The precautionary principle and management of uncertainties in EIAs – analysis of waste incineration cases in Finland

Abstract

This article examines whether and how the practice of environmental impact assessments (EIAs) serves the idea of the precautionary principle. The article provides an empirical examination and description of how uncertainties and risks were dealt with in conclusive outputs of EIAs of chosen example cases of Finnish waste incineration projects. The analysis focuses on the time when several incineration projects were in the planning phase in Finland and the effects of this development were seen as being controversial. The findings of the analysis are evaluated against a legislative and theoretical framework. The challenges and strengths of the practice are identified. The article suggests that EIA as an environmental policy tool can promote the pursuit and the application of the precautionary principle. In addition, uncertainty analysis and risk communication in EIA could benefit from a shift towards more collaborative knowledge-making.

Keywords::

• the precautionary principle
• environmental impact assessment
• uncertainty
• risk

Introduction

The problem with risk is that it is an abstract concept that refers to the future. It is not entirely here at the present moment; instead, it depends on a multiplicity of choices, which are made at the present moment. It is never entirely concrete and it always leaves room for different interpretations and debate. Parties involved in these debates, such as scientists, experts, journalists, lay people, power companies, or social movements avail themselves to different resources in an attempt to establish their views over the views of others, but looking from sociological perspective they all take part in social construction of risk, which is an ongoing process full of inconsistencies and contradictions. (Litmanen 2001, p. 45)

The precautionary principle can be defined as environmental protection based on precaution, even where there is no clear evidence of harm or risk from an activity (Bell & McGillivray 2006). As a term, ‘the precautionary principle’ is familiar from legal contexts, and many different definitions in the literature exist (see e.g. Comest 2005). Instead of understanding the principle as a rule or code for action, Renn (2008) and Tannert et al. (2007) remind us that the principle should be considered as ‘a state of mind’ which helps decision-makers to be more sensitive to uncertainties, ambiguities and ignorance related to development. The principle does not provide clear guidance for practice, and therefore it is sometimes criticized for being administratively inconvenient (Calman & Smith 2001; Comest 2005). In fact, from a philosophical perspective, Sandin (2007) has concluded that there is no precautionary principle, in the sense of a principle that can help in solving problems, but that it can bring the decision, or decision horizon, into focus. Earlier, O'Riordan and Jordan (1995) suspected that the precautionary principle may not exert the influence it deserves because the concept does not show much coherence other than the spirit of challenging the past course of development.

The environmental impact assessment as an environmental policy tool reflects the precautionary principle in part because it seeks to identify and reduce the uncertainties and negative impacts associated with development (De Sadeleer 2002). It can also be argued that the environmental impact assessment (EIA) should encourage or demand a precautionary approach to development where needed. Within the range of environmental management tools, the EIA is ideally the procedure where uncertainty factors and possible environmental risks should receive constant attention (Lawrence 2003). The EIA system functions as a regulatory framework for actions that should aim to create a more detailed and analysed picture of the situations in which proposed development projects are planned. An EIA procedure is a framework for conducting the mixture of research and information, knowledge gathering and analysis to promote more environmentally friendly decisions. In fact, the EIA could be considered a form of ‘post-normal science’ (for more about the concept, see Ravetz 2004), which has an organized idea of its purposes but has no methodological discipline or possibility to provide testable answers to its ultimate questions. Overall, the EIA is a tool to reduce and/or make visible the risks related to a planned development project. All information gathered and presented in an EIA increases the knowledge about a project and its possible effects. In this way the information reduces the uncertainty surrounding the plans and its effects, and aids in risk evaluation. On the flip side, it has been noted that this reduction is often flawed in practice, leaving socio-interactive aspects of uncertainties unexamined (Duncan 2008, 2013) or, more generally, EIAs are seen to fall short in descriptions of uncertainties or their management altogether (Tennoy et al. 2006).

In this article we examine whether and how the EIA serves the idea of the precautionary principle. We believe that provision of information may lead to better decisions, but as is apparent later in the text, we believe that EIAs need space for constructivism and that focusing on decisions can be a pragmatic way to promote communication. We also understand that EIAs are inescapably concerned with power. Methodologically the main focus is on the empirical document analysis, which describes how risks and uncertainties in EIAs were dealt with in chosen example cases of Finnish EIA practice. The descriptive analysis is provided by looking at conclusive EIA outputs for waste incineration projects from the time when the development was seen as being controversial in Finland but when interest in incineration was on the increase.1 The observations are mirrored against the legal and theoretical framework, which is intended to clarify the research philosophy adopted in this article. EIA implementation is not directly compared with some of its idealized models, but rather the implementation is considered as balancing between different rationales of efficiency and precaution. Legal and theoretical elements of the framework show the tensions between these rationalities. Our framework does not form direct criteria for comparison; instead the aim is to increase the understanding of the problems noted in practice (of the approach, see Snell & Cowell 2006). Then, based on the findings of the document analysis, the article discusses whether and how EIAs could have been used more extensively as a precautionary environmental policy tool in the example cases.

