A new opera house: combining entrepreneurship,garbage can features and windows of opportunity

Instrumentality and garbage can features combined

An instrumental or rational analytical model stresses that decision-making processes are characterized by control and rational calculation (Dahl and Lindblom, 1953: 57), meaning in our case that political and administrative leaders exert tight control over the decision-making process and that their organizational thinking is unambiguous – i.e. they have clear intentions/goals, the problems and solutions are well defined, they have insight into the preferred consequences and achieve the desired effects. This model comes in two versions, one hierarchical, the other negotiational (March and Olsen, 1983). In the latter version, more actors are involved and leaders sometimes disagree over problems and solutions. In contrast to the hierarchical version, this then leads to negotiations and compromises, potentially producing less clear solutions but more legitimacy because of broader participation (Mosher, 1967).

A garbage can model emphasizes that participants, problems and solutions are selected and coupled in unpredictable ways, as organized anarchies or loosely coupled systems, and decisions are therefore difficult to understand and explain (Cohen et al., 1972; March and Olsen, 1976: 27–32). These features are caused by a combination of two factors: (1) owing to problems of capacity and attention, actors are only part-time participants in any decision-making process or choice opportunity; (2) decision-making processes are characterized by ambiguous stimuli. Often actors come and go in front of decision-making opportunities, problems and solutions are defined and redefined in ambiguous ways, decisions are characterized by the use of symbols, ‘superstitious learning’, etc. Conventionally, one would expect problems to look for solutions, but in garbage can processes, solutions may be ready-made and look for problems.

In garbage can processes, timing and contextual factors are extremely important. If actors seek to act rationally in such processes, this is best characterized as individual or local rationality and may lead to problems with collective or organizational rationality because local actors have a narrow perspective and insufficient influence, attention, and knowledge to take a broad view of the larger decision-making context and hence to coordinate or control it (see Allison, 1971).

The garbage can model, which emerged in the early 1970s, can be seen as combining two trends in organization theory. First, it can be seen as an extention and elaboration of the model of bounded rationality introduced by March and Simon in the late 1950s (March and Simon, 1958; Simon, 1957), which was further elaborated in the seminal work of Cyert and March (1963) on negotiation features in decision-making. The garbage can model goes one step further and indicates that formal structure and organizational rationality are less important, meaning that individual choice and temporal context play an important role.. Second, the symbolic aspects of the garbage can model are rooted in social constructivism, but were developed from the 1980s onwards into a broader neo-institutional approach by March and Olsen (1995), adding to the pool of organizational theories pertaining to broad cultural and social processes.

Kingdon (1984) presented a modified version of the garbage can theory and applied it to agenda-setting and decision-making in a variety of policy areas. He retains the framework of problems looking for solutions and solutions looking for problems, but adds the role of policy entrepreneurs and windows of opportunity. This implies more structured and instrumental decision-making processes and increases the ability to predict outcomes (Aberbach and Christensen, 2001: 413–414).

This model is temporal or contextual in nature, that is, there are many preconditions or contexts that have to be combined in a dynamic way to ‘open a window of opportunity’ and possibly to keep it open for some time. Kingdon's (1984) argument also alludes to a more culturally inspired theory, where traditions and historical trajectories may be punctuated by critical junctures and shocks (Baumgartner and Jones, 1993; Selznick, 1957), leading to a new path. This was very typical of the reforms in New Zealand (Gregory, 2003).

This model moves the focus back more from individual choice and rational, atomized actors to a scenario where decision-making processes are structured by policy entrepreneurs acting on behalf of public organizations; in other words, it is possible to strengthen the organized part of the decision-making anarchies and modify the impact of temporal context (March and Olsen, 1989). The instrumental element is also seen in the modification of cultural path-dependency in decision-making processes.

Based on the models presented, we will analyse the decision-making surrounding the building of a new opera house in three ways, or according to three competing and/or supplementary interpretations. The first is rooted in the rational or instrumental model, which assumes that the decision-making processes surrounding such a large public project should be dominated by central and municipal political and administrative leaders. We will also ask whether the process was characterized by hierarchical control, unambiguous problem definitions and stable solutions, or by different interests from inside and outside the cultural sector and by compromises.

