Multiplayer governance and democracy: convergence or divergence?

IPSA 2016 - Politics in a World of Inequality Session CS07 Public Policy and Administration Panel: Issues of Governance and Public Administration: Stakeholders' Participation, Compliance and Trust Building Tuesday, July 26 th, 2016 Multiplayer governance and democracy: convergence or divergence? Isabelle Lacroix, Ph.D., Prof. Université de Sherbrooke Abstract Since the 1990s, Western states have generally increased multiplayer governance within their administrations (Gaudin, 2006). One expectation behind this trend was to increase, through the participation of all stakeholders, the effectiveness of decisions, the accountability of the various participants and the democratization of decision-making processes (Giroux, 2006; Paye, 2005; Godbout, 2001 Paquet, 2001). What really happened in those administrations? The issue at the core of this new model of organizations is the distribution of effective power (Squashes, 2012; Bhérer, 2011; Vergez and Caddy, 2002; Thibault, Lequin and Tremblay, 2001). Does multiplayer governance promote equal distribution of power in decision-making bodies? Do multiplayer-making processes increase equality between actors involved in the public management of collective resources? Using five case studies, based on direct observation of the functioning of decision-making of several bodies active in different areas of state intervention and a series of interviews made with public organizations practitioners (Lacroix, 2012; Lacroix and Dostie-Goulet, 2013, 2014, 2015; Grégoire, 2015; Grondin-Landry, 2016), we describe what kind of practices result from the implementation of multiplayer governance and help to understand what changes it implies, in relation to issues of efficiency and democratization of organizations. 1

Introduction A few years ago, through an extensive review of literature, we proposed the following definition of multiplayer governance: the rules and collective processes, formal or informal, by which the actors are involve in the making and implementation of public actions. These rules and processes, such as the decisions reached, are the result of constant negotiation among multiple actors involved. This negotiation facilitates the sharing of responsibility between all actors involved, each with some form of power, and guides decisions and actions (Lacroix, 2012). This definition has the great advantage to echo both in the scientific literature and in the reality of practitioners we met at that time. This definition still seems relevant today and we used it to study the multiplayer governance and the democratization of collective management in modern states. Is the deepening of the multiplayer governance go hand-by-hand with a deepening of the democratization of public action? This communication looks at this issue by putting together several case studies produced in Québec in recent years. We first present a few central concepts and theoretical grids that have guided our approach, then we present various case studies from which we draw, in the final section, general findings that should enable us to draw our first empirical multiplayer governance. 1. Governance and Democracy The definition of governance presented in the introduction is based on three fundamental elements for understanding this reality; the concepts of networks, participation and change (Lacroix St-Arnaud, 2012). Multiplayer governance embodies a network of actors who cooperate to ensure the production of public actions. Within these networks, power relationships tend to be more horizontal, although they generally remain unequal, in order to generate positive decisions and ensure their full implementation. These networks are based on the participation of mobilized individuals. This participation comes in three main actions: information (participants receive information resources from one or more other members, in a position of authority or not), the consultation (participants share opinions and suggestions using their limited power of influence) and participation (participants contribute directly to the decision and its implementation through shared formal authority) (Caddy and Vergez, 2002). These three levels of participation are related to a variable distribution of power and responsibilities among the participating members, 2

information is the level where the members have less space to exercise their power, and participation, the level where they formally have more power (Bhérer, 2011). The multiplayer governance represents a change in the "pilotage de l action publique" called pragmatic by Gaudin (2002). The late 1990s saw the emergence of a polycentric and fragmented political environment while the state is no longer an actor who can simply imposes its will on other stakeholders (Gaudin, 2002; Moreau Defarges, 2003; Smouts 2003; Giroux, 2006; Côté, Lévesque and Morneau, 2007; Paquet, 2010). The environment is now made up of constraints and varied obligations that directly affect the culture of regulation, giving rise to new social and political expectations. The result is an even greater demand for democratic deepening of public decision making and collective action, particularly in terms of expected transparency related to decision and implementation of public policies (Lacroix St-Arnaud, 2012). In regard to democracy, we believe that the concept of multiplayer governance requires to be investigated from two complementary angles. First, from representative democracy where the actors are recruited through elections. In this case, members of governing body acquire their legitimacy to participate in the management of the organizations concerned from the citizen choice or peer choice. This method of appointment is generally strongly framed by very specific rules. Second, the notion of participatory democracy is often associated with multiplayer governance while being involved in the mission or in the actions of the organization can be considered sufficient to be able to contribute to the management of the organization (CSE, 2006). In this case, the rules framing the recruitment of participants are generally more flexible and sometimes entangled lower decisionmaking power. For many public organizations, new management practices explicitly imposed this deepening of participatory or representative democracy, now considered as good practices (CSE, 2009). As a result, several practitioners consider the spread of multiplayer governance within public organizations as a response to the largely shared expectations of greater democratization of public management (Lacroix, 2012). 2. Some theoretical grids, some concepts The multiplayer governance is at the junction of four classic theoretical grids in political science which, taken in complementarity, offer a tenfold explanatory potential. In this text, we selected a few key concepts that, we believe, can offer an interesting understanding potential. Thus, like Easton (1974), we can conceive public governance networks as political systems that ensure their persistence and stability over change. The exchange between the system and its environment (input - output 3

