Public participation in EIA and attitude formation

Abstract

Although public participation in EIAs aims for public acceptance, it does not guarantee the project (of which the EIA is part) is accepted. Moreover, when citizens participate in EIAs, they develop attitudes towards the project itself, and towards the broader policy. This paper addresses the complex relationship between participating in EIA studies and accepting policies. We first discuss what attitudes can develop towards a project, and how they may influence the perception of the broader policy sector and the attitude towards policy processes. We distinguish four attitudes and show they are present in a dike-relocation case in the Netherlands. We further show that different experiences in the EIA led to different project evaluations and this impacted stakeholders' attitudes to participation, projects and policy-making. We found four aspects of the EIA process of this project that partly explain the arisen difference in project attitudes. These aspects are important to consider in the design and management of participative EIAs.

Keywords:

• EIA
• public participation
• project and policy acceptance

1. Introduction

Public participation is institutionalized in environment impact assessment (EIA) procedures and in various governmental policy-making processes, for instance on water and flood management (Héritier 2001, Scott and Trubek 2002, Wood 2003). Public participation is also based on the idea that those who are influenced or affected by a decision have a moral right to be involved in the decision-making process, and often includes the promise that the public's contribution will influence the project's1 outcome (Webler et al. 1995, Sinclair and Diduck 2001, Halvorsen 2003, Vanclay 2003, Creighton 2005, André et al. 2006, Enserink and Koppenjan 2007, IAPP 2007, Pahl-Wostl et al. 2007).

The aims of public participation in EIA vary from improving the quality of plans and projects, improving implementation by preventing litigation and costly delays and meeting legal requirements, to improving active citizenship, complementing democracy, protecting individual rights and creating acceptance for the project's outcome (Klijn and Koppenjan 2000, p. 109). It has been argued that public participation can help broaden the information base of ecosystem assessments (Fabricius et al. 2007), and might be a possible remedy for Not-In-My-Backyard (NIMBY) sentiments as it may lead to acceptance for new policies (Beierle and Cayford 2002), policy innovations and tough decisions among the public (Dietz et al. 2003, Irvin and Stansbury 2004). The merits of the process are often seen as decisive for the success or failure of public policies.

1.1 Acceptance

Stakeholder participation is used to gain acceptance of projects. This goal legitimatizes (democratically) the spending of resources and energy for organizing participation processes (Woltjer 2002). However, public participation does not guarantee acceptance, as both proponents and critics of public participation have argued (Monnikhof and Edelenbos 2001, Enserink and Monnikhof 2003, Innes and Booher 2004). In some cases, public participation has contributed to project acceptance; in other cases public participation has negatively influenced project acceptance, especially when the system prevents substantive participation in the first place (Carnes et al. 1998, Roberts 2004, Dietz and Stern 2008). Participatory EIAs can fail to generate acceptance when participation is considered a formal procedural requirement. In such cases public participation is reduced to a pro-forma exercise, and public concerns are not truly included in environmental decision-making (Shepherd and Bowler 1997, Wilkins 2003, Hartley and Wood 2005).

1.2 Public policies

Attitudes towards projects and policies are interrelated (Ostrom 2005, Wüstenhagen et al. 2007). However, empirical studies on this relationship are rare. We expect that public participation in project EIAs not only leads to acceptance or rejection of these projects by the stakeholders but also to acceptance or rejection of the broader policy. Stakeholders negative experiences in a project EIA may lead them to disengage and withdraw from future projects. Since good governance obviously is a main motive for public participation, disengagement of stakeholders is counterproductive. Therefore, this paper addresses the complex relationship between participating in EIA studies and accepting policies. We first discuss what attitudes can develop towards a project, and how they may influence the perception of the broader policy sector and the attitude towards policy processes. Four attitudes are distinguished that can develop in a single project and we demonstrate their presence in a dike-relocation case in the Netherlands. We will show how different experiences in the EIA led to different evaluations at the project level and how this impacted stakeholders' attitudes to participation, projects and policy-making.

2. Project attitudes

A variety of actors or stakeholders are involved in any project, with different attitudes towards the project. These differences can be attributed to differences in position, interest and adherence to good governance norms (Hardin 1998, Blair 2000, Woltjer 2002, De Bruijn and Ten Heuvelhof 2008).