The second analytical approach is based on the garbage can theory. Were the decision-making elements in this process – i.e. problems, solutions, actors and decision-making opportunities – ‘dumped’ in the ‘decision-making can’ in a more or less unsystematic and unpredictable way? Did actors find it difficult to consistently stick to definitions of problems and solutions, let alone prioritize systematic and stable participation? An important aspect of this is that problems and solutions can be connected and disconnected in an unpredictable manner. Applied to the opera house project, this might mean that the opportunities for making decisions lay in being able to connect relatively disparate problems and solutions, or in temporarily decoupling or disconnecting some of these, or then recoupling them. Such processes are potentially extremely complex, and the possibilities for controlling them are small.

A third possible interpretation, combining the first two, is that such unpredictable processes are possible to control in part by opening windows of opportunity and capitalizing on them, i.e. utilizing the temporal context. This can happen when a political or administrative leader (or group of leaders) consciously acts as an entrepreneur and tries to connect or disconnect problems and solutions in order to achieve goals. There can also be a hidden logic in many seemingly paradoxical twists and turns of events in a decision-making process, where problems, solutions, actors and decision-making opportunities are continually decoupled and recoupled. Another option in this alternative interpretation is that the process is characterized by negotiation: different actors or coalitions of actors try to influence which problems and solutions are connected to or disconnected from the process, and how windows of opportunity are used.

The relevance of the potential weakness of the cultural sector

As well as using organization theory as our point of departure, we also examine the importance of how strong sectors are, or their ability to influence others. Jacobsen (1965) focuses on the dynamics between sectors in public decision-making. He takes an empirical look at the fisheries sector and agricultural sector in Norway. His perspective is that strong sectors have a dual characteristic: (1) they are good at setting premises (preconditions) for decision-making in other sectors, or in getting other sectors to accept their views; and (2) they are good at protecting their own sector from premises set by other sectors. This means that they are ‘strong’ in comparison with other sectors. The relevance of Jacobsen's perspective for analysing the opera house project is that, historically, the cultural sector has been relatively weak compared with other sectors; it has struggled to obtain funding and its boundaries have been relatively permeable with respect to influence from other sectors. How did this influence the decision-making process related to building a new opera house? What other factors influenced the decision-making process? How important were the territorial aspects of the process in this respect?

It is worth mentioning that the cultural sector in Norway, though weak relative to other sectors, in a comparative perspective, is probably doing rather well. ¹ For a long time now there has been a political consensus that the government should support all the cultural sector's activities, meaning that funding theatres, ensembles, groups, artists, libraries, etc. is a deeply rooted aspect of public policy. This attitude has changed to some extent in recent decades, and the argument has come to prevail that – as was the case with the opera – the concept of culture should be broadened and more sectors and interests should be connected to culture in order to increase the legitimacy of public spending on large cultural projects.

A gradual increase in the number of problems and solutions

The idea of a Norwegian opera with its own opera house first arose in the 1920s. Initiatives taken immediately prior to and after the Second World War emphasized the need for a separate opera and a separate opera house, but even the former was not realized until 1958. The opera rented space in the Folketeateret building (FT) in the centre of Oslo, and for the next 50 years the Oslo opera remained an opera without an opera house. This might reflect the cultural sector's historical weakness with respect to financial resources, and perhaps also that opera enthusiasts were not influential in public decision-making processes. For a long time it was impossible to convince the wider community that staging opera in the FT building was an unsatisfactory solution.

The process that finally led to the building of the opera house actually began after the Western Railway Station (WRS) in Oslo closed down in 1988–89 and talk began of building an opera house at the same location, so in a sense it was triggered by an external event. This was a typical garbage can feature: two circumstances which actually had nothing to do with each other became connected. The Norwegian National Opera commissioned a report in 1989 with a view to developing a land-use programme and site analysis for a separate opera house, and supported the report's finding which recommended the WRS site. The site therefore became the main contender. The opera house project thus appeared on the agenda without there having been strong pressure for a solution. In 1990 and 1991, two inter-ministerial committees led by top civil servants from the Ministry of Culture discussed a future opera house and after reviewing 20 different sites, eventually favoured the WRS site. Hence, having started with temporal coupling, a garbage can feature, the process continued with instrumental initiatives from actors in the cultural sector.

At the same time, cultural institutions throughout Norway led by the then Labour minister of culture began to lobby in favour of the WRS site. This triggered a process of counter-lobbying by members of the Labour Party and the group called ‘Socialist Left Party for Opera in Oslo-East’: thus the Labour party split on the issue of where to build the opera house and an alternative site was proposed. This signalled clearly that actors in the process were looking beyond the cultural aspect and saw the building of an opera house as part of city development (territorial element) with symbolic features, since Oslo east, a less affluent city district, had traditionally been a Labour Party stronghold.

