The Bergen approach to public administration and political organization

SAMMENDRAG Dette essayet tar utgangspunkt i arbeidet med å bygge opp faget Offentlig administrasjon og organisasjonskunnskap i Bergen på slutten av 1960-tallet og ut over 1970-tallet, med Knut Dahl Jacobsen som “the founding father”. Arbeidet med å utforme et forskningsprogram og et utdanningstilbud tok sikte på å skape en teori-basert, empirisk tilnærming til studiet av offentlig administrasjon og politisk organisering. Dette innebar utvikling av en relativt unik kombinasjon av organisasjonsteori og demokratisk teori. Arbeidet hadde en solid forankring i statsvitenskapen, men det tok likevel form av en oppbygging av et eget fag og etter hvert et eget institutt. I ti punkter diskuterer jeg noen av de grunnleggende premissene – teoretiske ideer, forskningsspørsmål, normative standarder og faktiske antakelser – som skulle rettferdiggjøre denne spesielle tilnærmingen til studier av offentlig administrasjon og politikk, til forskjell fra det statsvitenskap, sosiologi, økonomi og jus hadde å tilby. Mye er forandret i løpet av det halve århundre som har gått siden oppstarten. Det gjelder både studieobjektet, organisasjonsteorien og demokratisk teori. Faget står derfor nå overfor både teoretiske og praktiske veivalg. For det første, utpensling av en faglig identitet og en akademisk profil som gjør det mulig å videreutvikle fagets teorigrunnlag og rettferdiggjøre et eget institutt. For det andre å frembringe kunnskap praktikere finner nyttig, også i situasjoner hvor det er betydelig uenighet om hva “nyttig” betyr og impliserer. Mitt siktemål med essayet er dels å bidra til den faglig hukommelsen, og dels å bidra til en debatt om mulige veier fremover for et spennende faglig prosjekt. Mer enn noen gang er det viktig å forstå hvordan politisk-administrative organisasjon og organisering kan bidra til demokrati og sivilisert sameksistens i en turbulent verden.


AN OCCASION FOR LOOKING BACK
This essay is an edited version of my remarks at the inaugural March-Olsen Honorary Lecture given by Professor W. Richard Scott, Stanford University, at the University of Bergen, 23 May 2018.I used this occasion to look back at the roots of public administration as a research program and an educational subject in Bergen in the late 1960s and early 1970s: its study object, the main questions asked, and its theoretical and empirical assumptions and aspirations.The Bergen approach gave primacy to theory-based empirical studies of the actual working of formal organizations and political-administrative institutions.The approach developed some foundational ideas that explained and justified the academic identity and scholarly profile of the new program and outlined what made it reasonable to establish a specific academic entity differentiated from sociology and comparative politics, as well as law and economics. 1 I have left out my thanks to the department and its chair, Jacob Aars, for honoring James G. March and me in this way, and my thanks to W. Richard Scott for making the long journey in order to give the inaugural lecture.I have also left out how, half a century ago, I first met Jim March when I was a visiting scholar at the University of California, Irvine, where he was the Dean of the School of Social Science.I moved from Irvine to Bergen in 1969, where I spent twenty-four exciting years.Jim came to Bergen in 1970, together with Michael Cohen and Jim Glenn, a visit that was of great importance for me, for the department, and for the Scandinavian community of organizational scholars.Important parts of the garbage can model (Cohen, March and Olsen 1972) and Ambiguity and Choice in Organizations (1976) were written in Bergen.
I believe that the ten points presented below capture important aspects of the vision of the founding father of public administration in Bergen, Knut Dahl Jacobsen.They also reflect some fairly widely shared ideas about the way forward for the program, and also what actually took place in the early phase after Public Administration was formally established as an academic discipline in 1967-first located in the multi-disciplinary Department of Sociology, and, from 1980, in the Department for Public Administration and Organization Theory.
Nevertheless, I am well aware that this is my story, based upon my cooperation with Jim March, on my work at the department and with the first Power Study (NOU1982: 3), characterized as "the great leap forward" in Norwegian political science (Kuhnle 1986: 61).I use an institutional approach, a specific way to understand the interdependencies, interrelations, and interactions between political institutions, political agency, and the societal contexts within which political institutions and actors function.When it comes to references, I give priority to publications of mine that develop further the arguments presented below.