An attitude can be defined as a ‘psychological tendency that is expressed by evaluating a particular entity with some degree of favor or disfavor’ (Eagly and Chaiken 1993, p. 1, cited by Johnson and Boynton 2010, p. 19). Attitudes can stem from, and can be expressed through, three sources: cognition, affect and behaviour (Oskamp and Schultz 2005). Thus, a project attitude refers to how a person thinks and feels about a project, and a policy attitude refers to how (s)he thinks or feels about a policy. Although traditionally attitudes were seen as relatively stable, learned dispositions, recent scholars have argued that the stability of an attitude is an empirical rather than a definitional issue (Zanna and Rempel 2010). The stability and longevity of an attitude are dependent on how important the person finds the attitudinal object, and on the source on which the attitude is based. Experience tends to create stronger attitudes than mere cognitive sources such as the mass media (Fazio and Zanna 1981, Eagly and Chaiken 1993). Attitudes have been studied in three different ways: through description, through measurement and through polling. The analysis of attitudes in this paper rests mainly on description and some polling.

2.1 Four attitudes

Among others Scharpf (1997) and Abelson and Gauvin (2006) argued that support for a project is the product of support for the project's content and support for the project's process. Thus, an actor whose interests are not served by a project may not (completely) reject it, because this actor makes a positive evaluation of the project's process, takes his or her loss and/or accepts government's ruling in this decision. Designing ‘better’ processes to gain acceptance of projects is based on the premise that actors do accept negative outcomes if the decision is based on an accepted and legitimate decision-making process (Lind and Tyler 1988, Gross 2007). Problematic in this reasoning is that it implies that acceptance of projects is a one-dimensional concept to which process support or content support contribute, and that process support and content support compensate each other. This reasoning has led scholars to question whether process support or content support contributes most to project acceptance (e.g. Tyler and Lind 2002).

However, we wonder whether an actor with a favourable evaluation of a project's content, but a negative evaluation of the project's process, judges in the same way as an actor with a negative evaluation of the project's content, and a positive evaluation of its process. Do these two different evaluations lead to similar attitudes towards the project and the broader policy sector? We prefer to consider the actor's perspectives on the process and on the content as two separate dimensions of his or her attitude towards a project (see Figure 1). If an actor's attitude towards a project has two dimensions, four possible attitudes can develop, which we call support, acceptance, rejection and ambivalence.

Figure 1 Project attitudes.

Support is the result of an actor's positive evaluation of both the content and the process of a particular project. This stakeholder is content with the decision, will support the project's implementation and may even offer resources for doing so. Moreover, this actor will be positively motivated to support or participate in future policy-making processes.

Rejection means that an actor negatively evaluates both content and process of a project. Such actors will try to obstruct the implementation of the project. If an actor with this attitude believes his or her active resistance is useless, (s)he might lose interest in the project and withdraw from it. This actor will be negatively motivated to participate in future policy-making processes.

Acceptance results when an actor has a negative evaluation of the project's content, but accepts the way the decision came about. This attitude is the desired result of participatory processes when not all actors can benefit from the outcome, and some may be negatively affected by it. Even though the project's outcome is not welcomed the outcome is seen as authoritative or legitimate. The accepting actor will not provide any resources to implement the project, but will not obstruct it either. This actor will be positive about participation in future policy-making processes.

Ambivalence is the ‘opposite’ of acceptance. An ambivalent actor is positive on the project's content, but rejects the project's process (i.e. the way the decision came about). An ambivalent actor will be less willing to participate in future projects, because of the negative experiences with the project's process.

2.2 Project and policy

EIAs are embedded in projects, and public participation is only one element of an EIA. Likewise, projects are not isolated events; usually they are connected to a broader policy and other projects. First and foremost, the project at large and the policy sector need public acceptance, not just the EIA or the public participation process of the EIA. Therefore, public participation in EIA has to be studied as a part of the project that is implemented in the context of a policy (see Figure 2).

Figure 2 Nestedness of EIAs, projects and policies.

In practice, for most of the public, a project EIA is the first and often the only extensive experience with policy-making and with project implementation. Since personal direct experiences are crucial in forming attitudes, it can be expected that participation in EIA is an important factor for developing project attitudes for most of the public. Furthermore, social psychology leads us to expect that an actor's project attitude will affect his or her attitude on the policy as a whole, since the public is usually involved in only one single project in a policy area (e.g. Horowitz 2007, Chen et al. 2010).

Lind and Tyler (1988) demonstrated that attitudes on processes influenced the attitude towards the broader policy. Therefore, we hypothesize that project attitudes are not merely relevant for the present, but potentially have an important impact on the actor's attitude towards the broader policy and thus on comparable future projects. We especially expect both rejection and ambivalence to negatively influence the willingness of stakeholders to participate in future projects. It is in light of these future effects that the project attitude ambivalence gains importance. Although the current project might not be hampered by an actor's ambivalent attitude, we expect this actor to be less willing to participate in future projects. In contrast, accepting actors may be of little use in the current project, but we expect them to be willing to participate in future projects.