- feedback) ensures this change, the dynamic adaptability of the system to its environment. Within this system, working actors are constrained by institutions that limit their actions and have, simultaneously, a margin of freedom (marge de liberté) they may or may not used. According to Campbell (2004), the observable institutional changes are the result of constraints innovations. These can occur by bricolage or by translation, but they are always the result of an institutional rearrangement built from the existing internal architecture and the external influences that contaminate the space available for change. According to Crozier and Friedberg (1977), even if institutions are critical, a margin of freedom always persist, defined by the areas of uncertainty (zones d incertitude), where actors mobilize resources to implement efficient strategies regarding their relations of effective power. We believe that this strategic exercise frequently takes the form of mutual partisan adjustment, as defined by Lindbloom (1965), which take place in the interactions between actors working in democratized organizations. Thus, decisions would reflect a constant adjustment formed by the incremental evolution of the relations between the actors, ensuring their relevance in the current governance and would represent what Lindbloom called "The Intelligence of Democracy." 3. Case studies 3.1 The cases The origin of these case studies is as varied as their scope. They were developed as part of investigations conducted with colleagues and graduate students since 2010. In this text, we selected 20 organizations studied trough five separate research projects. Voluntarily, we have chosen very different cases because we wanted to cover a wide range of practices associated with multiplayer governance to be able to highlight common practices. Case 1. In 2012, we studied the governance practices of two school boards of Québec. We studied six topics specific to these organizations, from direct observations and semi-structured interviews 1. The data collected covered both observed practices, prescribed practices and perceptions of the actors involved. At the head of these organizations there is a Board of Commissioners (Conseil des commissaires) composed of elected officials, parents and members of management. Only elected commissioners have the right to vote and the Board shall hold public decision-making meetings. 1 Under the direction of Guy Pelletier, Ph.D. and André Brassard, Ph.D., Lacroix, Isabelle. 2012. Les pratiques de gouvernance de commissions scolaires du Québec. PhD thesis, Université de Sherbrooke. 4

Case 2. From 2012 to 2016, we observed the activities of decision-making bodies from 15 public organizations in seven different sectors (education, youth, childhood, environment, regional development, economic development and health) 2. The data collected were only related to observed practices and the prescription of formal operating rules. The decision-making body at the top of these organizations is composed of a variable number of players, but they all have in common to be multiplayer. Case 3. In 2013, we attended the governmental forum on higher education organized by the Parti Québécois government to end the social crisis ensued from the pronounced increase of tuition fees 3 for university students. It was a temporary governmental forum and the participants were invited by the government, so they did not have any formal decision-making power, but they all had some influence power, though variable, depending of the different actors. The data collected mainly relied on direct observation of the five meetings held and the rules associated with the brief operation of that organization. In this case, it is therefore observed practices. Case 4. In 2015, a study of the governance relationship between actors of municipalities, regional county municipality (MRC) and civil society - including private contractors related to the aborted project Estrie Enviropôle 4 (a large-scale biogas project) was conducted. In that case, power relations within these instances, the legitimacy and representativity of the actors involved and their influence capacity were questioned. This survey was based mainly on semi-structured interviews with stakeholders, including policymakers. The data collected relied on the perceptions of the respondents involved and the formal prescribed practices of these organizations. Case 5. In 2016, the entire citizen consultation process led by the City of Sherbrooke to conceive and adopt their planning and development plan was studied 5. Essentially, this approach was based on citizen consultations to broaden participation of the community in the production of this policy, but only elected had a real decision-making power within the City Council. Like the previous case, this 2 With the collaboration of Eugénie Dostie-Goulet, co-investigator. 3 With the collaboration of Daniel-Pierre Roy, research assistant. This data collection was one of the first projects of the Observatoire des politiques publiques de l Université de Sherbrooke, recently created. 4 This survey was produced by Pierre Grégoire, graduate student. I was the research director of the student throughout the conduct of his study: GREGOIRE, Pierre. "The municipal governance: the case of Estrie Enviropôle". Master s thesis directed by Isabelle Lacroix (18 March 2015). 5 This survey was produced by Alexis Grondin-Landry, graduate student. I was the research director of the student throughout the conduct of his study: GRONDIN-LANDRY, Alexis. "Citizen participation in the elaboration of the planning and development plan of Sherbrooke city." Master s thesis directed by Isabelle Lacroix (24 May 2016). 5

