1. Introduction

Local participatory processes that enable citizens and civil society organizations to directly express their demands to municipal authorities are justified normatively as a way of enhancing democratic legitimacy in political decision making. However, municipal authorities can undermine the legitimacy of participatory processes in many ways, for example, by limiting the agenda or selectively adopting proposals (Font et al. 2018; Newman et al. 2004). One set of practices that can affect legitimacy, but has received surprisingly little attention, is the manner in which experts who participate in the process apply their policy expertise in judging which of the proposals can be justified on legal, economic, technical or other grounds. In normative terms a significant difference exists, for example, between those participatory processes where such judgements are made in a way that is open to participants and where such judgements are made behind closed doors, after the initiative has ended and where there is no public scrutiny.

The application of specialized knowledge resonates with concerns within the extensive theoretical discussions of the democratization of expertise (Bäckstrand 2003; Brown 2009, 2014; Collins 2014; Edelenbos et al. 2011; Fischer 1993; Forester 1999, 2009; May et al. 2016; Nez 2011; Petts 1997; Sintomer 2008; Yearley 2000). However, within that body of work there is a remarkable lack of attention to the specific ways in which different participatory processes at a local level deal with expert knowledge in practice: are the assessments on the feasibility of citizen proposals incorporated in a transparent or obscure manner? A fundamental aspect of democratic legitimacy in these processes is the extent to which they allow the participants to oversee and, if necessary, challenge what we call ‘expert assessments’ so that the final proposals emerging from the process still reflect their genuine will.

Participatory processes vary greatly in their aims, scope, format, sponsors, and institutionalization, among other characteristics (Fung 2003, 2006; Nabatchi 2012; Smith 2009). Thus, for example, participatory budgeting generally integrates expert assessments in specific stages during the process where bureaucrats, consultants, citizens, and associations exchange information, arguments, and criticisms (Baiocchi & Ganuza 2014). Other participatory mechanisms follow different logics, from cases where expert considerations play a decisive role to cases where they are absent or ignored.

To be clear, our interest in this paper is not in expertise in general, but in a particular form of expert knowledge: the inputs of officers and/or external consultants who are empowered to evaluate the feasibility of proposals that emerge from local participatory processes aimed at influencing public policies. These actors often hold a privileged position of epistemic authority in local participatory processes, based on their professional training and credibility. Their judgements can determine whether a proposal is selected, abandoned or substantially reframed in the final set of recommendations to municipal authorities. A crucial distinction thus emerges as to whether these assessments are carefully incorporated as part of the participatory design or whether there is no opportunity for participants to consider and challenge them.

The literature on participatory processes has typically focused on design concerns such as the type of participants or the rules for decision making. Less attention has been devoted to the timing, style, and specific impact of expert assessments. Thus, two research questions arise: Are the expert assessments of local officers and/or consultants on the proposals raised in local participatory processes incorporated in a transparent or obscure manner? Is it possible to develop a typology of processes regarding their management of expertise? We argue that the diverse modes of incorporation of specialized knowledge have relevant consequences for the democratic legitimacy of participatory processes, that is, whether or not the participants have oversight and, if necessary, are able to challenge the impact of these interventions on the final outcomes of the process. To address this question, our study draws on a mix of qualitative and quantitative data from a set of 38 local participatory experiences that were implemented in the Spanish regions of Andalusia, Catalonia, and Madrid between 2007 and 2011. The diversity of this sample – permanent and temporary, expensive and relatively costless, complex and simple initiatives – allows us to analyze how the type of participatory design conditions the way in which this form of expert knowledge is incorporated. Although the role of experts varies in accordance with the objectives of each process, the degree of transparency of their inputs on the proposals that emerge has significant impact on their democratic quality.

The paper begins by drawing insights from the theoretical debate on the democratization of expertise and the role of institutional design in participatory processes, clarifying the type of expert knowledge that is the object of our study. The next section describes the methods through which the Spanish cases were selected and analyzed. Third, we address our research questions, distinguishing three broad types of participatory process according to how expert assessments are integrated. We offer illustrative cases from our sample. Finally, we discuss the implications of the different ways of dealing with expertise for participatory governance.

2. Expertise and Participatory Processes

Public policies can be understood heuristically as a match between policy goals (the problem) and policy means (the tools to achieve goals). This match is performed across a techno-political process: on one hand, the technical dimension corresponds to the search for the optimal tools that help to resolve the problem; on the other, the political dimension emphasizes the disagreement between actors on the nature of the policy problem and potential impacts of the available tools (Howlett et al. 2009). According to this traditional policy-making perspective, these two dimensions are the domain of two different sets of actors. Elected representatives manage the political work to prioritize values and understandings of reality, drawing on social demands and their own beliefs. Bureaucrats, with their specialized knowledge and skills, seek the optimal match between policy goals and means.

However, the multiplication of partnerships, participatory processes and new forms of collaboration disrupts the traditional view of decision making not only in the more obvious political dimension, but also opening up technical aspects (May et al. 2016). As think tanks, interest and advocacy groups, citizen committees, lay citizens, and the like take a more direct role in the policy process, we witness the multiplication of ‘policy experts’ and the diversification of the location of expertise (Craft & Howlett 2013; Talpin 2011).

In this context, Edelenbos et al. (2011) usefully demarcates three broad forms of knowledge: scientific, bureaucratic, and stakeholder (also called alternatively lay, practical, or non-professional) knowledge. The most common distinction is drawn between scientific and lay knowledges: the former based on objective measurements and tools defined and reviewed by communities of peers, who share common theories and techniques (Maiello et al. 2013; Petts & Brooks 2006); the latter drawing on experiential knowledge, based on people’s common sense and everyday practices (Edelenbos et al. 2011; Fischer & Gottweis 2013; Nez 2011; Petts & Brooks 2006; Sintomer 2008). The literature on the democratization of expertise has focused specific attention on challenging the boundaries that have been drawn between these two sets of knowledge, in particular showing how citizen participation can contribute to scientific inquiry by complementing or correcting its universalistic principles with contextual information (Bäckstrand 2003; Yearley 2000).

A third type of knowledge that is distinguished from scientific and lay knowledge is that of bureaucratic actors. Such bureaucratic knowledge is related to understanding the appropriateness of ideas to solving policy problems with regard to administrative standards, procedures and rules (Edelenbos et al. 2011: 675). It is derived from a mix of knowledge about policy and administrative processes and particular professional and scientific training. Again, it would be a mistake to draw a tight distinction between bureaucratic and other forms of knowledge. As Hunt and Shackley stress, ‘contrary to the linear model of science policy relationships… the civil service is not solely a recipient or user of knowledge, but is actively engaged in constructing and formulating knowledge’ (1999: 144). Bureaucrats are also facilitators of public learning and empowerment (Forester 1999, 2009), combining insights from a variety of actors and perspectives with their own knowledge of procedures and professional expertise (Feldman & Khademian 2007).