The conceptual scheme in Figure 1 can be used to plot the actors' attitudes towards a project. This can be a useful tool to interpret how the development of the actors' attitudes is related to events that take place in a project (e.g. resolution or aggravation of conflicts) and how these attitudes might influence future projects. Analysis of the case study, dike relocation Nijmegen-Lent, illustrates how the scheme can be used. The case study shows that both content and process are important for forming project attitudes, and shows how public participation in project impact assessment can also affect the attitudes towards the policy.

2.3 Case study methodology

A case study was selected to explore the four attitudes in their context (Yin 2003): the design project (2002–2008) about the dike relocation near the village of Lent (the Netherlands). The case study draws primarily on 17 open interviews with participating stakeholders in the project, and on a small survey with non-participating inhabitants of the Lent area. Both interviews and survey were held between October 2008 and March 2009. At that time, the national decision to implement the dike relocation had been taken and the municipality was preparing the development of the detailed plan. This data is supplemented with primary and secondary project documents, such as previous studies, and project reports. Respondents to the interviews represented specific interests – flood management, urban planning, citizens' concerns, and held specific positions in the project – citizens (6), interest group representatives (2), civil servants (6) and decision makers (3) respectively. The interviews started with a personalized description of the project, with a focus on content, process and stakeholders, and ended with an evaluation of the project and expectations about future projects. The interviews were fully transcribed and analysed through a sequence of interpretive coding, based on grounded theory principles.2 The survey was held among non-participating inhabitants of the Lent area, with respondents being selected based on the proximity of their house to the project area. A total 80 inhabitants were asked to participate and 23 useable surveys were returned. Half of the respondents lived facing the project area, and half of them lived nearby. The survey was designed for this case study, and included questions about what the respondents knew about the project, whether and why they participated, how they evaluated various aspects of the project and what they expected from future flood management projects. The survey was administered face-to-face to allow recording any relevant extra comments of the respondents. Survey results were documented in SPSS and analysed using descriptive analysis tools, such as frequency tables and cross-tabulations. The project documents were used to reconstruct the project's process and to corroborate the events mentioned in the open interviews.

3. The dike relocation Nijmegen-Lent

In December 2006, the new flood management policy ‘Room for the River’ (RfR) was formally approved by the Dutch Parliament (Min. V&W 2007). Part of this policy was the controversial project ‘dike relocation Nijmegen-Lent’. This project can be divided into five periods:

1.	From 1993 to 2000, the option of widening river beds appeared on the national flood management agenda including some initial plans to take spatial measures in the Nijmegen area. At the same time the city of Nijmegen developed a city expansion plan along the north shore of the River Waal.
2.	Between 2000 and 2002, technical studies were executed into possible spatial flood management measures in the Nijmegen area. Simultaneously, negotiations took place between the state and the municipality on the conditions for cooperation in designing and implementing a dike relocation.
3.	From spring 2002 till the end of 2004, the project dike relocation Nijmegen-Lent was one of the pilot projects of the new policy to-be-developed, ‘Room for the River’. In this period, an EIA was initiated. At the end of 2004 the draft EIA report and the advice were sent to the Vice-Minister of Water Management.
4.	From November 2004 to February 2007, the national decision-making process on the policy RfR took place, which was accepted by parliament in February 2007. The project dike relocation Nijmegen-Lent was embedded in this policy.
5.	In 2008, the city of Nijmegen took the lead in the further plan development of the dike relocation, after a year of negotiations with the Ministry on the conditions for managing the project.

The first four periods will be described in more detail. The focus will be on the EIA process (period 3), which turned out to be crucial to how non-governmental stakeholders developed their project attitudes.

3.1 The pilot project

In 1993, the city of Nijmegen, a municipality in the east of the Netherlands situated on the south bank of the River Waal, made a large city extension plan for the north bank of the river. To implement this plan, Nijmegen incorporated the village Lent on the north bank of the River Waal. Meanwhile, in 1995, large low-lying areas near Nijmegen had to be evacuated due to high river discharges. This event, together with the near-flood of 1993, led to short-term measures and the new policy for flood management called ‘Room for the River’. This policy was designed to simultaneously prevent floods and add to landscape quality, through undertaking spatial measures, such as dike relocations, conveyance channels and flood storage areas, rather than technical measures, such as dike enforcement (Wiering and Arts 2006, Wolsink 2006). Formally, the decision-making process on RfR started when the Vice-Minister announced the new vision in 2000. The policy included projects, subject to normal parliamentary decision-making, and local pilot projects with pre-funding for immediate execution.