survey was based mainly on semi-structured interviews with stakeholders, so data collected here are also related to the perceptions of those involved and the prescribed rules governing their operations. The objective of this text is to emphasize what is common to all of these organizations, to be able to create an empirical basis for a concept used by both practitioners and academics. From what we know, the concept of multiplayer governance has never being anchored in a review to bring out fundamental criteria from the study of the governance of pubic organizations. We throw here the first bricks of our theoretical generalization approach. We are well aware of the methodological limitations posed by the way we use various cases. Although all of the case studies used similar research tools - either interviews or direct observations, or both - each was targeting slightly different research objectives than the others. They were all interested in multiplayer functioning of these organizations, but some with a more pronounced orientation for the decision, for consultation or for the realization of the organization mission. Considering those limitations, we restricted, for the present communication, the number of components studied to those for which we had data collected directly in each study. We have excluded several components that, although they could have possibly brought an interesting perspective, had not been systematically studied in each case and whose findings could not be generalized to all. These significant limitations invite us to be humble in our conclusions. In this sense, we do not claim to provide a complete picture of the multiplayer governance nor to generalize our findings to all public organizations operating under similar guidance. However, we believe that the diversity of the cases studied presented, and the strict guidelines we have followed, allow us to believe that our findings have certain potential for transferability. We also believe that this is only the first step and further studies will be needed to deepen and confirm our results, and if possible, our conclusions. 3.2 Components In this research we used three components: 1) the decision-making practices related to the cases studied, 2) recruitment of actors involved and 3) their role. These components allow us to cover the organizations and their participating members. Each of them includes three sub-components used to categorize our observations on these components, as summarized in Table 1. 6

Table 1. Observed components of multiplayer governance Decisional practice Role Recruitment Procedures Statut Provenance Structures Ressources Designation Strategies Influences Legitimacy In the case of decisional practice, we considered both the observed practices (direct observation), the prescribed practices (documentary observation) and the practices perceived (interviews with respondents). We seek to qualify the practice (consensual, adversarial, collaborative...) from the procedures defined by the actors, the formal structures in place and the strategies implemented by the various actors involved. This component is more focused on the organization. The role is a component rather based on the actors, but the rules of the organization are influential in this case. For this component, we considered the observed roles (direct observation) and the perceived roles (interviews with respondents). We based this component on to the formal status of the actors involved, the resources owned/mobilized/missing and the influences mainly related to the collective decisions studied. Finally, the component of recruitment is based on the provenance of the different actors involved, the method of appointment used in their organization and the legitimacy attributed to them. This component is also more focused on the actors. For the analysis of this component, we considered the formal recruitment procedures (documentary observation) and the perceptions of the actors involved (interviews with respondents). Those components are the result of several inductive case studies put together. We believe that this pooling is meaningful regarding our understanding of this political phenomenon whose implementation tends to multiply, considered in a more comprehensive way. 3.3. The results The five cases presented in the previous section encompass very different, but complementary, realities of the multiplayer governance. They reveal the scope and complexity of this multifaceted and multidimensional concept. This overused concept covers particularly varied realities of governing, the five cases listed tend to prove it and should have a comprehensive value because of this correspondence to 7

equally varied situations. We believe that our study offer a significant generalization potential, though limited to the multiplayer space. These five cases represent to us various realities of the same orientation: in all these cases there is a multiplayer type of governance implemented with the participation of many actors, unequal as to their respective powers. Beyond their own characteristics, what about the presence of the targeted components? If we take each briefly presented case, it is possible to draw conclusions from each component. We must remembered that each case was the subject of a different research project. But it is interesting that, despite the wide variety of cases, certain trends emerge from the comparison of selected components. Table 2 provides a summary of our comparative approach. Table 2. The observed and perceived multiplayer governance, from each case Cases / Components Decisional practice Role Recruitment Case 1 - School boards Case 2 - Decisionmaking bodies Case 3 - Government Forum Case 4 - Relations municipalities, MRC and private contractor Case 5 - Municipal planning and development plan Consensual, with the presence of various oppositions Consensual, almost complete absence of opposition Sought consensus, presence of opposition little discussed Confrontational, importance of temporary coalitions Consensual, presence of opposition Varied resources, unequal influence, importance of individual Varied resources, unequal influence, importance of experts Varied resources, unequal influence and domination of one actor (government) Varied resources, coalition of opponents contributes to the failure of the project Varied resources and domination of one actor (municipality) Legitimacy of elections, expertise and involvement Legitimacy of expertise and involvement Legitimacy of elections and involvement Legitimacy of elections and involvement Legitimacy of elections and involvement If the observation of a wide variety of resources resulting in a very unequal influence among actors and sometimes by a domination of the official public authority (government or municipal) is not surprising, two observations seem more unexpected. First, with regard to recruitment, the legitimacy of involvement seems much more influential than the legitimacy of expertise or the elective legitimacy. 8

