A Study in the Dynamics of Emergent Change - A Structuration Perspective

A Study in the Dynamics of Emergent Change - A Structuration Perspective

Extreme-case analysis on role development and institutional change

Master thesis, MSc BA Change Management University of Groningen, Faculty of Economics and Business

May 2016

Robbin Jan Elbers

s1865676 Aquamarijnstraat 419 9743PK Groningen (+31) 628529961 r.j.elbers@student.rug.nl __ Ad van Dorpen s2597144 Liza Jongbloed s2016362 Reinder van Sluis s2590441

First Supervisior: Dr. C. Reezigt/University of Groningen Co-assessor: Dr. H.C. Bruns/University of Groningen

Abstract

This study, comprising four theses, investigates the processes through which emergent change comes about. Therefore, it contributes to our understanding of managing chan ge in complex and uncertain environments, providing more explanatory research within an immature theoretical field. This study draws on a qualitative, extreme-case analysis and includes findings from interviews and observations conducted at the municipalit y in Groningen, the Netherlands. Emergent change processes in newly developed social systems were studied using Gidden‟s Structuration Theory. Examples show how actors from different contexts are constituted by (overlapping) social structures while they simultaneously replicate or revise them in order to enable new activity. Furthermore, by looking at how actors develop roles within an emergent social system, and the dynamics through which they realize institutional change, a more comprehensive understandin g is given of how actors create activity within an emergent social system. Results imply the importance of sensemaking and social interaction, as actors collectively have to deal with ambiguity in order to take roles, construct meaning and exercise agency. By means of conscious

experimentation, actors were able to improvise, make sense of the world around them and create structural change. Communicating role expectations was found to be crucial in order for different actors to collectively construct meaning and take an active role. The analysis and discussion resulted in a set of nine propositions that can stimulate future research in scientific fields such as emergent change, institutional change and role theory.

Preface

This thesis is part of a larger project which comprises four separate theses. Consequently, each thesis has a certain overlap with the others and sections such as the introduction , theory or methodology are present in all theses. Besides being a group project, this thesis was also an individual graduation project in which I focused on a separate (yet related) topic in order to study the dynamics of emergent change. In light of this unique nature, we strove to highlight both parts equally well and thus provide a rich context, both in theory and results, in order to help increase the understanding of the reader. We think that the end product, although comprehensive of nature, accomplished this goal.

Acknowledgement

I would like to take this opportunity to thank my fellow researchers as we extensively co -operated the past months and constantly discussed and compared our interpretations. Furthermore, I would like to thank all the involved respondents for their time and effort. Studying the Changelabs of the Municipality was an insightful and challenging experience and allowed us to acquire knowledge on topics such as bottom-up initiatives, the municipal system and the regulatory body. More importantly, it allowed us to learn about doing scientific research. Lastly, I would like to thank my supervisor for his meetings in which he inspired and guided us through a n extensive literature field and his personal feedback.

Table of Contents

Abstract ... 1

Preface ... 2

Note to the reader ... 2

Acknowledgement ... 2

Table of Contents... 3

1. Introduction ... 5

2. Literature Review ... 7

2.1 The Emergent Change Process... 7

2.2 Structuration Theory... 8

2.2.1 Structure of signification...9

2.2.2 Structure of domination. ... 10

2.2.3 Structure of legitimation. ... 11

3. Methodology ... 13

3.1 Research Design ... 13

3.2 Case Description... 13

3.3 Data Collection ... 14

3.4 Data Analysis ... 15

3.5 Coding and Categorising ... 16

4. Initial results... 18

4.1 Initial Analysis... 18

5. Role development within an emergent system ... 20

6. Within-case analysis... 24

6.1 Case A: Innovation and Social Security ... 24

6.2 Case B: Sports in Hoogkerk ... 26

6.3 Case C: Happy Neighbourhood ... 28

6.4 Case D: Safety ... 30

7. Extreme-case analysis... 33

7.2 Sensemaking... 33

7.3 Communication standards ... 35

7.4 Resources in order to experiment, take roles and create activity... 35

8. Discussion... 37

8.1 Dealing with ambiguity ... 37

8.2 Sensemaking... 37

8.3 The dynamics through which actors realize institutional change... 38

8.4 Theoretical implications ... 40

8.5 Managerial implications... 41

9. Conclusion ... 42

9.1 Limitations and further research... 42

10. References... 44

1. Introduction

“Change is not only the product of engineered effort, nor solely the result of completely free improvisation of organizational players, but a complex process that happens somewhere on the edge between order and chaos”. -

(Maimone & Sinclair, 2014)

This quote displays the complex and ambiguous nature of organizational change. While the planned approach to change depicts organizations as stable entities that move from a current unsatisfactory state to a desired future state (By, 2005), the emergent approach, on the other hand, sees organizations as entities that are continuously adapting to their ever-changing environment (Burnes, 1997). While the importance of emergent change processes intensified over the years, few empirical studies regarding this change have been conducted (Burnes, 2014; Higgs & Rowland, 2011). To understand why emergent change works, it is important to identify factors contributory to this emergent process (Ford & Ford, 1995; Pettigrew, Woodman, & Cameron, 2001; Porras & Silvers, 1991). According to By (2005), it is too difficult for senior managers to effectively identify, plan and implement the necessary organizational responses from the emergent approach perspective. Therefore, the responsibility for organizational change has to become increasingly devolved (Burnes, 2014; By, 2005). This claim is supported in literature; research has demonstrated that employees play a major role in the success or failure of change initiatives (Shin, Taylor, & Seo, 2012; Van Knippenberg, Martin, & Tyler, 2006). In the case of emergent change, it happens when actors continually improvise by making sense of and acting coherently in the world around them (Orlikowski, 1996). Consequently, emergent change should be considered as a social process. Besides being a social process, emergent change is also a political process, since different actors in a social system struggle to protect or enhance their interests and must work together and make sense of the ambiguity inherent in the process due to its complex nature (Burnes & By, 2012). A social system is defined as the patterned series of interrelationships existing between individuals, groups and institutions and forming a whole (Luhmann, 1995). It is through interaction that actors adjust their web of beliefs and habits of actions to obtain new experiences (Tsoukas & Chia, 2002). This implies that to enhance this

sensemaking process to bring about emergent change, actors should regularly interact within the context of change. A high level of activity could, therefore, be an indication of a high level of interaction. Activity within this research was defined as “doing in order to transform something” (Kuuti, 1995, p.22). Activities are generated by actors in a social system, they are aimed at achieving change and they stimulate an interaction process.

under study. More specifically, how these social practices are related to an increased level of activity is important within this study. Therefore, the question this research aims to answer is:

What is the relationship between social practices which actors create, share, negotiate and inst itutionalize through interaction, and the level of activity within emergent change initiatives?