Because of the conflicting interests with the city expansion plans of Nijmegen, the Vice-Minister decided that the Lent dike relocation should be a pilot project. She wanted to prevent the city expansion inhibiting future spatial flood management measures (Min. V&W 2000). The Vice-Minister's plans surprised the regional and local governments. By then the city expansion plans had developed to the point where contracts had been signed with various project developers. If these projects were to be stopped, Nijmegen would face large damage claims from the project developers. Following the announcement by the Vice-Minister, the alderman of Nijmegen entered into negotiations on the conditions for Nijmegen's cooperation with the state to implement the new spatial flood management measures. Nijmegen agreed to cooperate with the dike relocation plan under the following conditions: full state guarantees for eventual damage claims from project developers, the construction of a second bridge across the Waal, and the start of an EIA in which other alternatives would be considered and local stakeholders would have a voice.

3.2. Studies

Parallel to this negotiation process, the Vice-Minister commissioned two studies. The first of these was a quick scan, which compared three spatial alternatives: (1) a large conveyance channel to the north of the village of Lent, (2) a conveyance channel closer to the village of Lent, and (3) a dike relocation on the north bank of the River Waal (see Figure 3). The local village council, a non-elected citizen group, was informally consulted on the quick scan and expressed a preference for a large conveyance channel. The authorities preferred the dike relocation.

Figure 3 Three alternatives (in black) in the quick scan (Source: RWS 2005, p. 39).

Second, a commission of ‘wise men’, led by the well-known former mayor of Tilburg, Mr Brokx, and supported by a group of specialists, developed the dike relocation alternative preferred by the government in more detail and provided judicial advice on the implementation to the Vice-Minister.

Their so-called ‘Plan Brokx’ consisted of a landward relocation of the levee on the northern bank, and the development of a side channel, which would function as a river arm in times of high water levels. In this way, a peninsula would be created between the side channel and the river, connected to the mainland by a bridge in times of high water discharges (see Figure 4). To implement this plan, 50 currently inhabited houses would have to be demolished and 800 planned houses relocated.

Figure 4 Artist's impression of Plan Brokx (Min. V&W 2003, p. 7).

At a citizen information meeting in Lent in September 2000, the alderman announced that the city extension project ‘the Waalsprong has been saved’. However, since most citizens in Lent had not been informed of the flood management plan, this announcement came as a surprise. The citizens were not convinced that all alternatives had been considered, and disputed the need and necessity of such invasive measures at this point in time. They were supported in their views by the professor-emeritus of civil engineering, Van Ellen, a resident of a nearby area. In reaction to the event and supported by Van Ellen, the citizens designed their own alternative to the dike relocation. Their alternative became known as ‘Lentse Warande’ and was presented in December 2000 at a city council meeting.

The main idea of the Lentse Warande was to dig a side channel in the floodplain and make a spatial claim on a land area near the dike for eventual future dike relocation (see Figure 5) (RWS 2005, p. 27). The side channel would allow handling of water discharge levels up to 16,000 m³/s at Lobith, the current Dutch flood management norm for the River Waal. The citizens' plan was adaptive in character; if climate change and/or developments upstream led to higher norms, additional measures such as a dike relocation could be enacted in due course. Until then, the reserved area could be used as a city park or for non-permanent housing.

Figure 5 Artist's impression of Lentse Warande (Min. V&W 2003, p. 9).

Civil servants from Rijkswaterstaat,3 the province and the water board,4 though appreciative of the initiative, were not convinced by the citizens' alternative. They pointed to negative visual impacts and to the years of uncertainty that house owners in the area would face, knowing their house might have to be demolished at any moment. Moreover, implementation of the Lentse Warande would lead to invasive flood management measures upstream, such as implementing emergency flood storage areas. In April 2002, the governmental actors (Ministry – including RWS, province, water board, region and municipality) signed a contract on starting the project Lent dike relocation. The first step in the project was to start an EIA project, in which the two alternatives, Plan Brokx and Lentse Warande, would be evaluated.

