Constructing democratic participation in welfare transitions: An analysis of narrative interactions

Abstract Objective This article provides insights into the democratic character of local enactments of welfare reforms by analysing narrative interactions about changes in care. We analyse processes of storytelling that are part of the interactions between citizens giving, receiving and organizing care and the policymakers governing welfare reforms. We also study how narrative interactions shape understandings about changing care practices and what types of narrative interactions support democracy in care. Background Stories about recent welfare reforms include messages about citizens’ care, citizen participation, citizens’ powers and revitalization of democracy. However, researchers have cast doubt on their emancipatory and democratic character. Research setting and methodology We conducted research of four initiatives and municipal policy settings in the city of Maastricht that organized social care in lifeworlds connected to arts, crafts, farming and entrepreneurship during welfare reforms. Using narrative ethnography, we analyse narrative interactions between the master narrative of welfare reforms about ‘lifeworld care’ and ‘citizen powers’, and small stories told by participants in new care practices. Results We identified two types of narrative interaction: idealizing and pragmatizing. Idealizing narrative interactions were strategic for care initiatives in finding support and for policymakers in proving that a so‐called ‘participation society’ works. Pragmatizing narrative interactions gave expression to insights into the everyday practices of social care experiments and included a greater variety of stories. Conclusions We conclude that pragmatizing narrative interactions adjust the master narrative about welfare reforms and replace ideals of independency with ideals of active participation in webs of dependency and care.


| INTRODUC TI ON
Political stories about transitions from welfare state to 'participation society' in the Netherlands or a 'big society' in the UK have promoted the reallocation of care work and various vulnerable groups from bureaucratic professional organizations to the everyday lifeworlds and volunteers. These transition narratives include an emancipatory storyline about liberation from bureaucratic dominance that will revitalize democracy through new forms of citizen participation. 1,2 Nonetheless, the democratic legitimacy of these welfare reforms has been questioned. [3][4][5] In the UK, for example, the influential transition narrative of big society claimed to be about empowerment, emancipation of vulnerable groups, and civil society. Bulley and Sokhi-Bulley argue, though, that it was in fact instrumental to management and control. 3 It was about 'the power to cope with neoliberalism rather than challenge it' (p. 465). 3 The decline of civil engagement during the big society programme also suggests that the revitalization of democracy has failed. 4 In the Netherlands, meanwhile, scholars like Horstman and Putters have pointed out that welfare reforms were carried out in a context of declining turnouts during elections, especially by the groups most affected by these reforms. 5,6 They argue that more emphasis is needed on participation in political decision making and less on participation forms that are instrumental to predefined government goals. Bovens On the other side, critics of welfare regimes point to the democratic potentials of reform. 9,10 Even if research about big society shows a decline in democratic participation, reforms may promote participatory democracy in other settings. Jim Diers, for example, promotes 'asset-based community development', an approach based on critiques about 'welfare colonialism', and argues that selforganizing communities in Seattle have revitalized participatory democracy. 11 In the Dutch context, researchers have predicted a strengthening of participatory democracy when welfare institutions relinquish some of their power. 12 These researchers see the old bureaucratic welfare institutions as a hindrance to citizen-initiated change. During Dutch welfare reforms, research institutes and the government discussed citizen engagement in new forms of care in terms of 'do-democracy', which can be seen as a form of participatory democracy. 13 The government saw this as a route for residents to have a say in the shaping of society by engaging with concrete issues in the public domain. 14 To stimulate the do-democracy, municipal and national governments and research institutes issued many publications about successful citizen initiatives with stories demonstrating the power of citizen collectives. 10,15 However, scholars analysing the new forms of public participation or do-democracy have pointed out that this may increase inequalities and thereby undermine democracy. 16,17 Calhoun, for example, points out that the growing participation ideology is a mixed blessing: it has managed to involve citizens in large-scale political processes in new ways, but it also disguises the continued reproduction of inequalities. 18 In the Netherlands, a number of scholars who have analysed do-democracy are not convinced about its democratic character. 2,19,20 According to them, citizen actions are often initiated top-down and do not arise spontaneously in a bottom-up process. Moreover, most influential citizen initiatives consist of high-income citizens and not of vulnerable low-income groups.
Other experiments with democratic renewal during welfare reforms worked with deliberative forms of democracy. In 'citizen summits' or 'mini publics', citizens were selected based on random or purposeful selection and invited to deliberate about public issues. 5,8,21 In well-organized conversational settings supporting respect and listening, citizens deliberate and may form and alter their opinions without the pressures of the media and electorate faced by elected representatives. The outcomes of such deliberations have been embedded in various ways in institutionalized political processes as recommendations or starting points and support for new citizen initiatives. While analyses of such deliberative democracy show that citizens engage in respectful dialog and bridge social and cultural divides, such deliberative approaches also have an exclusionary character. As Polletta points out, 'deliberators' are expected to support their arguments with reasons that are acceptable to others. 22 People in marginalized positions often do not have the discursive resources to convince others, however. 'The danger, then, is that already marginalized speakers and positions are further marginalized' (p. 224). 22 These debates about democracy in the contexts of care and inequality show the gains and limits of citizen participation in dodemocracy and deliberative forms of democracy. In spite of their potential, different forms of democratic renewal in the context of welfare reforms are at risk of reproducing or aggravating social inequalities. Moreover, the debates highlight that social and political inequalities cannot be separated: in stratified societies, there are no ideal communicative settings in which citizens can deliberate as equals regardless of existing social inequalities. 23 In this article, we aim to contribute to this debate by analysing the discourse that emerges within narrative interactions in the new care practices that developed during welfare reforms. This means that we follow the processes of storytelling between citizens giving, receiving and organizing care and the policymakers governing welfare reforms.
When compared to deliberations as a specific type of argumentative discourse, narratives form another type of discourse in which people organize and connect experiences and events. 24 In the context of debates about democracy and welfare reforms, narrative interactions are interesting because they include fragmented, incomplete, personal and small stories as well as more articulate and coherent storylines about the past and future of care practices. Moreover, in narrative interactions, narrators make sense of their own experiences and observations by connecting to other stories. Narrative interactions therefore have the potential to include marginal perspectives and experiences in the formation of public understandings about welfare reforms.
We analyse narrative interactions about new practices of care with a focus on two questions: (a) How do narrative interactions shape understandings about changing care practices? (b) What types of narrativ1e interactions support democracy in care? We first discuss scholarship about narrative interaction in relation to democracy in care. Next, we introduce our case, welfare reforms in a Dutch city.
Subsequently, we present our findings about narrative interactions in new practices of care.

