different participants, issues, and ways of making, implementing and justifying decisions. While the concept "political system" suggests an integrated and coherent institutional configuration, Weber observed that institutional orders are never perfectly integrated and that modernisation inevitably produces imbalances, tensions and collisions between spheres.  For that reason, students of institutional dynamics need to examine the patterns and tempos of separate institutions, including the conflicts around them, and analyze the interaction of different spheres (Orren and Skowronek 1996: 121).
The political order of the European Union makes inter-institutional tensions and collisions likely. First, they are likely because of the lack of agreement on the normative principles and ends according to which the European polity is to be integrated and governed. Secondly, they are likely due to the lack of a clear and stable allocation of powers between levels of governance and institutional spheres.  Consequently, a key to understanding the changing patterns of power in the emerging European polity, is to study how institutions collide and penetrate each other and how reformers may exploit such institutional dynamics.
Large-scale collisions may take place when institutional striving leads to "overstreching" one ideal (like competitive markets, democratic politics or judiciary power) and imposing principles and codes outside their traditional legitimate sphere of activity. The significance of institutional collisions can also be understood through studies of how institutions, after they are formally and legally established, learn their place in an institutional order. While institutions may be killed through a formal decision, or starved through reduced resources, "living institutions" cannot be imposed on the world full blown by decree. Enduring practices, understandings and resource allocations evolve as actors learn from experience - mundane encounters in everyday life as well as historical battles (Olsen 1997b: 175). For example, the European Court of Auditors, as a new institution, had to "chart the difficult waters of interinstitutional relations", establish its credentials, and find its place in a larger institutional order, through a search and learning process (Laffan 1999). Similar processes are likely in a variety of other EU-settings as new agencies are set up.
in complex and dynamic worlds
The history of the European Union illustrates that institutions evolve and change in a number of ways. It is a history of founding acts and deliberate institution-building as well as informal and gradual institutional evolution where common practices have been codified into formal-legal institutions. The development is consistent with the view that institutional arrangements are contingent and malleable, yet not necessarily in a voluntaristic way.
Like many democratic systems, the EU has combined a belief in institutional engineering with the experience that comprehensive reform is difficult to achieve. Still, the long term development has been in a consistent direction. There has been a gradual increase in the powers of common European institutions and in intergovernmental cooperation across an expanding agenda. The EU has moved towards "an ever closer Union", in spite of periods of European sclerosis.
A simple model of institutional engineering, assuming predetermined political will, understanding and power will be most relevant in fairly stable, coherent, comprehensible and controllable worlds. The model is less likely to capture processes of comprehensive reform in complex and dynamic political orders like the EU. This does not deny that human will, understanding and power are important for understanding institutional dynamics and that developing political institutions of self-governance is a first-order political process.
Setting up new formal-legal institutions, in particular if they are well funded and staffed, is likely to provide routinized attention and support for specific concerns, interests and principles. However, comprehensive formal-legal reforms are not likely to produce precise and stable policy outcomes. Many factors are intervening between institutional form and substantive results, leaving the precise nature of the influence to be determined. Over time institutions tend to gain some autonomy. They also adapt to other sources of power as they try to find their place in the larger order.
Reformers may have success in implementing legitimate principles such as democratic participation, transparency or the rule of law. An appropriate task for democratic governance may be to maintain and develop the conditions for various institutional spheres and to influence the mix of institutional structures and processes in order to avoid democratic perversions (March and Olsen 1995). The nature of the principles on which a polity is founded is important for the identity of the political community. Yet, again the precise impact on behavior and substantive results is uncertain.
Still, the institutional approach used in this paper suggests that there are several options for deliberate intervention in existing structures. Confronted with demands for comprehensive reform due to an expanding agenda, territorial enlargement and globalization, EU-reformers may both reduce the need for reforms and make reform more feasible.
The need for comprehensive reform may be reduced by improving ordinary processes of learning and adaptation. Somewhat paradoxically, the need for comprehensive reform may also be brought down by strengthening reform capabilities. This is so because institutionalized capabilities will make it easier to break up large-scale reform into smaller consistent reforms, digestable for the political system.
Furthermore, successful reform is more likely if a shared reform vocabulary evolves in the EU and there is a convergence in causal and normative beliefs and identities. A precondititon for such a development is that reform is understood as occasions for interpretation and opinion formation as much as decision making. Finally, comprehensive reform will become more feasible if institutional engineers understand ongoing inter-institutional collisions and learn to use their dynamics intelligently.
Ackerman, B. 1991, We, the People. Cambridge, MA.: Harvard University Press.
