Introduction

Meeting global targets that maintain temperatures at 1.5 °C above preindustrial levels while adapting to the growing impacts of climate change requires significant and rapid changes in technological and behavioural domains, energy systems and structures, governance, and in the worldviews, cultures, norms and beliefs underpinning high carbon societies (IPCC 2018). Within this context, there has been growing attention to building community resilience to shocks and stressors and as a forward looking process to addresses underlying challenges associated with climate change (Pelling 2011; Pelling and Manuel-Navarrete 2011; Bassett and Fogelman 2013; Revi et al. 2014). Community resilience is relevant to many fields including community development, disaster risk reduction, fuel poverty, vulnerability, health, education and sustainability (Magis 2010; Wilson 2013; Aldrich and Meyer 2015; Berkes and Ross 2016; Mulligan et al. 2016; O'Donnell et al. 2018). Although there has been extensive research on conceptual aspects (Chandler 2014; Cretney 2014; Choudhury and Haque 2016), there has been much less on understanding how to build community resilience in practice (Fazey et al. 2018a).

Communities are groups and individuals connected by common values, norms and/or interests, and sometimes a geographic place, that shape a shared sense of identity (Kenter et al. 2015; Barrett 2015). In a context of climate change, the issues facing communities are diverse. They include direct impacts to lives and livelihoods, such as from floods, droughts or shifting markets (Jacob et al. 2010; Fazey et al. 2016; Rahman et al. 2018; Currenti et al. 2019; Doloisio and Vanderlinden 2020). Many of these climate impacts intersect with, or are exacerbated by, other issues such as poverty, ill-health, food insecurity, obesity, gender inequalities (Friend and Moench 2013; Jordan 2018; Currenti et al. 2019; Doherty et al. 2019), or with continually evolving stressors that affect everyday life (Câmpeanu and Fazey 2014; Rahman et al. 2018). The pathways through which these challenges are experienced as a community also depend on interplays between different ethnic, cultural or socioeconomic groups (Fazey et al. 2016); through relationships with local and regional governments and other actors (Woodruff et al. 2018; Revell and Dinnie 2020); and through differing contexts of social power (Ramcilovic-Suominen and Kotilainen 2020; Revell and Dinnie 2020; Haverkamp 2021). Some communities face existential threats from climate change, ranging from challenging communities sense of culture, value or identity (Jacob et al. 2010; Jaakkola et al. 2018; Doloisio and Vanderlinden 2020; Espeso-Molinero and Pastor-Alfonso 2020) to threats to actual existence, such as experienced by those having to relocate in the face of major coastal land loss and sea level rise (Connell 2016; Bisschop et al. 2018).

Community resilience can thus be considered to be the ability of a community to adapt to different kinds of interconnected social, environmental and economic change and in ways that promote further change towards healthy community functioning (Magis 2010; Wilson 2013; Fazey et al. 2018a). Recent reviews highlight there are diverse approaches to building or enhancing community resilience (Ross and Berkes 2014), including: capturing stories about place to inspire change; building intangible social assets to reduce vulnerability (Burnell 2013); and using community participation, assessment or planning (Pfefferbaum et al. 2015). Resilience building is, however, predominantly a social process involving different actors working and learning collaboratively (Hahn et al. 2006; Sitas et al. 2016; O'Donnell et al. 2018). It requires, for example, building relationships and trust, creating organizational linkages, boosting social supports, developing decision-making capacities and trusted sources of information (Norris et al. 2008; Cavaye and Ross 2019). Many resilience initiatives require customised approaches for each unique community (Pfefferbaum et al. 2015) and support from a proactive local government as well as shifts at wider regulatory and political levels (Revell and Dinnie 2020; Schneider et al. 2020). To be effective, resilience building tends to require going beyond working with symptoms to also addressing underlying drivers of risks and vulnerabilities, such as particular socioeconomic trends and inequalities (Bassett and Fogelman 2013; Choudhury and Haque 2016). Given the rapidity of social, environmental, economic and other change, building community resilience is thus likely to require transformational thinking and approaches (Cutter 2020).