Statsbygg (a public construction and property management company) carried out an analysis and presented its results in 1994. This envisaged three alternative solutions: the WRS, FT or Bjørvika (BV), the last being a site in the eastern inner city along the bay, with a major road nearby. The range of possible solutions thus became established and the fronts were now well defined: Norway's cultural institutions and the minister of culture supported the WRS; the Oslo city council together with the Labour Party, the Socialist Left and the Directorate for Cultural Heritage supported the BV; and some actors with less clout supported the FT site. The preservation order issued for the WRS site in 1995 restricted the options for this alternative – in other words, a new random garbage can-like event influenced the process, weakened the the arguments for the WRS alternative and brought on board potential conservation interests.

At this point, those who supported the new opera house project were worried that the issue was still not high enough on the political agenda, i.e. it did not have enough political backing and needed someone to open a window of opportunity. Various governments in the 1990s were relatively noncommittal in their support. While the mood looked set to swing in favour of the opera house, it was clear that prioritizing a new opera house had to be balanced against other important considerations, both within the cultural sector and relative to other sectors.

The breakthrough, however, came with the Bondevik I government in October 1997, because the incoming minister of culture was unequivocal in stating that the government would submit a report on building the new opera house. She seemed to have relatively broad support for this initiative and was definitely the policy entrepreneur needed. On the basis of a number of studies, the government presented a proposal in February 1998 recommending the building of a new opera house. A peculiar feature of the rationale for the new opera house was that it diverged from a traditional, more narrowly defined argumentation informed by cultural policy and instead followed a much broader line of argumentation, whereby the government wanted a wider audience, indeed a new audience, to become involved in opera. Following such a tack endowed the project with wider social legitimacy and countered the arguments of those who claimed that opera was an exclusive, elitist cultural activity. The symbols used here had a garbage can flavour to them and were one factor responsible for the decision-making premises being removed from a cultural footing. It also created the impression that an opera house would be economically profitable, which has since turned out to be unsubstantiated, for the opera house has had a large running deficit ever since its opening; opponents had often warned of this danger during the decision-making process. Importantly, the parliamentary proposition disconnected a potential argument for the BV option by emphasizing that comparable experience had shown that the opera house was unlikely to have a ‘locomotive effect’ – i.e. to revitalize eastern Oslo. This implied that the WRS was the best option and undermined the broader-based BV alternative.

When the government recommended the WRS site, it played down the limitations imposed by the preservation order, while heavily emphasizing the opportunities for rapid progress and linkage to existing art and other cultural institutions. This latter argument, which was tinged with cultural elitism, was a mélange of elements of cultural policy and a co-location argument. The BV option failed to win the majority vote because it was linked to road reconstruction and the ensuing costs. The uncertainty of progress was also a decisive factor in rejecting this option.

By the time the case was presented to parliament it had been tampered with. The Labour Party and the Socialist Left Party supported the BV site, citing the argument that the location would give the building a monumental presence and would drive the development of the eastern part of the city. This argument was ignored in the government's proposition, but had a broader appeal, meaning that for these actors the opera house was more important as a symbol and vehicle for city development than as an opera house as such. The governing parties, the Christian Democrats, the Conservative Party and the Centre Party supported the government's recommendation. Parliament's treatment of the proposal, which led to there being no decision in principle to build a new opera house, was dominated by a tactical game played by the Conservative Party, which wanted to kill the BV option. In reality, a large majority agreed that the time was ripe for a new opera house, but none of the alternatives received a majority vote because the main groups of actors differed in their decision-making premises and interests. Thus the matter had to return to the government. The government had, in short, not done enough political preparation or secured support prior to presenting the proposal; if it had this might have ensured a majority vote in favour of the government's recommendation of the WRS site.

After failing to win a majority vote on the issue, the government's strategy seemed to be to engage in a new round of detailed examination of the three alternatives, especially the BV option, with its possible decoupling from transportation problems. The government also worked closely with the City of Oslo to this end. In March 1999, a new parliamentary proposal was presented by the government, based on much the same premises as the previous year's proposal. In contrast to the previous year, the government now argued that the BV option could contribute positively to urban development, thus undermining its main priority, the WRS option. It claimed, however, that the opera house could either be built as part of a comprehensive infra-structure development project or as an independent project: the first option was advised against while the latter option was recommended. This implied decoupling the BV alternative from the broader context – i.e. a garbage can feature – which, by the way, was seen as important for the whole opera house project, thus opening up a different option for the project's realization. These twists and turns in the decision-making process definitely bore garbage can characteristics.