The research framework

Legislative demands: the precautionary principle in EU and Finnish EIA law

The precautionary principle has had a growing role in environmental law and it has been incorporated into several international environmental conventions.2 European Union law also reflects the approach in several ways. The precautionary principle is anchored in Article 191(2) of the Treaty on the Functioning of the European Union. In addition, the principle is explicitly or implicitly incorporated into various environmental directives. One of those is the EIA Directive,3 which is implemented through the Act (468/1994) and Decree (713/2006) on Environmental Impact Assessment Procedure in Finland.

Both the EIA Directive and Finnish EIA legislation contain two concrete elements of the principle. Firstly, a precautionary approach is included in the screening requirements. Under EIA Directive (Art. 2(1)), the assessment must be carried out if the project is likely to have significant effects on the environment. The wording indicates that clear evidence of a significant impact is not needed to trigger an EIA. In addition, Article 4(3) of the directive requires that the national screening system must take into account the selection criteria set out in Annex III. These criteria contain, among others, ‘the risk of accidents, having regard to particular substances or technologies’. Correspondingly, the Finnish EIA legislation requires that the project must be subject to an assessment if it is probable that it will have significant environmental consequences (Section 4 of the EIA Act). The selection criteria for screening are also congruent with the directive (Section 7 of the EIA Decree and Pölönen 2007).

The precautionary approach is also included in the content requirements for the environmental impact statement (EIS). The information to be included in it is specified in Annex IV and Article 5(3) of the EIA Directive. The specification contains a description of the likely significant effects of the proposed project on the environment and an indication of any difficulties (technical deficiencies or lack of know-how) encountered by the developer in compiling the required information. The wording of the directive does not refer to consideration of uncertainties or risk assessment as such, and it indicates that the environmental impact statement need only cover the technical deficiencies or lack of know-how, which the developer has determined.

In addition, the Finnish legislation uses flexible terms in defining the content of an EIS. However, it goes further than the EIA Directive in terms of precaution. By virtue of Section 10 of the EIA Decree, an EIS must contain, on a sufficient scale, an estimate of the environmental impact, any deficiencies in the data used and the main uncertainty factors – including an assessment of the possibility of environmental accidents and their possible consequences.

It is noteworthy that the European Court of Justice has not had an active role in addressing the quality issues related to EIA documents (Commission 2003; Pölönen 2007; Commission 2010). In addition, in the Finnish system the courts have had a very limited independent role in forming the minimum requirements for EISs. The Finnish Supreme Administrative Court has given significant weight to the opinion of EIA authority while reviewing the adequacy of the environmental impact statements (Koivurova & Pölönen 2010).

As a result, the EIA administration plays a crucial role in defining how the uncertainties should be addressed in the EIA at the minimum in Finland. So far the legal duty to provide information related to uncertainty factors has not led to situations where an EIS would have been considered inadequate (see also Pölönen 2007). This indicates that the administration does not interpret the EIA legislation very rigorously in terms of risk assessment.

In all, as a point of departure for the analysis in this article, the EU legislation seems to represent rather a weak version of the precautionary principle. Although the Finnish EIA legislation goes further with its requirements, altogether the legislation leaves significantly room for the actual realization of precautionary considerations.

Theoretical concepts: uncertainties and the idea of risk

Failing et al. (2007) have noted that sound management of uncertainties in environmental risk management is a part of quality assurance. In the context of EIAs, Lawrence (2003, p. 427) has suggested that the issue of uncertainties is not well handled: ‘the problem is a combination of confusion regarding the nature of uncertainty and the related concepts of risk and health effects and ambivalence concerning the most appropriate approach for managing uncertainties in the EIA process’. In EIAs, countless uncertainties can be identified and it can be argued that there is no one right way of action in deciding how uncertainties and risks should be addressed in EIAs (Lawrence 2003). Thus, it seems that, although the EIA itself has been developed to provide a more organized way of addressing these, the practice is often criticized for not answering this call.4

Looking from perspectives of definitions and characteristics of risk and uncertainties, the reasons for these recognized difficulties and claims can be outlined. First, both risks and uncertainties are polymorphous concepts and their significance changes according to the frame and perspective taken. Second, although the idea of deliberation is understood to be the key to wider considerations of them, there is no absolute line to cross for general satisfaction. In holistic EIAs, frames vary and different ways of knowing come into play, and when there is no readily a common way or language to talk about them, even finding one is a challenge.