According to Hendriks, in any given policy arena different actors promote their particular ‘participatory storylines’, that is, ‘narratives on who the public are, and how they should be involved in the policy issue’ (Hendriks 2005: 15). Hence, the prevailing storyline in a given matter legitimizes both the participatory mechanism and whether key actors are characterized as consumers, corporatist stakeholders, or lay citizens, for instance. The literature on democratization of the policy process suggests that a common storyline amongst bureaucrats is one of resistance towards the integration of lay knowledge in the policy process (Güemes & Resina 2019; Hendriks & Lees-Marshment 2019). Bureaucrats often perceive lay knowledge as unreliable with regard to problem solving. Citizens are taken to be incapable of dealing with complex policy issues either because of their lack of specialization and/or their partiality. In comparison, bureaucrats often identify themselves as offering an impartial perspective that privileges their assessment of appropriateness with regards to standards and rules (Ganuza et al. 2016; Maiello et al. 2013; Petts & Brooks 2006; Yang 2005). For sceptics, the democratization of expertise is often little more than a one-way relationship or a top-down approach through which bureaucratic knowledge is transferred to the public (Petts 1997). Citizens are seen as peripheral actors who lack the relevant knowledge to influence the technical dimension of the policy-making process: they provide inputs to bureaucrats who act automously on the final output (Migchelbrink & Van de Walle 2019).

Participatory processes are thus of particular interest as they are a political space in which a variety of expertise is potentially present (Brown 2008, 2014; Nez 2011; Sintomer 2008). In principle, the democratic legitimacy of such processes rests on the manner in which knowledge co-production takes place: the nature of the exchanges of facts, interpretations, assumptions, and causal relations from different knowledge domains (Edelenbos et al. 2011: 675–677). Here, the role of local officers and the policy professionals or external consultants that organize participatory processes becomes critical. These actors often share backgrounds, interests and prejudices (Edelenbos et al. 2011: 683) and, when designing participatory spaces, they have an implicit control over how scientific, lay, and bureaucratic knowledge is brought to bear (Bherer et al. 2017).

Among the diverse knowledges that usually converge in a participatory process, our specific interest in this paper is to pay attention to the application of expert assessments by local officers and/or external consultants, which can lead them to select, abandon, or reframe the participants’ proposals. Their professional training, bureaucratic knowledge and expertise potentially gives such actors a privileged position in shaping the final recommendations that emerge. In this sense, the consideration of ‘expertise’ in our research focuses on the very practical question of how a particular form of epistemic authority is exercised in a participatory process. In this inductive approach, we are interested in the function of the actor or actors with responsibility for, first, design and conduct the process and, second, assess the feasibility of proposals generated. Since a broad range of participatory spaces exist at the local level, these functions can be performed alternately or simultaneously by different figures. Thus, in our research, such expertise can be exercised by the local officer (bureaucrat, technician, etc.) who organizes the process, facilitates its meetings and synthetizes the final proposals, as well as, for example, the municipal architect who evaluates the feasibility of a sports center proposed in the participatory budget. Likewise, when the local authority outsources the process (commissioning a strategic plan from a consulting firm, for example), the participatory practitioners hired to carry it out exercise similar expertise. Hence, our aim is to understand how the presence and timing of these expert considerations on proposals affects the transparency of and democratic control over the process. We are concerned with the way expertise is integrated into the architecture of a participatory process.

This approach resonates with an influential perspective in research on participatory processes that attributes an explanatory role to institutional design (Bua & Escobar 2018; Delli Carpini et al. 2004; Fung 2003, 2006; Gastil & Levine 2005; Landwehr & Holzinger 2010; Nabatchi 2012; Nabatchi & Leighninger 2015; Smith 2009), whether in terms of how design characteristics affect the realization of democratic and institutional goods or particular outcomes. What is striking about this literature is that it has relatively little to say about our particular research question. Much of the work on process design has concerned itself with analysis of the effects of participants selection, the form of communication and the extent of influence on decision-making (Fung 2003, 2006; Johnson & Gastil 2015; Nabatchi 2012; Smith 2009). A paucity of attention is apparent in this literature on either the ways in which the design of participatory processes impact on when and how expert assessments are brought to bear, and the effect it can have on the democratic quality of these experiences. Fung (2003, 2006) and Smith (2009), two of the more influential authors on the democratic implications of design choices in participatory institutions, are rather limited in their consideration of these issues.

In distinguishing between the different types of participatory process – educative forums, participatory advisory panels and participatory problem-solving – Fung (2003: 340–341) implies that the diverse goals pursued by different participatory processes has an effect on the role played by expert assessments. But it is not a line of argument that he develops. In his later work on the ‘democracy cube’ (Fung 2006), experts and professionals are one potential group of participants but the way in which they interact with lay citizens remains underdeveloped, in particular how this impacts on the dimension of ‘communication and decision’. He claims that decisions in participatory processes are not always determined through deliberation or aggregation of citizen preferences, ‘but rather through the technical expertise of officials whose training and professional specialization suits them to solving particular problems’ (Fung 2006: 69). While the democracy cube recognises the significance of the integration of technical expertise to the design of participatory processes, it gives us little sense of the different ways that such expertise can be integrated in practice.

In Smith’s analytical framework for democratic innovations (2009), the role of expertise is recognised as a potential element in realising the democratic good of considered judgement. The effect expert knowledge can have on different designs – assembly-based, randomly selected mini-publics and direct legislation – is explored to some extent, but like Fung, insights on the relationship between participatory design and expertise are not developed systematically. Fung, Smith, and other authors are liable to the criticism that they tend to focus on exemplary participatory processes (Spada & Ryan 2017) where the democratic consequences of expert contributions have been carefully considered. This does not capture the way that more common and everyday local participatory processes incorporate specialized inputs.

To conclude, previous studies have recognized the significance of the exercise of expertise in participatory processes and how this can raise significant questions about the democratic legitimacy of citizen engagement in the policy process. What we are missing is any systematic understanding of how the design or architecture of participatory processes affects the way in which public officials or external consultants are able to exercise expertise and thus influence the proposals that emerge.

3. Methodology

Our analysis is based on a set of 38 participatory processes implemented in the Spanish regions of Andalusia, Catalonia, and Madrid. The main reason to select Spain was pragmatic, as we had access to a previous database covering hundreds of diverse participatory processes in these three Spanish regions: large and small, successful and failed cases, including their basic descriptive characteristics. Spain is reasonably representative of the Southern European model of institutionalised local participation, which is more politically oriented than in other parts of Western Europe (Talpin 2011). At the same time, these three regions represent a diverse set of socio-economic and political contexts with different traditions of participation (Sintomer & Del Pino 2014). Selection of cases from a single polity ensures a constant legal context, at the same time allowing contextual variation through the selection of diverse municipalities and regions.