Resilience building is complementary to community development (Zautra et al. 2008; Ross and Berkes 2014; Henfrey and Giangrande 2017; Cavaye and Ross 2019). Community development provides important insights about how to achieve endogenous or sympathetically facilitated ‘bottom-up’ iterative processes and enhance confidence and capacities for collective working towards practical needs and to learn from successes and set backs (Bhattacharyya 2004; Matarrita-Cascante and Brennan 2012). This includes insights about empowerment, improving well-being of the more disadvantaged, and challenging structural relationships that keep such people disempowered (Brown 2016; AlWaer and Cooper 2020; AlWaer et al. 2021). Such work emphasizes the need for building capacities for self-reliance while also enhancing linkages, support and partnerships at higher social scales (Chapin et al. 2016; Ayal et al. 2017; Henfrey and Giangrande 2017). Social–ecological strands of resilience, on the other hand, highlight the need to work in integrated ways with systemic and interconnected issues, such as across sectors (e.g. adaptation and mitigation, water and food), social scales (e.g. from families, communities to governments), and time (past, present, future) (Berkes and Ross 2013; Wilson 2013, 2014; Chandler 2014). Community development and resilience both highlight the need for capacities for effective navigation of diverse norms, values, perspectives and interests (Voß et al. 2007; Fazey et al. 2011; Kenter et al. 2011; Brown 2016). Yet, despite important insights from both social–ecological strands of resilience and community development, there still has been limited cross fertilization of ideas (Matarrita-Cascante and Brennan 2012; Berkes and Ross 2013).

This research aims to understand the social dynamics involved in resilience building to inform how it can be improved in practice. The work analyses the dynamics of the Scottish Borders Climate Resilient Communities (SBCRC) project in the UK, which applied an action-oriented, participatory, systemic, and relationship building approach and combined insights from resilience and community development. Our research assumes that developing knowledge about resilience building requires learning from trying to do it in practice. Rather than providing an evaluation of the project’s outcomes, the focus is to understand how resilience–building unfolded, and the implications this has for enhancing future efforts. The paper first explains the context and change process applied, followed by methods used to understand the social dynamics, the findings and discussion. The paper is important because it: focuses on the ‘how’ of community resilience building rather than the ‘what’ of concepts and problems; includes a novel process for resilience building; and presents a new grounded theory of the social dynamics involved.

Materials and methods

The focal project

Aims and context

The SBCRC project, which brought together insights from resilience and community development, aimed to enhance resilience of three communities to shocks and stressors of climate change by working with interconnected issues and challenges. This involved operationalising ten essentials of resilience (Fazey et al. 2018a), such as working with interconnections across issues like housing, land use, inequalities, flooding, adaptation and mitigation; working vertically across governance scales; working with different normative goals, envisioning and enacting new futures and by encouraging learning and capacities for adaptability.

The 15-month project (May 2015–September 2016) worked with three communities (Peebles, Hawick and Newcastleton) in the Scottish Borders, a region south of Edinburgh on the east coast of Scotland (Table 1). The region includes a network of small market towns that historically developed along the rivers, with the water used to drive mills for a once thriving textile industry (McLean 2016). The region has an ageing demographic and is known for rural recreation (game fishing, mountain biking, tourism), large-scale farming and commercial forestry. Its most immediate climate related challenge is flooding, with eight areas being acutely disadvantaged in terms of flooding, including some of the larger towns and rural villages (Kazmierczak et al. 2015). Flooding is increasing (Werritty and Sugden 2013) and is combining with other issues, such as potential rising food prices and low employment, thereby increasing stress in communities. The sector most involved in working with climate change has been the emergency services and local government, such as through establishing local community resilience groups. This work has primarily focused on enhancing capacities to respond to shocks rather than being framed as working with climate change.

Table 1 The three communities involved in the resilience building project
Full size table

The three communities were chosen because they had a history of growing impacts of climate change (mostly flooding), but also to provide different contexts and challenges (urban regeneration, commuter town, rural development) (Table 1). The project was implemented on a small budget (£100,000) with funds provided by the Joseph Rowntree Foundation’s Climate Justice programme. The team included the University of Dundee, the Scottish Borders Council, Tweed Forum, Southern Uplands Partnership, the International Futures Forum and the Scottish Association for Marine Science. A dedicated project officer was embedded in the Scottish Borders Council. In total 284 individuals participated in the project. 219 attended workshops with 166 being community-based participants and 53 coming from established organisations, such as public bodies (e.g. Scottish Environmental Protection Agency, the Forestry Commission, and Scottish Borders Council); local organisations (e.g. Housing Associations and a large local Estate); and from other NGOs with a focus on rural development, energy, and environmental management.