I find this acceptable in the context of the March-Olsen Honorary Lecture.It is beyond doubt that others will have different stories to tell, and I apologize for not giving more attention to important themes such as local and regional administration and government and to the relations between professions and their clients.However, many can do this better than I can.The different stories will use different approaches.Some participants or on-lookers will, for example, give explanatory primacy to individual decision-makers and rational choice explanations.Others will, as many liberalists and Marxists do, give primacy to economy, technology, and societal structures and processes, rather than to political institutions.They will also present other references than I do.
In sum, the essay is not meant to be an exercise in self-glorification.It is, rather, an invitation to others to come forward with their versions and thereby contribute to a fruitful debate about the upstart of the Bergen program, what has happened over the last half century, and possible roadmaps for the future academic study of public administration and political organization, a project close to my heart.

THE ROOTS OF THE BERGEN APPROACH
The aim of the essay is modest.I have no intention of writing the history of the starting phase of public administration as an academic discipline at the University of Bergen.I simply wish to add a few points to earlier discussions of this special approach to public administration and political organization, an approach combining foundational ideas from organization theory and democratic theory (Bleiklie 1985;Kuhnle 1986;March 1997;Egeberg and Laegreid 1999;Christensen and Laegreid 2004;Olsen 2007a;Sverdrup and Trondal 2008;Roness and Saetren 2009;Christensen 2015).
(1) Organization matters A basic, starting premise was the perceived importance of formal organizations-deliberately established, goal-directed, and formally organized entities-as a special and increasingly important form of social organization, and as the dominant agent in administrative and political life.Another premise was a belief in the explanatory power of organization theory.Many of us read-with enthusiasm-Herbert Simon's Administrative Behavior (Simon 1945).Two building blocks of the reading lists were Organizations (March and Simon 1958) and Formal Organizations (Blau and Scott 1962), together with publications by Crozier, Etzioni, Mayntz, Perrow, Selznick, Thompson, and of course Max Weber, among others.
For many, there was a developing trend: from the study of formal organizations as frameworks for micro behavior, towards organization and organizing as general politicaladministrative phenomena.Attention was directed towards the effects of various forms of political-administrative organization, the causal processes through which the effects materialized, and the processes through which different organizational forms emerge, are maintained, transformed, or eliminated in different historical-spatial contexts.Organizations, and what are called institutions in everyday language, could be interpreted as technical instruments for leaders, or as institutions with an inherent value and dynamics of their own (Selznick 1957).Macro political orders were arrangements of institutions, and a key issue was how the organization of the political order, institutional developments, and reform proposals could contribute to or obstruct political legitimacy, citizens' well-being, and help a population to achieve the aim of living together in a civilized manner.
(2) The organizational basis of governmental institutions In contrast to mainstream international organizational studies, the Bergen approach was solidly anchored in political science.The core aim was to understand the role of organizational factors in political-administrative life, based upon theories of democracy and power as well as on organization theory inspired by the Carnegie Tech school.The combination of political science and organization theory was somewhat special as the two disciplines have had "parallel agendas but mutual disregard" (Olsen 1991).Primacy was given to studies of political organization, activities within governing institutions, and attempts to affect the performance of public authorities.It was emphasized that public administration is public.Agents have powers derived from and authorized by the public at large.They are acting on behalf of-and they are ultimately accountable to-the citizenry, making public administration different from private administration in fundamental ways.
The public administration was seen as the backbone of the state, providing an organized capacity for preparing policy-making, making decisions, settling disputes, and getting things done.The public administration was not a neutral instrument in the hands of shifting governments, but an important agent in its own right.Administrative organization was not a technical, apolitical issue solely concerned with economy, efficiency, and management.All institutional arrangements create a special "mobilization of bias" (Schattschneider 1960).The public administration influences public policies and its competence and capacity are of great importance for the welfare and life chances of citizens.
Democratic quality could not solely be linked to the input side and various forms of public participation and representation.It had also to take into account the working of the state organization.A book from the Power study, Byråkrati og beslutninger (Laegreid and Olsen 1978), based upon the 1976 survey of all civil servants in Norwegian ministries with more than one year of experience, documented the explanatory power of institutional belonging, formal organizational position and profession for administrative thinking and behavior.These are findings that have been supported by similar surveys in 1986, 1996, 2006 and 2016, then also including directorates.This is, indeed, a remarkable time series of administrative data analyzed in a book by Tom Christensen, Morten Egeberg, Per Laegreid and Jarle Trondal: Sentralforvaltningen. Stabilitet og endring gjennom 40 år (2018).