Fully representative frames of participatory processes are uncommon, but this dataset ensured a rich diversity of processes in the initial sampling frame.1 The main condition for processes to be included in our research was that they ended up issuing policy proposals to public authorities. Having proposals does not guarantee an equal level of policy impact among our cases, but at least it helps us to discriminate processes that frame participation as a way to increase the fairness and efficiency of policy making from other processes with merely informative, educative, or tokenistic goals. For the final selection of processes we adopted a stratified random sampling strategy, taking type of participatory process, municipality size, region, and participatory tradition of the municipality as selection criteria. Whenever choice was possible, the final selection of cases within each strata was random.

Our cooperation rate was a healthy 81.3 percent, with less than one third of the selected cases substituted by similar processes, either due to their ineligibility for the research (process not completed, out of the study’s time frame, lacking policy proposals) or to the lack of cooperation from local officials. Our initial sample was formed by 40 cases, but for one of the processes we discovered that there were no records too late, so that substitution was not possible. Table 1 shows the final sample composition.

Table 1

Final sample composition.

N %
Nº of participatory processes in the municipality
    Three or more 24 62%
    Less than three 13 33%
    No info 2 5%
Process design
    Participatory budget 8 21%
    Strategic planning 14 36%
    Other permanent 8 21%
    Other temporary 9 23%
Municipality size
    Less than 10,000 inh. 11 28%
    10,000 to 50,000 inh. 12 31%
    More than 50,000 inh. 16 41%
Region
Andalusia 19 49%
Catalonia 10 26%
Madrid 10 26%
  • Source: Own elaboration. Percentages rounded do not always add to 100.

One of the 39 cases has been excluded from the analysis because the information about our central variable was not reliable enough. Therefore, our unit of analysis is each of the 38 local participatory processes developed during the period 2007–2011 in three Spanish regions that generated proposals for which we have sufficient information. We selected the period 2007–2011 (from one local election to the next) to ensure that there had been time enough for at least the initial implementation of the proposals, but also that memories and administrative records of the process were recent enough to be tracked.

After a pilot study in one of the municipalities, the fieldwork developed from April to December 2014 to collect information about each of these processes on three main dimensions: local context, participatory process, and process proposals. Information came from a variety of sources, from municipal web pages and official documents (constitutive rules, technical reports, official minutes, and others); to in-depth interviews with local officers, government and opposition politicians, and civil society participants. In total, we conducted 162 semi-structured interviews with informants, of which 43% were local officers,2 16% government politicians, 17% opposition politicians, 17% civil society participants. The information provided by local officers was cross-checked through interviews with civil society actors, as well as with government and opposition politicians for those issues liable to partisan bias. The mean of informants per process was 4.6.

Although a relevant part of the information collected comes from the analysis of official documents, the interviews were essential to triangulate and complement findings. On the one hand, they allow us to go beyond the formal design of processes and verify that the documentary information on their design characteristics corresponds to their actual practice. On the other hand, because simple and temporary processes generate little or no documentation (especially in small municipalities), they tend to be excluded from systematic studies. In these cases, interviews with key actors – the politician who drives the initiative or the officer in charge of organizing the process, for example – are essential to grasp its basic characteristics, operating rules and the final list of proposals. Finally, the interviews provide qualitative information that help us to better understand the less visible aspects of the process and the fate of its proposals, especially in the less transparent cases. In turn, the diverse profiles of the interviewees allow a more objective approach to the case as compared with exclusively institutional or politically biased sources. Thus, both sources, official documents and in-depth interviews, inform both the quantitative (summarized in Table 3) and qualitative (description of representative cases) analysis.

The documents and interviews provided information for variables related to local context, process design and proposals. These were coded by a team of four fieldworkers (social science doctoral students).3 To access relevant data on the role of expertise, the interviews included a specific question on how and when the participatory process integrated expert assessments from local officers or external consultants (when they were hired to manage the process). In addition, the research team produced fieldwork journals for each case with the aim of elucidating the dynamics that went beyond its formal design and rules.

The form of expert incorporation on the part of local officers or external consultants is our expertise variable. Of the 38 cases with available information, there are 9 cases in which there was no incorporation of specialized knowledge. In the remaining cases, we draw a distinction between what we term ‘accountable’ and ‘non-accountable’ incorporation of expert assessments. What we understand by ‘accountable’ in this study is fairly limited: we aim to capture the difference between processes in which participants are in a position to scrutinize the application of expertise on their proposals and those where such scrutiny is absent. In our sample, this distinction can be drawn in a fairly straightforward manner. Even if the timings and procedures were quite diverse, there was a clear distinction between processes in which participants could scrutinize and, if necessary, challenge expert interventions on their proposals (either because they were sitting at the same table or because they could verify them at an specific stage of the process), sharply contrasting with those cases where, after the participatory stages, bureaucrats and/or external consultants could modify the citizens’ proposals without any oversight.

In an initial quantitative analysis of available data, we undertook a bivariate analysis (V Cramer tests to look for significant associations between categorical variables) between our expertise variable and several process characteristics. Some of the variables in our process level dataset have too few occurences to provide significant results,4 but we were able to analyse the relationship with 10 characteristics of the participatory processes. Several showed no significant results5 and the text reports on the four variables that showed a significant relationship with the form of incorporation of expert knowledge. Table 2 provides a summary of these variables and the response categories used.

Table 2

Description of variables.

Variables Operationalization: response categories Mean (standard deviation)
Form of expert incorporation 0 = technical criteria accountable
1 = technical criteria not accountable
2 = no expert incorporation
Nominal
Type of participatory process 1 = participatory budgeting
2 = strategic planning
3 = other permanent processes
4 = other temporary processes
Nominal
External funding for implementation 0 = no
1 = yes
0.34
(0.474)
Level of information 0 = low: no information or only general introduction
1 = medium: technical report from municipality
2 = high: external experts invited
Nominal
Type of participants 1= only citizens
2 = only associations
3 = citizens plus associations
Nominal

The quantitative analysis foregrounds our more extensive qualitative analysis that draws on official documents, interviews, and fieldwork notes. This allows us to present illustrative cases of the three prevailing approaches to dealing with expertise. The selection of these cases combines two criteria. First, we give priority to processes that are broadly representative of the variation in participatory design in the same category. Second, we choose well documented processes that illustrate the typical challenges that expert incorporation, or its absence, entails.

4. How Expertise Is Incorporated

We use participatory process as the unit of analysis to describe the prevailing dynamics of the integration of expert knowledge. From our sample, we can distinguish three broad approaches to integration. The first is where no expert considerations are introduced: in nine participatory processes participants suggested proposals without any specific assessment of the feasibility of their ideas. A second approach corresponds to 15 participatory processes where expertise was introduced in a transparent way during the process. As a result, participants were active in the weighting of these considerations. For the other 14 cases, expert assessments were incorporated in a way that was not accountable to citizens and associations, often once the engagement process had ended. In sum, most processes introduce expert knowledge in one way or another (29 out of 38) and almost half of them did so without direct oversight by participants, thus leaving room for discretion and opportunities to ‘cherry-pick’ or modify proposals.