This may seem logical in organization with a functioning based on the participation of actors involved in networks, but it still raises questions about the expectations and anticipated results, especially when it is related to the wide variety of resources and influences of those actors. Second, the spread of consensual practice, even in cases where there is a presence of opposition, surprises. It is interesting to note that the only exception to this practice is found in case 4, where a multiplayer governance practice was implemented because of a conflictual situation. Beyond these general observations, we wish to address each component. 3.3.1. Decisional practice Regarding the decisional practice, the component more focused on the organization taken as a whole, four of the five cases have adopted a consensual practice (cases 1, 2, 3 and 5). It was even an often stated objective that justify the use of a governance type of operation and this kind of practice was considered by many respondents as "the good practice". However, despite this trend, the opposition is not excluded from such practice. In case 1, even if the decision-making practice generally ended with an observed consensus or unanimous formal vote, some issues raised the expressing of sharp opinions, and some oppositions were sometimes even affirmed. The perception related to these episodes of opposition was generally polarized and for some participants it indicates a malaise of operation. In case 2, we observed a very important space occupied by the transmission of information, a lower space occupied by the opinions and an almost missing space occupied by the opposition. Although the bodies were all decisional, the function of transmitting information was paramount and the undertaking more of a consultation exercise, sometime even a pedagogical one, appeared to correspond to an expected practice. The decisions generally endorsed the presentation made of the issue by the actor responsible. In case 5, the study of the consultative procedures highlights the differences between the expectations of organizers and participants, their evaluations of the approach and the contribution of the consultations. Thus, if elected officials and municipal administrators made a very positive reading of the approach and the contribution of consultations, participant citizens judged it severely. The consultations were not decisional and it resulted in a strong impression, on the part of citizens, that the expression of opposition had not been considered or treated by established consultative procedures. The other two cases (3 and 4) are interesting to deepen the component of decisional practice, because they present a conflict situation. So in regard to case 3, throughout 9

the preparatory meetings and the forum itself, governmental actors - the Prime Minister at the time and the Minister of Higher Education - repeated constantly that they all had to create consensus and convergence. The panel presentations generally sought to enrich the reflections, but the committee work was essentially oriented to build consensus, which were then presented at the plenary session. The context of conflict was certainly very influential in the choice of this orientation, but it had two observable consequences. First, the positions of confrontation and opposition were not or very little discussed. Some protester actors have even boycotted the final meeting (the student association ASSE) after having participated in the preparatory sessions. Second, several players showed, in informal discussions, skepticism about the contribution of this kind of operation, but said they wanted to "play the game". At the end, the consensus made by government actors was not shared by all the actors involved 6, some even officially expressed their dissidence (one leader of a national student union even refused the final group photo), but all stakeholders at the forum demanded that the government still maintain their presence and their participation in all proceedings following this forum. And, as for the case 4, the conflict situation has led the actors to try to reach a consensus by using procedures associated with multiplayer governance. At the heart of the confrontation, the main municipality involved set up an inter-citizen committee to study the project and to make recommendations to enable the acceptability of the project. The work of this committee was never recognized, opponents have never recognized the credibility of it, the project was never completed and was ultimately withdrawn. In sum, if the consensual trend seems well established and matches what is expected to be a consequence of the introduction of a multiplayer governance type of functioning, difficulty or inability to process oppositions seems equally important. Thus, the cases reveal that decision-making processes produced recognized consensus, leading to an absence of opposition from the "inside" of the governance system or a lack of treatment thereof. Some conflict situations even presented the development of a network outside the formal system to allow the expression of opposition. Does the multiplayer governance allow functional decision-making processes only in a consensual context? The cases lead us to raise the issue they reveal of the gap between goals and virtues associated with multiplayer governance and the observed reality. 6 At the end of the final forum, the Prime Minister confirmed the decision announced by the government prior to the forum: they were maintaining the cancellation of the raise of tuition fees announced by the previous government and establish the indexation of these fees, frozen for several years. 10