(3) From formal-legal to living institutions The study of governmental institutions has been a founding pillar of political science (Rhodes, Binder and Rockman 2006).Yet, in opposition to the then-dominant formallegal institutional approaches emphasizing constitutions and the legal order, the US-led behavioral revolution in political science had its focus on theory-based empirical-behavioral research and how government and politics work in practice.The main tendency was to see political institutions and constitutions as less important.Priority was given to the role of socio-economic factors in actual political rule and as a prerequisite for democratic order (Dahl 1956).
The organization theory-inspired neo-institutional turn in political science brought institutions and an interest in the organizational basis of government and politics "back in".Institutions were seen to have explanatory power of their own.They could not be reduced to an artifact of broader societal forces or the decisions of individual actors.However, focus was now on "living institutions" rather than formal-legal arrangements.That is, neo-institutionalism attended to organizational practices and informal norms, rules and routines embedded in meaning structures and resources structures, and not solely formal-legal structures and rules (March andOlsen 1984, 1989;Olsen 2010).
In line with the international development, the Bergen approach gave little attention to the written Constitution and the courts.Constitutionalism, understood as an ideology of limited government, division of powers, and an active role of courts in preventing arbitrary and tyrannical use of public power2 was not high on the political agenda in Norway.The public interest and individual and minority rights and freedoms were most often seen to be protected primarily by the people's elected representatives, accountable and controlled through competitive elections.The Norwegian political "climate" in the middle of the 1970s, and the relative importance of Rechtsstaat issues, may be illustrated by a late president of the Parliament (Stortinget), telling me that the Norwegian Grunnlov (Constitution) was perfect.It had never prevented parliamentarians from doing what they wanted to do.
(4) Beyond the decision-making framework and rational choice A key theme for the Bergen approach was the relationship between organization and decision-making processes.Interpretations of agency, and how individual actors were portrayed, were from the start solidly embedded in the bounded rationality tradition of the Carnegie Tech School of organizational research (Simon 1945;March and Simon 1958;Cyert and March 1963).Focus was on the cognitive constraints on human rationality and the importance of formal organizations-their rules, roles, routines and standard operating procedure-in forming decision-makers' attention, their models of the world, values, interests, resources, behavior, and performance.Yet, studies of decision-making increasingly went beyond the (bounded) means-end rational framework.They called attention to both what happened in what was usually called decision-making processes and to how decisions were made in practice.It was observed that what was called decision-making processes in everyday language involved more than decision making.It was also observed that decision-making did not always follow the pattern prescribed by rational choice approaches.
What was called decision-making processes sometimes involved ritual acts (Olsen 1970) and sense-making (March and Olsen 1976) more than deliberate choice.Dominant ideas about rational choice and calculation of expected utility, borrowed from economics and statistical decision theory, were challenged by ideas about bounded rationality, aspiration levels and a search for satisfactory alternatives (Simon 1945, March and Simon 1958, Cyert and March 1963), and later by ideas about garbage can processes, temporal structures and chance elements in decision-making (Cohen, March and Olsen 1972).Alternatives to a logic of consequentiality and behavior driven by material incentives were explored.The traditional emphasis on legal rule-following was supplemented by an interest in the logic of appropriateness (March andOlsen 1984, 1989) and rule-following guided by internalized democratic identities and roles, and by identity-and role-prescriptions embedded in specific institutions or professions.
(5) Normative principles: democracy, the sovereign people and the sovereign individual An effect of the behavioral revolution, with its clear distinction between normative and empirical-analytical dimensions and its emphasis on the latter, was that normative theorybuilding, prescribing what morally good and rightful democratic rule implied, did not play a central role in the early phase of the Bergen project.Norms were studied empirically, but not as part of building normative theory (Jacobsen 1964).
The political climate in Norway worked in the same direction.The 1960s were the hey-day of the Scandinavian welfare state and social democracy.There was belief in democratic (majority) governance of society.There was also economic growth, a rapidly growing public sector, allocation of slack resources rather than austerity and budgetary cuts, a fairly homogeneous society, and a relatively low level of societal conflict.Probably, such conditions contributed to a democracy-that-is-us attitude, with modest incentives for public deliberation about the desirable normative principles of the political order.Normative principles remained fairly ambiguous, not necessarily consistent, and they were often loosely coupled to actual behavior (Olsen 2014).Only slowly, as the political and the academic climate started to change, was more attention given to the normative assumptions empirical studies built upon and also an explicit interest in normative theory-building (Blichner and Sangolt 1994;Eriksen and Fossum 2000).