Aristotle (384-322 B.C.) 1985, Nichomachean Ethics (transl. by Terence Irwin). Indianapolis,IN: Hacket.
Brunsson, N. and J.P. Olsen 1998, Organization theory: Thirty years of dismantling, and then...? In N. Brunsson and J.P. Olsen (eds.): Organizing Organizations: 13-43. Bergen: Fagbokforlaget.
Búrca, G. de 1999, The institutional development of the EU: A constitutional analysis. In P. Craig and G. de Búrca (eds.), The Evolution of EU Law: 50-81. Oxford: Oxford University Press.
Commission of the European Communities 2000, The Commission's Work Programme for 2000. Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions. Brussels 9.2 2000 Com(2000) 155 final.
Commission of the European Communities 2001, The Commission's Work Programme for 2001. Communication from the Commission to the Council, the European Parliament, the Economic and Social Committee and the Committee of the Regions. Brussels 31.1 2001 Com(2001) 28 final.
Cowles, M.G. and M. Smith 2000, Risks, reform, resistance, and revival. In M.G. Cowles and M. Smith (eds.): The State of the European Union. Vol. 5. Risks. Reform, Resistance, and Revival: 3-16. Oxford: Oxford University Press.
Curtin, D.M. 1997, Postnational Democracy. The European Union in Search of a Political Philosophy. The Hague: Kluwer.
Cyert, R.M. and J.G. March 1963, A Behavioral Theory of the Firm. Englewood Cliffs, N.J.: Prentice?Hall. Second edition 1992, Oxford: Blackwell.
Dahl, R.A. 1989, Democracy and Its Critics. New Haven: Yale University Press.
Dahl, R.A. 1998, On Democracy. New Haven: Yale University Press.
Dehaene, J-L., R. von Weizsäcker and D. Simon 1999, The Institutional Implications of Enlargement: Report to the European Commission. http:/law.harvard.edu/Programs/Jean Monnet/dehaene.htlm.
Egeberg, M. 1987, Designing public organizations. In J. Kooiman and K.A. Eliassen (eds.): Managing Public Organizations: 142-157. London: Sage.
Eriksen, E.O. and J.E. Fossum (eds.) 2000, Democracy in the European Union. Integration through Deliberation? London: Routledge.
Etzioni, A. 1964, Modern Organizations. Englewood Cliffs N.J.: Prentice-Hall.
European Union Consolidated Treaties 1997, European Union Consolidated versions of The Treaty on European Union and The Treaty Establishing the European Community. Luxembourg: Office for Official Publications of the European Communities.
Finnemore, M. and K. Sikkin 1998, International norm dynamics and political change. International Organization 52 (4): 887-917.
Gerth, H.H. and C. Wright Mills, 1970 (paperback ed.), From Max Weber. London: Routledge and Kegan Paul.
Friedrich, C.J. 1951, The political theory of the new democratic constitutions. In. A. J. Zurcher (ed.): Constitutions and Constitutional Trends Since World War II: 13-35. New York: New York University Press.
Goodin, R.E. 1995, Institutions and their design. In R.E. Goodin (ed.) 1995: The Theory of Institutional Design: 1-53. Cambridge: Cambridge University Press.
Grieco, J.M. 1988, Anarchy and the limits of cooperation: A realist's critique of the newest liberal institutionalism. International Organization 42: 485-507.
Habermas, J. 1996, Between Facts and Norms. Cambridge MA: The MIT Press.
Hooghe, L. and G. Marks 2001, Multi-Level Governance and European Integration. Forthcoming (Rowman and Littlefield).
Jachtenfuchs, M. and B. Kohler-Koch (eds.) 1996, Europäische Integration. Opladen: Leske and Budrich.
Joerges, C., Y. Mény and J.H.H. Weiler (eds.) 2000, What Kind of Constitution for What Kind of Polity? Responses to Joschka Fischer. Florence and Cambridge Ma: European University Institute and Harvard Law School.
Keohane, R.O. 1982, The demand for international regimes. International Organizations 36 (2):141-171.
Keohane, R.O. 1984, After Hegemony. Cooperation and Discord in the World Political Economy. Princeton: Princeton University Press.
Kochanek, S.A. 1971, Perspectives on the study of revolution and social change. Comparative Politics 5 (3): 313-319.
Kohler-Koch, B. 2000, Framing: the bottelneck of constructing legitimate institutions. Journal of European Public Policy 7 (4): 513-31.
Kohler-Koch, B. and R. Eising (eds.) 1999, The Transformation of Governance in the European Union. London: Routledge.
Laffan, B. 1999, Becoming a "living institution": The evolution of the European Court of auditors. Journal of Common Market Studies 37 (2): 251-268.