The results of this study contribute to a better understanding of the processes through which emergent change comes about. Since it is important to enhance our understanding of managing change in complex and uncertain

environments, this study might be of managerial interest. Specifically, it may provide practitioners some guidance on how to initiate and facilitate emergent change, by proposing guidelines on which factors to stimulate and eliminate during the process to create activity. Furthermore, this study has theoretical significance by providing more explanatory research within an immature theoretical field.

2. Literature Review

2.1 The Emergent Change Process

Advocates of the emergent approach to change argue that due to more turbulent and continually changing environments in which firms operate nowadays, the emergent approach may be more suitable for modern organizations (Bamford & Forrester, 2003). They argue that internal practices and behavio ur must be continually and synergistically adapted, in real time, to changing external conditions (Beer & Nohria, 2000). Within emergent change literature, scholars take on different perspectives regarding change initiatives. Analytical-Processualists such as Dawson (2003) and Pettigrew and Whipp (1993) reject prescriptive approaches to change and instead focus on the interrelatedness of individuals, groups, organizations and society. They share the opinion that emergent change needs to be viewed holistically and contextually (Mintzberg & Westley, 1992). Kanter, Stein, Moss, and Jick (1992) and Kotter (1996) take on a more consultant-oriented view emphasizing the role of managers using political skills to overcome resistance and prescribing recipes and checklists for successful change. However, according to Burnes (2014), there is a general agreement among scholars about the main t enets of emergent change. These tenets are that change (1) is continuous, (2) is best achieved by small-to medium-scale changes, leading to reconfiguration and transformation over time, (3) is a multi-level, cross-organization process, (4) is not an analytical-rational process, (5) involves an important role for the manager in shaping the long -term process of change, and (6) requires three organizational activities to operate successfully, which are information -gathering, communication, and learning. Orlikowski and Yates (2006) recognize change as a political process with different groups struggling to protect or enhance their own interest, underlining the importance of social processes within emergent change. Emergent change is an open-ended, cumulative and unpredictable process, which consists of a continuous sequence of autonomous, local initiatives that pop up within the organization (Kickert, 2010). This view is in line with Weick (2000), who sees these local autonomous initiatives as occurring in an unplanned, unforeseen and unexpected fashion. These on-going accommodations, adaptations, and alterations produce fundamental change without a priori intentions to do so. From an emergent perspective organizational transformation is an on -going improvisation enacted by organizational actors who continually try to make sense of and act coherently in the world (Orlikowski, 1996). Recurrent and reciprocal variations in practices over time create the conditions for further breakdowns, unanticipated outcomes , and innovations that subsequently lead to more variations. Variation or enactment as Weick (1969) mentions, is a starting point of the sensemaking process. Gioia (2006) sees variation as the human action that creates the ambiguous contexts and situations that people need to make sense of or act on. Variation is a necessary condition to create a change in the frame of reference and constitutes the information input for the processes of selection and retention (Weick, 1969). These variations are also on-going with no beginning or end point during the change process (Orlikowski, 1996).

system but as the result of on-going improvisation by multiple actors within interconnected social systems. To study how unanticipated variations in practices occur, by looking at activities within an emergent change process, the change process itself cannot be analysed exclusively on an individual, group or organizational level. Structuration Theory offers a solution to this problem; it provides concepts that effectively bridge these lev els of analysis, thus building a more complete social theory (Hartman, 1988; Robey & Markus, 1998). It is a theory to investigate how and why activity within an emergent change system is constituted, interpreted, shaped and institutionalized.

2.2 Structuration Theory

The structuration theory, created by Giddens (1984), aims to explain how humans interact in society. Within this theory, all dualisms of social theory are rejected (MacKay & Tambeau, 2013). The most important notion of the theory is the „Duality of structure‟, which sees “structure as the medium and outcome of the conduct it recursively organizes; the structural properties of social systems do not exist outside of action but are chronically implicated in its production and reproduction” (Giddens, 1984, p.374). Social structures provid e the contextual rules and resources that allow human actors to make sense of their own acts and those of other people (Orlikowski & Robey, 1991) and condition social practices. Social practices refer to the activities within a so cial system that produce, reproduce or change this system and involve interaction between humans, which is situated temporally and contextually (Orlikowski & Robey, 1991). Furthermore, social structures established by prior human action come to define and shape individuals‟ interactions, which subsequently recreate this social structure anew (Orlikowski and Robey, 1991). Enacted social structures however, cannot be viewed in a vacuum, because interaction within a social system can be enabled and shaped by social structures that might be the product of prior human interaction within another social system. Therefore, the process of structuration operates at multiple levels of analysis: individual, group and social system (organization and society) (Orlikowski and Robey, 1991).

Figure 1. The three pillars of structuration

2.2.1 Structure of signification

From an emergent perspective, organizational transformation is an on -going improvisation enacted by organizational actors who continually try to make sense of and act coherently in the world (Orlikowski, 1996). Within emergent change, important concepts such as sensemaking, interpretive schemes and/or frames of references can be linked to the first pillar within Structuration theory, i.e., the structure of signification. The structure of signification consists o f all the social rules that are formed by interaction and communication and which subsequently constitute and enable further interaction. These social rules are also mediated by the individual‟s and group‟s frames of reference (interpretive schemes). These frames are viewed as templates that help organizational members understand and interpret events (DiSanza, 1993; Weick, 1995). However, frames, i.e. ways of looking at the world, can differ between actors. As a result, frames can be (partially) appropriated, edited, but also resisted by different employees (Chreim, 2006). Furthermore, when these frames emerge in organizations, whether by managers or employees, they are seldom self-contained. They resonate with, or are derived from the wider institutional environment and

potentially shape discourses within this environment. Therefore, individual and organizational frames, or interpretive schemes, are intertwined with wider discourses that go beyond the individual and the organizational meanings (Chreim, 2006).

and the reflexive monitoring of action. Giddens (1984) defines reflexive monitoring of action as: “the purpo sive, or intentional character of human behaviour, considered within the flow of activity of the agent” (p. 376). Through this reflection and creation of meaning, all humans are able to consciously shape and create social systems and structures (Fuchs, 2002)

The understanding and meaning creation of organizational change is mediated by the individual‟s context, way of thinking and interaction with colleagues (Balogun & Johnson, 2005). Cornelissen (2012) mentions the importance of language in this process, as it is a resource for individuals to understand novel situations, to articulate a role for themselves, and to cope with the voiced or perceived expectations of other actors in the organization. Through sharing of accounts, which are “descriptive constructions of reality embodying possible interpretations of events”, collective construction of meanings is created (Maitlis, 2005; Stigliani & Ravasi, 2012). Therefore, redirecting continuous change is to be sensitive to discou rse, and to understand that people‟s interpretations differ and so their attitudes towards a change initiative can differ as well. This on -going cyclical sensemaking process can lead to the revision of old frames or development of new frames, leading to po ssible institutionalisation of new ways of working within a social system (Weick, 1995).