Four-tiered process for community resilience building

A community resilience–building process was implemented that included eight main features, conceptualised as four interacting tiers (Figs. 1, S1). The first tier focused on developing relationships and trust between the project team, community members and representatives from different organisations (e.g. statutory agencies, local government and nongovernmental organisations). This was then supported by the second tier that focused on enhancing capacities to work with interconnections across issues, social scales and time and to enhance agency for more holistic approaches to resilience. The third tier then supported the first two by guiding how the project was implemented. This included: three workshops in each community and a policy synergy workshop; actions implemented in collaboration with different stakeholders in each community between the workshops; and implementation of principles of community development. These aspects helped ensure the process enhanced participation, co-learning and capacities for collaborative action. The fourth tier provided action-oriented research to iteratively support the project as it unfolded and to elicit insights for future initiatives (this paper). The ‘research’ was delivered in a participatory way to complement other tiers, including co-learning and relationship building. Full details of the resilience–building process are outlined in S1.

Fig. 1
figure 1

The four-tiered community resilience building process (See also S1)

Full size image

Importantly, while it was recognised from the outset that the way resilience was conceptualised mattered, the project deliberately did not seek to directly draw out interpretations and framings of the concept of resilience. This was to ensure the main focus was on building trust and relationships with different actors and ensure the project was immediately relevant to them, which would have been hampered by too much focus on such a contested term and to which any consensus was unlikely to be achieved. Instead, underlying aspects of what was already known as being key to resilience were explored, such as drawing out understandings of the interconnections between many facets of climate change, exploring how these related to the dynamic and localised nature of each community, and then how this connected to those disadvantaged within them. This then provided opportunities to support learning and gradually build collective understandings of the complex nature of climate change, helping avoid superficial responses and actions. As such, it was through gradual exploration and attempts at building resilience that participants developed and shared their understandings of community-based change, and thus a deeper understanding of both the nature of community resilience and how it comes about.

Research methods

Approach

The research aimed to develop ‘how to’ knowledge for resilience building. ‘How to’ knowledge has generally received limited attention compared to problem analysis in academia. Doing so requires different approaches that can elicit and develop practice-oriented knowledge, such as embodied techne and phronesis, as well as more abstract episteme (Rolfe 1998; Flyvberg 2001; Fazey et al. 2018b). To advance ‘how to’ knowledge, this study was thus framed as second order research (Umpleby 2016). This rejects the often implicitly held assumption that researchers can and should be independent of what they observe. Instead, second order scientists actively work ‘as if from within’ the system being studied (Umpleby 2016; Fazey et al. 2018b) with knowledge generation and action viewed as being closely intertwined and being more aligned to issues of community importance (Greenwood and Levin 2007; Umpleby 2016).

A second order approach enabled a co-creative process where different actors played diverse and complementary roles. For example, ‘researchers’ acted as facilitators, project managers, and knowledge brokers and ‘practitioners’ acted as researchers, such as by helping collect and make sense of data and by being critical observers. It included a primary ‘pracademic’ (E. Carmen)—an embedded researcher and practitioner—who was the central actor and meaning-maker in the project. Validity was then assumed to have been enhanced because participant researchers were closely aligned to the messy real-world of action, enabling them to learn from it rather than being divorced from its practice. This work was, however, also supported by systematic processes that encouraged reflexive monitoring about how being embedded influenced insights as they emerged (Arkesteijn et al. 2015) (S1).

Data collection and analysis

While researchers were embedded in action, to understand the social dynamics of resilience building, four explicit data sources were collected:

1. A reflexive diary kept by the primary pracademic: This provided detailed reflections on discussions, workshops, meetings, local policies, practices and initiatives, how the project was unfolding and why, emerging storylines and narratives, and critical learning and observations about the social dynamics involved. Diary entries were made throughout the project, such as after significant meetings and workshops or when faced with obstacles or challenges.