3.3 An environmental impact assessment

In January 2003, Rijkswaterstaat presented the EIA scoping document, featuring the dike relocation and the citizens' alternative, as the two alternatives that would be compared. To guarantee input from interest and community groups, the EIA project organization was structured in three groups:

•	A civil servants' ‘project group’ was responsible for the day-to-day development of the EIA. It was chaired by Rijkswaterstaat and included representatives of the municipality, water board and province. They got technical support from a consortium of two large consultancy firms to write the EIA documents.
•	The ‘steering group’ was formally accountable and consisted of politically responsible officials from the regional governments and a high-ranking civil servant from Rijkswaterstaat. They had to deal with any political issue that might appear during the EIA and had to give a regional recommendation to the Vice-Minister (as input for the national decision-making process). The steering group met irregularly.
•	The ‘advisory group’ consisted of invited interest and community groups having a stake in the project. The chair was independent and a non-voting member of the steering group. In the advisory group local issues could be raised and in this way the public support for the plans could be assessed. The group commented on draft versions of the EIA report, was involved in the further specification of the citizens' alternative by the project group, and drafted a recommendation to the steering group on which alternative to implement. During the EIA process the advisory group met frequently.

Two issues played a central role in the EIA: the river discharge norms and seepage in the case of dike relocation:

•

As part of the RfR policy, the flood management norm had been heightened to 16,000 m3/s at Lobith. Anticipating climate change effects, the RfR even had a policy outlook on a future (year 2100) safety norm of 18,000 m3/s. Because of the perceived increasing difficulty to implement spatial measures in Lent in the future, the government chose to use this heightened 18,000 m3/s norm in the project Lent dike relocation. The need for this heightened norm was disputed by the community groups of Lent. They emphasized that their side channel lived up to the current norm of 16,000 m3/s and stressed the big uncertainty of climate change effects. They argued on behalf of the adaptive character of their alternative; additional measures could be taken if, and only if, circumstances require them. And, initially they did not stand alone. Since intrusive flood management projects had been justified by this 18,000 m3/s norm, the province and the bordering regional German government had a discharge levels study executed themselves. This study showed that the maximum discharge level possible at Lobith would be 17,000 m3/s (Provincie Gelderland et al. 2004). In light of this outcome the province ceased its resistance to the 18,000 norm arguing that discharge levels over 16,000 m3/s were possible. In contrast, the community groups interpreted the same result as a strong argument not to accept the 18,000 m3/s norm, since that discharge level was not considered realistic. They argued to stick to the 16,000 m3/s norm and to include options to take more ‘invasive’ measures only if higher discharges became reality.

•

The area affected in the dike relocation alternative has a complex straticulate soil structure. Consequently, it runs a high risk of seepage. Seepage is a known problem in the current situation too, but the current houses and public infrastructure are adapted to these circumstances. Community groups were worried that the seepage problem would migrate landwards after dike relocation and reach areas with houses not designed to be resilient to seepage. After an explorative study with a non-decisive outcome on the occurrence and effects of seepage, Rijkswaterstaat included a minor budget reservation in the dike relocation plan for eventual and not yet known solutions for this problem. A more extensive study to find a definitive solution was deemed not necessary at this stage. The community groups, however, interpreted this ‘solution’ as sheer arrogance of Rijkswaterstaat. They felt that their concerns were not taken seriously, since no answer was provided on how the plan would solve this issue.

3.4 National decision-making process

Shortly after the regional recommendation was received by the Vice-Minister, she decided to incorporate the decision on the Lent dike relocation in the policy document Room for the River, which was up for decision in parliament. This was done for two main reasons: (1) due to delays, the decisions on the Lent project and the RfR policy nearly coincided, and (2) the choice for a dike relocation could be explained more easily in the context of the RfR policy than by itself. As a consequence, the EIA described above was disrupted.

In the build-up to the decision-making process in parliament both citizens and one of the involved consultants spoke to members of parliament on the Lent dike relocation. In January 2007 parliament approved RfR, including the Lent dike relocation. Following this policy decision, the Minister requested the municipality to take responsibility for the further development of the dike relocation.

4. Evaluating project attitudes

The above outline mentions a wide variety of actors involved, all pursuing their own interests and participating in different ways. During the project, all four possible attitudes towards a project – ambivalence, rejection, acceptance and support – developed among the involved stakeholders (see Figure 6). The different attitudes, derived from the interviews, are described below and in Box 1.

Box 1. Project attitudesThe EIA process was key for the development of the attitudes of citizen group representatives towards the project. In contrast to the expectations of the civil servants, the EIA fuelled the negative feelings about the project. In particular, the relation between the project group and the representatives of the citizen groups became strained early on, which led to an atmosphere of distrust. The community groups perceived the process as unfair, because they felt that:

The decision for the dike relocation had already been taken, which they saw reflected in the dominance of the Brokx alternative, the heightened norms, the name of the project and the a priori preference of the civil servants.