There was, however, a tacit understanding that in Norway, as a well-functioning democracy, "the people" are the ultimate source of public authority.The "great Norwegian narrative" is about the sovereign people, the sovereign nation state, and the sovereign parliament (Olsen 2014).Nevertheless, belief in sovereignty-implying one single, ultimate center of public authority-could not hide that there were competing conceptions of "the people" and their role in democratic governance.Representative democracy, for example, involves enduring tensions between the sovereign people and the sovereign individual, public authority and private autonomy, lay-men and elected leaders and experts, political community and solidarity and self-interest, and citizenship and competing identities and belongings (Olsen 2014)  The Bergen approach was based upon the idea that the internal organization of public administration matters.Still, it was not enough to study the internal organization.Public administration was analyzed as part of a larger political-democratic order and system of governance, and in relation to the policy sector and part of society it was supposed to administer.Many were inspired by David Easton's system thinking about how binding decisions are made and executed, and the need to understand the interdependencies and interactions of the components of a political system as they try to cope with changing environments (Easton 1953(Easton , 1957(Easton , 1965)).The Power Study also benefitted from the book Politics, Economics and Welfare and its focus upon key mechanisms of rational calculation and social control in modern democracies (Dahl and Lindblom 1953).
The normative prescriptions of parliamentary government, with a clear and simple hierarchical principal-agent chain of authorization, command, control, responsibility and accountability-from the people as voters to elected representatives and majority government, to the executive and the public administration, and to the people as subjects-was seen to capture neither the normative nor the empirical complexity of modern democratic polities (Olsen 1978: Ch. 1).Likewise, the normative distinction between politics and administration, with voters, political parties and legislatures as the primary political actors, was seen to be unlikely to capture the actual role of public administration in governmental processes (Christensen et al. 2014, Christensen et. al 2015).Rather, it has been argued that good government depends more on a competent and neutral public administration than upon majority-based democratic institutions (Rothstein 2011).
There was no support for the idea that all conflicts are resolved through a constitution or by the decisions of democratically elected policy-makers.Rather, like other organizations (Cyert and March 1963), public administration was seen to live with unresolved conflict and over time to discover their interdependencies with the environment, as well as their own capabilities and capacities (Simon 1953, Selznick 1957).The central administration, and even each administrative role, was interpreted as a meeting place for a variety of premises: political, legal, administrative, economic, professional, technical etc. (Simon 1945;Kaufman 1956;Jacobsen 1960Jacobsen , 1964)).In simple situations, a role could prescribe appropriate behavior precisely.In more complex and politically interesting situations, roles were more likely to have only a guiding function.They would suggest what premises to consider, how to frame problems, and where to search for information, legitimacy and support.Many actors also have several roles and loyalties.They may act on the basis of a logic of consequentiality or a logic of appropriateness, and on a strong or weak resource basis.
Neither did empirical observations support the pluralist normative and empirical claim, that all interests had their own watchdog.Rather, an important issue was: what values, interests, worldviews and resources are built into administrative structures and routinely attended to by full-time professionals?It was asked, what issues are organized in or out of politics (Schattschneider 1960), and what issues and groups are regulated or protected by administrative units, with different degrees of autonomy and discretion, and with different capacities for organized action in terms of professional staffs and budgets?(Jacobsen 1960(Jacobsen , 1964(Jacobsen , 1965;;Egeberg 1978).There was also focus upon what groups of citizens had the resources and capabilities to actually use different public services (Jacobsen 1964(Jacobsen , 1965)).
Growing political agendas in modern democracies have contributed to institutional differentiation and functional specialization, more administrative autonomy from and less accountability to majority institutions and public opinion.Administrative agencies have instead claimed legitimacy based upon expertise, integrity and peer (professional) accountability (Olsen 2017).Administrative actors can identify with a specific organization, institution, profession, or client group.The image of the ideal civil servant will vary among groups and over time.New role definitions emerge, including ideas about what are the legitimate tasks and responsibilities of administrators, elected politicians and citizens (Jacobsen 1964). 3Civil servants are participants in the processes through which identities, values and interests were defined, shaped, mobilized, given priority, or ignored.Initiatives, analysis, decision-making, and getting things done were not necessarily tightly coupled in a formal chain of parliamentary governance (Olsen 1978).Formal decisions by public authorities are not always implemented, or implemented as intended by the decision-makers (Saetren 1983, Strand 1985).