2. Formal interviews conducted by an evaluator (J. Rao-Williams): This included 47 face-to-face or telephone interviews of 20–30 min, mostly with the same participants at different project stages (27 different participants, with 9 from the project team, 9 from participating organisations: and 3 from each community). Interviews focussed on project progress, what was or was not being achieved, challenges and opportunities, and assessed tangible, capacity building and learning outcomes. Findings were fed back into the project as well as being used in later analysis.

3. Surveys to evaluate the ten workshops to understand participants’ perceptions of project progress and their learning.

4. Three formal reflective meetings and a final reflective workshop with the project team: This helped surface new insights, confirm or challenge emergent thinking, and consider why particular outcomes and findings were occurring.

The data were triangulated and analysed to develop a grounded theory of the social dynamics of resilience–building through three iterations (Fig. 2a). This included: coding and development of project timelines and the different influences on emerging outcomes (Fig. 2b); refining timelines with team members; further refinement and development of a causal loop diagram (CLD) (Sterman 2000) to show key influences and feedbacks and enable wider systemic insights to be derived (see full details in S1). Ethical clearance was obtained through the University of Dundee.

Fig. 2
figure 2

Analysis: a The iterative process of data analysis and b an example of a timeline produced for each community during analysis. Yellow: Project activities; Pink = Learning outcomes; Blue = Tangible and capacity outcomes; Green = External influence; Orange = Opportunities

Full size image

Results

The resilience–building unfolded in different ways in each community (Table 2, Fig. 3). In Hawick, the project capitalised on the existing work of the local government and engineers who were designing a new flood scheme to bring in community members to re-shape designs so the scheme contributed to wider systemic issues. In Peebles, effort focused on building relationships between diverse actors, which eventually led to establishing a community-based resilience group to work collaboratively with local emergency services. Finally, in Newcastleton, new relationships between the community and local authority were formed, leading to new collaborative working groups and community-based initiatives (Table 2, Fig. 3).

Table 2 How projects unfolded in each community
Full size table
Fig. 3
figure 3

Visual overview of how projects unfolded in each community. Yellow boxes are the issues around context of each community; pink boxes the project activities; green boxes external influences and blue boxes key outcomes. Numbers refer to the workshops, which provided structure around which other activities occurred

Full size image

From a critical stance the tangible outcomes may seem limited. Yet, while we do not seek to over-claim success, in a short period the project also laid important foundations and capacity for collective working across different sectors and governance scales from which long-term systemic interventions were beginning to become possible. Had the project received long-term support, more transformative and holistic outcomes and action are likely to have emerged. The grounded theory emerging from the analysis explains how the different aspects of the social dynamics led to these important foundations (Fig. 4). These dynamics included five main reinforcing loops (R1–R5) and four key influencing factors.

Fig. 4
figure 4

A grounded theory of the social dynamics of resilience–building represented as a causal loop diagram (Sterman 2000). Arrows indicate a causal relationship between variables with the direction of the arrow indicating the influence of one variable on another. Polarities explain the nature of that relationship. Positive polarities indicate that when an initial variable increases or decreases, the variable it influences changes in the same direction. ‘R’ labels indicate reinforcing feedback loops. That is, the direction of change (increase or decrease) in one variable affects other variables so as to reinforce the direction of change in the initial variable. Solid arrows are influences supported by empirical research findings. Dashed arrows appear to be occurring, but the precise evidence has not been able to be fully substantiated

Full size image

Reinforcing loop 1: strengthening relationships enhanced collaborative action

The first reinforcing loop, relationships and collaborative action (R1, Fig. 4) represents how enhanced strength of relationships and trust from relationship–building efforts (e.g. enhancing connections between actors, convening dialogue, mediation between groups) led to greater willingness and ability to wield power collaboratively. Social power had been expressed in the project in different ways, such as how participants: shared or withheld information; enabled or allowed meetings; chose to be absent or present; held on to, or chose to apply scarce human, monetary or physical resources; or in the way they made different decisions with or without certain people. Power was thus not simply wielded by those in local government because they could make decisions across wider geographical scales and provide support over a large number of communities with limited resources. Instead, the need for local government and wider actors to provide equitable support across a large geographic area and for many people meant that these actors were highly dependent on cooperation by community members who also wielded their own kinds of convening power, such as in choosing who and when they engaged. As greater willingness to collectively wield power emerged, however, this then increased likelihood of effective collaborative, systemic action.