They needed to battle the civil servants in order to be heard or even gain access to essential information. For instance, the advisory group initially was denied access to the financial information on the plan. A compromise was reached only after a threat by some citizen groups to leave the advisory group, when one member was granted access to the information.

Civil servants twisted information favouring ‘their’ preferences. For instance, the advisory group's recommendation to the Vice-Minister was accompanied by annotations from the project group, in which they explained what they had done to address the issues mentioned in this recommendation letter. This annotation was interpreted by the citizen groups as an attempt of the civil servants to manipulate the outcome. They felt they were not taken seriously.

Civil servants, on the other hand, felt that citizens were biased in handling information; stressing what was favourable to them, while dismissing information that was unfavourable to the citizens' alternative. Some civil servants acknowledged that in the atmosphere that developed it was tempting to do the same.

5

Figure 6 Project attitudes of the interviewed stakeholders.

The civil servants who participated in the project and the decision-makers of the various local and regional governmental bodies and the representatives of two interest groups supported the project. Civil servants saw the project as necessary, interesting and challenging. The chosen alternative was seen as the most robust, the most environmentally friendly, and providing the best opportunities for improving ‘spatial quality’. The input from the citizens, including the citizens' alternative, was regarded as exceptional, high quality and valuable; nonetheless, the alternative was seen as second best. One civil servant explained: ‘The difference between the two alternatives is that the Lentse Warande is an optimization measure, Plan Brokx is a structural improvement. Lentse Warande would be implemented, if you couldn't do anything else anymore’.

The civil servants especially enjoyed the cooperation with other civil servants from other governmental bodies and the friendly atmosphere in the project group. As mentioned above, most civil servants appreciated the input of the advisory group and regretted the breach of trust that developed over time. Some civil servants did not speak to members of the advisory group during the project, and thus learned of citizen attitudes from the newspapers only. They had a negative impression of the citizens' attitudes and behaviour in the project, qualifying them as NIMBY (Not-In-My-Back-Yard).

In hindsight, the interviewed decision-makers – the aldermen and high-ranking officials – were satisfied with the project's process and most of them are happy with the outcome. All decision-makers accepted the content of the project. The aldermen acknowledged the necessity of public involvement and was satisfied with the result of the negotiations. The director of the water board and the provincial authorities supported the outcome because they were convinced of the plan's flood management qualities. The other interviewed decision-makers were generally satisfied with the process; they only regretted the project's delay.

The interviewed representatives of the interest groups were quite positive towards the project and its outcome. One group was indifferent to the two alternatives; the other favoured the initial plan, because they supported the RfR policy. The latter complimented the flood managers for their decisiveness in this project, indicating he had had bad experiences with other interactive processes. Admittedly, this latter interest group representative did acknowledge that his evaluation of the project might have been different if they had favoured the other alternative. Two representatives indicated having decided to become less involved in the advisory group during the process because of the intensity of the meetings and the extreme level of detail when discussing the alternatives.

Two citizens developed an ‘acceptance’ attitude. These two had been involved in other policy projects before, and felt it was important to stay on the ‘bandwagon’. One of them felt that at least some of her input had been taken into account. She commented that those citizens who had put all their efforts into promoting the Lentse Warande would be less satisfied. Four other citizens developed a ‘rejection’ attitude; they had the impression that the real decision had been taken before the public participation in the EIA. As a result, they felt powerless, manipulated or used, and therefore they questioned the usefulness of participation. They showed a declined interest in participating in such projects, although at least one of them continued ‘against better judgment’. With respect to the content, they questioned the necessity of the measure and thought the negative impacts of the plan were too big to justify implementation. Also, they were disappointed in how their input had been used in the project. Except for one of them they abandoned active resistance towards the project. Yet the interviews revealed that the citizens did not withdraw because they accepted the policy, but rather because the opponents became acquiescent, grounded in a feeling of powerlessness.

The accompanying survey showed that 65% of the respondents preferred the citizens' alternative, and 79% indicated that the governmental actors made insufficient use of the citizens' input in the project. Only 35% of the respondents expected that in future projects the input of stakeholders would be used correctly. These results indicate that the negative perspectives of the interviewed citizens were broadly shared in the community. Moreover, contrary to the hopes of civil servants and decision-makers, few non-participating citizens seem to whole-heartedly accept the project.

