The role of social capital in participatory arts for wellbeing: findings from a qualitative systematic review

ABSTRACT Background Social capital is often cited as shaping impacts of participatory arts, although the concept has not been systematically mapped in arts, health and wellbeing contexts. In wider health inequalities research, complex, differential, and sometimes negative impacts of social capital have been recognised. Methods This paper maps of social capital concepts in qualitative research as part of the UK What Works for Wellbeing evidence review programme on culture, sport and wellbeing. Results Studies often cite positive impacts of bonding and, to a lesser extent, bridging social capital. However, reported challenges suggest the need for a critical approach. Forms of linking social capital, such as reframing and political engagement to address social divisions, are less often cited but may be important in participatory arts and wellbeing. Conclusions Future research should further specify dimensions of social capital as well as their nuanced effects in arts, and wellbeing contexts.

Social capital; sociology; qualitative data analysis; public health; music

Background
In recent years, wellbeing has come to the fore in research and policy, and subjective wellbeing is increasingly measured at population level in many countries. Wellbeing is a broader concept than health, suggesting a positive state shaped by subjective feelings as well as social experiences. Wellbeing is intrinsically valuable and is associated with many desirable social outcomes, such as those relating to health, education and employment (Huppert, 2017). There is no universally agreed measure of wellbeing, but most indicators encompass hedonic dimensions, such as feelings of happiness or anxiety, as well as eudemonic dimensions focused on perceptions of the extent of meaning and purpose in one's life. (Daykin et al., 2020). As the field of wellbeing studies has expanded, the challenge of loneliness has been identified in research and policy. Loneliness is a complex phenomenon, not to be conflated with social isolation or solitude. There are different types of loneliness, but it is generally viewed as a negative experience that is increasingly widespread and associated with poor health and wellbeing outcomes .
The connections between participation in arts and culture and wellbeing have been demonstrated in a growing literature. The field of arts, culture and wellbeing spans a wide range of domains and activities, including targeted interventions for specific population groups, arts psychotherapies, actives in healthcare settings and in museums and galleries, as well as everyday creativity (Ewbank, 2020;Fancourt, 2017). A prominent theme has been the potential of arts engagement to exacerbate or reduce health inequalities (All Party Parliamentary Group on Arts, Health and Wellbeing, 2017;Arts Council England, 2018;Daykin et al., 2018;Fancourt & Finn, 2019;Mansfield et al., 2018).
The connections between participatory arts and wellbeing have been explored in a four-year evidence review programme undertaken by the current authors with the UK What Works Centre for Wellbeing (https://whatworkswellbeing.org) A series of studies have examined impacts, processes and concepts relating to wellbeing in a range of art forms including music, dance, visual arts and leisure activities Mansfield et al., 2018Mansfield et al., , 2019. This current paper is based on a recent review of qualitative research on wellbeing and loneliness across domains of participatory arts and sport. The review was extensive, spanning three volumes that are available elsewhere (L. Mansfield et al., 2020aMansfield et al., , 2020bMansfield et al., , 2020c. Here we develop a focused discussion, developing the key themes arising from the participatory arts studies included in the review. The study sought not to measure effects but to understand key processes shaping wellbeing and loneliness outcomes. This reflects a wider trend towards understanding the underlying mechanisms and processes that support positive outcomes in participatory arts, rather than simply measuring effects (Fancourt et al., 2014). This is a challenging task in a field that is essentially multidisciplinary, and attention has often focused on biological and physiological mechanisms in individuals, reflecting a degree of medicalization within the field (Daykin, 2019). However, reports of participatory arts projects often cite social impacts such as connection, shared identities and social support (Argyle & Winship, 2018;Morse et al., 2015;Pearce, 2017). The notion of social capital is often cited, although the concept has seldom been critically examined in participatory arts contexts.
Social capital has been extensively reviewed in wider health research (Kawachi & Berkman, 2014;Moore & Kawachi., 2017;Villalonga-Olivesa et al., 2018). Putnam's work, in which social capital is a positive resource comprising trust, norms and networks that can through "bonding" and "bridging" foster social relationships and cohesion, has been influential in public health (Coleman, 1990;Putnam, 1995Putnam, , 2000Putnam, , 2002. Bourdieu's work, which connects social capital to economic and cultural capital through a multiplier effect, has also been widely cited (Bourdieu, 1984(Bourdieu, , 1986. For Bourdieu, social capital is generated by the established power and status of the dominant class and class fractions that can control the "conditions of a certain life-style, through the mediation of the habitus" (Bourdieu, 1984: 260). Cultural capital encompasses formal assets such as educational attainment and ownership of cultural goods, and embodied states incorporating values, skills, knowledge and tastes that function symbolically to shape experiences, social standing and hierarchies. Social and cultural capital operate both at macro and micro levels, entrenching "in-group" and "out-group" distinctions (Tajfel & Turner, 1979) and sometimes negatively affecting health and wellbeing (Pinxten & Lievens, 2014).
In summary, social capital is a complex, multidimensional concept with differential effects that are sensitive to contexts (Shiell et al., 2018). This suggests a need to explore the role of social and cultural capital in shaping wellbeing outcomes in participatory arts, including enhancing individual and community wellbeing, but also perhaps segmenting communities and reinforcing existing inequalities. To date, few studies of participatory arts and wellbeing have examined these processes in depth (Daykin, 2013). This paper develops analysis of data from a review of qualitative research of participatory arts that sought to identify key processes shaping wellbeing and loneliness outcomes