A core idea has been that no organizational form works well for all purposes and under all conditions.In modern democracies, there is no single right way to organize the public administration-be it an ideal form of a Weberian bureaucracy, or a stylized private enterprise model-independent of the tasks, goals, resources available, and the historical-institutional contexts within which the administration is acting.Thus, political-administrative organization is a mixed order that has to be analyzed in relation to the various and shifting tasks the public administration is expected to cope with, the various and shifting normative standards, interests and values it is expected to meet, and in relation to who controls relevant resources.Public administration influences and is influenced by its societal environment and, in the 1960s and 1970s, functional representation through corporative arrangements were of special importance among the many channels between polity and society, authorities and citizens, and professions and clients.In contrast to the international literature portraying organized societal interests as pressure groups and lobbyists, the Bergen project was attending to corporatist arrangements and how organized interests were integrated into the political order and public policy-making.Inspired by Stein Rokkan's claim that "votes count, but resources decide" (Rokkan 1966), focus was upon the interdependencies, interrelations and interactions of democratic-representative institutions, public administration, professions, and organized interest embedded in corporative structures of representation (Olsen 1978(Olsen , 1983;;Egeberg 1981).For example, studies of organizational society and the "segmented state" (Egeberg, Olsen and Saetren 1975) analyzed coalitions between specific administrative units, committees in the Storting, and organized interests.These were fora usually delegated considerable discretion and power by framework laws.For a small country with an open economy, and thus highly dependent on global markets, the tripartite cooperation between state, labor and capital was by many seen as a precondition for national economic competitiveness and prosperity.
3. To the best of my judgment, Jacobsen's book Teknisk hjelp og politisk struktur (Jacobsen 1964)-an historical study of the dynamic relations between majority and non-majority institutions and elected and non-elected actors-would have been an important part of the international literature if it had been written in English and not in Norwegian.
Gradually, corporative arrangements of representation came under attack.In Norway, and even more so in other countries and in the European Union, lobbying became more important and more professionalized.New Public Management reforms defined new "stakeholders", and new power-relationships and channels of influence emerged.There was a development from government to governance, and from intra-organizational to inter-organizational approaches, and new forms of cooperation across public-private boundaries (Christensen andLaegreid 2001, 2002); and the new types of public-private partnerships and non-hierarchical networks had implications for the diffusion of ideas and for administrative editing and translation processes (Røvik 2007;Christensen et al. 2015).Some networks were national.Yet, many transcended national borders.
(8) Beyond the state as the dominant political entity At the start of the Bergen project, the dominant context for studies of public administration was the nation state, including its local and regional levels.There was little doubt that the state, as a specific form of political organization, was the most important political entity.Key questions were related to the power of the state and the power over the state, and to a lesser degree to the organization of the relations between states.The study of comparative administration and international administration were relatively weakly developed and not given much attention.
However, when Morten Egeberg came to the Power project in 1974, he insisted that we had to go beyond the nation-state framework and intergovernmental arrangements and take "the fourth level of government" seriously (Egeberg 1980).To make sense of how public administration operates in modern democracies, it was important to take into account not only its relationships to elected, majority-based institutions and actors, and to societal organized interests integrated into public policy-making-it was also necessary to take into account the public administration's relations to institutions and regimes outside national borders (Egeberg 1989: 52).
Since then, Egeberg has certainly been proven right by European and international developments.He has also given important contributions to theorizing European institution-building, the role of the European Commission, and the interaction between national and European administrations.The Norwegian public administration has increasingly been integrated into the emerging European executive order.It has become "double-hatted"-a participant in two executive-administrative orders, the national and the European (Egeberg 2006;Trondal 2010;Egeberg and Trondal 2018).The development of European-level institutions has become a process of great historical significance, challenging nation-state focused approaches to democratic-political order and change (Olsen 2010(Olsen , 2017(Olsen , 2018)).
(9) History and context matter From the very beginning, Knut Dahl Jacobsen emphasized the importance of history and the problems of universal, non-contextual explanations not analyzing the conditions under which organizational factors were likely to have explanatory power.Different public administrations are located differently in time and space and there is a need to study the relations between the historical development of problems and conflicts to be coped with, the possible solutions attended to, and organizational forms and processes.As a consequence, theory-building at the meso-level, taking into account specific historical-institutional contexts, are likely to be more fruitful than generic, universal explanations.
The Bergen approach was largely consistent with Stein Rokkan's careful balancing between, on the one hand, theoretical generalization, discovering patterns and building models, and, on the other hand, historic explanations of specific political phenomena.The program was also consistent with studies of the historical development of European nation states-the formations and transformations of states and nations-and with Rokkan's interest in Europe as a region with a special history and development (Kuhnle 1986: 54).Different states have different histories.They have taken different routes and ended up with different institutional arrangements (Rokkan 1999).For some nation-states, institutional developments have involved critical junctures in terms of war, civil war, revolution, coups, and destructive conflict.The history of others is characterized by peaceful political competition-balancing consensus and contestation, what to agree about and what to disagree about-embedded in public debate, compromises, and incremental change.