2.2.2 Structure of domination.

An emergent change process is a political process with different groups struggling to protect or enhance their own interests (Orlikowski & Yates, 2006). Through the mobilization of allocative and authoritative resources, different groups can exercise power to transform social structures, while at the same time these resources are structural elements that influence the allocation of power (or structure of domination). These resources are almost always asymmetric, and only when this distribution is challenged, can the existing structure of domination be modified (Orlikowski & Robey, 1991).

process. The role of managers and their power positions can have an influence on the change process since they often try to control or shape sensemaking processes (Filstad, Geppert, & Visser, 2011). Managing change should, therefore, be seen as a political process, invested in power structures and interrelated with knowledge (Filstad et al., 2011).

2.2.3 Structure of legitimation.

Actors derive the legitimacy of their actions from the norms and rules of the social system they associate with. To facilitate emergent change these norms and rules need to encourage experimentation, divergent thinking, rule -breaking and foster a climate in which people are free to own their own power, think innovatively, and operate in new patterns (Bechtold, 1997). Strong norms and beliefs will become shared throughout the group due to

conformity; individuals change their own norms and beliefs to be able to continue identifying themselves with the group (Vonk, 2007). The structure of legitimation entails the higher-level morals that determine how people act or interact, through the use of norms. These norms constitute legitimate or appropriate behaviour, on which interaction is sanctioned. Moral sanctions are used among people to condone inappropriat e behaviour. At the same time, human interaction shapes the norms and rules that determine how people act or interact. Therefore, norms are both

produced and/or reproduced rules that entail legitimate or appropriate behaviour (Giddens, 1984; Orlikowski & Robey, 1991).

The three pillars of structuration are all interrelated and reciprocal within an emergent change process, and should therefore not be considered in isolation. For example, power exercise becomes legitimate and more

successful when it is in line with organizational norms and values. Additionally, within an interconnected social system, actors might derive the legitimacy of their actions from another social system. Therefore, overlapping social structures have to be taken into account. Figure X shows a visual representation of overlapping social structures. An actor (A) interacts within the social system of the organization or department (X) he works in, where context -specific structures influence that actor‟s actions. At the same time the actor (A) collaborates with another actor (B). This collaboration (Z), is new and ambiguous, and so actor A and B makes sense of their new social context (Z) by drawing upon what they know from their own social system (X and Y).

Figure 2: Overlapping social structures and sensemak ing processes

3. Methodology

To research the process of an emergent change, a theory development approach was used. The first two steps of the empirical cycle: observation and induction help to develop new and expanding theoretical insights in the academic field of emergent change. This approach results in a set of propositio ns that can be tested in follow-up research (Aken, Berends, & van der Bij, 2012, ch. 2). A case study method is used, because the intimate connection with empirical reality in individual cases increases the likelihood of generating novel theory (Eisenhardt, 1989). The research design of the study consists of the following steps (taken from Eisenhardt, 1989):

First, the research focus was determined, to prevent becoming overwhelmed by the data. The research

focus based on Giddens‟ Theory of Structuration (1984) resulted in the following research question: „What is the

relationship between social practices which actors create, share, negotiate and institutionalize through interaction, and the level of activity within emergent change initiatives?‟.

Second, multiple cases were selected to enhance the reliability of the study. An extreme case approach was

chosen to show extremes in the level of activity present in the emergent change context. This approach to case analysis illustrates the contrasting characteristics that highlig ht the differences under study, and enables the recognition of patterns (Voss, Tsikriktsis, & Frohlich, 2002). The cases were selectively chosen based on the following criteria: 1) the case is considered as an emergent change process, 2) the level of activity within the case is either high or low and 3) the time interval between activities is short or long.

Third, multiple data collection methods, including interviews, observations, and secondary data sources

were used to remedy the shortcomings and biases of each instrument (Aken, Berends, & van der Bij, 2012, ch. 13). Four researchers were involved in the data collection and analysis, serving researcher bias. The researchers‟ multiple insights and perspectives contributed to the „creative potential of the study‟ (Eisenhardt, 1989, p.538), and enhanced the quality of observations, resulting in more confidence in the findings.

Fourth, the researchers entered the field. In doing so, field notes and observations were documented to

create consciousness of what is happening in the research. To serve researcher bias, the interviews and observations were conducted by at least two researchers. Interviewees with different backgrounds and functions were selected (e.g. employees, consultants or citizens) with approximately the same number of respondents per case, to account for subjectivity and bias.

Fifth, the researchers coded and categorized all documentation for the within - and cross-case data analysis

in search for patterns. The sixth step involved shaping the propositions from the discovered themes, concepts, and relationships in the case analysis. In the final step, propositions were linked to existing literature. Conflicting and similar literature was discussed to enhance the theoretical level of our research and to strive for theoretical saturation.

3.2 Case Description

introduced a new initiative called “Changelabs”, in line with the coalition agreement „For Change‟, presented by the City Council for the upcoming years. The Changelabs initiative targeted the goal of the coalition agreement: to bring the municipality and its citizens closer together. For a period of nine months, eight Changelabs, with each a different topic chosen by the City Councillors of a specific domain, were set up to experiment with new ways of working together.

Public organizations like the municipality have been said to have bureauc ratic structures and tight control, to ensure values such as accountability, legality and reliability. Unlike this rigid structure, Changelabs were

encouraged to ignore all bureaucratic principles and experiment freely, and not to worry about deadlines and results. Each Changelab was supposed to have two people facilitating the entire process: the external and internal driver. The external driver, from this point on considered as the „External Change Agent‟, is an external consultant hired to apply his expertise in managing change to the project. The internal driver, henceforth described as the „Internal Change Agent‟, is a civil servant employed by the municipality, who is appointed to ensure an organizational learning effect and to bring the citizens and municipality closer together. From these criteria, it can be derived that the change process is emergent in nature, as there are no predetermined steps and decisions made are unpredictable and evolve over time. Another actor within the Changelabs is the Cen tral Coordination Team (CCT), which is a group of civil servants, some of whom have introduced the Changelab initiative, who were appointed to make sure that everything runs smoothly within the Changelabs. The CCT see themselves as facilitators within the Changelabs, in that they know what is going on and can help when needed, but they are not active participants or controllers.