| S TUDYING DEMO CR AC Y IN C ARE THROUG H NARR ATIVE INTER AC TI ON S
We build on the work of Gubrium and Holstein in narrative ethnography to analyse narrative interactions and their meaning for democracy in care practices. 25 Advocates for democracy and emancipation look with suspicion at stories told about vulnerable groups or individuals and prefer to include stories told by people in subordinated, exploited or otherwise disadvantaged positions. 23,26 Nonetheless, it is common practice within care and policy-making to recycle and retell stories about groups such as 'the unemployed', 'the homeless', or 'impaired people'. 27 Moreover, in care practices the telling and retelling of stories about needs, vulnerabilities, recovery and resilience are a vital part of caring interactions. 28 Even if the ideal is that people in care tell their own stories, we also need a better understanding of the democratic meaning of such narrative interactions where people retell stories of and about others in different settings.
We use the concept of master narratives as explained by Lindemann Nelson. According to her, transition narratives about welfare reforms can be seen as 'master narratives', stories 'that serve as summaries of socially shared understandings' (p. 6). 29 Master narratives play a role in shaping identities and relationships, and their plotlines guide images of change. Master narratives are often resisted in counter-stories oriented at changing specific relationships or identities. Lindemann Nelson gives the example of hospital narratives about 'technical' doctors and 'touchy-feely' nurses. 29 This presentation of the division of labour is nourished by master narratives about rational men and emotionally subservient women. From many small stories, anecdotes and histories, nurses constructed a counterstory that moved beyond this rational-emotional dichotomy and showed that nurses were skilled professionals. However, often small stories and adjustments to master narratives are often not explicitly told as counter-stories and remain unheard by broader publics. 16  In narrative research, several authors have argued that narratives do not develop in a vacuum but are part of everyday narrative interactions in diverse settings and relationships. 24,25,30 Gubrium and Holstein call this the 'narrative environment': the intimate and distant relations, organizations and social settings in which stories are told or remain untold and are connected to other stories. 25 Intertextual connections between stories told in different narrative environments can be traced through ethnographic study. While much narrative research has focused on the internal structuring of narratives, recent contributions have paid more attention to what narratives do, to their performance and 'external organization'. 25,31 Narrative interaction thus refers to intertextual connections between small stories, master narratives and the performance of