In order to clarify the concept of risk, Renn (2008) has stated that all concepts of risk feature the distinction between possible and chosen action. The risk refers to the future: the idea is that potential negative or positive consequences may occur and they are all associated with options for taking actions. This is also the core of the EIA: to describe alternatives for the planned development and to evaluate their sensibility. Eräsaari (1997) has noted that, rather than a definable concept, risk is an idea that is useful when we try to square the controversy between knowing and uncertainty. According to him, risk offers us a chance to talk about environmental problems, for example.

Risks are created and selected by people and are mental constructs. They can be approached from several perspectives, for example, the technical (probability), the economic (monetary value), the social (values) or the psychological (fears, willingness to take risks). However, risks are also real. The reality is that, for certain reasons, human lives may be lost, health impacts or other actual harm can be observed or the environment is damaged, for example. This reality and the experience of it are the basis for the thoughts that something, usually undesirable, might happen, for a perception of risk. (Renn 2008).

The perception of risk is both individual and collective. Litmanen (2001) suggests that risk perception is a social process where norms and rules govern social behaviour affecting individuals' judgements on what dangers are threatening, what risks are worth taking and how they are confronted. In addition, risks can be both material and abstract, and while there are no unanimous risk perceptions, they also tend to change over time. It is essential that, in order to discuss the acceptability of a risk, the threat it poses needs to be clarified for or constructed by those discussing it. It can also be argued that, in the EIA, a prerequisite for a meaningful risk discussion is the successful and convincing formulation of the threats.

However, the effects of threat formulation can also be unexpected. Eräsaari (1997), has declared that political rhetoric of risk with its problem framings only reduces a large amount of uncertainties into a seemingly calming illusion. On the other hand, Litmanen (2001) reminds us that, by formulating and exposing a risk, we can also cause a monster capable of causing social unrest. These views encapsulate some pivotal problems EIAs face when their usefulness and functionality are considered. If the purpose is to manage risks in a holistic way, both reduction and imaginative speculation are needed, but their outcomes may not necessarily serve the purpose of ‘being more sure of the consequences’. Moreover, it does not help the situation that the ‘illusion’ of risk cannot often be judged to be correct or incorrect.

A claim of risk is always associated with uncertainty. In fact, in an EIA, risk offers a perspective of significance to uncertainty analysis. Yoe (1996, p. 5) has defined that ‘uncertainty describes any situation in which we are not absolutely sure’. However, in decision-making situations not all uncertainty counts. Failing et al. (2007) suggest that, in the deliberative decision context, it is only those uncertainties that are linked to actual choices that matter; others are usually of little use.

In order to evaluate whether an uncertainty matters, it must be recognized. Different typologies may help in identifying different uncertainty types and provide guidance on what type of tasks should be performed in evaluation. For example, De Jongh (1988) has suggested that the approach to dealing with uncertainties in EIA should consist of at least both the techno-scientific and the socio-scientific elements. The former should focus on identifying and reducing the defined uncertainties mainly through predictions, while the latter should focus more on the subjective elements of EIAs, such as relevance or significance. A more thorough attempt to create help for evaluation can be found from der Sluijs et al.'s (2003) guidance on uncertainty assessment and communication to be used in environmental assessments. In their concept they emphasize that, when selecting an appropriate approach, the purposes need to be considered. In their view, problem framing is essential and typologies and uncertainty matrixes are suggested to provide a common language for viewing uncertainty. The typology (modified by der Sluijs et al. 2003) that they suggest to be useful in uncertainty analysis consists of five dimensions for identification: (1) ‘location’ describes where uncertainties occur; (2) ‘level’ describes where uncertainty manifests itself on the spectrum between deterministic knowledge and total ignorance; (3) ‘nature’ describes whether uncertainty stems from knowledge imperfection or if it is a direct consequence of inherent variability; (4) ‘qualification of knowledge base’ refers to level underpinning and packing of the information (e.g. data, theories, models, methods, argumentation) and characterizes their reliability to a certain extent; and (5) ‘value-ladenness of the choices’ refers to the presence of the values and biases in the various choices involved.

Moreover, a slightly different approach can be drawn from Tannert et al. (2007), who have proposed a taxonomy of uncertainties and decisions, which can be used for analysing the nature of uncertainty and evaluating responsibilities and the opportunities to reduce it. Their idea is that objective uncertainty can be divided into epistemological and ontological uncertainties and subjective uncertainty to moral and rule uncertainties. Epistemological uncertainty is uncertainty that can be reduced by research and is thus a moral responsibility. Ontological uncertainty is caused by stochastic features of situation, usually involving complex technical, biological and social systems. These are characterized by nonlinear behaviour, which makes it impossible to resolve uncertainties by reasoning or research. In these cases it is impossible to make entirely rational decisions, but past experience and probabilistic reasoning can provide some guidance on how the systems may react. The subjective uncertainties in part are often the real reasons behind societal anxiety or conflict. The moral uncertainty refers to situations where the decision-maker has a lack of applicable moral rules to follow and they have to follow more general moral rules. The rule uncertainty refers to situations where the decision-maker has to rely on intuition rather than knowledge or moral rules (Tannert et al. 2007).