The government concluded once again that WRS was the best alternative, but this time, however, the BV was recommended as an alternative solution. It might seem rather surprising that the WRS option continued to be recommended even while the way was being paved for BV, especially since the government could not control whether the BV was eventually chosen. The parliamentary committee recommendations followed the same party-political lines as before, but when the vote was taken, BV was chosen. It was also important for the final result that several other actors were now more open to BV, in particular, the Norwegian National Opera, Oslo's city council, several Conservative Party politicians in the parliament, and a former minister of culture from the Labour Party, who was now running for mayor in Oslo. So the BV alternative served as a kind of decision-making garbage can where different actors could dump their diverse interests.

Building a winning coalition

What characterized the actor patterns early in this process, from the inter-war years through the 1980s, was that they were restricted to actors from a small, limited cultural sector. The failure of these original initiatives reflected the relative weakness of the cultural sector, and did not provide the premises for other sectors or raise a sufficient level of interest in a new opera house for the project to be put on the wider political agenda. Yet it was eventually these actors, mainly from the Norwegian National Opera itself, and also from the somewhat contemptuously labelled ‘cultural elite’, who managed to instigate a process around 1990 that helped to open a window of opportunity. Notwithstanding, they did not play a decisive role with respect to the outcome or the location, other than agreeing that an opera house should be built, i.e. they were not the important actors utilizing the decision-making window of opportunity. These groups' primary cultural-political argument for a new opera house did not feed into the final result because the cultural elite, understood in broader terms, all rooted for the WRS option, at least until near the end of the process, and all underscored more culturally elitist arguments.

Even after the opera house project was put on the political agenda around 1990, government interest in it remained rather lukewarm, although the then minister of culture strongly supported the WRS alternative. The key actor to address the issue and to lead it towards a solution was the minister of culture in the Bondevik I government. Given that the government, on two occasions, in 1998 and in 1999, recommended one location – the WRS – and lost both times, this must be seen as a political weakness. The first time the proposal (bill) was voted on, the three alternatives had not been equally examined, whereas by the time of the second proposal they had been. It may seem paradoxical that the government presented the BV alternative as more attractive the second time around, even though it did not itself prefer this option. A sympathetic interpretation of the matter would be that the government acted as a facilitator for the final BV choice, i.e. ‘piggybacking’ the supporters of the BV option, but was not itself prepared to shoulder the burden of responsibility for that choice, because the WRS alternative looked better, in terms of both progress and cost. The drift towards BV was, however, evident in the fact that several actors supported that alternative who had previously supported WRS, which was also an important part of the picture.

The parliament eventually assumed a central role in the case. Labour and the Socialist Left mostly supported the BV alternative. Particularly in the Labour Party, this required a balance between the cultural elite, who had supported the WRS for a long time, and the parliamentary representatives from Oslo, who supported the BV. By then it had become important to reassure district representatives who were concerned about the costs of road building associated with the BV. The government helped in this matter by insisting on a separate BV opera building project, which endowed it with increased legitimacy, but without necessarily reducing the cost of the BV. So a technical ‘decoupling’ of the opera project from the expensive road project (removing the traffic from where the opera house should be built) helped to garner public support, even though everybody knew that the road development costs would reappear, i.e. the building of the opera house and the road project would eventually be recoupled.

The City of Oslo was also a key actor here in negotiating with the government and backing central actors in parliament. It had a strong interest in building an opera house, but perhaps even more in the knock-on effects such a building would have, not only as a tourist magnet, but also because it would pave the way for massive city development from which both local public and private actors would benefit from the creation of business and apartment space. Hence, from the perspective of urban development, the BV stood out as a more attractive option. This alternative therefore gathered actors with different underlying interests (Lindblom, 1959). Table 1 sums up the main features of the decision-making process.

OPEN IN VIEWER

Table 1.