Research process and methodology
The review question was developed through discussion with our project stakeholders, who are drawn from a wide range of policy, research, service delivery and service user interests across culture and sport sectors in the UK. This iterative process has guided the review programme from the beginning, enabling the team to build on successive findings over four years, during which time the focus has shifted from general questions about wellbeing outcomes to more specific and honed questions about how outcomes are achieved and how processes play out in specific contexts. As well as wellbeing and loneliness, our stakeholders were interested in questions of space and place, referring not just to physical environments, localities and community assets, but to the emotional, social, cultural and political meanings attached to spaces and how these are shaped by participatory experiences that reflect personal, local, national and global connections. This interest in space and place reflects that fact that many participatory projects form part of programmes designed to enhance social capital in deprived geographical communities (Brownett, 2018;Hampshire & Matthijsse, 2010;McConnell, 2016;Murray & Crummett, 2010;Vella Burrows et al., 2014). Linked with this, our stakeholders were also interested in the role of intangible assets such as shared language and heritage in mediating wellbeing and loneliness and in the role of volunteering, both of which feature in current policy priorities. These priorities shaped the review approach and the question, which was: How are space or place, intangible assets and volunteering conceptualised in reported qualitative research findings on participatory arts and sport/physical activity for enhancing wellbeing and alleviating loneliness across the adult life course (16+ years)?
The protocol for this review was registered on the PROSPERO International Prospective Register of Systematic Reviews (Registration number CRD42019142558), available from https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=142558. The review followed PRISMA guidelines for reporting systematic reviews. The review included healthy and unhealthy adults (16+ years) participating in the arts or taking part in or watching sport or physical activity. The included arts activities were music, drama, dance, visual arts, literary arts, engagement with public art and creative art classes. Excluded activities were professional arts activities and arts undertaken as part of clinical treatment. The review included qualitative research published between 2009 and June 2019.
The systematic literature search encompassed 12 electronic databases ( Figure 1) and additional hand searching to target qualitative studies published between 2009 and 2019 linking wellbeing or loneliness with participatory arts or sports/physical activity with adults.
Search results were independently screened by two authors and discrepancies were resolved by consensus. The search strategy is presented in Figure 2. A total of 103 qualitative studies were included from 1160 identified. Of these, 43 of these were focused on arts and 60 on sports. Data extraction and quality assessment followed standard approaches detailed in the project's published methods guide (Snape et al., 2019). Data analysis proceeded inductively (Noblit & Hare, 1988), using a data extraction tool to identify second order constructs proposed by authors as key processes connecting participation to wellbeing. These were then organized thematically within groupings discussed and agreed with the team (Toye et al., 2014). A narrative synthesis of these themes (Campbell et al., 2011) revealed key processes working to affect wellbeing and loneliness across culture and sport.