In what ways are these strategies related to other design characteristics of the participatory processes? While recognizing that we are dealing with a small number of cases, Table 3 presents four process characteristics that have significant associations with these three broad approaches to expert integration. First, the type of participatory process. Participatory budgeting appears more often (47%) amongst accountable processes, whereas strategic planning is strikingly prevalent among those processes where the incorporation of expertise is not accountable to participants (86%). Second, the level of information provided to participants during the process also has a clear association: where participants are able to hold local officers or external consultants accountable, they will only rarely face this challenge with a low level of information6 (13%), whereas this situation appears in 78% of the processes where expert considerations were absent. Third, non-accountable processes tend to be supported by external funding from other supra-municipal administration (69%),7 a characteristic which appears rarely among processes where there are no expert assessments (11%). On the other hand, accountable incorporation of expertise appears almost equally among both externally and non-externally financed processes. Finally, the type of participants also exposes relevant differences. All the participatory processes characterized by non-accountable incorporation of expertise within our sample combine the presence of associations8 and individual citizens. On the other hand, designs with accountable expertise are more mixed in their composition: citizens as the only participants is more common than in other approaches (33%); but a mix of citizens and associations is the most common (53%). Processes without expert assessments tend to include only associations (44%) or citizens and associations (44%). Thus, who participates becomes an important design factor that appears to condition how expertise is incorporated: the challenge of combining individual and associative participation (different degrees of involvement, expertise, or perceived legitimacy, for example) seems to create the conditions for external expert involvement, but, in many cases, without any clear democratic scrutiny.

Table 3

Process characteristics and incorporation of expertise (% column).

Accountable incorporation Non accountable incorporation No expert incorporation
% % %
Type of participatory process
Participatory budgeting 47 7 0
Strategic planning 7 86 0
Other permanent 20 0 56
Other temporary 26 7 44
External financial help*
Yes 53 69 11
No 47 31 89
Level of information
Low 13 50 78
Medium 27 14 11
High 53 21 0
No info 7 14 11
Type of participants
Only citizens 33 0 11
Only associations 13 0 44
Citizens plus associations 53 100 44
Total 100 100 100
(n) (15) (14) (9)
  • * One case excluded due to missing information. Percentages rounded do not always add to 100.

    All variables significantly related (Cramer’s V below 0.05) to expert incorporation.

These associations between variables suggest that there are relevant relationships between aspects of process design and the different strategies for dealing with expert knowledge. To shed more light on these issues, the qualitative description of cases that follows provides a richer sense of the dynamics at play in the three broad strategies identified within our dataset of participatory processes.

4.a. No expertise involved

The first design option is a participatory process where expert considerations from local officers or external consultants are completely absent. Broadly speaking, these are processes organized to deal with relatively simple issues through assembly meetings where anyone can make proposals. Expert inputs are not so relevant and participants are assumed to be equally politically competent: expertise is equated to lay or experiential knowledge (Fischer & Gottweis 2013; Nez 2011; Petts & Brooks 2006; Sintomer 2008), a form of judgement derived from citizens’ everyday experience.

An example is a Council of Wise Women that gathers together older women in a small town (between 5,000–10,000 inhabitants) near Barcelona (Catalonia) to debate and make recommendations directed to municipal authorities and other local and neighbourhood entities (Camprubí 2007: 41–43). Established in 2007 with the support of the municipal government, the council is a permanent participatory mechanism open to any woman from the municipality aged sixty or older who wishes to participate. It works as a discussion group aimed at improving the town’s quality of life. A facilitator selected by the council members organizes the agenda of the meetings and ensures fairness in contributions. Beyond this facilitation there is no expert intervention and participants are free to make any proposal. The name of the mechanism – ‘Council of Wise Women’ – itself assumes experiential knowledge among the participants. However, its recommendations are not binding and there is no formal integration with the policy-making process. When the local authority responds to the council’s recommendations it often issues technical explanations, but this is not a formal stage of the participatory process.

Similar to the Catalan council is the Municipal Immigration Forum that was held in May 2007 in a small town (between 5,000–10,000 inhabitants) in the province of Granada (Andalusia). This participatory process was an initiative of the local government aimed at integrating effectively the high volume of immigrants living in the locality. Aside from resident immigrants, the Forum had a strong presence of representatives of several local associations and unions. The nature and structure of the event made the incorporation of expert knowledge unnecessary: it was a single-issue process, held on just one day and focused on making simple proposals on possible training courses and workshops to improve the integration of immigrants in the local community. Since the documentation for this case was lacking, interviewing the local officer who organized the process was crucial to understand its design and final proposals. As in the Catalan case, there was an assumption of the validity of participants’ experiential knowledge and, thus, the level of information provided by the organizers was low. Also, the Granada council was not integrated formally within the local policy process and its proposals were non-binding. The cost of both processes was low and they did not receive any external funding.

4.b. Expert assessments held to account

The second approach is participatory processes in which expert assessments are democratically accountable: their integration constitutes an explicit stage of the process, with participants actively overseeing the application of expert judgements. As seen in Table 3, this is common for participatory budgeting, which typically has explicit and detailed procedures that enable internal transparency. The role of experts is clarified through democratic means.

The participatory budgeting process in a small town (between 5,000–10,000 inhabitants) in the province of Girona (Catalonia) illustrates this dynamic. Since 2003, this process has engaged a significant portion of its population in deciding around 50% of the local budget (Bou 2011: 183). The budgets from 2003 to 2011 included binding proposals on matters as diverse as environment, education, health, social welfare, urban planning, or local festivities. Participation was organized around eight neighborhood and 10 thematic assemblies open to any person aged 16 or above who was already registered in the municipality. These assemblies were constituted every year in the autumn to debate local needs and generate up to 10 prioritized proposals. Additionally, each assembly selected two representatives for the Council of Citizens, the central institution of the process. The Council of Citizens evaluated the proposals coming from the assemblies and decided – through consensus or majority vote when needed – the final list of items to be included within the municipal budget.

Expertise was introduced at two specific stages of the process, both enabling citizen oversight. Firstly, a series of thematic councils were added to the budget cycle in 2007, working as consultative bodies where the actors involved in a specific field – city councilors and bureaucrats, representatives from the corresponding assembly and from local associations – could debate and generate detailed recommendations for the neighborhood and thematic assemblies. The aim of these councils was to increase the quality of strategic policies, which was difficult to achieve in assemblies with hundreds of individual participants (Bou 2011: 163–164). In this sense, they offered a space for cooperation and communication among local officers and politicians, on the one hand, and associated citizenry and civil society organisations, on the other hand. In any case, these thematic councils were advisory, with assemblies retaining the final decision on the 10 proposals to go forward to the Council of Citizens. Also, at a second stage, the process established a Technical Office with the task of advising the Council of Citizens during the preparation of its budgetary proposals. This office, also integrated by city councilors and local bureaucrats, could recommend changes to the Council’s initial proposals, but the final decision was kept in the hands of the Council’s citizen members.