An example was in Newcastleton, where community members already had extensive motivation and considerable capacity for action, but were initially deeply concerned about loss of autonomy if they worked more closely with the local authority, in which the community had limited trust. Extensive effort was then made in the project to enhance working relationships, with the project officer and those from a third-party organisation often acting as mediators. Once initial levels of trust had been developed, multiple actions rapidly flowed (Table 2, Fig. 3). For example, as collaborative working increased, then so did relationships and trust, closing the feedback loop (Fig. 4, R1). In all cases collaborative working continued for at least two years after the project ended. In Newcastleton dialogue continued through partnerships to examine land use (Table 2) and in Hawick a community group established in the project had made significant progress in developing a renewable energy scheme. Enhanced collaborative action between communities and local government also directly enhanced effectiveness of responses to flooding and extreme weather events of winters 2016/17 and 2017/18. Thus, the extensive focus on relationship and trust building proved critical for long term, novel and collaborative outcomes.

Reinforcing loop 2: learning about systemic aspects enhanced likelihood of collaborative systemic action

The interviews with participants highlighted the process stimulated greater learning about what constitutes disadvantage, the systemic nature of climate change, processes and skills in facilitation and community engagement, and the diverse expertise of others (e.g. about flood risk, renewable energy, community development, poverty initiatives) (Fazey et al. 2017). This learning enhanced the way participants understood interconnectivities between issues, with the enhanced holistic understanding increasing likelihood of collaborative and systemic action (R2, Fig. 4). In Hawick, for example, as the flood scheme began to be viewed more widely and in holistic ways, other working groups emerged that focused on the arts, education, economic development, and recreation, enhancing possibilities for systemic action. Further, while not fully evidenced, there were also suggestions that engaging in more systemic-oriented aspects reinforced holistic understanding (R2, Fig. 4). Future initiatives seeking to enhance ability to work with interconnections between issues therefore need to provide the right conditions, and increase the opportunities, for doing so.

Reinforcing loop 3: learning to understand the understanding of others enhanced relationships and collaborative action

The extensive work on engagement exposed participants to diverse perspectives, mindsets, values and expectations of others. Many of those involved expressed how this helped them develop understanding of the understanding of others, which in turn increased strength of relationships and trust, leading to more collaborative action, closing the feedback loop (R3, Fig. 4). In the project, the loop was enhanced by actively surfacing and working with different mindsets, assumptions and action logics. For example, there were different perceptions of how change was expected to come about and who had power in decision making. In many cases, the different perspectives stemmed from adherence to traditional notions of governance and control. For example, those from local government often implicitly assumed a more top-down approach because they could not see how devolving decision making would enable them to retain sufficient control to ensure statutory obligations were met, such as ensuring public safety in relation to floods where quick and decisive action is needed. While many such issues were difficult to surface and challenge, when it did occur, substantial breakthroughs occurred. For example, in Newcastleton mediation between opposing perspectives unleashed possibilities for a new kind of relationship, unlocking possibilities for a number of different initiatives to emerge. (Fig. 4, R3). Overall, ability to enhance understanding of the understanding of others and work with this in constructive ways proved important for systemic and collaborative outcomes to occur.

Reinforcing loop 4: Holistic learning and understanding of others

Enhancing understanding of the understanding of others also enhanced holistic understanding and appreciation of the interconnected nature of climate resilience (R4, Fig. 4). For example, in Hawick participants’ different orientations to the future were explored in a workshop to shape the design of a new flood scheme. The workshop, by chance, occurred five days after a major flood that caused significant damage, with some residents losing homes. While tensions in the workshop were extremely high, the futures approach helped participants to work simultaneously with different future-oriented mindsets (managerial, entrepreneurial and visionary). Surfacing and legitimising all three mindsets completely changed the atmosphere from one of anger and fear to one of hope, motivation and greater trust, and ultimately to formation of highly motivated working groups. This opened up space for greater appreciation of the holistic nature of the challenge, increasing understanding of the understanding of others, thereby closing the feedback loop (Fig. 4, R4). The key lesson here was that enhancing understanding of the understanding of others through surfacing and working constructively with different orientations to the future enhanced willingness and capacity to work with interconnected issues in more holistic ways.