The Norwegian experience has been of the second type, a combination of peaceful co-existence and "revolution in slow motion" (Olsen, Roness and Saetren 1982).Institutions have developed through historical processes of trial and error, successes, failures, and compromises, rather than through large-scale confrontations and coercion.Historically, new groups have become politically mobilized, but the legitimacy of institutions and actors have had their ups-and-downs (Olsen 1983).There have been periods of political contraction and detraction.Initiative and decision-making authority have been concentrated in the parliament, or dispersed to fairly autonomous bureaucracies (Jacobsen 1964(Jacobsen , 1966)).Still, history matters.An example is the great compromise reached during the 1930s between state, labor and capital, which has had enduring effects upon the Norwegian political and societal order.For example, the criticism of corporate arrangement notwithstanding, the Norwegian government's new strategy for Europe emphasizes the importance of also protecting this form of state-society cooperation in the future (Utenriksdepartementet 2018).
Since the start of the Bergen project, much has been written about the temporal dimension: how the past affects the future; how institutions provide an imperfect and temporal order; the role of institutionally embedded memories; and how public administration and formal organizations learn (or not) from experience and adjust to varying and changing environments.Organizations change through institutionalized routines (March 1981).Yet there is a variety of competing interpretations of the role of institutional inertia and sclerosis, historical inefficiency, the legacies of the past in terms of shared experiences and memories, path dependency in historical change processes, and the difficult balancing of institutional order and change (March and Olsen 1989;Olsen 2010).
In spite of claims that the European nation-state and nationalism have caused wars and homicide, the nation-state and the nation have shown considerable resilience and robustness.Nevertheless, today Europe and large parts of the world are in an unsettled historicalinstitutional situation that calls attention to a dynamic, and not a static, view of demo-cratic-political order.There is a need for a better understanding of what factors contribute to institutional flexibility and effective adaptation, or the opposite.What are the specific effects and implications of different organizational forms?Do citizens have the political order and the system of government they want?How are new organizational forms, practices, and identities institutionalized and old ones transformed or eliminated?Which factors can be manipulated, and by whom?
Understanding the quest for political order and the possibility of breakdown of order are fundamental problems in political theory (Wolin 1960).Likewise, the vision of political science as an architectonic, applied discipline is one of the oldest ideas in political thinking (Aristotle 1962;Wheeler 1968).The Bergen approach assumed that analyzing the conditions for deliberate institutional design requires attention to both the demand for order and predictability and the demand for flexibility, adaptation and innovation.Making sense of the shifting importance of deliberate institutional design, historical drift, and routinized change processes involves questions of knowledge of cause-effect relations as well as questions of power.Successful deliberate institutional design thus requires realism with regard to the actual information and analytical capacities of would-be designers, as well as their relevant resources and action capabilities (Roness 1979(Roness , 1995;;Olsen 1982Olsen , 1988Olsen , 1997Olsen , 2009Olsen , 2010;;Egeberg 1984Egeberg , 1989)).
Whereas formal-legal institutional arrangements can be difficult to change, deliberately designing and reforming living institutions and achieving predicted and desired results is under many conditions even more problematic.However, deliberate change in politicaladministrative orders by starving institutions and scaling down the public sector may be a more "effective" process than deliberately designing new institutions and making them do exactly what reformers want them to do, thus achieving predetermined and desired results.
As the faith of the 1960s in rational forecasting, planning and control dwindled, even in Norway, academic studies observed that processes of rational design and competitive selection did not guarantee that the most effective and efficient organizational forms were discovered or chosen.In the political milieu, however, there was increasing trust in governing based upon evaluation and learning from experience, and today academia is again called upon to help solve foundational societal problems through studies of experiential learning.Evidence-based policy-making and innovation in the public sector have become favorite slogans of reformers.Yet, such efforts are usually focused primarily upon public serviceproduction, competitiveness, practical issues, and borrowing from market ideology and the private sector.Less attention is given to theoretical insights and empirical facts about how a well-functioning democratic society depends on a well-functioning political-democratic order and public administration, and the huge challenges now facing democracies, as key institutions and actors are losing legitimacy and public support, and new political and societal dividing lines emerge.