| Research setting and data collection
We studied a specific Dutch case, namely welfare reforms in the city of Maastricht, a city of about 120 000 inhabitants in the Netherlands. Welfare transitions in Maastricht were particularly challenging as this post-industrial city faces a high burden of disease and unemployment compared to other regions. 32 We analysed tran-  Table 1). 35 The first author, supported by two students, conducted participant observation for 20 months in three categories of 'narrative environments': first, the daily work of participants in workplaces, farming, art workshops or other social meetings; second, in regular meetings of the leadership of each social care initiative; and third, in municipal policy meetings, including meetings between leaders of social care initiatives and municipal policymakers (see Table 2). To collect stories from municipal policy environments, we selected landmark documents dating from the years 2015 to 2017, during which the Maastricht welfare policy was developed and explained. We also attended political meetings and conversations with municipal policymakers that funded the four social initiatives. These three types of narrative environment were systematically included. However, stories told by the professionals of referring institutions were also included occasionally. In each of the four initiatives, we interviewed at least ten and at most 21 people for whom participation in the social care initiative was part of a personal struggle with health, employment or social reintegration. We held interviews with 12 leaders of the social care initiatives, three municipal policymakers who were responsible for supervising the funded initiatives, and four consultants at the social security office who were responsible for enacting the Societal Support Act and the Participation Act and who referred people to the social initiatives.
Respondents were asked to talk in their own words about the meaning of the social care initiative in their lives and in the field of care and support. The interview approach further enabled the study of 'intertextuality' as respondents were asked to comment on stories picked up in other conversational settings.
We chose a participatory research approach to support collective learning processes about the transition of welfare and care arrangements. 36 Design of fieldwork and results was discussed with leaders of the initiatives and the municipality. Participatory ethnographic design was guided by the code of conduct in anthropology. 37 We were clear about our research aim in all interactions in the field, using information letters and oral explanations.
The focus of analysis was on the external organization of stories: on the types of formal or informal encounters in which stories were told and the intertextual connections between small stories and master narratives told by participants and leaders of the social care initiatives. As a demonstration of this approach, here is the story of a participant of initiative C, Mr Pamuk, which was told in different narrative environments. Mr Pamuk said in an informal setting that he is angry about being in the Netherlands because he would much rather live in Turkey, his country of origin. In a meeting that the leaders of C have with the unemployment agency, they retell this story while at the same time connecting it to other stories about medicalization, labelling and the power of genuine personal contact.

One leader says:
We believe that something big happens in a real conversation in which you can talk about what you really want, and we are very critical about labeling.
For example, we talked to a young Turkish man who received the label of autism. But we believe that he is angry because he does not live in Turkey.
In this narrative interaction, the leader of C makes intertextual connections and expects policymakers at the unemployment agency to make their own connections and consider stories about autism in different ways. Moreover, by retelling this small story in the narrative environment of an unemployment agency, the leader of C also gives the small story the power to challenge the bureaucratic system of the agency, which includes medical labels. These narrative interactions can include complete stories, but they also include small and fragmented utterings, like Mr Pamuks expression of anger, that acquired additional meaning in interconnected narratives. In our analysis, we trace these types of narrative interactions.  Two students, working on their bachelors thesis for social work and for health sciences contributed to the data collection of this project. They had both followed training in qualitative research-methods and their fieldwork was supervised by the first author. The students formulated their own research aims and question, however their questions also contributed to the overarching research project.

| RE SULTS
We identified two types of narrative interaction between master narratives about the transitions and small stories told in newly developing care practices. The first type can be characterized as 'idealizing narrative interactions', in which small stories about care were selectively retold to support the policy ideal of caregiving within informal lifeworld relationships. The second type of narrative interaction gave more prominence to small stories that did not fit the master narratives. We refer to them as 'pragmatizing narrative interactions'. Here, policymakers and citizens used small stories told in care practices to make adjustments to the caring roles and relationships imagined in the transition narratives. Figure 1 gives an overview of the two types of narrative interaction and adjustments to master narratives. Below, we first outline idealizing narrative interactions, and then we describe pragmatizing narrative interactions. In the discussion, we reflect on the democratic character of these narrative interactions.