Thus, this participatory design integrated expertise in two successive steps: prior to the formulation of proposals by the assemblies, and prior to the final budgetary proposals adopted by the Council of Citizens. In both cases the citizens’ level of information was high, the intervention of specialists was transparent and the final decision on whether to follow or not their recommendations was kept in the citizens’ hands. This picture resonates with the literature on participatory budgeting that highlights the capacity of the process to generate internal transparency, professional support to lay citizens and accountability of the public officers involved in the process (Smith 2009: 176).

4.c. Expert assessments not held to account

The third category of participatory process is one in which expert considerations lack democratic oversight. The consequence is an unrestricted intervention of experts and an opaque link between the will of participants and the process outputs. Some of the cases of strategic planning in our sample show how an opaque application of expertise can lead to the ‘cherry-picking’ of proposals. In these designs, the participatory stages are usually considered a source of ideas that can be more or less freely filtered by local officers or external consultants during the drafting of the final plan.

An example is the Plan for Gender Equality developed in a town of less than 5,000 inhabitants located near Granada (Andalusia). This plan, developed in 2010 with the external funding of the Granada provincial government, was the result of a participatory process over three months, where two hired consultants launched several initiatives – questionnaires to neighborhoods and local institutions, workshops, interviews with women’s associations and municipal staff, discussion groups – seeking inputs for the final document. The final Plan included 24 recommendations on gender equality for the local administration.

Similar to other strategic plans in our sample, this process aimed to provide the external consultants with a diagnosis of the local context and ideas for action as well as to legitimize their drafting of the final document. As one of the consultants interviewed explained, the process design did not include a specific stage for monitoring the application of specialized criteria when drafting the plan, because it was assumed from the beginning that the external consultants should work with a degree of flexibility. Therefore, although the plan was presented as the result of a participatory process, it is not easy to track a clear connection between the plan’s recommendations and the proposals that emerged from participants.

A second example is the participatory process developed in 2008 in a large town (more than 50,000 inhabitants) located near Madrid that aimed to incorporate citizens’ suggestions into the municipal budget. Rather than a participatory budget, this process was a citizen consultation to distribute a relevant budget line. The annual cycle had two major moments. In the spring phase, there was a large-scale collection of proposals from citizens and local associations through diverse channels including website, email, postal mailing, and ballots distributed in open meetings. The suggestions were grouped and sent for assessment in the relevant municipal departments, which generated a reduced list of proposals to be prioritized by citizens in the autumn meetings. Thus, the 900 proposals initially submitted in the spring of 2010 for the 2011 budget were reduced to a final list of 81. Here, the interviews with the local officers opened the ‘black box’ of the process, enabling us to discover that this filtering was undertaken by the municipal bureaucracy without any explanation, so that proponents remained unaware of the specific reasons for the removal of their suggestions from the final list presented for the citizens’ vote.

In other cases, we find intermediate situations in which the participatory process enables a collaborative elaboration of proposals among participants and specialists but, ultimately, local officers or external consultants can make changes behind closed doors. Here, expertise is introduced during the participatory stages but also at the end of the process, allowing room for ‘cherry-picking’. The Participation Plan of a big Andalusian city (more than 50,000 inhabitants) illustrates this dynamic. Between 2009 and 2010, the local government launched an ambitious participatory process aimed at approving a Citizen Participation Plan (2010). In order to diagnose the situation and collect proposals the process developed through separate stages: interviews with local officers; work tables with politicians, municipal staff, and associated and non-associated citizens; a deliberative forum with citizens selected through stratified random sampling; and a specific website to collect suggestions from individual citizens. Since the process incorporated different stages and types of participants – politicians, municipal staff, associations, individual citizens – it is difficult to track each group’s specific influence on the approved Plan: its final drafting was internally undertaken by the local officers. Consequently, the 95 measures included in the final document could respond either to citizen proposals or to bureaucratic (and maybe political) imperatives.

The need to reconcile inputs from different participatory methodologies and different types of participants is a common situation in participatory governance, which raises clear difficulties from the point of view of democratic accountability when the bureaucrats have to take a final decision on which proposals should be included in the final list of recommendations to the authorities. In this case, our documentary analysis shows that the Citizen Participation Plan included a number of modifications to citizen proposals as well as additional proposals coming from other actors, including municipal officers. When there is no accountability in the drafting process it is difficult to know whether proposals have been rejected or modified for being too generic, technically unacceptable or politically challenging, among other considerations.

To sum up, the 14 cases in which expert assessments were not accountable show a different profile from the previous two categories. We are dealing here with municipalities that put in place a sophisticated participatory design with several opportunities for citizens to be heard but without a formal stage that provided for a transparent integration of the multiple inputs. Furthermore, in these cases, municipalities often use strategic planning processes and mechanisms that usually combine the participation of individual citizens with associations and stakeholders, making the interaction between different types of participants and specialists more challenging.

5. Discussion

This paper has addressed the different ways in which participatory mechanisms deal with expert assessments on citizen proposals and whether this alters the democratic quality of the process. Our analysis offers an initial characterization of three broad types. The first is a set of quite simple participatory processes where there is no explicit incorporation of expertise. Such designs are quite common and tend to have low costs and no formal links to political decision making. Precisely because of their limited impact, external funding, and complexity, such designs do not receive much attention from the academic community. However, these processes are a relevant part of the reality of participatory experiences in the South of Europe, even if they are much less visible in research dominated by exemplary cases or more sophisticated experiences (Font et al. 2014; Spada & Ryan 2017). The absence of expert considerations does not necessarily undermine the quality of these processes, as in some cases the citizens’ lay knowledge is sufficient to make good judgements. Thus, an advisory panel such as the Catalan Council of Wise Women prioritizes the participants’ experiential knowledge over the application of expert criteria. That could partially explain why 78% of the processes that did not incorporate expert assessments also provided participants with low levels of information, assuming that they could rely on their lay knowledge.

Secondly, we find a set of participatory processes that incorporate expert assessments on proposals in a way that enables a degree of democratic control by participants. The most common type of such accountable processes is participatory budgeting, in which local officers and other policy practitioners participate in one or more steps of the process, but with oversight and the final decision remaining in the hands of participants. Our focus here is not on the systemic role of bureaucracy, but on its impact on particular sensitive stages of participatory processes. Local officers may well exert significant power in shaping the broader structure and reception of participatory processes but, in what we term ‘accountable process’, a moment exists within which participants are able to scrutinize and challenge expert interventions on proposals. This finding chimes with evidence on the translation of participatory budgeting from Brazil to Europe. While in many cases translation watered down elements of democratic control, forms of ‘empowered’ participatory budgets that retained elements of the original Brazilian design are more prevalent in Spain compared to other parts of Europe (Baiocchi & Ganuza 2014; Sintomer 2008).

The third approach is one in which experts typically play a critical but unobserved role in producing the final recommendations of the process. They filter, order, and modify citizen proposals – in other words, the process is open to ‘cherry-picking’ through the application of expert criteria. This occurs for many strategic planning processes in our study, where citizen engagement is often framed as a way of sourcing multiple ideas to be filtered at a later stage by local officers or external consultants hired by the local administration. While it was a consistent finding in our study, it is not necessarily the case for strategic planning in compared research, which makes necessary to further analyze the ‘participatory storylines’ taking place within this mechanism.