Main features of the opera house decision-making process

	Definition of problems	Main actors and support for solutions/alternative	Important events influencing decisions
Instrumental perspective	The need for a new opera house	Few supporters for reviving the FT alternative Cultural institutions and ministers of culture support the WRS option Oslo City Council, Labour Party, Socialist Left, Directorate for Cultural Heritage and public/private business interests support the BV alternative	Government twice fails to secure majority for preferred WRS option Formation of a coalition of supporters of BV manages to win majority in parliament
Garbage can perspective	Non-cultural problems like city development and infra-structure changes	Group of actors supports BV alternative	Closure of WRS in 1988–89, Preservation order for WRS in 1995, government makes BV option more acceptable, formation of coalition for city development
Revised garbage can perspective	The non-cultural problems most important, but the cultural ones also part of the equation	Group of actors supports the BV alternative	Opera house on the agenda by cultural actors and active minister opened the window of opportunity Coalition actors jumped through the window – acted as policy entrepreneurs Decoupling of infra-structure project

A winning coalition takes advantage of opportunities

If we use an instrumental perspective to understand the decision-making process related to building a new opera house in Oslo, we can take as a point of departure the importance of hierarchical leadership (March and Olsen, 1983). Overall, hierarchical control by political executives was rather limited in the process. Getting the building of an opera house put on the agenda was not the initiative of the government, but of actors from the cultural sector. Support from the government for the project then remained lukewarm for some years, before the minister of culture finally in 1997–98 played a key role in launching the phase that finally led to the decision. However, the fact that the government twice supported a different location from the one eventually decided on in parliament testifies to limited hierarchical influence.

The explanatory power of a negotiation version of the instrumental perspective is greater given that different coalitions had fluctuating influence over the process (Cyert and March, 1963). Diverse actors from the cultural sector, after having rather little influence over the process for decades, eventually succeeded in getting the building of an opera house put on the agenda in the early 1990s. This group, however, supported the WRS option and therefore did not succeed in getting their interests taken into account in the final decision. The group was also more characterized by internal conflicts when the time came for a final decision. The political leadership was instrumental in bringing the process on track for the final stretch, but was unable to secure a winning coalition with the cultural actors. Instead, the winning coalition primarily consisted of local actors from several parties, but most importantly the Labour Party and the Conservative Party, which out-manoeuvred the government, and also eventually succeeded in bringing on board central actors from the cultural sector. The main reason for this was their deployment of a strategy that presented a broad justification for this option as a potential ‘locomotive’ of development in the eastern inner city. In this endeavour they were actually helped by the government, who made the BV alternative more and more attractive, not least by decoupling the road project from the process. So overall, the negotiative elements of the instrumental perspective seem to explain central aspects of the final stages in the decision-making process (see Lindblom, 1959).

Returning to the framework for analysis, we can then ask: to what extent does the opera house decision-making process display garbage can features, regarding a loose coupling of problems, solutions and actors, decoupling and recoupling of these ‘streams’, and symbolic elements? Over a long period, until the late 1980s and early 1990s, when the process gained a certain momentum, the case revolved around definitions and solutions rooted in cultural policy, i.e. it was a quite stable coupling of problems, solutions and actors. The opera had a temporary building and the main issue was whether this was appropriate and whether a new building was necessary. But the WRS alternative became linked to the opera house project rather coincidentally, through the closure of the railway station, which also fits into garbage can thinking. From 1992 three location options were in the running, yet without the government having done any thorough research on the real need for a new opera house. So the solutions seemed to be presented before any profound analysis of possible problems had been carried out, another classic garbage can feature. The BV option was launched by Labour and the Socialist Left in Oslo via another external premise – the desire to enhance urban development in eastern Oslo, something that historically had had very little to do with building a opera house. This shows that this alternative was a side effect of another, broader process, the political desire, espoused primarily by the Labour Party to develop the city centre on the eastern, least affluent side of Oslo – a coupling which also tallies with garbage can thinking.

When the Bondevik I government initiated the final phase in 1997–98, the decision-making premises were broadened still further to include a social profile, namely, that the opera should appeal to a broad audience. This social profile did not differentiate between the alternatives but merely provided a broader justification for the need to use public resources for the project. This broadening and coupling, just like the city development one, indicated that narrow cultural arguments were not enough to get support for the new opera house, i.e. the project had to be justified on a number of different decision-making premises that were loosely coupled. Or, to put it in a garbage can terminology, several different decision-making elements were ‘dumped’ into the garbage can in a somewhat disorganized and temporal fashion.