The two interviewed consultants were ambivalent. Although they ardently supported the chosen alternative, they were not very satisfied with the process that led up to this choice. They were especially critical of the process design of the public participation process. They had advised differently and would have preferred a process with a different role for the participating citizens and interest groups. Rather than involving the citizens in the debate on the most painful issue – dike relocation or not – they suggested it would have been better to have made a clear policy choice in the beginning and then, in cooperation with the stakeholders, make the designed the plan. The project leader (civil servant) acknowledged this process design would probably have been better suited for this type of project. One of the consultants showed a decreased interest in participating in future projects, if organized in a similar fashion. However, he had also had other more positive experiences and thus his decreased willingness to participate was limited to certain types of projects and did not extend to all projects in the policy sector. Although his colleague did not show decreased willingness to participate, he did develop a more negative view of Rijkswaterstaat.

The participative EIA process obviously did not lead to project acceptance. The contrast between the evaluations of the civil servants and decision-makers on the one hand, and those of the interviewed citizens on the other, is noteworthy. The cynicism of most interviewed citizens regarding the usefulness of participation raises concerns for future policy processes, and especially public participation in this area. This concern is reinforced by the results of the small survey, which indicate that the critical attitude of the participating citizens was shared by many non-participating citizens in the area.

4.1 Explaining project attitudes

We have shown that all four possible project attitudes (support, rejection, ambivalence and acceptance) arose in our case study. Support was found with the civil servants, decision-makers and interest groups; rejection was found with most of the interviewed citizens; two citizens though showed acceptance and the experts were ambivalent. What factors can explain these diverse attitudes? We hypothesize that three specific features of this particular EIA process can be considered partial explanations of the diverse attitudes of the stakeholders.

From their stories it seems actors developed different attitudes because they experienced a different process. Apart from a number of public information meetings, the EIA was the only phase in which the citizens were invited to participate actively, while the civil servants and decision-makers were involved in the overall project and/or negotiating with the national government. As a consequence, the EIA process was key to citizens, while for the civil authorities the ensuing national political process constituted the bulk of the process. Consequently the decision-makers, politicians and high-ranking officials stressed the importance of the negotiations before the start of the EIA and of the studies preceding the EIA, rather than the EIA itself; to them the EIA was merely just a step in the process. Once the citizens became aware that the decision-makers attached less importance to the EIA than they [the citizens] did, they felt they were not being taken seriously by the decision-makers. So the citizens' evaluation of the process of the project was primarily an evaluation of the EIA process, whereas the governmental parties evaluated a broader and longer decision-making process on the project level.

Second, it seems the project was a part of a policy shift that was not yet accepted. The different evaluations of the content can be understood from the policy dynamics. This EIA became a platform for a debate on the assumptions underlying the water safety norms, the speed of climate change and the need for spatial solutions. Whereas traditionally flood management policies were a reaction to actual water levels and catastrophes, RfR was developed to handle future events that might never occur. As a consequence, the necessity of this project was motivated differently and the need heavily disputed. Unintentionally, the EIA resulted in a clash between different interpretations of ‘adaptive’ policy. Supporters, like the Ministry, its agency Rijkswaterstaat and the water board not only anticipated on future discharge levels (the 18,000 m³/s norm) but also on the urbanization of the area. Their rationale for a landward dike relocation now was that otherwise spatial measures might become impossible in the future. Moreover, the new policy was seen as a window of opportunity to get rid of the river's bottleneck for shipping and discharge for once and for all. Consequently, the problem formulation of Rijkswaterstaat was tied up with the proposed solution. The citizens did not readily accept Rijkswaterstaat's bottleneck as the problem-to-be-solved. In short, the content of the project was disputed via alternative plans, but essentially the debate was about policy objectives, norms and political considerations. These results are in line with Kane and Bishop's work (2002), which showed that participative approaches can lead to polarization of viewpoints. In our case this happened, because no consensus existed on the problem definition of the project and the approach taken to deal with it.

Third, the participating citizens' main input was their alternative to the plan of Rijkswaterstaat. During the EIA process, these citizens focused on elaborating their own alternative and the EIA was seen as a tool to compare both alternatives. As a result of this process design these citizens could only win or lose everything; their input could either be accepted or completely rejected. The other citizens and the experts could win some or lose some on partial issues and in debates during the overall process. From their perspective, the process was much more satisfactory, since they saw some of their input reflected in the chosen solution. The frustration of the losers was increased by the seepage issue not being solved during the EIA despite their strong concerns. This strengthened the impression among citizens that their input was not taken seriously.