A key distinction that emerges from our study is whether or not the participatory design incorporates formal stages for holding experts to account, especially when there are several venues of participation in the same participatory process, what Spada and Allegretti (2017) call ‘multiple channels of engagement’. This is the case in participatory processes which are specifically designed to target different segments of the population. Here, different participatory settings are planned in the same decision-making process in order to reach different objectives. Although the diversity of participatory arrangements is often seen in the literature on democratic innovations as a positive way to enlarge citizen engagement (Bherer & Breux 2012), focusing on expert incorporation reveals the difficult challenge of dealing with multiple proposals coming from very different kinds of participatory settings (Spada & Allegretti 2017).

From the analysis of processes such as the Participation Plan in the big Andalusian city, it becomes difficult to ascertain whether an obscure style of expert assessment is a strategic attempt to reduce the direct influence of citizens or a pragmatic solution to combine the outputs from several participatory sources and methods. The sceptic would find evidence to bolster the broader critique of bureaucratic actors in the literature on the democratization of expertise that we discussed earlier in the paper. On this reading, many participatory processes are seen as tokenistic forms of co-option, with elitist manoeuvres undermining the will of participants in the interests of the administration (Arnstein 1969: 217; Bua & Escobar 2018: 132; Newman et al. 2004). In a significant number of participatory processes, ones that usually have external funding (69% of cases in our sample) and deal with more complex and costly issues, it is difficult to attribute the emerging proposals to the will of those participating. That participants do not seem to object or protest against this lack of transparency may be evidence of fairly low expectations or emotional commitment to a process in which they are peripheral actors, or an acceptance that it is legitimate for decisions to be made elsewhere.

A more charitable reading is that the larger the scale of the participatory exercise, the less direct control participants are able to exercise (Dahl 1998: 110): the final intervention of bureaucrats or other policy practitioners is an efficient way of dealing with a large number of inputs. For example, in the Plan for Gender Equality described earlier, citizens’ preferences are derived from a range of sources, including questionnaires, workshops, interviews, and discussion groups. Such processes typically mix individual citizens and associations (in all the cases of our sample) and the public authority is faced with the complex task of synthesizing the diverse contributions of these actors. Here, the filtering of proposals is the only option for the local authority when the sensitive issue of citizen oversight has not been planned in advance. The counter-example from participatory budgeting shows how a process that engages large numbers of participants can be carefully designed to enrich citizen proposals with expert assessments in a transparent manner (Smith 2009: 176). These designs ensure the provision of higher levels of information (medium or high levels in 80% of cases in our sample) to participants. The implication is that a more knowledgeable citizenry will be more effective – both democratically and epistemically – in enacting oversight and their proposals will be in less need of unrestricted scrutiny by specialists.

What we find then is that the role and oversight of the application of expertise is a key dimension in institutional design and one that can have a profound effect on the democratic qualities of participatory processes. Our study shows that the transparency and accountability of expert interventions varies significantly – from those where the application of such expertise appears less relevant to those where design choices have meaningful effect on the capacity of participants to scrutinize and oversee expert interventions on their proposals. What is clear is that, resonating with Fung’s framework, this a design choice. A number of cases of participatory budgeting in our sample show that this can be done in a transparent manner. Compare this with the opacity of most of the strategic planning cases. The question we are left with for those where expert intervention is opaque is whether these are design choices that represent a lack of imagination or a strategic decision to avoid accountability.

6. Conclusions

The democratization of expertise has generated substantial normative attention but there is still a lack of empirical specificity when it comes to actual participatory processes. This paper argues that more careful attention is required to the manner in which participatory processes deal with the expert knowledge that impacts on the realization of the will of participants. It deserves as much attention as more standard design concerns such as the democratic impact of the mode of recruitment of participants, presence of facilitation or the form of decision-making, to mention a few characteristics. It is not enough simply to state whether or not expert assessments occur, rather the extent of democratic oversight of these interventions is a key dimension for understanding the democratic quality of participatory processes.

As is often the case in an exploratory research, we faced several challenges. Our categorisation of participatory forms in relation to expert intervention is potentially limited by the characteristics of the 38 cases analyzed. For instance, our sample did not include randomly selected mini-publics (Smith 2009). Such a design would fit within the ‘accountable’ category in the sense that experts are able to present to, and be questioned by, the selected citizens, but then the recommendations are left to the participants. While this participatory design is widely discussed in academic circles (Brown 2014; Fung 2007; Gastil & Richards 2013; Johnson & Gastil 2015), particularly amongst deliberative democrats, they are rarely organized in the Spanish context in comparison to other forms of participatory process (Alarcón & Font 2014). Our findings concerning participatory budgeting and strategic plans may thus be generalizable beyond the Spanish case, but the impact of expert interventions in other participatory designs and contexts requires further empirical research. Here, the study of Roberts et al. (2020) on the role of expert witnesses in citizen’s juries constitutes a relevant contribution.

We also decided to limit our analysis to a specific understanding of expert knowledge: the expertise of local officers and/or external consultants in assessing the feasibility of the proposals generated during a local participatory process. There are other types of participatory processes that are not included in our analysis in which having specialized knowledge (scientific, bureaucratic, or practical) is an explicit condition for being an active member of the process. This is often the case with advisory councils, in which most of their members are selected on account of their professional experience or because they represent important stakeholders that have relevant knowledge. Their sectorial scope – immigration, youth, health, economy, women – and the specialized profile of their members implies that there is no need for further external assessments.

Even with this caveat, our findings resonate with previous research that considers institutional design as a key explanatory factor in understanding the outcomes of participatory mechanisms. But our analysis suggests that there is work to be done, either by incorporating expertise within existing analytical and explanatory frameworks, or by developing new frameworks that deal more effectively with this design characteristic. Certainly we can see how the manner of integration of expertise could enrich Fung’s (2003, 2006) distinctions between the objectives of processes or Smith’s (2009) account of how democratic goods are realized. Furthermore, the way expert knowledge is incorporated can be seen as a relevant proxy to evaluate the manner in which citizen proposals are connected to formal decision making. Preliminary quantitative analyses of our sample show that incorporating expert considerations – accountable or not – into participatory processes seems to facilitate the acceptance of their final set of proposals on the part of local authorities. These are promising lines of inquiry for a more nunced account of how expertise is managed in participatory governance.