Because the Changelabs operate independently from each other and are guided by change agents with different personalities, they provide a diverse population to select cases from. The cases were selected following the aforementioned criteria. Four cases were chosen as more would be too time -consuming within the specific time frame, and less would not allow for valid and reliable research. The four case descriptions from Changelab A, B, C and D, will be further explained in appendix I.

3.3 Data Collection

The data were obtained with multiple research instruments; interviews, observations and secondary resources. Interviews were conducted in three rounds, to get insights on the research topic, process and variables influencing emergent change. Interviews can help researchers by contextualizing qualitative findings (Frels & Onwuegbuzie, 2013). Within all the interview rounds, contact with the interviewees was made through various ways (e.g.

telephone, e-mail, Facebook). Confidentiality and anonymity were guaranteed to the interviewees at the start of the interviews. The protocols of the two interview rounds and the list of interviewees can be found in appendix II, III & IV. Also, a list of additional relevant actors is provided in appendix V.

Agents). The goal of these interviews was to develop an understanding of the process in order to establish the research topic and determine the level of activity to select the cases.

After selecting the research topic and the four cases, another interview round was held. In this second interview round, consisting of twelve semi-structured interviews each lasting between 60 and 90 minutes, the interviewees were chosen regarding their level of involvement within the four chosen cases. Next to different functions like the Internal- and External Change Agents and the Central Coordination Team; involved citizens were interviewed. The focus of this interview round was finding concepts that are relevant within the emergent process of these Changelabs.

The third interview round consisted of ten interviews, which lasted between 30 and 70 minutes. Semi-structured interviews were used to gain more insights in the specific chosen topics and to en able deeper probing into the responses of the interviewee (Cooper & Schindler, 2014). The interviewees were selected because of their involvement within the cases, and most of them were also interviewed in the previous round(s). The third round of interviews focused on the four research topics of the researchers in order to find patterns and mechanisms within these topics. The topics were established from the data from the previous two interview rounds.

Next to these three interview rounds, observations were held to create insights in the process of the Changelabs and to enrich the data of the conducted interviews. This resulted in observa tions of different types of meetings (e.g. gatherings about the Changelabs, planning workshops and reflection sessions). Observations were done by making field-notes during the meetings. The focus was mainly based on behaviour, roles of the participants and the content of the meetings.

The third data-gathering instrument is analysing secondary sources (e.g. written documents). These resources were used during the whole research to get a better understanding of the process within the Changelab, to clarify the start of the Changelabs. Documents were provided to us by the municipality directly, obtained through Facebook groups, the Internet or provided by some of the participants.

3.4 Data Analysis

coding scheme, as displayed in appendix VI. The information derived from the second round led to four concepts of interest that were further researched within the third round of interviews.

Immediately following the interviews, the researchers used self-recordings to capture the most important moments during the interview and their general impression. The observations of the meetings, like the field notes, were also coded inductively and deductively. In a similar way , documents were analysed to find relevant aspects and patterns regarding the formulation and the process of the Changelabs. Furthermore, by documenting all activities within a logbook, a detailed overview is given on how this research is structured and carried out, therefore ensuring controllability.

3.5 Coding and Categorising

Below, an overview is provided of the steps taken to develop the final coding scheme for the analysis of the interview transcripts and field notes. Step 1 and 2 took place after the preliminary data collection, which were explorative interviews that were not transcribed. Step 3 to 11 regard the coding, categorization, and initial analysis of the second interview round, resulting in emergent structures for further investigation. Steps 12 through 16 involve the coding, categorization and analysis of the elaborated t hemes emerging from the second round of interviews. After these 16 steps, the within case and a cross case analysis were conducted. In the following chapter the initial results will be addressed, where step 9, the frequency analysis, will be presented.

First phase

Step 1. Early pattern recognition from first interview round Step 2. Development of deductive coding scheme

Second phase

Step 3. Data collection: second round of interviews

Step 4. Coding of 12 interviews from second interview round [and field notes]

Step 5. Adding codes to inductive coding scheme and modifying deductive coding scheme Step 6. Critically compare coded interviews and discuss differences with fellow researchers Step 7. Modify and finalize coding scheme

Step 8. Coding of 12 interviews from the second interview round by a different researcher Step 9. Frequency analysis of codes from second interview round

Step 10. Elaborated themes and questions developed Step 11. Development of deductive coding scheme

Third phase

Step 12. Further data collection: third round of interviews

Step 13. Coding of 10 interviews from the third interview round [and field notes]

Step 15. Critically compare coded interviews and discuss overlapping themes with fellow researchers Step 16. Meetings in which coding is discussed with the fellow researcher to reduce biased coding

Fourth phase

4. Initial results

This section covers the important findings from the analysis of the data acquired in the second phase of this study. Below, an overview is given of the initial data analysis after the second round of interviews, using frequency analysis. From this frequency analysis, individual topics per researcher were selected for further investigation.

4.1 Initial Analysis

Like explained in the methodology section, a finalized coding scheme was developed, which consisted of 73 codes, falling under 15 categories. In appendix VI an overview of these deductive and inductive coding categories is presented. The goal of this analysis was to select four individual topics for further research. When codes in the data were more frequently mentioned than others, this indicated a stronger presence of accompanying topics over others in the data, and thus interesting for further investigation. A ll codes were counted in order to do the frequency analysis. The number of times each code was mentioned per Changelab, as well as the number of interviews in which that specific code was mentioned (depth), were counted. A minimum threshold of 50 was used for the frequency of mentioning the code and 10 for the number of interviews in which that specific code is mentioned. Only codes equal to or above the threshold are taken into account for the selection of the individual research topics because they indicate a high frequency and thus a strong presence of certain topics. The top 16 codes satisfying these criteria, including the results of counts can be found in appendix VII.

The most frequently mentioned codes indicated four interesting phenomena for further investigation. These are Roles, Power, Sensemaking and Ownership. These phenomena will be separately investigated by each researcher in the final round of the interviews. The influence of change agents and politics can be related to the role of power within the emergent change system. Often, interviewees described situations indicating that there was something that restricted or inhibited the change agent from doing everything and involving everyone. And in other cases, interviewees described situations where they did not regard these restrictions as restraining to them. W hat this is, and how it works, was expected to be related to the structure of domination present, and should be researched more in-depth in order to understand its relationship with the level of activity and social practices in emergent change initiatives.