Results
Qualitative synthesis revealed overlapping themes pointing to processes by which participatory arts and sport operate to enhance wellbeing and/or alleviate loneliness. By processes we mean emotions, responses, actions and behaviours that lead to changes in wellbeing. The notion of connection through bonding and bridging served as a cross cutting theme across the domains of space and place, volunteering and intangible assets. Many studies reported feelings of belonging in processes of identity formation and development in participatory activities. The notion of connection was discussed in relation to community and locality, with concepts of wellbeing through strengthened personal identity and interpersonal relationships extending to include the development of community spirit and civic participation. Processes of bonding and bridging were also discussed in the context of therapeutic and sensory spaces, through enhancing coping and support, although in these studies processes of distraction and escape from everyday stresses were also emphasised. Connection was also discussed in relation to the theme of safe spaces, which offer open and equal interactions and freedom from bias, discrimination and stigma.
This broad brush summary suggests that, across the domains of arts and sport, participatory activities can foster wellbeing by encouraging connection, fostering feelings of belonging, offering protection from hostile environments, providing tools for coping with difficulties, supporting personal development, and promoting broader community and civic awareness. The data highlight the centrality of processes of bonding and bridging social capital, apparent through social interaction and enjoyment of shared interests, experiences and activities, through availability and access to participatory opportunities and assets, and in experiences of diversity and social divisions, including the impact of physical and mental health conditions and the effects of migration and war. The data suggest some negative aspects of social capital. For example, bonding can create in-group identities and, by definition, can reinforce exclusion of out-groups and can alienate some participants. The evidence illustrates that participatory arts and sport participation fosters bridging through a number of processes, for example, by promoting learning, extending participants' access to places and spaces they would not usually encounter, sharing culture and heritage across demographic and social divides, and providing spaces and allowing for place-making in which political divisions and conflicts can be safely addressed.
Arts and sport are both broad domains and this broad-brush approach does not allow consideration of specific processes pertaining to different contexts. For this reason, the rest of this paper focuses in more detail on the studies of participatory arts. The centrality of the bridging and bonding themes suggest that it is useful to examine these data using a social capital lens. This development of the analytic process is a stepwise progression from our overarching thematic analysis. A further paper is being prepared that examines the data in sports contexts using a similar approach in order to identify similarities and differences. Here we focus on reporting of social capital themes in research on participatory arts for wellbeing.
Data from 40 papers inform this discussion (Table 1). Excluded from this subsample are studies of sports engagement and general leisure (n = 60) as well as studies of clinical arts interventions (n = 3). Here we examine data from these studies drawing on recent Some participants experienced discomfort in relation to some practices, e.g., the requirement to move in time to the music as well as singing. Kelson et al. (2017, Canada). Ethnographic study of social recreation groups involving public artworks for people with young onset dementia. Bonding and Bridging: Participants expressed enjoyment and curiosity and were able to access newly inclusive urban spaces. Kluge et al. (2012, USA) Ethnographic study: observations and interviews with 9 older female participants in a dance movement therapy class in a residential care setting. Bonding: integration and belonging in a new place of residence.
Bridging: Dance classes provided by a high calibre and empathetic instructor stimulated self-awareness, growth and created insights into life experiences that participants hadn't had (those of a professional dancer).

Bonding
The art class was offered in a community described by participants as overlooked, under-resourced and damaged by loss of traditional social practices and hubs such as the local pub. They reported feelings of rejection, separation, frustration and anger. In this context they responded with enthusiasm to the provision of an art class, reporting a sense of achievement, a feeling of creativity and social benefits including increased opportunity for social interaction and forming new friendships. The project fostered a sense of identity and belonging. Bridging The main impact of the project was in enabling participants to assert power and challenge negative external social representations of the community.
Most participants in the study were women, while the men enjoyed the art work it did not mesh with the more instrumental activities that are more important in male working class culture. Despite their enthusiasm for the project (bonding), it did not extend to forging wider connections (bridging). Participants were reluctant to take up opportunities outside the local area.
Pearce ( (1) Bonding (2) Feeling connected -to others in the group, to ancestors, to local area and landscape -connections previously 'fractured' as result of mental illness, experience of discrimination or stigma and resulting isolation.