Other central features of the final stage of the process, however, also show clear signs of garbage can features. The criteria used in the first parliamentary proposal were primarily technical in nature and lacked any clear cultural policy underpinning. Two subordinate considerations were thus connected to the process: (1) the preservation order on the WRS, and (2) road construction in Bjørvika; the first element undermined the WRS alternative, while the second one accidentally boosted the BV option. The first Bondevik parliamentary proposal disconnected the ‘locomotive effect’ argument, thus implying that the WRS was the best alternative. In the proposal the following year, cultural policy and artistic concerns became more apparent, yet without being explicitly elaborated. It is not easy to see why the WRS was the government's preferred choice based on the way these concerns were presented, i.e. the lack of rational analysis was evident. Far more important for the final result were the technical and systematic arguments, which, together with the decoupling of road construction and coupling of the urban planning arguments to the process, became decisive for the parliamentary majority's vote for Bjørvika. All the actors knew that the whole success of the new opera house depended on removing the traffic from the area, which was eventually brought about by constructing a sea tunnel for through traffic. All the actors also knew that this would cost a lot of money, without much effect on the environment. But the willingness of the rest of the country to shoulder the costs for both the new opera house and the traffic solution was enhanced by first decoupling the two projects, and then later recoupling them again, but this time including a major city development in the area around the opera house as well, which at least symbolically made the whole project look more legitimate. All this decoupling and recoupling are typical garbage can features.

Bringing in the last model, Kingdon's (1984) revised garbage can model, gives more depth to our analysis of the process. Did the executive leaders instrumentally manage to open a window of opportunity and/or take advantage of such an opening? One can say that the government and the various ministers of culture were for a long time rather passive with respect to promoting the new opera house. Even after a coalition of actors from the cultural sector managed partly to open a window of opportunity in the early 1990s, it is possible to say that through the passivity of the government, they participated in preventing the initiative from being put on the political agenda, meaning that an opportunity failed to be grasped.

A clear change occurred when the new minister of culture in the Bondevik I government took the first decisive action in 1997. She was definitely a policy entrepreneur opening a decision-making window, but failed to capitalize on it herself politically, and instead paved the way for an opposing winning coalition. What was striking about the Bondevik government's conduct was that it twice recommended an option, the WRS, but both times this option failed to win a majority vote in parliament. So the government was not able to utilize the decision-making window of opportunity it had itself opened, partly because of lack of political sensitivity, but also because of the broadening of the premise basis and the decoupling of the road/tunnel project from the opera house project, thus opening up for the BV option. Conditions for action are of course difficult for a minority government in cases where it needs to generate a majority, but this event does not testify to skilful political manoeuvring not does it score high on rational calculation.

The second and decisive round in 1998 can, however, probably be better understood by examining the process that characterized the instrumental negotiations. The government, which has a wider interest in the case than building an opera house, central actors in the Norwegian parliament from ‘the Oslo Bench’, especially the Labour Party, and actors from the City of Oslo seem to have entered into an alliance in order to push the case through parliament, i.e. they seized on the window opened by the minister. There were two crucial factors that enabled these actors to work together: (1) that the opera house was linked to the development of the eastern inner city, a powerful political symbol, and (2) that the BV option was defined as a separate, isolated opera project, decoupled from road construction. It is, however, remarkable that a majority decision was reached via this manoeuvre, because the argument of decoupling road construction from it was actually a fake argument, i.e. a kind of symbolic and temporary decoupling.

So rather than a policy entrepreneur capitalizing on a decision-making window of opportunity, as in the image used by Kingdon (1984), a broad coalition completed the job, which in this case makes the legitimacy basis more solid. The winning coalition also managed to obscure the coupling between their own intentions and the outcome (Wildavsky, 1979). In a way it was a kind of win–win situation that the government should have been able to grasp, since building a new opera house, diverting traffic through a sea tunnel and opening up the eastern inner city centre for broad development were obviously in the interests of many public and private actors.

Coming back to our second research question and the theory on the relevance of inter-sectoral influence patterns for the decision-making process, there seems to be evidence in this case that the cultural sector and its actors had problems influencing the decision to build a new opera house. It took the actors from the cultural sector several decades before they finally got the issue put on the agenda in the early 1990s, and even then they did not have a significant role to play in the final decision. For a long time they supported an alternative that was not chosen, they saw other non-cultural decision-making premises dominating and they experienced internal conflicts.