Fourth, from the start, the experts had differing ideas on how to design a participative process and saw the negative dynamics as a confirmation of their ideas on process management and the role of citizens in this particular EIA. The experts were frustrated by this neglect of their expertise. As a result, their willingness to work on such processes has decreased. In short, the strategy of Rijkswaterstaat to have the citizens proposing an alternative plan and to include that alternative in the EIA did not lead to closure, rather it deepened the distrust and the frustration with the participating citizens. Clearly for project acceptance the input of the various participants should in some way be reflected in the end result. The process design should cater for that.

5. Conclusion

Acceptance or even support for projects is considered to be a main goal of public participation, especially in impact assessment. It constitutes a democratic legitimatization for spending resources and energy for organizing participation processes. We have shown that next to acceptance and support, two other attitudes towards projects are relevant: ambivalence and rejection. The latter attitudes sprout from a negative evaluation of the process by which the participation in the project or in the EIA procedure was organized. Thus perspectives towards a project's process and a project's content should both be considered when assessing the attitudes of stakeholders towards projects.

We showed that projects and policy are interrelated through attitude formation, thus adding to the work of Ostrom (2005) and Wüstenhagen et al.(2007). We presented a case study, the Nijmegen-Lent project, in which, over time, all four attitudes developed in the stakeholder network of the EIA process. Furthermore, we showed that indeed these attitudes were not only relevant for this particular project but had consequences for the willingness of stakeholders to participate in future projects and their attitude with respect to the broader policy: a rejecting or ambivalent stakeholder will be less willing to cooperate in future projects. Somewhat distressing is the finding that the negative attitudes developed by the involved citizens are reflected by the attitudes of the broader citizens' community. Although the presented evidence does not allow bold verdicts, the latter adds to the responsibility of the process managers.

Our analysis revealed four factors that contributed to the development of these different attitudes in the EIA procedure. Although related to the Lent case some of these findings have a general character. First, for some stakeholders (the citizens) the EIA was the dominant phase in the decision-making process, for some the first and maybe only time in life to be part of such a process. For other (professional) stakeholders (the authorities and civil servants) it was just one step in a regional planning process or even a minor element in a national policy-making process. In our case, the authorities at least carry the burden of suspicion of not communicating the extent to which the citizens could influence the decisions to be made and allowing the citizens to get the impression of their concerns not being taken seriously. Thus conflict arose about the influence the EIA should have on the whole project. Second, because of its close connection to the national policy-making on Room for the River, the project EIA became a platform for a political debate in which the content to be assessed and the criteria which were to be used became contested. Third, by comparing two alternatives, each one backed by a different constellation of stakeholders, a win-all/lose-all situation arose. For those stakeholders whose alternative was not chosen, their invested time, energy and input were in no way visible in the end result. Fourth, the non-use of the expertise of the experts on process design led to frustration among the consultants. Future research will analyse the relevance of these factors in other cultural settings.

To ensure that the participation process in EIA will lead to increased acceptance of project and policy the design of a participative EIA process should take the above-mentioned aspects into account. For instance, attention should be paid to how the EIA is interwoven in the larger project and policy-making process. From the case studied it became apparent that not taking these aspects into account at the level of an individual project can negatively affect the broader policy and reduce willingness to participate in future policy-making. The legitimacy of policy-making and policy-makers themselves becomes challenged when the project attitudes ambivalence and rejection develop, especially if these negative attitudes propagate to a wider public. Clearly, careful design and management of public participation are essential for a successful project, and can contribute to the legitimacy of policy-making.

Acknowledgements

This research was financed by the Water Research Centre Delft.

We would like to thank all interviewed stakeholders for their time and stories. We would also like to thank Rijkswaterstaat-Oost and Gemeente Nijmegen for the use of their project documents (both published and unpublished) for this research. Finally, we thank our anonymous reviewers for their constructive comments.

Notes

^1. In this paper we use the term (policy) project, to indicate policy processes in which policies are developed, which when implemented directly influence the socio-technical system. In this sense, projects are Ostrom's (2005) ‘operational level’ of policy-making in action.

^2. All opinions and sentiments expressed in the interviews used for this analysis reflected personal opinions and sentiments and did not represent the official positions of the organizations the respondents worked for.

^3. Rijkswaterstaat is the executive administrative body of the Ministry of Transport, Public Works and Water Management in the Netherlands.

^4. A water board is a regional governmental body responsible for flood protection and other water-related issues.

^5. Anonymous transcripts and the codes can be made available for validation purposes.