Leaders of every political system try to create an image of a rational and meaningful order, in particular when their legitimacy is challenged.They try to establish and cultivate a belief in the legitimacy of the order and its binding authoritative rules, and thus acquire the voluntary support and seemingly deserved compliance of citizens (Weber 1978: 213).Still, success is not guaranteed.Governing through institution-building is a challenging process, and a successful "architectonic approach" has to acknowledge that experiential learning, like all available alternatives, is an imperfect change process.The past may be uncertain and difficult to interpret (March and Olsen 1975).Experience is a teacher-but not necessarily a good and reliable one-securing intelligence and effective adaptation to varied and shifting environments (March 2010).The attentions of reformers are easily distracted (March and Olsen 1983).Historical junctures and external shocks are likely to generate demands for comprehensive design and reform.However, "constitutional moments" tend to attract many participants and issues so that major deliberate design is difficult to achieve (Olsen 2007b: Ch. 9, 2010).
For students of public administration and political organization there is still much to learn about factors that facilitate deliberate institutional design and reform of political orders, as well as factors that drive institutionalization and de-institutionalization processes in general.The vision of public administration as an "architectonic discipline" is of great importance for both theoretical and practical reasons.Yet, students of public administration aspiring to contribute to realize the vision, face two democratic challenges.First, to explore under what conditions design processes are likely to help modern democracies move closer to the old vision that citizens as equals shall decide how they are to be organized and governed politically.Second, under what conditions design of institutions are likely to make it possible for democracies to fashion democrats-democratic identities-in a democratically legitimate way.

AN OCCASION FOR LOOKING AHEAD
The study object has changed considerably since public administration was established in Bergen.Holding the two power studies together calls attention to important changes in political organization, government and public administration in Norway and in Europe (NOU 1983: 3, NOU 2003: 19).Studies of the public administration of a small unitary state with a relatively homogenous population have been expanded to studies of multi-level, multi-centered and multi-cultural polities.
The name of the department has also changed.In 1985, Institutt for offentlig administrasjon og organisasjonskunnskap became Institutt for administrasjon og organisasjonsvitenskap.Public was no longer part of the department's identity, as expressed by its name, possibly indicating a wish for a more loosely coupled relationship to political science and studies of democratic government and power-relations.The initiative for the new name did not come from the faculty; nor did the new name, to the best of my recollection, have any immediate effect upon the work practices of most of the faculty.
The name shift nevertheless reflected international developments in practical-political discourses, with more negative attitudes towards democratic politics and public administration.From the end of the 1970s, the welfare state and the public sector came under attack from both the left and the right.But the main development was towards a more liberal order, a shorter social contract, and more individual responsibility (Dahrendorf 1988).Internationally, New Public Management-inspired administrative reforms from the mid-1980s gave priority to economics, markets, privatization, devolution, managerialism, efficiency, and incentive-driven behavior (Christensen andLaegreid 2001, 2002).It was also claimed that the nation-state, like the city-state before it, had become obsolete.For some, European integration and international institution-building was the solution to new challenges.
The new name also reflected an international drift in the study of formal organizations, with a growing attention to, and idealization of, private enterprises in competitive markets and private leadership, entrepreneurship and management, together with a more critical view of Weberian bureaucracies and democratic-political, majority-based institutions.Arguably, the move of mainstream organization studies to business schools (Augier and March 2011) also contributed to scaling down the differences between public and private administration, and to a growing hegemony for ideas from private sector in many academic quarters.
These developments have, in turn, been criticized by studies analyzing scholarly developments transcending New Public Management (Christensen and Laegreid 2007).The belief in self-regulating markets has dwindled.European integration processes have come under attack after a series of crises in the last decade.There is a perceived democratic deficit in the European Union, that key institutions of have lost legitimacy (Olsen 2017(Olsen , 2018)).Some are afraid of a possible breakdown of not only the integration project, but also of democratic order and democratic control over future developments.The European continent is seen by many to be at an historical juncture, facing a new era.As children of the Enlightenment, Europeans face a culture emphasizing human will, understanding, and control.In practice, Europeans face a world increasingly characterized by ambiguous and conflictual goals, uncertainty and fake news, and limited democratic control.