| Idealizing narrative interactions
In

| Narrative relocations of care
A first storyline in master narratives is dedicated to the relocation of care from professional institutions to citizens and volunteer initiatives. The vision of the municipal future agenda was to move caring activities closer to lifeworlds: 'We put the citizen and their direct environment at the centre. And not the system'. 38 The contrast between lifeworlds and systems was supported with metaphors borrowed from nature: We stimulate voluntary engagement and value volunteers. Volunteer initiatives have a natural way of making people look after each other. They are the unique DNA of our municipality, center, or neighborhood because they originated from the energy of our residents and not from blueprints. 38 In the context of this policy narrative, the social care initiatives were funded as they were considered to be locations of 'collective TA B L E 2 Narrative environments and central topics The fund will finance citizen initiatives that have the potential to take over a social task of the municipality and to subsequently execute this task cheaper and in a qualitatively better way. 39 Participants in the initiatives told many small stories about mutual care between volunteers that support these master narra-   With this story about his strengths and relentless job search, it was difficult to understand why Perry did not manage to find employment.

| Stories of citizen power
The leader of C explained in a private conversation that she did not manage to convince Perry that an unpaid position might also be an acceptable achievement for him. It seems that no one managed to put his 'strengths' in perspective, to reflect on the failing strategy of the ongoing job search or take away his financial worries. Like the story of the driver, this story also had the potential of a counter-story, challenging shared understandings about citizen power in welfare reforms.
However, this small counter-story about failure also had a similar lifecycle and remained an isolated case. It was not retold in meetings between leaders of initiatives and municipal policymakers.

| Pragmatizing narrative interactions
In The overburdening of staff and volunteers led to conflicts, intimidation, gossip and dropouts. Two volunteers said, for example, that they felt frightened by the stories told by another volunteer: We had a girl walking around here… she did everything for the farm manager, morning 'til evening, and that was sparking off a lot among volunteers, so much gossip and lies, no one dared say anything anymore.
Some conflicts between volunteers led to dropouts who expressed their anger in meetings with the municipality and partner organizations. Even if most volunteers found a valuable and supportive environment in the farm, the ideal of the 'natural caring environment' was shattered in these stories. The managing board of the farm invited a municipal official to take part in conversations about these developments. What was needed to improve care?
The board did not want to transform the farm into a professional care institute. However, the board and municipal official did con-

| Adjusting stories of citizen power
Pragmatizing

| D ISCUSS I ON
In Western welfare states, people who struggle with disease, disability and unemployment have seen dramatic changes in the forms of care and support available to them, and many discussions deal with the democratic character of these reforms. According to reform advocates, care is moving out of bureaucratic systems and into the lifeworld, and in the process, citizens are liberating themselves from colonizing professional regimes and labels. Furthermore, they did not retell the stories about the failures of participants to find employment in spite of great efforts and belief in their own powers. These stories remained powerless or even incomprehensible in a setting that celebrated lifeworlds and citizen power.
In pragmatizing narrative interactions, leaders of social care initiatives and municipal policymakers also retold the disconcerting small stories about experiences that did not fit the ideal of participation society. While still connecting to the master narrative, these stories also changed its meaning. 'Lifeworld care' and 'citizen power' remained central themes, but the conflicts that occur in lifeworlds and the struggles and vulnerabilities of people were taken more seriously. structures including clients and lay persons, our focus has been on the influence and translation of stories shared in daily interactions. [43][44][45][46] Making sense of chaotic, disturbing or isolated stories in the narrative environment of work floors, leadership and municipal policy helps to construct democratic routes in care, as introducing such stories in formal and informal settings implies that the experiences of participants not included in roundtable discussions will be voiced.

ACK N OWLED G EM ENTS
We thank Marit de Rijk and Alina Clames for the help they provided with interviews.

CO N FLI C T O F I NTE R E S T
The authors have no conflict of interests to declare.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are not publicly available due to restrictions, for example their containing information that could compromise the privacy of research participants.