Notes

  1. Four previous lists of participatory processes were used to build the sampling frame. Three followed similar research protocols and represent mostly medium and large size cities. The fourth list comes from Andalusia (and is the reason why this region is overrepresented) following a different research protocol aimed at collecting information from smaller municipalities. Previous analyses show that region and city size were not influential variables. The final sample does not include any case of randomly selected mini-publics, a reasonable result given that this participatory design is rarely organized in Spain as compared with other participatory processes (Alarcón & Font 2014). A more extensive explanation of methodological details appears in Font et al. (2016).
  2. As previously stated, within the category of ‘local officers’ we include local bureaucrats in charge of citizen participation or involved in the assessment of the process’ proposals (e.g., municipal architect, head of the local police, etc). In addition, when the local administration decided to outsource the participatory process we also include the external consultants/practitioners hired to organize it.
  3. The codebook is available at https://cherrypickingproject.files.wordpress.com/2015/06/codebook1.pdf. It was tested and improved in a pilot case study. Each case was coded by a single coder, and weekly team meetings were held to ensure the use of common coding criteria.
  4. For example, policy areas (environment, urban planning) and the methodological tools used (surveys, meetings) all have too few positive cases in each category to reach significant results.
  5. The non-significant variables included characteristics of the organizers of the process (inclusion of civil society actors and involvement of other public administrations), other characteristics of the participants (number of participants and an alternative dichotomous variable capturing whether associations had been invited) and other characteristics of the process (process cost and number of policy proposals generated).
  6. A process was coded as having ‘low information’ when no information was provided or only limited oral information from a single source was offered (see Table 2).
  7. This is more often department (‘provincia’) or regional funding, but it could be also a national or European project. The existence of external funding for the process may indicate more availability of resources, but also potentially a lower motivation (i.e., the process would have not developed in the absence of the external funding).
  8. Within the category ‘associations’ we include all types of organized groups in civil society, ranging from well organized corporatist groups, as unions or business associations, to many other organizations like NGOs, neighbour associations, women associations, and so on.

Acknowledgements

This research has benefited from financial support from the Spanish Ministry of Economy and Competitiveness (Grant CSO2012-31832). We thank the anonymous reviewers for their comments, which have contributed substantially to improving the article.

Competing Interests

The authors have no competing interests to declare.

References

1 Alarcón, P., & Font, J. (2014). Where are the boundaries of deliberation and participation? A transatlantic debate. Journal of Public Deliberation, 10(2), Article 11.

2 Arnstein, S. R. (1969). A ladder of citizen participation. Journal of the American Planning Association, 35(4), 216–224. DOI:  http://doi.org/10.1080/01944366908977225

3 Bäckstrand, K. (2003). Civic science for sustainability: Reframing the role of experts, policy-makers and citizens in environmental governance. Global Environmental Politics, 3(4), 24–41. DOI:  http://doi.org/10.1162/152638003322757916

4 Baiocchi, G., & Ganuza, E. (2014). Participatory budgeting as if emancipation mattered. Politics & Society, 42(1), 29–50. DOI:  http://doi.org/10.1177/0032329213512978

5 Bherer, L., & Breux, S. (2012). The diversity of public participation tools: Complementing or competing with one another? Canadian Journal of Political Science/Revue canadienne de science politique, 45(2), 379–403. DOI:  http://doi.org/10.1017/S0008423912000376

6 Bherer, L., Gauthier, M., & Simard, L. (Eds.). (2017). The professionalization of public participation. Taylor & Francis. DOI:  http://doi.org/10.4324/9781315637983

7 Bou, J. (coord.). (2011). Refundar la Democracia: Presupuestos Participativos en Contextos Diversos [Relaunching Democracy: Participatory Budgeting in Diverse Contexts]. Sevilla: Atrapasueños.

8 Brown, M. (2008). Fairly balanced: The politics of representation on government advisory committees. Political Research Quarterly, 61(4), 547–560. DOI:  http://doi.org/10.1177/1065912907313076

9 Brown, M. (2009). Science in Democracy: Expertise, Institutions, and Representation. Cambridge: MIT Press. DOI:  http://doi.org/10.7551/mitpress/9780262013246.001.0001

10 Brown, M. (2014). Expertise and deliberative democracy. In S. Elstub & P. McLaverty (Eds.), Deliberative democracy: Issues and cases (pp. 50–68). Edinburgh: Edinburgh University Press.

11 Bua, A., & Escobar, O. (2018). Participatory-deliberative processes and public policy agendas: Lessons for policy and practice. Policy Design and Practice, 1(2), 126–140. DOI:  http://doi.org/10.1080/25741292.2018.1469242

12 Camprubí, M. (coord.). (2007). Consell de les Donnes Savies [Council of Wise Women]. Begues: Begues City Council.

13 Collins, H. (2014). Are We All Scientific Experts Now? Cambridge: Polity Press.

14 Cordoba City Council. (2010). Plan Municipal de Participación Ciudadana [Citizen Participation Plan]. Córdoba: Area of Citizen Participation of the Cordoba City Council.

15 Craft, J., & Howlett, M. (2013). The dual dynamics of policy advisory systems: The impact of externalization and politicization on policy advice. Policy and Society, 32(3), 187–197. DOI:  http://doi.org/10.1016/j.polsoc.2013.07.001

16 Dahl, R. (1998). On Democracy. New Haven: Yale University Press.

17 Delli Carpini, M., Cook, F., & Jacobs, L. (2004). Public deliberation, discursive participation, and citizen engagement: A review of the empirical literature. Annual Review of Political Science, 7, 315–344. DOI:  http://doi.org/10.1146/annurev.polisci.7.121003.091630

18 Edelenbos, J., Van Buren, A., & Van Schie, N. (2011). Co-producing knowledge: Joint knowledge production between experts, bureaucrats and stakeholders in Dutch water management projects. Environmental Science & Policy, 14(6), 675–684. DOI:  http://doi.org/10.1016/j.envsci.2011.04.004

19 Feldman, M. S., & Khademian, A. M. (2007). The role of public manager in inclusion: Creating communities of participation. Governance, 20(2), 305–324. DOI:  http://doi.org/10.1111/j.1468-0491.2007.00358.x

20 Fischer, F. (1993). Citizen participation and the democratization of policy expertise: From theoretical inquiry to practical cases. Policy Sciences, 26(3), 165–187. DOI:  http://doi.org/10.1007/BF00999715

21 Fischer, F., & Gottweis, H. (2013). The argumentative turn in public policy revisited: Twenty years later. Critical Policy Studies, 7(4), 425–433. DOI:  http://doi.org/10.1080/19460171.2013.851164

22 Font, J., Della Porta, D., & Sintomer, Y. (2014). Participatory democracy in Southern Europe. London: Rowman & Littlefield.

23 Font, J., Pasadas del Amo, S., & Smith, G. (2016). Tracing the impact of proposals from participatory processes: Methodological challenges and substantive lessons. Journal of Public Deliberation, 12(1), Article 3.

24 Font, J., Smith, G., Galais, C., & Alarcón, P. (2018). Cherry-picking participation: Explaining the fate of proposals from participatory processes. European Journal of Political Research, 57(3), 615–636. DOI:  http://doi.org/10.1111/1475-6765.12248

25 Forester, J. (1999). The deliberative practitioner. Encouraging participatory planning processes. Cambridge: MIT Press.