Reinforcing loop 5: Dancing with tensions

Enhanced strength of relationships and trust also enhanced ability to work with tensions, which led to greater likelihood of collaborative action, reinforcing relationships and trust (R5, Fig. 4). Tensions are dilemmas that cannot easily be reconciled and for which there is no single right way to approach them (Höijer et al. 2006). Instead, continuous reflection and re-orientation is needed to ‘dance’ within an accepted bandwidth between the extremes of the different dimensions involved (Sharpe et al. 2016).

In this project, many tensions were experienced (Table 3). This included tensions between perceived need for achieving quick outcomes versus long-term goals and between the extent to which participation versus direction was needed, with some community members wanting more involvement in decisions and action while local government members struggled to see how this could be achieved. In some cases tensions were able to be managed through balancing others. For example, participation versus direction was partially foreseen and managed by balancing structure versus flexibility, where a skeleton of community workshops provided a degree of direction (Fig. 1) but which was then used flexibly depending on shifting and emerging needs stemming from participants. Achieving balance between participation and direction was further supported by balancing provision of support versus encouraging autonomy, where there was a continuous dance between assisting community activities and then stepping back to encourage ownership and autonomy. This, however, was complicated by independence versus vested interests of project team members, where the project team needed to be independent to provide genuine facilitation and participation but who also were accountable to their own institutions and project funders.

Table 3 Tensions experienced in community resilience building
Full size table

Overall, as relationships and trust were enhanced, those involved developed greater willingness to work with tensions in more constructive ways, leading to greater likelihood of collaborative and effective systemic action, and eventually closing the feedback loop (R5, Fig. 4). This was enhanced by anticipating tensions, which suggests that future projects could be enhanced by surfacing tensions at the outset and finding ways to work flexibly with them.

Additional influence 1: willingness or ability to engage in radical learning affected potential for transformational outcomes

In addition to reinforcing loops, there were four critical influences affecting how projects unfolded (Fig. 4). The first was willingness and ability of participants to invest in learning, which was key to exploring and breaking out of particular mindsets and assumptions (Fig. 4). Many participants, for example, missed critical opportunities for learning because, at short notice, they were required to respond to urgent duties elsewhere. New approaches that can quickly draw out implicit thinking, work with underlying assumptions, and enhance willingness of those involved to engage with uncomfortable and messy learning processes are therefore needed. Finding ways to do this was a significant challenge in the SBCRC project.

Additional influence 2: quality and quantity of support affected opportunities for relationship building, learning and working flexibly with complexity and tensions

Quality and quantity of support was critical to achieving collaborative and systemic action (Fig. 4). This support included high level skills in facilitation, participation, and working with systems and futures. While such skills are often highlighted as important (Hagmann et al. 2002; Sitas et al. 2016), they are also often overlooked, not sufficiently resourced, or available. In the SBCRC project the dedicated project officer was essential to support community-based projects while facilitation and futures methods and skills were key for working with tensions, encouraging holistic understanding, understanding others, and in strengthening relationships and trust (Fig. 4). Quality support also enabled flexible working. An example was when major floods struck each community, providing opportunities to galvanise action and form new resilience groups in two of the towns: Peebles and Newcastleton. Flexibility was also greatly enhanced by having an adaptable funder and a project focused on ‘softer’ process related goals of learning and relationship building as well as more tangible fixed outcomes. Overall, the quality and quantity of skilled support proved to be a necessity rather than a luxury.