The roles of change agents and employees of the municipality can be rela ted to the field of sensemaking. Interviewees shed light on the difficulty of discovering what their role was in the Changelab. How this role

5. Role development within an emergent system

Change is initiated and carried out by individuals in an organisation (Bartunek, 1984; Porras and Robertson, 1992) and is increasingly seen as part of every manager‟s role (Clarke, 1994). According to Buchanan et al. (1999), this change management role has become more widely dispersed in both public and private sector organisations, to involve staff from all organisational levels in substantive change design and implementation roles. This development is welcomed by the emergent school, who claim that this role should be the responsibility of everyone in the

organisation (Burnes, 2009). In order to generate some form of success within an emergent change system, there needs to be a certain amount of activity. Actors need to be proactive, which means they “identify opportunities and act on them, show initiative, take action, and persevere until meaningful change occurs” (Crant, 2000, p. 439). In other words, they need to take an active role.

First, „role‟ has to be defined more precise. The set of behaviors that others expect of individuals in a certain context is called a role (Friedman and Podolny, 1992; Katz and Kahn, 1978; Nandram and Klandermans, 1993). These expectations of a given role do not arise in a vacuum, but instead develop in the context of other interdependent behaviors and expectations that make up the social system (Floy d and Lane, 2000). In general, the more clearly expectations are articulated, the more likely that the individuals will perform a role conforming to those expectations (Fishbein and Ajzen, 1975; Stein, 1982; van de Vliert, 1981). Interactions between individuals become more predictable when these individuals interact within well-developed roles. This increases the level of trust within the organisation (Mayer, Davis and Schoorman, 1995). A role is always a position occupied by an individual in the context of a social relationship, making role definition a social process (Banton, 1965) based on interactions and expectations of a role holder and their role set. Role sets refer to those actors who, in the context of their work, interact with the role holder (Fitzgerald et al., 2006). However, agreement among actors regarding the appropriate roles to be played might differ, for instance because there is a lack of role clarity. This lack of clarity regarding a particular role is defined as role ambiguity (Fitzgerald et al., 2006). Ambiguous role expectations might decrease productivity and lead to reduced innovativeness, as actors are unsure of what is expected of them and others (see also Kahn et al., 1964). However, some amount of ambiguity can act as a challenging and empowering experience for some (Willcocks, 1994). Clearly, it becomes essential for actors to collectively deal with this ambiguity, which is inherent in a newly formed social system, in order for them to take an active role. This implies that group

effectiveness depends upon each member understanding the role expectations of other members in the social system, so that each is clear about his or her own role expectations (Solomon, Surprenant, Czepiel and Gutman, 1985). Communicating role expectations is thus crucial in providing role clarity. Whether or not a particular part (of expected behaviors) is acted appropriately, also called role enactment, is determined by reactions of fellow actors and observers. Confidence in the appropriateness of one‟s own role, leads to satisfaction with a particular

that must be faced by individuals who participate in a certain social system and less on (social) change (Biddle, 1979). As previously stated, key (emergent) change decisions within a social system evolve over time and are the outcome of political and cultural processes (Burnes, 2009). Change agents, as an active managers of the change process, maintain their own agenda which they seek to promote or impose by managing and shaping the perceptions of those concerned (Pettigrew, 2000; Weick, 2000). Developing one‟s role within an emergent change system is therefore a political and cultural process which cannot be viewed in is olation from the wider context and interconnected systems. Multiple actors with different interests, goals and positions simultaneously exercise a directing influence on the change process (Kickert, 2010). For this reason, especially in a newly formed emergent social system, actors face a serious amount of ambiguity surrounding the appropriate roles to enact. With high levels of uncertainty, actors seek to clarify their situation so that they can return to non -conscious responding (Weick, 1995). However, this sensemaking process is likely to differ between actors.

Weick (1995) notes, that due to the variety of contexts faced, for instance by managers, it is likely to cause them to interpret the same cues differently or to note different cues. The way in wh ich actors look at the world and interpret information, i.e. their frame of reference, is therefore influenced by the social system or context in which they operate. For instance, Leonard-Barton (1992) and Starbuck and Milliken (1988) note that, a manager‟s interpretation of events and their primary organisational roles are influenced by his or her functional expertise. When a new emergent social system is formed, in which all kinds of actors with different roles in different contexts come together, they bring along their social logic, traditions or knowledge bases from another social system. This is defined here as overlapping social structures. As a result, different actors are used to different social structures, which enable and determine their behaviors in a different way. Various environmental cues might therefore be interpreted differently, prompting managers to reach different conclusions as to what roles and actions are appropriate (Floyd and Lane, 2000). When an actor is communicated different role expectations, he or she might experience role conflict, which is defined as the tension experienced by an actor over which role to enact (Floyd and Lane, 2000). Furthermore, due to the inherent structural ambiguity within a newly formed social system, actors can interpret social structures differently and therefore come to different conclusions on the actions to be taken. These issues stress the importance of taking into account the different social contexts in which actors operate.

Studies of institutional change focus on changing social structures . However, the role of individual actors in realizing institutional change has received little attention so far (Zilber, 2002). Therefore, the roles of actors, and their respective development within these systems, must be highlighted in order to see how they both constitute (change) and are constituted by (constrained) certain institutionalized structures and ultimately impact the amount of activity or agency within an emergent change system. Institutions can be defined as “shared rules and

particular setting”. These scripts encode a certain social logic, which subsequently enables and guides future human interaction or agency. In this research, actors interact with each other and enact and validate their roles through the use of these scripts. Institutionalized aspects of an actors‟ role, which are shaped through prior human interaction, are the shared rules and typifications that identify the appropriate activities or relationships of the social actor. These aspects are the taken-for-granted scripts which define the constitutive expectations of actors (Beckart, 1999), e.g. role expectations.