From a practical point of view, one challenge is that it is difficult, if not impossible, to say what is precautious enough for the avoidance of harmful environmental impacts in the future. Risk-taking and uncertainty-ignorant practice may be sufficient and may not lead to any harmful consequences. On the other hand, there is no guarantee that relevant uncertainties are recognized, no matter how heavy and thorough an analysis is conducted, and it may be that no project is feasible if the lack of certain proof is considered sufficient grounds for rejection (see e.g. Foster et al. 2000; Lawrence 2003). In an EIA, people have to balance these ends. While studying scoping in EIAs, Snell and Cowell (2006) have noted that the contestability of these rationales of efficiency and precaution problematizes any search for deficits in implementation; therefore, rather than trying to compare implementation with some of its idealized model, it may be more worthwhile to try to explain practice in terms of its tensions between these different rationales.

In addition, in order for an impact assessment instrument to be effective, recent thinking has emphasized the role of impact assessments (IAs) as a socio-political process rather than just an informative knowledge-based process (Cashmore et al. 2008; Jha-Thakur et al. 2009; Partidario & Sheate 2013). This notion largely draws on the conception that, in the context of planning, meaningful knowledge is seen as a collaborative construct. It is considered that IAs should provide opportunities for all interested stakeholders not only to have a voice, but also to have the chance to learn and create new knowledge. The linear knowledge transfer from those who know to those who do not has been found to be problematic in IAs because, when it comes to the future, nobody knows for sure.5 When views are diverse and different things are seen as important, nobody should take the power and stance to determine what counts alone.

In our view, the desire for more collaborative IA processes does not mean that traditional expert-driven research is not seen as necessary, but a different approach to how information is gathered and evaluated is. The traditional way in EIAs, where knowledge is produced to be used by others, often ends up in criticism. This is also typical of traditional scientific research – argumentation holds until weaknesses are found. In the realm of EIAs, weaknesses are rather easy to point out, but whereas in science a weakness usually means that findings do not hold, in EIA the effect is not necessarily the same. For example, with regard to a critical voice, it is often suggested that the concept of uncertainty is weakly communicated in EIAs (Lawrence 2003; Tennoy et al. 2006) and sometimes a worry is expressed that decision-makers and other stakeholders in EIAs are often not appropriately made aware of inherent uncertainty that is unavoidable in the process (Tennoy et al. 2006). This worry is reasonable, but the paradox of balancing and the nature of the concept usually prevent us from saying how things should be done or even whether the lack of quality is significant.

Therefore, the wrong type of criticism does not necessarily improve outcomes. When a requirement for action is blurred, it can be more fruitful to concentrate on the positive aspects of the work of others than on the negative. Within IA, Sheate and Partidario (2010) and Partidario and Sheate (2013) use the term ‘knowledge brokerage’ to describe the approaches they believe are needed in assessments so that they would serve their intended purposes better. In general, they believe in different forms of participant engagement, positive planning and mutual learning, and remind us that we should also look beyond decision-making events – IA processes should be seen more as possibilities for learning than they now are. For example, Duncan (2013) has recently suggested that IAs need institutional spaces for uncertainty disclosure and that facilitating negotiation and deliberation could engender greater legitimacy and credibility for IA outcomes.

In this article the conceptualization for the risks and uncertainties functions as a research tool in recognition of the handling of uncertainty and risk aspects in the descriptive analysis of the conclusive outputs of the EIA cases. The legal and theoretical elements of the framework are summarized in Figure 1.

Figure 1 The research framework.

Analysis of EIA practice from the precautionary perspective

EIS review in the Finnish EIA system

In the Finnish EIA system, quality control for EISs is, in the first instance, a duty of the competent authority (Centres for Economic Development, Transport and the Environment). An EIS is the report where the developer collects the results of the assessment of a project's impacts. At the end of the EIA procedure, the centre, acting as a competent authority, reviews the adequacy and completeness of the EIS and gives its statement on the assessment. A competent authority's statement is the main output of the EIA process where the quality of an EIS is considered. Before the review, a draft EIS is often submitted to the competent authority for unofficial comment.