3.3.2 Roles Analysis of roles seems interesting, as it reveals a varied picture including also important trends. The expected roles exercised allow the understanding of the influences of the various players, by observing the strategies implemented and the resources deployed. Depending on the case, we treated the perceived roles (interviews) or the observed roles from the actions of the actors involved, including the use of the right to speak. Consistent with what is associated to the concept of multiplayer governance, all case studies reveal that actors possess varied resources. This reality is present in respondents' perceptions as well as in the observations that have been made. Possession of strategic information seems to be one of the most critical resources for a possible influence on participatory decision-making, and so, in both consensual and conflictual situations. This element seems so important in the observed processes, that we believe it requires a dedicated study. But we can already say that this wide variation in resources possessed and mobilized by the actors is also reflected, in all cases studied, by great inequality in the influences exercised 7. This verdict goes in agreement with the reality theorized by the concept of multiplayer governance as we earlier presented. Despite this strong trend, differences from the comparison of cases bring out the need for deepening, particularly in regard to whom are the actors advantaged in these power games. In case 1, a trend emerged from the perceptions gathered who seems to point out that the role played by the actors in governance structures are based on two complementary elements: first the important expectations created by the formal rules guiding the role exercised (formal role played by the management, the representative of the community, the employee representative, etc.) and also, some respondents said that it is even more important, the individual resources of the participants themselves who will be mobilized to increase their influence and, at the same time, the influence of the represented group. For example, respondents frequently spoke of the great communication skills of some relay 8 or the network of some actors as individual resources who can increase the influence of participants involved, beyond their official status. In case 2, observations confirm the large and important area occupied by the experts against elected officials, users or community representatives considered more 7 In this text, we simply associated the notion of influence to some form of power that actors can exercise. 8 By representative-relay we mean the individual members of a group of actors concerned by the mission of the organization and whose role in the body is, firstly, to make known the interests and the concerns of this group and, secondly, to transmit to this group information shared and decisions taken by the body. 11

broadly. They seemed to be more mobilizing the resources at their disposal, including the right to speak possessed by all present members. We had, in other recent communications, raised questions about the democratic exercise associated to these participatory bodies if the opposition was absent (Lacroix and Dostie- Goulet, 2015). Do very broadly consensual decision-making processes could allow the expression of divergent preferences, as one of the basic principles of a democratized functioning? We found an important similarity between cases 3 and 5 while the government actor has a dominant position within the network of actors, where everyone still has a real, but limited, power of influence. Thus, we considered the ability of government officials to impose their vision to the Forum, even more over critical decisions in the heart of the crisis, as being representative of the power relations between the actors involved because it was largely widespread. If in case 3 we may believe that the conflict situation that justified the creation of the temporary governance network could have accentuated the imbalance of influences, a similar observation is made for case 5 even if the starting situation was not, in this case, confrontational. In the fifth case, if elected officials and municipal administrators make a very positive evaluation of the approach and the contribution of consultations, participants have opposite perceptions related to the citizen consultations. Expectations appear here to be in the heart of this argument, but also the very unevenly distributed resources among the different actors could offer an another explanation for this divergence, such as knowledge, a certain control over the issue, the time available, the direct access to decision makers... This study questions the perceptions of participants in regard to the transmission of information, the transparency and clarity of the consultative process, the reach for cooperation, as well as what Gagnon and Gauthier (2013) call "early involvement", being the fact that the consultations are made before important decisions are taken, so that participants car exercise a real influence over decision. A bit like in case 3, but in a less intensive way, respondents have reported the domination of government actor - here the municipality. Both cases lead us to question the horizontal aspect of power relations within multiplayer governance networks. And finally, case 4 reveals a particular situation regarding the distribution of influences between the different actors. We observed the formation of two coalitions in opposition. Private developers received support from policymakers of the municipality directly affected by the project and a part of the population (known from public opinion poll). While opponents were a local environmental pressure group set up for the occasion, who supported the decision makers from other municipalities of the MRC against the project and some citizens (engaged through a petition). This seems, at first sight, in contradiction with the fundamentals principle of governance while the multiplayer network of actors seems to do not 12