Some found group activities very challenging
Fear for the future -did not wish project to end, worry about loss of structure and focus when project would end and loss of sense of belonging it had brought.

Imperial War Museum
North ( (1) Bridging: making connections between individual experiences and broader patterns from history, projecting ideas of history and heritage into their own futures.
Bringing groups together involves complex emotional labour. Bonding and bridging Positive experiences of friendship and awareness of improved local environment. Social capital theory suggests is that a cultural organisation's relationship with its stakeholders might engender reciprocity, co-operation and trust, and in so doing generate social capital for a local community. Participants expressed a need for fostering community engagement and cohesion and building on existing and developing new civic partnerships.
Barriers included: perceptions of classcentric offer, education, access to resources, effect of national arts provision policy typologies of social capital in health. The analysis was led by the lead author, with separate validation by a coresearcher and checking by all members of the research team. The analysis sought to expand social capital concepts and identify different dimensions of social capital in participatory arts for wellbeing.

Bonding and wellbeing in participatory arts
Forms of bonding are reported in all the studies, and two broad sub-themes are connection/emotional support and fostering belonging/shared identity.

Connection and emotional support
All the studies attribute processes of connection and emotional support to arts engagement (Table 1). Arts projects are reported as bringing people together to share experiences, enjoy camaraderie, friendships and purpose. No negative instances of connection are reported, although one study acknowledges the complex emotional labour of bringing people together (Payson, 2018).

Bonding challenges and negative impacts
Challenges and negative impacts of bonding feature in some reports. For instance, some vulnerable participants find engaging in participatory arts very difficult (Heaslip & Darvill, 2018). There are concerns about the time needed to build relationships and trust with excluded communities (Elkins & Bailey, 2018) as well as concerns about sustainability, with some participants facing a void when activities come to an end (Baker et al., 2017;Heaslip & Darvill, 2018). Instances of exclusionary practices are reported within some projects, practices that reinforce social divisions such as class, ethnicity and gender Li & Southcott, 2012;Murray & Crummett, 2010;Southcott & Nethsinghe, 2019;Vella Burrows et al., 2014). Cultural capital is sometimes observed to reinforce hierarchies through aesthetic judgements and distinctions based on taste, repertoire, creative skills, accomplishments and experiences Li & Southcott, 2012;Pearce, 2017;Raanaas et al., 2019). It is noted that arts projects can project unrealistic, stereotyped and romanticized views of particular cultures or past cohesion, with nostalgia sometimes framed in terms of fears of loss of national heritage (Kiddy, 2015;McAvinchey, 2016).

Bridging and wellbeing in participatory arts
The bridging theme contains four sub themes of information/resources, extending networks of trust/safety, addressing social divisions, reframing and political engagement. In a small number of cases the bridging theme extends into linking social capital.

Information and resources
Facilitating access to information and resources is often cited as a key bridging process that supports wellbeing. Participants are reported to have increased their use of health services (Sun & Buys, 2013) as well as their knowledge, awareness and desire for learning and professional development following arts projects (Dunphy et al., 2014). Participants in a museum-based volunteering project for disadvantaged people in the north of England are reported to have increased their levels of volunteering as a result of the project (Imperial War Museum North, 2016).

Extending networks, trust and safety
Arts projects can encourage participants to gain access to new spaces and experience feelings of safety in unfamiliar or threatening ones (Imperial War Museum North, 2016;Kelson et al., 2017;Southcott & Joseph, 2015;Todd, 2017). In one project, choir participation for black students in a predominantly white college environment is described as a key mechanism through which academic success can be enhanced (Strayhorn, 2011). Arts participation can also increase access to symbolic assets including familiarity with artists and artistic lifestyles (Kluge et al., 2012;Meeks et al., 2018;Southcott & Joseph, 2015). Some projects also address the limitations of spaces and places by exploring local economic and political circumstances (McAvinchey, 2016;Payson, 2018;Veal, 2017).