The public (interested citizens and organizations), municipalities and other authorities also take part in the quality control of the EIS. They have an opportunity to make comments on the EIS and the competent authority must take these comments into account while considering the adequacy of the EIS. The EIA as an environmental policy tool reflects the recognition that information gathered and grounded with rational methods is not necessarily a sufficient basis for planning and for the decision-makers. One of the functions of public participation in EIAs is to broaden the knowledge base for planning and decision-making. Participation emphasizes the role of local knowledge and traditional ecological knowledge in the planning processes. It helps to ensure that different perspectives on planned actions and their potential impacts are accounted for. Ideally, participation should also contribute to the identification and, if possible, reduction of uncertainties associated with development. Consequently, it can be argued that precautionary thinking is strongly included in the concept of participation.

Usually, competent authorities attach all or most of the received opinions and comments to their statement. Courts have placed significant weight on the opinion of the competent authority when reviewing the adequacy of environmental impact statements. The case law of the Supreme Administrative Court indicates that it views the competent authority as the primary controller of the quality of EIAs (Pölönen et al. 2011).

The assessment procedure is completed when the competent authority delivers the EIS and its own statement and the other opinions on the EIS to the developer. The EIA is linked to the permit procedures, so that no consent according to any act can be granted unless the EIA has been carried out. The linkage between the EIA and decision-making is also strengthened by a requirement of the EIA act that relevant information gathered during the EIA process must be taken into consideration in the permit procedure.

As a whole, the competent authorities play a central role in quality control of EISs and in guiding the EIA procedure in Finland (see also Hokkanen & Jantunen 2012). Ideally, the procedure and the review system should take different perspectives of assessment into account and allow communication between different stakeholders. In the EIA guidance6 the authorities are encouraged to pay attention to and demand that developers clarify uncertainty aspects of planned actions.

Background for the cases: role of waste incineration in EU and Finnish waste policy

Waste management issues have played an important role in European Union environmental policy from its very beginning (see e.g. the first environmental action programme, 1973–1977). European waste policy is largely based on a ‘waste hierarchy’ (Table 1), which was clarified in the revised Waste Directive 2008/98/EC. According to this hierarchy, the prevention of the generation of waste is the first priority, followed by the preparing for reuse and recovery of waste, and finally, by the safe disposal of waste (Article 4(1) of the Waste Directive2008/98/EC). Priority amongst recovery operations should be given to re-cycling7 over other recovery operations such as energy recovery.8 This is based on the argument that material recovery has a greater effect on waste prevention than energy recovery (COM(96) 399 final; European Commission 2007).9 However, it should be noted that the hierarchy of recovery methods should be applied with flexibility, taking into account the options that deliver the best overall environmental outcome (see Article 4(2) of the Waste Directive).

Table 1 The waste hierarchy in EU waste policy (European Commission 2007).

1	Prevent waste in the first place	Best option
2	Reuse the product or components
3	Recycle the material
4	Recover the energy (waste incineration)
5	Dispose of the product (landfills)	Worst option

The waste hierarchy is implemented and promoted through several legal provisions which have direct and indirect effects on the role of incineration in EU waste policy. Among others, the Landfill Directive (99/31/EC) sets limits on landfilling and hence creates a need for waste prevention, reuse and recovery. It seems to have significant implications for the role of incineration, because it introduces substantial restrictions on the amount of biodegradable waste going to landfill. Article 5 of the directive requires that the waste strategies of EU Member States must ensure that the amount of landfill-disposed biodegradable waste decreases in weight to 75% from 1995 levels by 2006, to 50% by 2009 and then to 35% by 2016. In Finland, the recent waste policy seems to be directed by the assumption that these explicit demands are rather difficult to achieve without a significant increase in waste incineration levels (Koskinen 2006; Ministry of Environment 2007). This has raised the question of whether compliance with the Landfill Directive, through a considerable increase in waste incineration, leads to waste management contrary to the waste hierarchy. It is perhaps not surprising that one of the main arguments against energy recovery as a major waste management option is the argument that the existence of incinerators discourages waste prevention and recycling (see e.g. Suomen luonnonsuojeluliitto 2008).

In addition to the implementation of the Landfill Directive, the increasing interest in waste incineration in Finland can be explained by the low number of existing incineration plants. In the early 2000s, there was only one plant for mixed municipal waste in use in Finland, whereas similar plants were quite common in other European countries at the time (Meloni 2004; Koskinen 2006; Cewep 2007). One of the reasons for the situation was the former legislation on waste incineration (which set strict environmental requirements on incineration plants if the waste fuel exceeded 30% of the energy fed to the boiler). Because of the legislation, preference was given to co-incineration plants instead of plants for mixed municipal waste (Koskinen 2006). However, the situation has seen some change, and in 2010 there were three plants for mixed municipal waste, 10 co-incineration plants and one plant for hazardous waste in use in the country (Ministry of Environment 2010).