have succeeded in establishing a minimum cooperation 9 between actors to ensure its operation. In sum, the study of the perceived and exercised roles within organizations functioning under a multiplayer mode of governance reveals a distribution of resources and, therefore, influence capacity of the different actors in a variety of networks. This large variation sometimes allows the domination of an actor, usually in positions of authority, but could also lead to the fragmentation of the network in a situation of the formation of opposing camps. This could conduct to a breaking point in the multiplayer governance, observed by a rupture of cooperation or relationship. 3.3.3 Recruitment The last component, related to the recruitment of actors, is the one we least developed for this communication, but it seemed nevertheless essential to address some issues 10. Here, two important findings. First, all the cases have appointment methods (prescribed rules), operations in recruiting and decision-making process focusing on both elective legitimacy and legitimacy of involvement. Generally, the actors involved in governance multiplayer network are legitimate to exercise their influence, because they were elected or because they are concerned with some issues or the general mission of the organization. Case 3 seems particularly representative in this regard. At the beginning of 2013, the Quebec government established a forum on higher education in order to find a way out of the crisis arising from the increase of the tuition fees made by the previous government and which led to a broad movement of student strike, recurrent manifestations degenerating regularly in clash with police forces and ultimately conduct to a general election. This forum brought together, in addition to government officials 11, the actors only those invited by the authority involved in this issue so they can contribute to prepare the summit of February, during four preparatory meetings oriented on distinct and complementary themes 12. Also, in cases 4 and 5, networks were composed of municipal authorities responsible of the approach, citizens and group representatives involved in the organizations. Second, the study of the five cases leads us to question the legitimacy based on the expertise when it was the object of observation or perceptions validating their 9 By cooperation we designate, like Crozier and Friedberg (1977): "temporary and provisional acceptance of certain interdependence" (p. 107, note 33). 10 This component will be further developed in a text later. 11 For example, besides governmental actors, major university and college student associations, major unions and employer organizations attended the forum. 12 The themes were: The quality of higher education; Accessibility and participation in higher education; Governance and university funding; The contribution of institutions and research to the development of Quebec. 13

influential contribution only in cases 1 and 2. The observations conducted of the decision-making bodies of 15 organizations, as part of case 2, revealed that the space occupied by the experts (mainly management and professionals) was higher than the one occupied by the actors concerned (citizens, community representatives, specific users of the organization). More specifically, in case 1, it appears that methods of nomination and provenance of the actors are particularly varied and typically falls under the specificity of each case rather than under some rules prescribed or even defined by practice. This goes from a recruitment based on expertise to a formal election by the peer, or through recognition of the experience of involvement. The legitimacy accorded to actors also seemed more often based on the contribution of the individual involved than on the method of appointment that led him to become a member of the body. In sum, these findings seem to support the assertion of Gaudin who claimed in 2002 that the governance was, in fact, essentially a pragmatic approach seeking to facilitate the control and the achievement of the mission of public organizations. We understand, in this sense, the tendency of organizations to include actors in authority and those who are already involved. However, more studies will be needed to explore those results. 4. Discussion 4.1 Some theoretical reflections The multiplayer processes we have studied reveal a trend that we associate with governance as it takes form in practice. Although analysts say it is a contemporary reality, theoretical tools, some of them developed decades ago, offer us an interesting light on this practice. For us, the issue of consensus must be questioned at first. We believe that this is the most compelling illustration of what Lindbloom (1965) called mutual adjustment. The fact that the multiplayer governance is associated with a discourse oriented toward consensus and the fact that it is largely reflected in the observed practice is a result of this functioning based on a network of actors in constant negotiation and whose inter-influence is constant. One adjusting to the other, it reveals more consensual governance relationships based on effective cooperation in which everyone sees his interest considered. Is this a consequence of the "intelligence of democracy" implemented? We will return to this notion in Section 4.3. The corollary of this consensus culture is the limited space given to the opposition. For us, an avenue to explore this, is the twinning of the application of a logic of 14