Additional influence 3: framings of resilience shaped community resilience building

The appropriateness of narratives and framings of resilience affected likelihood of collaborative working and potential for systemic action (Fig. 4). As outlined in the methods, given the highly contested nature of resilience, the project did not initially open up conversations about the concept. Instead, it assumed that both resilience building and shifts in understanding about the concept had to be part of a wider learning journey, developed over time and shifting with experience and exposure to the perspectives of others. Nevertheless, different framings of resilience clearly had an influence on how different parts of the project unfolded. For example, in Newcastleton, resilience was framed by the community as coming from building capacities to work with longer term stressors such as employment and limited access to infrastructure. This was in contrast to those from the local government who framed resilience as enhancing efficiencies and ability to respond to more immediate shocks. Lack of alignment of framings slowed collaborative efforts and detracted from working towards longer term challenges, such as those associated with climate change. Thus, while levels of trust were not sufficient at the outset to tackle contested framings of resilience, opening up conversations about the differences as part of a follow-on stage of the project is likely to have been helpful. This would, for example, have allowed for further exploration of deeper implicitly held assumptions about social power, such as who needed to wield power and when to enable change to happen. In Newcastelton, this would have been helpful in establishing how they could be further empowered to take action for the longer term while also ensuring the local authority also retained control so they could meet statutory obligations, such as ensuring safety of people when they were at immediate risk from flooding.

Additional influence 4: the policy landscape influenced community resilience initiatives

A key influence on community resilience building was the policy environment, including wider political, legal and decision-making contexts in which communities and local governments were embedded (Fig. 4). Four key aspects were identified by policy professionals that limited effectiveness of community resilience building: (1) tendencies to favour economic growth over longer term climate resilience in cost–benefit decisions; (2) limited community capacity to work with climate challenges; (3) poor coordination across policy domains that limited opportunities for multiple gains; and (4) limited approaches for working with highly interconnected challenges. These policy environments tended to be outside of the control of both communities and the local authority, but had an important influence on what was deemed possible.

Overall dynamics

The combined ‘system’ of the social dynamics of resilience building (Fig. 4) has important implications. First, all of the components described were important for resilience building, and together provide ten lessons to guide future initiatives (Table 4). Importantly, however, when these aspects were enacted together and efforts were put in to help make connections between them, they began to stimulate a beneficial reinforcing process. That is, once critical thresholds in relationship building and trust had been achieved, and with support from working with other elements in the system, a more regenerative process began to emerge. In effect, rather than just enhancing what existed, a new systemic structure had been created, allowing a new set of dynamics to occur. This suggests that, at least theoretically, major change may be possible in the long term if there is an appropriate focus on systemic forms of change, the capacity to work with it, and sufficient attention to capitalise on the reinforcing nature of the social dynamics involved. Thus, while wider sociopolitical scales are important, and achieving deeper transformational change is difficult, communities in this project did at least begin to appear to be developing agency and capacity for change, helping them to become the drivers of change. It did, however, require extensive work and specialised support, and it was not clear how much and for how long support would have been needed for longer term outcomes to be realised.

Table 4 Ten requirements for emergence of a self-reinforcing process of community resilience building
Full size table

Discussion and conclusions

This research sought to help meet growing demand for knowledge about how to do community resilience–building in practice by learning through action. The findings highlight the complexities of resilience–building, requiring those involved to work with the messy and often intangible world of mindsets, assumptions, desires, relationships, power and trust. They show that this, in turn, requires a complex array of resources and capacities and being ready to capitalise on different opportunities as they emerge (summarised in S2). In particular, it involved being able to hold ‘spaces’ where contestation could be surfaced and worked through while also making sure those involved felt the process was of relevance, while also bringing long-term issues to the fore. Such resilience building is a ‘practice’ which, like the subconscious and embodied action of driving a car, requires learning by doing, and not just focusing on the development of abstracted epistemic knowledge.

Much of the challenges in the project related to inability of those operating at a single community scale to influence wider social contexts, including sociopolitical conditions, governance, fiscal and legal structures, framings and narratives. It also emerged from different needs and different framings of resilience. Some framed resilience through a narrow yet important lens of needing to uphold their statutory obligations to save lives in response to rapid events like floods, and across a large number of communities. This led to a more specified form of resilience, and attention to enhancing efficiencies. For others, it was viewed more in relation to empowerment and enhancing autonomy, but then the same individuals also at other times felt it came from external provision of support. In many ways, resilience was thus a contestation that revolved around who had power, for what and when, with the notion of resilience also changing for different issues. The contestation often limited possibilities for more collective endeavours, including limiting possibilities for a focus on deeper underlying systemic challenges facing communities or long-term issues associated with climate change.