A new national waste strategy for Finland was accepted in spring 2008. According to the strategy, energy recovery from waste will be increased. Waste incineration is considered an acceptable management tool as long as recyclable materials are not incinerated (Ministry of Environment 2008). Regional waste and land use planning, including SEAs, project-specific EIAs and environmental permits, will solidify the broad frameworks in the national waste strategy. These processes will play a vital role in the future of waste incineration, waste prevention, reuse and recycling.

The empirical analysis in this article focuses on the early 2000s, when interest in waste incineration as a waste management option was on the increase in Finland. At the time, different stakeholders saw the project type and its role amongst other management options as controversial, and discussions about its effects often resulted in conflicting views; the projects were seen as both harmful and beneficial (Meloni 2004; Koskinen 2006; Saarikoski 2006). Saarikoski (2006) has suggested that opponents and proponents of waste incineration had a fundamental difference in focus, the former wanting to address the question of waste minimization, with the latter already focused on incineration as an appropriate waste management option. As such, the situation could be considered a classic example of a situation where the whole justification of a project is questioned. In these types of cases, EIAs often face criticism of and dissatisfaction with their focus, which by definition is at project-level. In addition, Jalava et al. (2010) evaluated the quality of these projects' EIAs that were completed between 2001 and 2005. In their study they used the European Commission's guidance and review checklist (European Commission 2001) to evaluate the fulfilment of the content requirements of the projects' EISs and also the quality of the contents. From that viewpoint, they deemed the outputs to be more positive as regards quality than negative. The EISs showed that project developers had considered the required contents, but from a critical perspective it was noted that the EISs proved to be more descriptive than analytical. This relative finding, however, did not generate many practically useful conclusions. In general, the study concluded that outputs and speculation about impacts could have been improved through better communication amongst stakeholders. In the following sub-sections, these example cases are looked at from a slightly different perspective – the empirical examination aims to describe how risk and uncertainty issues were presented in these EIA processes' conclusive outputs in order to give effect to the precautionary principle. The research material consists of 15 environmental impact statements and competent authorities' statements about the EISs.

Consideration of uncertainties and risks presented by the developers in EISs

It appeared that the developers and the consultants had shown recognition to both uncertainties and risks in the studied EISs. ‘Uncertainties’ and ‘risks’ as concepts were understood and presented separately by the developers and were not mixed together to confuse readers. However, we did not find any clear evidence of extensive uncertainty analysis or risk analysis in any of the statements. Instead, the developers described examples of risks related to proposals as well as different uncertainty forms and sources that may affect the results of the assessment.

What was common for the risk descriptions was that they were described as situations with something (possibly) happening with possible or probable negative consequences. The risk descriptions were qualitative and mostly there were two types of risks explained. First, a distinct group of risk descriptions focused on different incidents that may occur during the construction phase or operation phase of a plant. For example, accidents such as fires, chemical leaks and methane explosions were mentioned as being possible. The consequences of such events were usually not specifically outlined. Instead, some prevention measures against these situations were listed, including, for instance, alarm systems, staff training, technical safety solutions and surveillance guidance.

The second type of risks found to been explained were the risks related to the predicted impacts. These explanations showed that different types of environmental risks were considered. For example, threats to scenery, air quality, surface and ground water, species on site and human health were evaluated. Typically the air emissions were described as being the most significant cause for probable impacts and it was usually predicted with calculations that the emissions for different components will remain below boundary limits and therefore they will not cause harm. If these impacts were described, they tended to be mentioned at a general level – for instance, by describing how emission components may cause health problems to humans.

Altogether, these findings indicate that the risks were explained in the EISs mainly using the same expressions as used in the content requirements of the legislation. At the same time less effort was accorded to handling of other risk types or to considerations with changing perspectives.

The common way developers described uncertainties was to explain that assessment always includes assumptions and generalizations, and it was usually noted that uncertainty exists in EIAs at many levels. It was also acknowledged that a lack of data and information can cause uncertainty and inaccuracy in assessment results.

In addition, in all of the EISs, the significance of uncertainties related to specific evaluated impacts was considered. In practice, this meant that an opinion about the reliability of information presented was announced. In 14 statements an overall statement was given: despite the fact that it was considered, uncertainty does not significantly affect the general EIA conclusions. Table 2 represents our interpretation of those instances where a developer clearly expressed that a certain issue had been considered in relation to some form of uncertainty. The table shows that in only four of the EISs (numbers 1, 2, 3 and 7) were these considerations explicitly explained under the heading ‘uncertainty’ for more than three topics (in addition to the main conclusion).

Table 2 Uncertainty considerations related to certain topics presented by the developers in the examined EISs.

In many cases choice-dependent uncertainty was recognized, but it was not seen to affect the evaluation or its results significantly (see the dark grey boxes in Table 2). Typically this uncertainty was seen to disappear after a choice was made. For example, in some instances it was mentioned that, owing to an early planning phase, the outlook and the size of a power plant building could not be described exactly. In some cases it was announced that the choice of the gas cleaning system would have an effect on the cleaning results.