Lake, D. A. 1996, Anarchy, hierarchy, and the variety of international relations. International Organization 1: 1-33.
Levinthal, D.A. and J.G. March 1993, The myopia of learning. Strategic Management Journal 14:95-112.
Lipponen, P. 2000, Speech given by Prime Minister Paavo Lipponen at the College of Europe, Bruges, Belgium November 10 2000. http://www.vn.fi/vn/english/speech/20001110e.htm
Machiavelli, N. 1950 (intro. by M. Lerner), The Prince and The Discourses. New York: Random House, The Modern Library.
Maduro, M.P. 1997, Reforming the market or the state? Article 30 and the European Constitution: Economic freedom and political rights. European Law Journal 3 (1): 55-82.
March, J. G., 1981, Footnotes to organizational change. Administrative Science Quarterly 26: 563?577.
March, J.G. 1991, Exploration and exploitation in organizational learning. Organizational Science 2: 71-87.
March, J.G. 1994, The evolution of evolution. In J. Baum and J. Singh (eds.): Evolutionary Dynamics of Organizations: 39-49. New York: Oxford University Press.
March, J.G. and J.P. Olsen 1975, "The Uncertainty of the Past: Organizational Learning Under Ambiguity". European Journal of Political Research, 3:147-171.
March, J.G. and J.P. Olsen 1983, Organizing political life: What administrative reorganization tells us about government. American Political Science Review 77 (2): 281-296.
March, J.G. and J.P. Olsen 1989, Rediscovering Institutions. New York: Free Press.
March, J.G. and J.P. Olsen 1995, Democratic Governance. New York: Free Press.
March, James G. and Johan P. Olsen. 1998. The Institutional Dynamics of International Political Orders. International Organization 52 (4): 943-969.
Mearsheimer, J.J. 1994, The false promise of international institutions. International Security 19: 5-49.
Merton, Robert. 1938. "Social Structure and Anomie". American Sociological Review, 3 (October): 672-682.
Meyer, John W. and Brian Rowan 1977, Institutionalized Organizations: Formal Structure as Myth and Ceremony. American Journal of Sociology 83: 340-363.
Nakrosis, V. 2000, Assessing governmental capabilities to manage European affairs: The case of Lithuania. Florence: European University Institute: EUI working papers RSC No. 2000/58.
North, D. C. 1990, Institutions, Institutional Change and Economic Performance. Cambridge Ma: Cambridge University Press.
Nystrom, P.C. and W.H. Starbuck 1981, Designing and understanding organizations. In P.C. Nystrom and W.H. Starbuck (eds.): Handbook of Organizational Design (Vol.1: Adapting Organizations to their Environments: ix-xxii. Oxford: Oxford University Press.
Offe, C. 1997, Micro-aspects of democratic theory: What makes for the deliberative competence of citizens? In A. Hadenius (ed.): Democracy's Victory and Crisis: 81-104. Cambridge: Cambridge University Press.
Olsen, Johan P. 1997a. Institutional Design in Democratic Contexts. Journal of Political Philosophy, 5 (3): 203-229.
Olsen, J.P. 1997b, European challenges to the nation state. In B. Steunenberg and F. van Vught (eds.), Political Institutions and Public Policy: 157-188. Dordrecht: Kluwer.
Olsen, J.P. and B.G. Peters (eds.) 1996, Lessons from Experience. Experiential Learning in Administrative Reforms in Eight Democracies. Oslo: Scandinavian University Press.
Orren, K. and S. Skowronek 1994, Beyond the iconography of order: Notes for a "new" institutionalism. In L. Dodd and C. Jillson (eds.): The Dynamics of American Politics: Approaches and Interpretations: 311-30. Boulder CO: Westview.
Orren, K. and S. Skowronek 1996, Institutions and intercurrence: Theory building in the fullness of time. In I. Shapiro and R. Hardin (eds.): Political Order: 111-146. (Nomos XXXVIII). New York: New York University Press.
Pérez-Díaz, V.M. 1993, The Return of Civil Society. The Emergence of Democratic Spain. Cambridge Ma: Harvard University Press.
Peterson, J. and E. Bomberg 2000, The EU after the 1990s: Explaining continuity and change. In M.G. Cowles and M. Smith (eds.): The State of the European Union. Vol. 5. Risks. Reform, Resistance, and Revival: 19-41. Oxford: Oxford University Press.
Pitkin, H.F. 1972 (1967), The Concept of Representation. Berkeley: University of California Press.