Figure 3, from Barley and Tolbert (1997)

Research on institutional entrepreneurship tries to investigate how actors can co nsciously pursue projects of institutional change (Fligstein, 1997; Colomy, 1998; Maguire et al., 2004). Koene (2006) addresses the fact that it is important to take into account the interaction between the context, in which institutional change takes plac e, and the behavior of the actors effecting the change, when trying to understand the nature and outcomes of institutional change. As Barley and Tolbert (1997) noted, the modification of an institution is more likely to require conscious choice, than does its maintenance or reproduction. This means that in order to revise a current script, the actor must first be triggered to engage in a sensemaking process. However, many authors still avoid the essential paradox of institutional entrepreneurship. This is the fact that institutional entrepreneurs operate within an institutional context that already defines meaning and individual action, thus also affecting the understanding and behavior of the entrepreneur. Topics such as sensemaking and mental (re-)mapping seem to be neglected in this field (Koene, 2006). Due to the explicit focus on the structural and behavioral, the framework of Barley and Tolbert (1997) is far less sensitive to the role of cognitions and interpretive frames during the institutionalization process, and therefore a more systematic exploration of the relative importance of both behavioral and interpretive phenomena is required (Barley and Tolbert, 1997).

6. Within-case analysis

Since no one knew what innovating within social security meant, the role of the changelab within its wider socio -political context remained uncertain at first. The civil servants faced role ambiguity due to their unfamiliarity with organising such a process. They remained passive and expected of the external change agent, William, to take a leading role as „project owner in order to make sure that something was done (A2) ‟. These expectations soon became grounded within the social logic of the changelab. William‟s knowledge on how to organise such a process was crucial for guiding the civil servants during the first months and for teaching them how to organise meetings themselves.

“At that moment I had the idea that the external change agent was the owner (…) everything that I said there and the questions I asked, they coloured how the changelab started (A1)”.

The role of the changelab within its wider socio-political context finally became clear when Matt, the city counsellor, and William had an informal conversation. This triggered a sensemaking process in which both actors shared their intentions.

“… only after we talked for one and a half hour we finally had enough time to talk about the challenge within the social domain that the changelab should focus on. It became clear that we actually want to build a social security focused on treating humans with more dignity. Consequently, all pieces of the puzzle fell in place (A1)”.

The vision of the changelab reduced role ambiguity fo r actors and enabled new agency. Five core values were formulated on which citizens were invited to share their ideas [pitches] and on which discussions were held. After several meetings, the external change agent noted that the civil servants remained pas sive and kept expecting of him to take the initiative. However, he was not able to live up to these expectations because he „had the least time, did

not work within the municipality and k new nothing about the topic itself (A1) ‟. This lack of resources triggered a

sensemaking process, in which his old frame of reference [I am the owner] was replaced by a new one [someone has to become active who is „working at the municipality and has knowledge about the content (A1) ‟]. William sounded the alarm at the responsible manager, warning that „things would not speed up (A1) ‟. Consequently, Debbie stood up as the internal change agent while William started to play more of a role in the background.

Another group of actors involved in the changelab was „Band‟. This was a group of independent entrepreneurs who helped with brainstorming and it was expected of them to take a leading role by „organising

things (A2)‟. However, communication soon proved difficult and at a certain moment Band quit. Their withdrawal

“I noticed myself approving of what he did. However, that‟s not good because William is not here for that. I am being a real typical civil servant again, and I actually should be able to become active myself (A2)”.

Debbie showed pro-active behaviour after that, which resulted in the civil servants organising new meetings. The general belief amongst both William and Debbie was that providing a stage for pitchers and facilitating them by bringing them together was enough. This frame of reference incorporated clear role expectations towards both parties. When both change agents organised one-on-one meetings with every pitcher, „just as an experiment to see if

that results in more movement (A1)‟, they discovered that some pitchers had not made much progress. Some had felt

discouraged from sharing the problems they encountered during the group meetings. This triggered a sensemaking process in which both parties reciprocally [both ways] addressed their expectations. This role communica tion process was characterised as open, reciprocal and based on equality and resulted in both pitchers and Debbie taking an active role in order to enable activity.

“Those were very open conversations, in which you look on what you need and can bring a s a pitcher, and what we need from William and Debbie and what they can bring. You‟re in there on an equal basis. That‟s nice, they do that extremely well (A3)”.

“The past period I‟ve been more of a connector inside the organisation, for instance towards p oliticians. So that‟s less about offering a stage for pitchers, but more in explicitly helping them realize their ideas (A2).”

Since pitchers expected of Debbie to take this new role, which they communicated as well, she experienced role validation and believed strongly in re-enacting this role. This led to the shared understanding that pitchers „have

much more need of someone that tak es them by the hand for a short time, a k ind of ambassador (A2) ‟.

Debbie could not have enacted this new role without her resources, i.e. her contextual knowledge [about municipality and domain], proccessual knowledge [how to accomplish something] and network. These resources, which she acquired in the social system of the municipality, enabled her to enact this new role and help pitchers in a way William could not.

“If Debbie isn‟t there i won‟t be able to help pitchers, because I don‟t know where to find certain civil servants and because I don‟t know anything about municipal policy (A1)”

In conclusion, both Debbie and citizens reinvented their roles within a new way of working together based on reciprocity and equality through sensemaking. This resulted in a new institutional logic.

“… the changelab is not something the municipality arranges with the help of some citizen s. It‟s something you do together and in which you make conclusions and prepare things together (A2)”

6.2 Case B: Sports in Hoogkerk

The first meeting between the external change agent, Ricardo, and the civil servants was characterised by role ambiguity surrounding each other‟s role. Both expected of the other to take the lead and provide role clarity.

“… they looked at me, and I looked at them. They said „alright tell us‟. And I replied „why don‟t you ask me the question?‟ That was contradictory because while they thought I was going to do something and change something, I had the idea that they had a problem which I should help them with (B1)”.

Most civil servants did not appear keen on working with Ricardo and became „really passive (B1)‟. With no internal change agent to help him yet, Ricardo decided to „get a better understanding of what should be done (B1) ‟ in order to reduce his role ambiguity. When Jantien was assigned as the internal change agent, Ricardo did not involve her due to her lack of contextual knowledge, proccessual knowledge and a network. This also made it difficult for Jantien to become active herself. Consequently, Ricardo had no role expectations towards her while Jantien was waiting for expectations to be communicated in order to reduce her role ambiguity. Therefore, she did not take an active role.

“I won‟t involve her because she can‟t add any value within this content … While other civil servants can show you where the interesting persons are and k now the whole system; she is just a trainee … (B1)”

“… but I still don‟t have my own network . This makes it really difficult for me and that‟s why I take so little action (B3)”

Ricardo could not make any progress due to the fact that both parties had conflicting role expectations of each other. While the sport associations expected of the municipality to solve the problem and „pay for it and realize it (B2)‟, the municipality could not live up to these role expectations and wished for the sport associations to take a new more self-sustaining role. Because the sport associations weren‟t „open minded (B2)‟ and „stuck in their own structures

(B2)‟, this problem could not be solved. Subsequently, Ricardo re-enacted his familiar role as independent

consultant and conducted many one-on-one conversations in order to develop his vision on the problem to be solved and reduce his role ambiguity.