Bridging challenges in participatory arts
Bridging challenges are also reported, including feelings of discomfort and heightened awareness of unequal relationships. Sharing stories can be empowering but can also be painful for participants . Feelings of embarrassment can arise in response to perceived pressures to perform in choir projects (Joseph & Southcott, 2017;Li & Southcott, 2012). Further, tensions can arise between members of community arts groups, for example, over choice of performance repertoire that reflects competitive processes framed by external political conflicts (Southcott & Nethsinghe, 2019). Bridging processes are not always successful therefore in overcoming marginalisation. One study reports that participants remained reluctant to take up opportunities outside of their small geographical area even though they had enjoyed the project (Murray & Crummett, 2010). Additional barriers include lack of funding that threatens the sustainability of arts projects (Dunphy et al., 2014).

Bridging and linking social capital: reframing and political engagement through participatory arts
As well as horizontal bridging, some reports identify vertical processes, such as using arts to challenge stigma, to (re)frame experiences and to empower as well as (re)represent disadvantaged communities (Kelson et al., 2017;Lazaroo, 2017;Murray & Crummett, 2010;Payson, 2018;Southcott & Joseph, 2015;Sun & Buys, 2013). One report describes the way in which personal stories emerged as a powerful tool to reframe Maori women's experience of abuse and collectively challenge colonial legacies of brutalization (Lawson-Te . Some projects fostered awareness of citizenship in participants and, in some cases, awareness of their rights and political engagement (Imperial War Museum North, 2016;Lazaroo, 2017;Li & Southcott, 2012;Raanaas et al., 2019;Southcott & Joseph, 2015;Vella Burrows et al., 2014). Linking is affected by the bridging challenges discussed above, including issues of sustainability and resources. Further, engagement with local power holders can be constrained by cultural and political factors such as perceptions of a classcentric offer and the effects of national arts provision policy (Vella Burrows et al., 2014).

Discussion
These studies suggest that Putnam's notions of "bonding" and "bridging" social capital as well as Bourdieu's more critical appraoch can both contribute to an understanding of the pathways through which arts and cultural participation can positively and negatively affect wellbeing (Bourdieu, 1984(Bourdieu, , 1986Putnam, 1995Putnam, , 2000. Bonding, in the forms of emotional support, belonging and shared identity, is reported more often than bridging. The findings support the notion that participatory arts can offer important community resources and help to connect people who share a common identity, situation or characteristics, contributing to improvements in health and wellbeing (Poortinga, 2006). However, arts are not exempt from problems with bonding, which can place demands on disadvantaged people without addressing their limited resources, restrict access to information, exclude out-group identities, level down aspirations, increase exposure to unhealthy behaviours, and further marginalize individuals who find it difficult to fit in with highly cohesive communities (Villalonga-Olives & . Those involved with planning and delivering participatory arts need to be aware of the benefits and difficulties associated with bonding social capital. Care needs to be taken to address the needs of vulnerable participants, to build trust relationships with excluded communities, and to challenge and limit exclusionary practices based on social divisions, hierarchies and cultural stereotypes. Bridging social capital may overcome some of these limitations, enabling boundary crossing, extending networks and sharing resources across different socio-economic groups (Whittaker & Holland-Smith, 2016). There are many instances of bridging in participatory arts, such as extending access to information, providing new opportunities, opening up social and cultural spaces and making them safe for excluded groups, and linking communities and generations. However, attempts at bridging can also draw people into unequal relationships that may be difficult to navigate successfully (Hampshire & Matthijsse, 2010). The data highlight bridging challenges in participatory arts including feelings of discomfort and embarrassment expressed by some participants, as well as competition and heightened awareness of unequal relationships.
Bonding and bridging concepts are often restricted to a focus on horizontal relationships, and both have been criticized for overlooking the role of the state in fostering social capital through a broad range of policies and infrastructure (Agger & Ole-Jensen, 2015). In contrast, "linking" social capital refers to the building of vertical relationships, connecting people to formal sources of power and authority and reinforcing norms of respect and networks of trust (Moore & Kawachi., 2017, Woolcock, 2015. Some linking processes were reported in the data on participatory arts, including challenging stigma and representations of disadvantaged communities, reframing experiences in sociopolitical and historical contexts, and fostering awareness of rights and political engagement. However, these processes can be constrained by local, national and international economic and political forces. This suggests that research on bridging social capital should do more to address the institutional and political boundaries that surround interventions (Shiell et al., 2018).
Questions about social capital are relevant in the context of recent trends towards social prescribing and asset-based approaches to health and wellbeing (Chatterjee et al., 2018;Daykin, 2019). These draw on social movement theory, viewing health oriented social movements as assets that can be harnessed to develop shared solutions to common challenges (Burbidge, 2017;Del Castillo et al., 2016;Kapilashrami et al., 2015). Social movement theory challenges public health institutions in Western neo-liberal societies, regarded as ill-suited to fostering social relatedness and trust, to become more agile (Edmondson, 2003). Harnessing social capital through participatory arts is not a straightforward process, as bonding and bridging are contingent on many factors, including participants' responses and the extent to which people view local community assets and networks as representative of their needs (Campbell & McLean, 2002). Nevertheless, social movements can potentially utilize social capital processes to successfully make demands on power holders regarding service provision, resources and support (Campbell, 2020).