In two statements uncertainty related to a certain evaluation topic was said to have affected the evaluation in a way in which a proper assessment considering the issue was not possible (see the black boxes in Table 2). In one case it was acknowledged that the health impacts of power lines were not properly evaluated because there was no clear evidence of their effects. On a general level this issue was also referred to in other statements. In the second case it was recognized that the effects of ash could not be properly evaluated because the amounts and qualities of ash being generated were unclear.

Regarding the consequences of the whole project and despite recognized uncertainties, the main conclusion expressed in every EIS examined was that the proposed project is feasible. However, in two different projects, one alternative in both was deemed unfeasible. In the first unfeasible case, the planned location was deemed inappropriate owing to public resistance and cultural values. In the second case the alternative was a coal plant, which attracted more public opposition than the main alternative – a bio fuel plant (wood, peat and construction waste as fuel). Thus, it could be argued that in these cases the conclusions were affected by the formulated threats: the first option that was not recommended posed a risk to landscape and the second one presumably to climate.

The developers did not question the feasibility of the proposed project because of strategic reasons (or because of uncertainty or risks related to strategic choices) in any of the EISs – the projects were considered to be in accordance with both regional and national waste plans in 14 statements. In one statement no clear position was taken.

In all, combining the findings from all the EISs, the developers' uncertainty considerations covered a variety of uncertainty aspects. The significance of uncertainty aspects was also evaluated. However, there were remarkable differences in the extent of presentation of uncertainty considerations between the different statements.

Consideration of uncertainties and risks by the public

The competent authorities considered the EISs to be legally sufficient, although in four statements they acknowledged that some extra studies would be required later in the planning process. Public comments presented in the competent authorities' statements show that supporters of projects had usually not spoken out, but anti-project sentiments were common – including frequent questioning of suggested project locations. There were a total of 174 public opinions summarized by competent authorities in their 15 statements. The issue of uncertainties was often dealt with in comments, and notions about insufficient data and reliability problems with assessments were widely addressed in the opinions.

Citizens were mostly worried about incineration emissions to air and water. It was noticeable that the public often expressed doubts about emission calculations and estimations. People questioned the information presented to them in terms of reliability and bias. Another common theme was doubts about the effects of transport during both the construction and operation phases.

In addition, a concern about the impartiality of EIA practitioners, and indeed those questioning the idea and principles of waste incineration, was regularly expressed. The former underlined the distrust between pro-project practitioners and others, usually by criticizing the technology-oriented thinking and by questioning the trustworthiness of the presented information. The latter type of opinions reflected criticism that citizens had towards waste management solutions; incineration was seen as the wrong option, and not as a potential energy or waste management solution. This indicates that those who criticized this issue also criticized the planning by saying that the question should have been solved when planning the strategy.

The competent authorities had taken into account the opinions of citizens in their statements, although often only by listing them. There was no evidence of extensive analysis of public opinions and concerns. However, the issues raised during participation were given weight in a few of the final statements. The inference of public participation can be seen in two ways. First, the competent authorities brought out some shortcomings in the EISs based on opinions, and second, they acknowledged that for some projects extra studies on impacts would be needed in later planning. As a whole, the public opinion took stands both by questioning objective features of assessment and by questioning the problem frame as presented in the EISs. The public participation also seemed to bring to light subjective uncertainty aspects relating, for example, to fears and anxiety experienced in affected communities both individually and collectively.

Concluding remarks

In this article, the legal requirements functioned as a point of departure for analysing the EIA practice from the precautionary perspective. The EU legislation on environmental impact assessment represents a rather weak version of the precautionary principle. The content requirements for environmental impact statements do not refer to consideration of uncertainties and risk assessment as such. The Finnish EIA legislation goes further by requiring information related to uncertainty factors, but it leaves the question of how uncertainty analysis should be conducted open to a significant extent.

Based on the document analysis of this article, it can be noted that the developers had encountered risk and uncertainty aspects in all the environmental impact statements and that the competent authorities considered the outputs as being legally sufficient. The public opinions summarized by the competent authorities clearly broadened the scope of issues and approaches. It was apparent that opposing perspectives on waste incineration and concerns about impact assessments were not excluded from the documents. The reliability and trustfulness of expert opinions were questioned and this was reported. According to some opinions, the framing of developers' scrutiny was also challenged. In this way, the public participation seemed to bring new risk and uncertainty aspects into the view. In some cases these concerns were also reflected in the competent authorities' statements. However, the public opinions and questions were usually not answered in the documents.