26 Forester, J. (2009). Dealing with differences: Dramas of mediating public disputes. Oxford: Oxford University Press.

27 Fung, A. (2003). Recipes for public spheres: Eight institutional design choices and their consequences. The Journal of Political Philosophy, 11(3), 338–367. DOI:  http://doi.org/10.1111/1467-9760.00181

28 Fung, A. (2006). Varieties of participation in complex governance. Public Administration Review, Special Issue, 66–75. DOI:  http://doi.org/10.1111/j.1540-6210.2006.00667.x

29 Fung, A. (2007). Minipublics: Deliberative designs and their consequences. In S. Rosenberg (Ed.), Can the People Govern? (pp. 159–183). New York: Palgrave. DOI:  http://doi.org/10.1057/9780230591080_8

30 Ganuza, E., Baiocchi, G., & Summers, N. (2016). Conflicts and paradoxes in the rhetoric of participation. Journal of Civil Society, 12(3), 328–343. DOI:  http://doi.org/10.1080/17448689.2016.1215981

31 Gastil, J., & Levine, P. (2005). The Deliberative Democracy HandBook. Strategies for Effective Civic Engagement in the Twenty-First Century. San Francisco: Jossey-Bass.

32 Gastil, J., & Richards, R. (2013). Making direct democracy deliberative through random assemblies. Politics & Society, 41, 253–281. DOI:  http://doi.org/10.1177/0032329213483109

33 Güemes, C., & Resina, J. (2019). ‘Come together?’ Citizens and civil servants dialogue and trust. Australian Journal of Public Administration, 78(2), 155–171. DOI:  http://doi.org/10.1111/1467-8500.12377

34 Hendriks, C. M. (2005). Participatory storylines and their influence on deliberative forums. Policy Sciences, 38, 1–20. DOI:  http://doi.org/10.1007/s11077-005-0870-3

35 Hendriks, C. M., & Lees-Marshment, J. (2019). Political leaders and public engagement: The hidden world of informal elite–citizen interaction. Political Studies, 67(3), 597–617. DOI:  http://doi.org/10.1177/0032321718791370

36 Howlett, M., Ramesh, M., & Perl, A. (2009). Studying Public Policy: Policy Cycles and Policy Subsystems. Oxford: Oxford University Press.

37 Hunt, J., & Shackley, S. (1999). Reconceiving science and policy: Academic, fiducial and bureaucratic knowledge. Minerva, 37(2), 141–164. DOI:  http://doi.org/10.1023/A:1004696104081

38 Johnson, L., & Gastil, J. (2015). Variations of institutional design for empowered deliberation. Journal of Public Deliberation, 11(1), Article 2. DOI:  http://doi.org/10.16997/jdd.219

39 Landwehr, C., & Holzinger, K. (2010). Institutional determinants of deliberative interaction. European Political Science Review, 2(3), 373–400. DOI:  http://doi.org/10.1017/S1755773910000226

40 Maiello, A., Viegas, C. V., Frey, M., & Ribeiro, J. L. D. (2013). Public managers as catalysts of knowledge co-production? Investigating knowledge dynamics in local environmental policy. Environmental Science & Policy, 27, 141–150. DOI:  http://doi.org/10.1016/j.envsci.2012.12.007

41 May, P. J., Koski, C., & Stramp, N. (2016). Issue expertise in policymaking. Journal of Public Policy, 36(2), 195–218. DOI:  http://doi.org/10.1017/S0143814X14000233

42 Migchelbrink, K., & Van de Walle, S. (2019). When will public officials listen? A vignette experiment on the effects of input legitimacy on public officials’ willingness to use public participation. Public Administration Review, 80(2), 271–280. DOI:  http://doi.org/10.1111/puar.13138

43 Nabatchi, T. (2012). Putting the ‘public’ back in public values research: Designing participation to identify and respond to values. Public Administration Review, 72(5), 699–708. DOI:  http://doi.org/10.1111/j.1540-6210.2012.02544.x

44 Nabatchi, T., & Leighninger, M. (2015). Public Participation for 21st Century Democracy. Hoboken, NJ: Wiley. DOI:  http://doi.org/10.1002/9781119154815

45 Newman, J., Barnes, M., Sullivan, H., & Knops, A. (2004). Public participation and collaborative governance. Journal of Social Policy, 33, 203–223. DOI:  http://doi.org/10.1017/S0047279403007499

46 Nez, H. (2011). Natures et légitimités des savoirs citoyens dans l’urbanisme participatif. Sociologie [online], 4, 2. DOI:  http://doi.org/10.3917/socio.024.0387

47 Petts, J. (1997). The public-expert interface in local waste management decisions: Expertise, credibility and process. Public Understanding of Science, 6(4), 359–381. DOI:  http://doi.org/10.1088/0963-6625/6/4/004

48 Petts, J., & Brooks, C. (2006). Expert conceptualizations of the role of lay knowledge in environmental decisionmaking: Challenges for deliberative democracy. Environment and Planning A: Economy and Space, 38(6), 1045–1059. DOI:  http://doi.org/10.1068/a37373

49 Roberts, J. J., Lightbody, R., Low, R., & Elstub, S. (2020). Experts and evidence in deliberation: Scrutinising the role of witnesses and evidence in mini-publics, a case study. Policy Sciences, 53, 3–32. DOI:  http://doi.org/10.1007/s11077-019-09367-x

50 Sintomer, Y. (2008). Du savoir d’usage au métier de citoyen. Raisons Politiques, 31, 115–134. DOI:  http://doi.org/10.3917/rai.031.0115

51 Sintomer, Y., & Del Pino, E. (2014). The national and regional contexts of participatory democracy. In J. Font, D. Della Porta & Y. Sintomer (Eds.), Participatory democracy in Southern Europe (pp. 21–36). London: Rowman & Littlefield.

52 Smith, G. (2009). Democratic Innovations: Designing Institutions for Citizen Participation. Cambridge: Cambridge University Press. DOI:  http://doi.org/10.1017/CBO9780511609848

53 Spada, P., & Allegretti, G. (2017). Integrating multiple channels of engagement in democratic innovations: Opportunities and challenges. In M. Adria & Y. Mao (Eds.), Handbook of Research on Citizen Engagement and Public Participation in the Era of New Media (pp. 20–37). IGI Global. DOI:  http://doi.org/10.4018/978-1-5225-1081-9.ch002

54 Spada, P., & Ryan, M. (2017). The failure to examine failures in democratic innovations. PS: Political Science & Politics, 50(3), 772–778. DOI:  http://doi.org/10.1017/S1049096517000579

55 Talpin, J. (2011). Schools of Democracy. How Ordinary Citizens (Sometimes) Become Competent in Participatory Budgeting Institutions. Colchester: ECPR Press.

56 Yang, K. (2005). ‘Public administrators’ trust in citizens: A missing link in citizen involvement efforts. Public Administration Review, 65(3), 273–285. DOI:  http://doi.org/10.1111/j.1540-6210.2005.00453.x

57 Yearley, S. (2000). Making systematic sense of public discontents with expert knowledge: Two analytical approaches and a case study. Public Understanding of Science, 9, 105–122. DOI:  http://doi.org/10.1088/0963-6625/9/2/302