“I like to play a role in which in not dependent on anyone in order to form my vision and take some concrete steps (B1)”

This was a continuous sensemaking process in which he frequently had to disregard old assumptions in order to make new ones and in which he frequently heard „another side of the story (B1)‟. His newly developed vision shaped his way of looking at the world, i.e. his frame of reference, and incorporated new role expectations towards both sport associations and municipality within the wider socio -political context. Sport associations should not just wait for the municipality to come to them, but make a social contribution within the community in order to

enthusiastic about his vision and thus willing to change their roles. This resulted in role validation for Richard and subsequently him wanting to re-enact his new role.

Communication with other actors had decreased because Ricardo was unhappy with how others enacted their roles, e.g. with Nicole who was increasingly „pushing for results (B1)‟. With no (role) communication, Ricardo was not sure what Nicole expected of him and decided on not sharing his intentions.

“… when Nicole told me that the changelab is focused on the region of Hoo gkerk, and does not have to become bigger than that, I decided to look for support elsewhere, while not telling anyone about it … (B1)”

Furthermore, the municipality was independently investigating a solution for the problem [as traditionally

formulated] by looking into new ways of budgeting. However, according to Ricardo this „pragmatic solution‟ was not the future and not in line with how the municipality should act. Both civil servants and Ricardo did not see each other play a crucial role anymore. The legitimacy of the role of Ricardo deteriorated even further when Patrick, the city counsellor, acted towards one of the sport associations without consulting Ricardo.

“… he made some commitments there and triggered some movement, about which I knew nothi ng. And in my experience, this had nothing to do with the changelab anymore (B1)”

Because of the formal communication standards the city counsellor enacted, Ricardo experienced this

communication as unpleasant and therefore did not feel free enough to con tact Patrick himself. Consequently, he expected of Nicole to do this. However, due to the lack of role communication, these expectations were not communicated.

“If Nicole had been more of a fan of me, she would have let me work together with Patrick much more (B1)” “If you want a meeting with the city counsellor you would have to tell me; otherwise I wouldn‟t know (B4)”

Ricardo‟s role was not seen as relevant within the intentions of the other actors, which was reaffirmed when the soccer association send an angry letter directly to the city counsellor without informing Ricardo. He decided to use his vision as a mental resource in order to build a coalition and „create results elsewhere (B1)‟. By doing this, he avoided a confrontation with both Nicole and Patrick which he „could not hope to win (B1)‟.

Since municipality and sport associations were not brought together, „it remained at individual opinions

(B4)‟ and conflicting role expectations. Ricardo did not possess the political skills to let both part ies listen to and

understand each other‟s language.

“… I did sit along the same table with both parties and then I really hear the difference in language they speak, which makes it difficult for me to intervene. I don‟t know if I can accomplish this witho ut offending people. I try to do that in a subtle way, at a different moment… (B1)”

himself and he re-enacted the same role throughout the process . This case was characterised by a low level of activity, no experimentation and high role ambiguity.

6.3 Case C: Happy Neighbourhood

The changelab started with the name „neighbourhood oriented work ing (C2) ‟, which was derived from a new espoused way of working with citizens. However, what this incorporated was unclear for the change agents. This resulted in a low level of activity during the first five months and uncertainty about the role of the changelab within its wider socio-political context.

“… It‟s difficult to find out what this changelab should focus on and how this is different from the neighbourhood oriented way of work ing that is already practiced at the municipality by the neighbourhood teams (C2)”

Jantien, the internal change agent, was unfamiliar with the process and relied on Enno to take a leading role in order to make progress due to his expertise. Consequently, she became passive.

“He‟s good in clarifying, choosing and giving direction. He‟s also got a lot of experience with leading change processes within organisations and therefore he has that influence (C2)”

However, Enno was hesitant to make concrete decisions and prematurely direct the changelab into a certain

direction. His actions reflected his own frame of reference. Enno, as co -founder of the changelabs, hoped to find out what role the municipality could take in this new way of working with citizens. The answer however, was dependent on what citizens expected of the changelab and their own role in it. Therefore, Enno expected of citizens to take the lead in giving direction or purpose.

“… It‟s a challenge to stay out of the concrete. You want things to have a chance to grow or originate. So you have to keep distancing yourself from deciding on the direction to be taken, because before you know it you‟re doing a planned change [going from situation A to B] (C1)”

It was not until the name of the changelab was officially changed into „happy neighbourhood (C1) ‟ that more energy was created, due to the fact that citizens „attached themselves to this new label (C2)‟, and a first group meeting was organised. However, because the intention of the changelab was still unclear, various citizens experienced role ambiguity, i.e. „not knowing what they have to do (C2)‟, and relied on Enno and Roelien to provide answers. On the other hand, Roelien and Enno were hoping that the citizens could resolve their uncertainties by taking the lead and communicating their expectations. Consequently, some participants became inactive and were not seen at

subsequent meetings. Even after a regular group was formed, opinions about what the changelab should entail differed strongly between citizens and „everybody had a bit of their own view (C2) ‟.

Later group meetings with other citizens were more successful since these citizens already had a specific idea that they wanted to realize. The changelab became a context in which citizens could get into contact with each other. Subsequently, the vision of the changelab became clear for Enno, which resulted in more role clarity for him.

“… to connect people and to help them progress (C1)”

Due to positive reactions, such as „you have awakened a latent need inside of me (C1) ‟, Enno experienced role validation and kept enacting the same role. Furthermore, he was convinced that municipality should stop here and not try to monitor every step citizens take with their initiatives.

“When they make this next step with someone else you lose sight of them. And how will you measure the result then? You shouldn‟t want to (C1)”

Since no other activities besides these group meetings were organised, both change agents unconsciously re -enacted their roles. There was no role communication that triggered a revision in the role expectations both had towards each other.

“My role is becoming clearer, for instance its organising these meetings … according to our skills it‟s logical that I do it and that Enno leads the meetings. We don‟t discuss, it just goes that way. Tha t‟s just how it is (C2)”

Although Roelien had relied on Enno and his vision throughout the process, she was now facing more role

ambiguity since she wanted to get some results and needed citizens to communicate their role expectations towards her. However, „there was not much coming from the group (C2)‟ which indicated no role communication during this phase yet.