In such a situation, do available analytical frames capture empirical observations and do they give guidance for future research on public administration and political organization?A challenge for the discipline's young lions is: Shall the Bergen project stick to its roots, emphasizing the search for theory-based empirical knowledge about public administration, political organization, governmental institutions, democracy, and power?Shall the project move in the direction suggested by the shift of name, toward emphasis on administration, organization and organizing in general, with less attention to the specifics of a public administration supposed to act on behalf of the public at large and according to democratic norms of appropriateness and accountability?Does what was once called public administration and organization theory need a new identity, subject of inquiry, mission and purpose?A reinterpretation of the old ones?Or something completely new, adapted to the current situation?Is the project still going to be a project closely connected to political science, and, if so, how much help is there to be had from political science as "a discipline divided" in several schools and sects (Almond 1990)?
Political science certainly needs a good understanding of public administration, political organization and organizing, and how problems, solutions, roles and power-relations develop and change over time and across societal contexts.The purification of ideas about public administration as an obedient tool for shifting democratically elected leaders, as an autonomous and competent rule-following entity, as neutral, technically means-end experts, or as a self-interested power group, are unlikely to capture how public administrations are organized, work and change in modern democracies.It is far from obvious what theoretical ideas from organization theory and democratic theory, if any, are able to tie students of public administration and political organization together in an increasingly specialized academic world.
However, to the best of my judgment, the future of the department is less than secure without an identity that unifies the various specialties and makes it clear what special contribution the Bergen program will contribute in the future to the understanding of public administration and democratic government, political order and change.Today the social sciences face demands for both new theoretical ideas and knowledge that practicians see as useful knowledge, even in situations where there is disagreement and struggle over what "useful knowledge" means and implies.
Furthermore, in the European context, there are public concerns about a European continent and a world in disorder.Disruptive forces are seen to threaten the existing order and it is uncertain whether established institutions are robust enough to secure order, rule-following, and predictability.Competing roadmaps are suggested: • Some advocate a New Deal: political programs combatting increasing inequalities in welfare and life chances through regulation and redistribution.This is a policy-focused return to the distributive aspects and a view of politics as Who gets what, when, how (Lasswell 1958) that requires attention to decision-making institutions.• Some believe in a New Narrative: developing new legitimizing ideas and politically integrating myths through deliberation, sense-making, political contestation, and negotiations.That is, explanations and justifications of the existing order, ongoing developments and reform projects, and thus who we are and who we want to be.This is a culture-and identity-focused approach that calls attention to educational and socializing institutions, settings where democratic citizens can form and develop their identities as democratic citizens-rulers and ruled.• Others, again, demand a New Political Order: calling attention to the whole arrangement of institutions and their interdependencies, interrelations and interactions.This is a political organization approach focused on shifting institutional power-balances between local, national, European and international levels of governance and between institutional spheres, i.e. between majority politics, public administrations, corporatist arrangements, courts, science, religion, markets, and civil society.A challenge is to analyze the shifting mix of founding normative and organizational principles and what are legitimate participants, issues, resources and power-relations in different contexts.
I think the current situation requires a political organization-focused approach.I also think that Norbert Elias (1939) was right in arguing that the civilization of Europe has involved civilization of both institutions and individuals.If so, future developments-including the possibility of a breakdown of the European integration project and democracy as the dominant principle of legitimation-depends on the qualities of institutions and the qualities of the citizenry.These qualities are now being tested, including the ability to learn from experience and to adapt intelligently to varying and shifting circumstances, while remaining democratic communities capable of civilized dispute resolution.There are no theories or empirical evidence that make it possible to predict in any detail the outcomes of ongoing contestations, change processes, and reform attempts.Today, reasonable people cannot ignore the possibility that the world will face a nuclear disaster, due to an accident or a misunderstanding.Furthermore, the Western hegemony when it comes to political thought, action and institutions-good government and public administration, good citizens, and the good society-are being challenged internationally.
Still, the situation is not without hope.There are analytical tools that may give constructive contributions to an improved understanding of political order and change of great theoretical and practical importance.I am also sure that studies of current European and global transformations will teach students of public administration and political organization important lessons about the varying and changing roles democratic institutions and citizens can have in political-administrative life under modern conditions.
I certainly agree with the title Dick Scott chose for the inaugural lecture: "Institutions matter: Now more than ever".I hope that the new generation in charge will do their part of the job of clarifying the roots and routes of the Bergen project, its contributions over the last half century, and possible road-maps for the future.I also hope they will get the help they need from friends in Norway and abroad.It is, now more than ever, important to understand how public administration and political organization may contribute to civilized co-existence for coming generations.

LITERATURE
. It is not always obvious what "the people" want, what normative criteria they use, under what conditions what the public want makes any difference, and whether end-results depend on how citizens participate or are represented in public policy-making.

( 7 )
Organized interests integrated into governmental institutions