mutual adjustment (Lindbloom, 1965), based on actors in constant interaction with each other, their margin of freedom strongly constrained by formal rules of the game (Crozier and Friedberg, 1977) and even more by informal rules for an idealized decisional practice. In this case, the area of uncertainty is strongly limited by the formal rules of operation focused on the search for consensus, a very strong expectation of the limitation of expression and opposition, and generally expected rallying framing practices. However, as defended by Crozier and Friedberg (1977), despite a very important constraint system, our study shows that strategic actors could still exercise their margin of freedom, as we have seen certain opposition in some organizations, conducted by a few individuals presenting a specific reading of their own interest. The issue of expectations can hardly be ruled out of the reflection about participation in collective management through multiplayer decision making in public organizations. As we have addressed this issue at the beginning of the text, the act of opening the door to the participation of stakeholders means for these participants to give them power. This reality, central to the concept of governance, necessarily leads to high expectations firstly on the part of these actors, they will be expected to have a noticeable impact on the processes in which they participate. But then, it also seems to bring general expectations for public management. This would create an expectation that is, if we can afford the expression, to be "institutionalized" and becomes a source of legitimacy for organizations. They have to comply. It is in this sense that multiplayer decision making practices become mandatory for the innovations (Campbell, 2004) in this matter, particularly with respect to their consensual aspect associated with "good practices". Despite this legitimation that seems necessary, the expectations of players, however, remain variable and regularly bring frustration in practice. We believe that our study also adds new insight to this frustration of participants, who has been repeatedly documented by investigations, in Quebec as elsewhere (Godbout, 2001; Brachet, 2002; Blondiaux 2001; Royo, Yenato and Acerte, 2011). If this dissatisfaction in regard to non-effective power, or to a perception of having no significant impact was real and was also reconfirmed in our investigation, we believe that the importance of the legitimacy of involvement has something to do with it. The fact that recruitment enhances significantly the contribution of stakeholders, often as a "counterweight" to the actors whose legitimacy is based on election or expertise, can make that the marginalization of their contribution or their interest will be even more unacceptable. This is in opposition to the notion of governance adopting the principles of mutual adjustment and may seem paradoxical. 15

What seems to be the root of this theoretical rupture would be the dispersion of resources 13 and the strategies they allow. For Crozier and Friedberg (1977), the strategic actor uses his strategic resources around existing areas of uncertainty to accentuate the margin of freedom of his own. For us, this is fully consistent with what was observed in the studied organizations. First, control around areas of uncertainty would explain the predominance of government actor seeking to limit uncertainty and to use its formal and legitimate resources to maximize its action. Similarly, the effective influence of experts could be understands by the fact that information seems the most valuable resource and they are the most credible actor concerning the information specific to those organizations. This inequality could be offset by practical transmission and highly institutionalized sharing practices based on the legitimate expectations of participation from those who are concerned and have the right, recognized, to access information necessary to assure their full contribution to the governance of the organization. It will be interesting to repeat this study in a few years to see what trend has emerged: the strengthening of the influence of experts on network governance or the empowering of the actors concerned. For now, we could not say. 4.2 The multiplayer governance: a movement of interdependence In the review of the cases studied, a concept has emerged from another angle, the concept of interdependence designed in logic of systemic loop, as developed by Easton (1974). The pooling of varied cases led us to conceive this interdependence as an additional limit to the scope of actors: the movement of each participant comes as an imposition to network actors, and influences the actions and strategies of all the others, "they should work from/with these movements." Observing these interdependencies, and their influence, should be possible by observing the functioning of the governance networks based on constant adjustment relating to areas of uncertainty. We even believe that the observation of dominant players, as governmental actors or experts possessing more power, could reveal the orientation of interdependence and obligations it entails, as we note their positioning and movement over the governance path created and the evolution of public management associated with the network of actors. Again, additional studies are needed. 4.3 And what about democracy? In 2000, in his book On Democracy, Dahl established five criteria for assessing the democratic aspect of the governmental guidance: 1) Effective participation, 2) Voting equality, 3) Enlightened understanding, 4) Control of the agenda and 5) Inclusion of 13 We refer here to the concept developed by Dahl in his book Who Governs?, in 1961. 16

adults (Dahl, 2000, 37-38). For Dahl, the criteria could be used to validate the extent of democratic functioning in a given situation according to the basic principles identified as essential: "And your constitution must be in conformity with one elementary principle: that all the members are to be treated (under the constitution) as if they were equally qualified to participate in the process of making decisions about the policies the organization will pursue "(Dahl, 2000, 37). The orientation chosen in Québec to include an even broader participation of multiple stakeholders" in all public governance processes seems not necessarily to result in a deepening of the democratic components of organizations where the extent of a consensual practice surprised. Persistent inequalities appear almost unavoidable in multiplayer decisional practice. Thus, our investigation leads us, like Dahl, to conclude to the existence of a certain law: "The law of time and numbers: the more democratic a citizens unit contains, the less citizens that can participate directly in government decisions and the more that they must delegate authority to others" (Dahl, 2000, 109). In sum, there would always be some limits to the deepening of democratization practices in public organizations, particularly with regard to participatory democracy practices, despite the real and effective implementation of multiplayer governance principles. 17