Thus, it could be argued that the stakeholders in these EIAs dealt with the concepts of risk and uncertainty, but the issues could have been more thoroughly discussed if emphasis on precaution was preferred with probable costs to scheduled efficiency. At least two suggestions can be made based on this study. First, more efforts could have been used in the analyses presented in the EISs – the described approaches used more extensively would have improved the overall quality of considerations. In particular, it seems that in these cases the developers and the consultants could have had a possibility to learn more from each other between the cases. Particular uncertainty and risk aspects were recognized in some of the projects, while in others the same aspects were left unaddressed, although apparently they would have been applicable. Second, more decision-focused communication would have probably improved the quality of the EIA outputs. It can be assumed that one initial step towards improving the uncertainty and risk analysis in these cases could have been for the developers or competent authorities to respond more precisely to the publicly declared concerns.

Moreover, a collaborative collection of different risk aspects and estimates of uncertainties related to them could offer valuable information for decision-makers. In the end, this is what the EIA is about – projections about the future and an evaluation of the effects of different choices on them. It does not mean that collaboration would need to find an agreement; rather, no one is overridden. In these cases and if simplified a little, the developers showed and believed with some certainty that their projects would not cause harm. In their opinion the minor risks they cause are worth taking. Their opinion was questioned in terms of reliability, and had they explained or responded to concerns or engaged with other stakeholders more thoroughly, they would have probably ended up with a more profound analysis and gained some extra respect. On the other hand, other stakeholders offered the view that, from a wider perspective, the cumulative effects of the projects may be, in an obviously uncertain worst case scenario, catastrophic if the whole waste management policy fails. Although it could be argued that this handling is beyond the scope of EIAs, from the uncertainty and risk perspective it is not – especially as it appeared that discussions related to risk at the strategic level were buzzing. In these types of situations, EIA could offer a valuable opportunity to create a knowledge base for strategic level discussions beyond a single project decision.

In all, according to our research framework, it was assumed that risk and uncertainty analysis are challenging owing to the fact that both concepts are polymorphous and have elements that escape definitions. Anyhow, it can be concluded that in these cases EIAs showed the potential to give effect to these considerations and expressed the spirit of the precautionary principle. We support Duncan's (2013) notion that uncertainty analysis could benefit from a shift towards collaborative knowledge-making. In practice, this could be contrary to the political climate, which seeks to speed up planning processes, but we believe that both during the processes and in retrospect, EIAs could be used more effectively for learning and understanding what kind of concerns or hopes a certain development may generate. Like Ravetz (2004) and recently Partidario and Sheate (2013), we also believe that respectful and positive attitudes towards different ways of knowing and not knowing may be a way forward in improving the practice of impact assessments.

Acknowledgements

The authors thank the Academy of Finland (project numbers 107880 and 138383) for the financial support. M.K. thanks the Jenny and Antti Wihuri foundation, EU IMPERIA-project (LIFE11 ENV/FI/905) and the University of Jyväskylä. P.H. thanks Academy of Finland for the Post-Doctoral research project no 131717. K.J. thanks the Maj and Tor Nessling foundation (project no 2009287) and the Department of Biological and Environmental Sciences of the University of Jyväskylä. The authors thank Associate Professor Kevin Hanna and the anonymous reviewers for their valuable comments.

Notes

^1. At the same time, in many other European countries waste incineration has been quite common. The context and background are described in more detail in the section ‘Background for the cases: role of waste incineration in EU and Finnish waste policy’.

^2. See e.g. Framework Convention on Climate Change (1992), Convention on Biological Diversity (1992) and Protocol on Biosafety (2000).

^3. Council Directive on the assessment of the effects of certain public and private projects on the environment 2011/92/EU. The European Commission has recently introduced a proposal (COM(2012) 628 final) for substantially amending the EIA Directive. The proposal has relevance to the precautionary principle, especially in the form of adapting the EIA to challenges of climate change and disaster risks.

^4. In the Finnish setting, for example, a recent survey showed that the EIA professionals named the issue as one of the general challenges still needing more attention in EIAs (Jalava et al. 2010).

^5. More strongly put, Cashmore et al. (2008) have suggested that environmental assessments do not actually operate within a conventional model of causation where analysis of consequences and publication of results should lead to informed consents and design decisions.

^6. For example, the European Commission's Guidance on EIA (European Commission 2001) and Finnish guidance for EIA review by Hilden et al. (1997) and EIA support material published on the Internet by the Ministry of Environment (2011).

^7. Recycling refers to the recovery operation by which waste materials are reprocessed into products, materials or substances whether for the original or other purposes. Article 3.117 of the Waste Directive.

^8. See also the preamble (point 29) of the Waste Directive, which points out that the Member States ‘should support the use of recyclates, such as recovered paper, in line with the waste hierarchy and with the aim of a recycling society, and should not support the landfilling or incineration of such recyclates whenever possible.’

^9. See also website of the European Commission (http://ec.europa.eu/environment/waste/).