Prodi, R 2001, The State of the Union in 2001. Speech, European Parliament Strasbourg, 13. February 2001. http://Europa.eu.int/rapid/start/cgi/g...ttxt=gt&doc=SPEECH/01/64|0|RAPID&lg=EN
Rometsch, D. and W. Wessels 1996, Conclusion: European Union and national institutions. In D. Rometsch and W. Wessels (eds.): The European Union and Member States: 328-365. Manchester: Manchester University Press.
Rose, Richard 1993, Lesson-drawing in Public Policy. Chatham, N.J.: Chatham House.
Sand, I.J. 1998, Understanding the new forms of governance: Mutually interdependent, reflexive, destabilized and competing institutions. European Law Journal 4 (3): 271-293.
Scharpf, F.W. 1999, Legitimacy in the multi-actor European polity. In M. Egeberg and P. Lægreid (eds.) Organizing Political Institutions: 261-288. Oslo: Scandinavian University Press.
Selznick, P. 1957, Leadership in Administration. New York: Harper & Row.
Simon, H.A. 1970, (paper back ed.), The Sciences of the Artificial. Cambridge MA: MIT Press.
Skowronek, S. 1995, Order and change. Polity 28 (1): 91-96.
Stinchcombe, A.L. 1986, Reason and rationality. Sociological Theory 4:151-166.
Stinchcombe, A.L. 1997, On the virtues of the old institutionalism. Annual Review of Sociology 23: 1-18.
Stone Sweet, A. and W. Sandholtz 1998, Integration, supranational governance and the institutionalization of the European polity. In W. Sandholtz and A. Stone Sweet (eds.): European Integration and Supranational Governance: 1-26. Oxford: Oxford University Press.
Strange, S. 1983, "Cave! Hic Dragones": A critique of regime analysis. In S.D. Krasner (ed.), International Regimes: 337-354. Ithaca, N.Y. Cornell University Press.
Sverdrup, U.I. 1999, Ambiguity and Adaptation. Europeanization of Administrative Institutions as Loosely Coupled Processes. Oslo, ARENA: Report No 8/2000.
Underdal, A. 1995, The study of international regimes. Journal of Peace Research 32 (1):113-119.
Weaver, R.K. and B.A. Rockman 1993, Institutional reform and constitutional design. In R.K. Weaver and B.A. Rockman (eds): Do Institutions Matter?: 462-481. Washington D.C.: Brookings.
Weber, Max 1978. Economy and Society. Berkeley: University of California Press.
Weiler, J.H.H. 1999, The Constitution of Europe. Cambridge: Cambridge Univerity Press.
Wheeler, H. 1975, Constitutionalism. I F.I. Greenstein og N.W. Polsby (red.): Handbook of Political Science: Governmental Institutions and Processes, vol. 5: 1-91. Reading MA: Addison-Wesley.
Yack, B. 1985, Community and conflict in Artistotle's Political Philosophy. The Review of Politics, 47 (1):92-112.
 According to Prodi, the White Paper will not present a ready-made "charter of competencies" listing the powers and responsibilities of the Union and the Member States. The paper will not take a position on constitutional question. Yet, it will propose a decentralization of the administration of the Union. Prodi also argues that the internal reforms of the Commission have to be brought to swift conclusions because no organization can live with so much stress in the long run (Prodi 2001).
 European Union Consolidated Treaties, Article 6.4 and Article 3.
 Prodi's "The State of the Union in 2001"-speech illustrates the claim that reform should start with substance, that is, what the European peoples want to have together and therefore the ultimate purposes of the Union. Then the institutional implications should be drawn and a "coherent and durable design for our enlarged Union" developed. Consistent with this "tool"-view of institutions, Prodi makes a distinction between the "eminently political" issues of substance, and institutional issues.
 Concepts like "historical inefficiency" and "path dependency" suggest that the match between reforms, environments, and institutional structure and performance is not automatic, continuous, and precise. Change processes depend to a large extent upon the internal constitutive characteristics of existing institutions. Institutions authorise and enable, as well as constrain, change. Therefore, there is a need for understanding how institutions may transform, modify, redirect and integrate, and not only aggregate, the demands, interests, and powers of societal actors and forces (March and Olsen 1989, 1995).
 Gerth and Mills 1970: 328-57, Weber 1978, also Skowronek 1995, Orren and Skowronek 1994, 1996. Often it is observed that reformers face institutional resistance. However, changes consistent with an institution's (or a network's) identity, tradition and dynamics are likely to be continuous and incremental. Changes in opposition to existing identities, traditions and dynamics are likely to be episodic and problematic and create conflict (March and Olsen 1989,1995).
 Curtin 1997, Maduro 1997, Sand 1998, Weiler 1999, Sverdrup 2000.