REFERENCES Bhérer, L. (2011). «Les relations ambigües entre participation et politiques publiques». Participations. 1(1) : 105-133. Blondiaux, L. (2001). Démocratie locale et participation citoyenne : la promesse et le piège. In Mouvements. 18 (novembre-décembre 2001), 44-51. Brachet, P. (2002). Démocratie participative : du slogan à la mise en œuvre. In Mouvements. 19 (janvier-février 2002), 115-119. Caddy, J. et C. Vergez. (2002). Des citoyens partenaires. Information, consultation et participation à la formulation des politiques publiques. Éditions OCDE. Document téléaccessible à l adresse : http://lnweb18.worldbank.org/essd/sdvext.nsf/66bydocname/descitoyenspartenair espublicationparl%e2%80%99ocde/$file/citoyenspartenaires.pdf Campbell, J. L. (2004). Institutional Change and Globalization. Princeton : Princeton University Press. Côté, L. et Lévesque, B. et Morneau, G. (2007). L évolution du modèle québécois de gouvernance. In Politique et Sociétés. 26 (1). Conseil Supérieur de l Éducation - CSE. (2006). Rapport annuel sur l état et les besoins de l Éducation 2005-2006. Agir pour renforcer la démocratie scolaire. Québec : Conseil supérieur de l éducation. Conseil supérieur de l éducation - CSE. (2009). Rappel des positions du Conseil supérieur de l éducation sur la gouverne en éducation. Québec : Conseil supérieur de l éducation. Crozier, M. et E. Friedberg. (1977). L acteur et le système : les contraintes de l action collective. Dahl, R. A. (2000). On democracy. New Haven and London : Yale University Press. Dahl, R. A. (1961). Who Govern? Democracy and Power in an American City. New Haven and London : Yale University Press. Easton, D. (1974). Analyse du système politique, Paris: Armand Colin. Gaudin, J-P. (2002). Pourquoi la gouvernance?paris: Presses de Science Po. Gagnon, M. et M. Gauthier. (2013). La participation du public dans les démarches municipales de développement durable: principes, conditions de réussite, enjeux et dispositifs. In Université du Québec en Outaouais. Centre de recherche sur la gouvernance des ressources naturelles et du territoire. http://w4.uqo.ca/crgrnt/00_fichiers/cahiers/1307%20- %20Fiches%20participation.pdf (Page consultée le 18 juin 2013). 18

Giroux, D. (2006). Observer l observateur Perspective épistémologique sur l usage de la notion de gouvernance pour l étude politique de l administration publique. In Revue gouvernance. 3 (1), page 39. Godbout, J.T. (2001). Gouvernance, participation et métarègle. In La démocratie à l épreuve de la gouvernance. Ottawa : Les Presses de l Université d Ottawa, p. 129-144. Grégoire, P. «La gouvernance municipale : le cas Estrie Enviropôle». Mémoire dirigé par Isabelle Lacroix, École de politique appliquée, Université de Sherbrooke, (18 mars 2015). Grondin-Landry, A. «La participation citoyenne dans l élaboration du schéma d aménagement et de développement de la ville de Sherbrooke». Mémoire dirigé par Isabelle Lacroix, École de politique appliquée, Université de Sherbrooke, (24 mai 2016). Lacroix, I. (2012). Les pratiques de gouvernance de commissions scolaires du Québec. Thèse de doctorat, Université de Sherbrooke. Lacroix, I. et E. Dostie-Goulet. (2015) «Decision-making process in multiplayer governance organisations». Communication livrée dans le cadre du Congrès annuel de la CPSA/ACSP, Ottawa, 2-4 juin. Lacroix, I. et P.-O. St-Arnaud. (2012). «La gouvernance : tenter une définition», Cahiers de recherche en politique appliquée, vol. 4, no 3, 19-37. Lindbloom, C. E. (1965). Intelligence of Democracy. New York: Free Press. Moreau Defarges, P. (2003). La gouvernance. Paris : Presses universitaires de France, collection «Que sais-je?». Paquet, G. (2001). «La gouvernance en tant que précautions auxiliaires». In Cardinal, L. et C. Andrew. La démocratie à l épreuve de la gouvernance. Ottawa : Les Presses de l Université d Ottawa Royo, S. Yenato, A. et B. Acerté. (2011). «Citizen Participation in German and Spanish Local Governments : A Comparative Study». In International Journal of Public Administration, vol.34, n.3, p.139-150. Smouts, M., Battistella, D., & Vennesson, P. (2003). Dictionnaire des relations internationales : Approches concepts doctrines. Paris: Dalloz. Thibault, A. Lequin, M. et M. Tremblay. (2000). Cadre de référence de la participation publique (Démocratique, utile et crédible). Québec : Conseil de la santé et du bienêtre/gouvernement du Québec. 19