“But that‟s still an issue; to what degree are you going to pull initiatives towards the municipality or are you receptive for initiatives and just let them happen? The latter is how Enno think s about it and what he radiates … but I feel like we should accomplish something now (C2)”

Furthermore, Roelien desired to take a new role by becoming „a bit more outward [municipality] oriented (C2)‟ and involving sceptical civil servants. She acknowledged that „you have to include people and create support. Otherwise

you can‟t create possibilities (C2)‟. However, she lacked the skills necessary and was „not used to doing that (C2)‟.

Furthermore, Enno did not see it as his responsibility, since he was the external change agent. Consequently, sceptical civil servants were not involved which resulted in lower activity.

Enno emphasized that a change agent needs to be „a certain type of person, who‟s not a typical civil

servants. It has to be a type of person who is enthusiastic, aiding, stimulating and sometimes provocating (C1) ‟. He

enacted his own skills by „creating an (informal) atmosphere (C1)‟ during meetings and by „inspiring, motivating

and letting people participate (C1)‟.

However, periods between meetings were of long duration. This had to do not only with „schedules and

limited time (C1)‟ but also with the limited amount of time Enno received.

“That‟s why there have been few meetings, if you want to have a meeting every week you can‟t manage with the amount of available hours. So actually, you can only organise something once a month (C1)”

experienced role clarity and role validation and was convinced of legitimacy of his own vision later during the process, Roelien was still facing role ambiguity and passively waited for citizens to communicate their role expectations. Furthermore, there was little experimentation and actors re-enacting the same roles and scripts.

6.4 Case D: Safety

Richard, the external change agent, could not start during the first months partly because he had to wait for a „green

light (D1)‟ from the city mayor. It took a month before the first meeting between the two. During this meeting it

became apparent that intentions differed and Richard elided the goal of the city mayor.

“… at the beginning the city mayor had this very determined idea that we should accomplish something in a very top-down driven way [implementing WhatsApp groups], which I did not agree with. Consequently, I took all k inds of measures so we did not have to do that (D1)”

This resulted in the city mayor taking an inactive role in the changelab. Furthermore, civil servants were all „busy

with other stuff (D1)‟ and „nobody got the time for this (D1)‟, which resulted in no internal change agent being found

and no involved civil servants. Since no civil servant was willing to become active, Linda one of the initiators, took up the responsibility for helping Richard primarily by discussing with him. Furthermore, „she never got the official

title as internal change agent, which gave her a lot of freedom to say things of discuss matters without there being consequences (D1)‟. Richard‟s established his intention of the changelab quite fast, which entailed making

Groningen a safer city, together with citizens, and by having citizens take an active role in this. For Richard, the changelab was „a context in which all kinds of persons do something together and momentarily forget that they are

work ing at the municipality or wherever (D1) ‟. He had experience with similar like bottom-up initiatives in

Groningen, and this knowledge strongly determined his way frame of what a changelab should do. Subsequently, he acted on this throughout the process.

“Safety… changelab… they [initiators] asked me to do something similar like Social Labs [work experience]. Well, this is it (D1)”

Furthermore, Richard believed strongly in an extreme decentralization of power. He enacted this resource „whenever

there‟s power influence at play (D1)‟. By trying to „make this power fall apart (D1)‟ he wanted to „guarantee a free atmosphere (D1)‟. Consequently, this gave Richard „a very free position (D1)‟, e.g. by being able to experiment, in

which he could not be influenced by others and in which he could „communicate more effectively (D1)‟. He started by conducting several experiments in order to find out what safety meant and in order to locate energy on which he could focus. This was a continuous sensemaking process in which he gained new insights but also had to try out new things, i.e. revise his script, whenever an experiment „was clearly not working (D1)‟. For instance, when „people

found it vague and experienced it as an anti-climax (D1)‟. During his third experiment, Richard invited citizens to

“Especially bicycle routes from the city centre towards the suburbs are frequently mentioned … That‟s a theme we didn‟t think of ourselves. Civil servants started to show interest again and suddenly there is more money to continue. It‟s alive among politicians as well, so there is movement (D1)”

He could not have created this much activity without the extensive enactment of his network throughout the process.

“That‟s why it [invitation] has been shared for instance by political parties and by at least two city counse llors. It does help that I k now so many people I think . People from the city council, I all k now them personally … (D1)” “Yes, he‟s amazing in the way he approaches it since you can reach different people in an easier way. But I don‟t know how you could approach this as municipality (D2)”

Furthermore, Richard used a „more human (D1)‟ language based on equality in order to avoid creating tension. This resulted in „very informal (D1)‟ and faster communication and more activity. By using his own communication platforms, and „not some vague website of the changelabs (D1)‟, he avoided creating the impression that the project was property of the municipality.

“I try to get everyone on board as citizen or human which results in innovative strength. Otherwise, people will return to their own frame, what you are allowed and aren‟t allowed to do (D1)”

“And when you label something @municipality.groningen.nl, some people will already start saying „that‟s bullshit‟ (D2)”.

After a few more experiments together with citizens, numerous solutions had been generated to improve safety. Furthermore, many could be carried out by participants themselves and with each other. Project groups were formed consisting of groups of citizens, each with their own approach to making tunnels safer. This resulted in a

communication process in which role expectations were reciprocally addressed. Citizens agreed on taking an active role by working on their projects, while the municipality was expected to realize a wish -list with topics citizens could not realize themselves. This enabled Linda to conduct an experiment of her own by spreading the wish -list throughout the organisation in order to see whether and where she would face obstacles. Linda and Richard hoped they could change how things currently go within the municipality „because it‟s ridiculous you can‟t achieve

relatively small things because of how the organisation functions (D1) ‟.

“And I told him: „if you do that to people, they won‟t be inclined to sit with you at your table the next time and have a good conversation. There are more who share my opinion.‟ Consequently, he chang ed his attitude and it turned out a very good conversation. I have the feeling contact is easier since that moment (D1)”

Richard also took a new role when he was triggered into a sensemaking process due to the role enactment of the management team of Groningen not being in line with his expectations. The GMT exercised their authority to influence the process, which Richard did not see as legitimate within this context. Consequently, he took a „much

more pressurizing role (D1)‟ towards the municipality. Both initiators were positive about this new role enactment

and encouraged him to keep on doing it. This gave him confidence in his new role and resulted in him being more inclined to re-enact it.

“I gained confidence from the path I was following. I believe there are people in the municipality that would approve of renewal and that things are not covered up again (D1)”