Conclusions
This conceptual review of qualitative evidence was prompted by the need for stronger understanding of the social processes that connect participation in arts and culture to wellbeing and loneliness. Here we have focused on social capital dimensions of bonding and bridging, as these emerged as warranting further attention from a broader review of evidence in culture, sport and wellbeing.
The studies reviewed here identify a range of social capital effects as well as complex interplay between social and cultural capital in participatory arts contexts. Studies tend to highlight positive impacts, but it is important to acknowledge negative experiences as well as the fact that the benefits of different social capital dimensions may not be shared by all participants equally. Salient challenges include negative enactments of cultural capital in the form of symbolic distinctions, exclusionary practices and aesthetic norms in participatory arts. This review supports the notion that bonding in the absence of sufficient bridging in arts projects can reinforce unequal social relations that are detrimental to health and wellbeing. Bridging processes are often reported in studies, most often in terms of breaking down horizontal divisions within communities. Bridging challenges are also reported. Instances of linking social capital, using participatory arts to foster vertical bridging to resist oppression and to successfully challenge dominant structures of power and control, are less frequently identified in current research. Nevertheless, the studies that did focus on linking highlight its relevance to understanding the impacts of participatory arts on dimensions of wellbeing, including loneliness. Reported barriers to successful bridging and linking in arts include lack of sustainability, the effects of broader social policies and the challenge of entrenched social divisions.
The breadth of social capital may reinforce the idea that this concept is too vague and elusive to offer a coherent framework with which to evaluate arts and wellbeing interventions. Social capital is indeed complex and multidimensional. Future research on this topic should specify its specific dimensions in arts contexts, addressing the differential impacts of these and the influence of institutional and socio-political contexts.

Study strengths and limitations
This paper further develops analysis of data from a subsample of 40 studies that reported themes relating to social capital in participatory arts and wellbeing. The data are drawn from an extensive culture and sport qualitative evidence review and encompass diverse population groups from many countries. Participants include rural and urban dwellers, people with and without identified mental or physical health conditions, people from different age groups, socio economic and ethnic backgrounds. Studies include a wide range of art forms including music, dance, theatre, creative writing, museums and heritage. The review is based on standardized procedures for qualitative evidence reviews, encompassing literature searching, screening, data extraction, thematic analysis, synthesis and quality assessments. The validity of the findings is to some extent dependent on the quality of the underpinning evidence. Our quality assessments revealed a range of issues with some high-quality studies and a small number of low-quality studies.
There is no quantitative evidence included and so it is impossible to hypothesise causal relationships from these data. We have focused on processes, and we have drawn on existing social capital theory. Further inductive analysis of primary research is needed to understand the workings of social capital in participatory arts contexts. For reasons of space and conceptual coherence we excluded studies of sports and physical activity from this paper. Further analysis is needed to map dimensions of social capital in sports contexts, and to identify similarities and differences across arts, culture and sport domains.

Disclosure statement
The authors report no conflicts of interest.

Funding
This research was funded by Economic & Social Research Council grant number ES/N003721/1, as part of the evidence review programme for the UK What Works Centre for Wellbeing.