Importantly, however, the results highlight that as deeper understandings of others, trust, relationships and appreciation of the systemic nature of challenges increased, then so did possibilities for collaborative and systemic action. This suggests significant potential for resilience building efforts to begin to shape social dynamics so they began to reinforce themselves in beneficial ways (Fig. 4). Had the project been able to continue, it would probably have been possible to take it to another level, including tackling deeper contested issues as trust and thus openness to explore differences increased. Importantly, while difficult to disentangle, it also appeared that a new system dynamic had been created, with new connections that led to new beneficial feedback loops. If this is indeed what happened, then the outcome was in line with the kinds of transformational approaches and outcomes that will be needed to work with extensive change (Cutter 2020).

A goal for future resilience–building would thus be to work with the ten key lessons (Table 4) to build self-reinforcing systems in line with emerging ideas of regenerative systems in other domains (Wahl 2016). This would involve reframing resilience building from being primarily about harm reduction (e.g. resilience to prevent loss and damage) to viewing it as a process of creating new system dynamics that reinforce restoration of critical community functioning, health and viability. This is a goal that should not be taken at face value as it would require significant effort to help create the pattern shifts needed for such outcomes and processes to be realised. Much of the insights arising from this project, for example, place a strong emphasis on learning, trust, relationship building, and working with highly contested issues and needs (Table 4). All of these are difficult to navigate at the best of times, let alone for major challenges such as those associated with climate change. Yet, this study suggests that positive outcomes are possible if the right kind of support over the right kind of timeframe is provided.

In terms of the practice of resilience–building, there is relatively little we would change with hindsight given the resources available and context of the project. However, in future projects we would place a much stronger emphasis on two aspects. First, we would seek to find new ways to work with underlying assumptions, particularly among the project team. This could, for example, be enhanced by using the tensions identified (Table 3) to focus dialogue around underlying assumptions that were held about how change was expected to come about. Second, we would seek to find better ways to engage with those more directly disadvantaged. In this project there was extensive engagement which, in many cases, went beyond the usual suspects. This was achieved because of the way the project built and encouraged a degree of ownership by those involved and ensured relevance to participants. This engagement, however, rarely included those from the most severely disadvantaged backgrounds, and the project failed in this regard. Doing so would require a different skill set and approach, including long-term engagement with disadvantaged individuals and tackling much more deeply rooted issues. For this project, such work was beyond the level of resourcing available.

In addition to practice, the project also had important lessons for research. Having researchers embedded in the action provided greater opportunities to develop ‘how to’ knowledge for resilience. This only worked, however, because it was accompanied by mechanisms that forced critical reflection on how embeddedness influenced meaning-making and provision of opportunities for stepping out of action, as highlighted as being important when a more critical stance is required (Arkesteijn et al. 2015). Further, the insights in the project were also made possible because of a subtle shift in emphasis in the analysis. The conceptual model specifically focused on the social dynamics of the resilience building not on the dynamics of the problem (e.g. of the existing relationships or dynamics between actors). This important shift enabled the work to go beyond critiquing problems or solutions to actively developing epistemic insights about resilience building as a practice.

Finally, development of ‘how to’ knowledge was ultimately only made possible by approaching the research as if from within the system being studied. This included starting from the premise that the primary goal was to stimulate action not to produce knowledge. This was important in helping shift towards the creation of a different kind of knowledge through action, allowing the development of a new kind of understanding about how such action can emerge. Contemporary knowledge production systems have been heavily criticised because they focus on producing knowledge about the world rather than production of wisdom about how to act within it (Maxwell 2007; Kläy et al. 2015). Shifting to an emphasis on the production of wisdom requires attending not just to empirical truths about the world but also to working with the personal truths people hold about what they feel is important or wish to see emerge. As experienced in this project this involves working with the messy world of emotions, values, and contestations around knowledge, from which most people, if given the opportunity, usually run a mile. Thus, in addition to new framings of resilience, if research is to help develop relevant ‘how to’ knowledge to support resilience building, then new kinds of research and capacities will be required.