Rethinking social capital in wildfire resilience: the case of central Portugal

Abstract

This article explores the role of social capital in wildfire resilience and is based on case study research in central Portugal. Given the recent revival of the concept across disciplines to explain how communities can cope with hazards, we critically analyse social capital through a social network analysis perspective, introducing an innovative methodological approach. By focusing on the socio‐cultural structures that influence preventive and recovery capacities, two rural communities were assessed and compared to understand the conditions under which social networks operate. The results confirm the importance of bridging and linking ties but reveal the need for shared collective norms and values to sustain preparedness and recovery initiatives, especially when institutional risk management systems fail. We argue that the context in which social capital is activated reflects the socio‐political dimensions of wildfires and sheds light on the overlooked efforts of rural communities to resist their relegated status as second‐class citizens.

1. INTRODUCTION

The prevalence of wildfires is rising globally, resulting in higher mortality, property loss, and environmental damage. It is widely acknowledged that the climate crisis will increase the frequency and severity of such events (United Nations Office for Disaster Risk Reduction, 2019).

Portugal, despite its small size, is one of the most affected countries in Southern Europe, with repeatedly the highest number of fires and the largest burned areas annually (Mira and Lourenço, 2019; Molina‐Terrén et al., 2019; San‐Miguel‐Ayanz et al., 2020). As Stephen Pyne (2021, p. 164) reflected: ‘Perhaps no country in the world had suffered proportionately with so many bad fires’.

The pattern of larger fires became more frequent in Portugal in the 1980s. In 2003 and 2005, its annual burned areas surpassed those of France, Greece, Italy, and Spain combined and civilian losses were also remarkably high (Silva et al., 2010). However, it was in 2017 that Portugal experienced its first deadly fires, during an exceptionally severe fire season. The Pedrógão Grande fire in June and the October mega‐fires set records for victims (116 deaths), forest damage (more than 500,000 hectares burned), and economic losses (500 companies, 4,500 jobs lost, and destroyed houses and infrastructure) (Viegas et al., 2019). The latter event was different and the first of its kind in Portugal and Southern Europe (Guerreiro et al., 2018).

According to Paulo Mateus and Paulo Fernandes (2014, p. 8), Portuguese fire risk management policies have been ‘reactive, inconsistent and short‐sighted’. Those implemented so far ‘have not been enough and … populations are increasingly vulnerable to both social and economic aspects’ (Rodrigues et al., 2022, p. 3). Additionally, there is a tendency to respond immediately to catastrophic phenomena by creating and restructuring programmes, laws, and institutional frameworks. In contrast to Australia, Canada, New Zealand, and the United States, fire suppression and emergency response in European countries have been prioritised while dismissing structural causes, such as land ownership, forest and land management, and social cohesion (Mateus and Fernandes, 2014; Moreira et al., 2020; Tedim et al., 2020).

This wildfire paradox, as it is called, worsens when fire‐related responsibilities are divided among different governments and agencies. This interventionist and top‐down approach, which relies on coercive and legal tools, neglects the socio‐ecological aspects of wildfires. Efforts to enhance wildfire resilience should, therefore, take account of the interconnectedness of fire, landscape, climate, and communities' values, attitudes, and behaviours (Tedim et al., 2021, p. 236).

Some authors have highlighted the significance of social structures alongside physical ones in the development of disaster management projects and policies. Understanding how communities cope with risks and mitigate them is equally essential, especially when existing official responses are inadequate or insufficient (Adger, 2003; Dynes, 2005; Cutter et al., 2008; Aldrich, 2012). Recent research has shown that social capital positively affects community resilience to environmental threats and disasters, underscoring the role of social ties and networks in fostering adaptive capacities (Carrasco, Ochiai, and Tang, 2024; Panahi and Moayerian, 2025; Zhao et al., 2025).

Mediterranean wildfire risk landscapes have been less examined with respect to social structures and dynamics (Górriz‐Mifsud, Burns, and Marini Govigli, 2019, p. 120), particularly the role of social capital in wildfire resilience (Zhao et al., 2025). The Portuguese context also lacks a deeper understanding of the social and cultural dimensions of fires, as well as qualitative approaches to grasp the phenomenon fully (Sousa et al., 2022).

The concept of social capital is not new and has deep historical roots across various disciplines; in fact, the term has experienced a revival in the literature (Portes, 1998; Portes and Landolt, 2000; Adam and Rončević, 2003; Meyer, 2018). Nevertheless, concerns about the concept should be raised and considered, as it is highly context‐specific, with numerous phenotypic applications, and there is confusion regarding its measurement (Adam and Rončević, 2003; Baycan and Öner, 2023). Consequently, this study aims to expand knowledge in the disaster literature by offering a sociological and critical approach to building a coherent social capital research programme applied to resilience, enabling its replication in other geographical settings and events.

Drawing on research conducted in two rural communities in central Portugal, we assess how and under what conditions social capital influences communities' ability to prevent, prepare for, and recover from wildfires. The article is structured as follows. The subsequent section offers a brief overview of the literature on social capital, disasters, and wildfire resilience, followed by a description of the research methodology used and an overview of the case study communities. The key findings are presented next, followed by an analysis and discussion. We conclude by providing new insights for future research on this topic, as well as policy implications for resilience‐building strategies within fire risk management in Portugal.

2. UNDERSTANDING WILDFIRE RESILIENCE THROUGH SOCIAL CAPITAL

The growing interest in social capital in the disaster research literature has sparked several conceptual discussions of the term (Meyer, 2018; Tierney, 2019, pp. 175–182). Numerous reviews exist, including Pierre Bourdieu's (1980) well‐known definition and James Coleman's (1990) perspective. For Bourdieu (1980, p. 2), social capital is ‘the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition’, while for Coleman (1990, p. 302), social capital is defined by its function: ‘Like other forms of capital, social capital is productive, making possible the achievement of certain ends that would not be attainable in its absence’. Meanwhile, political scientists have introduced a conceptual twist, in particular Robert Putnam, whose ideas are prominent; some have even called him the ‘patron saint of contemporary social capitalists’ (Adam and Rončević, 2003, p. 156). In Bowling Alone: The Collapse and Revival of American Community, Putnam (2000, p. 20) asserts that ‘social capital refers to connections among individuals – social networks and the norms of reciprocity and trustworthiness that arise from them’.

Despite the different proposals, there is a shared view that ‘social capital stands for the ability of actors to secure benefits by virtue of membership in social networks or other social structures’ (Portes, 1998, p. 6). In relation to disaster resilience, Russell Dynes (1969, 2005, 2006) was the first to argue that social capital explains why some communities effectively withstand catastrophes: ‘When new threats appear … we often miss the effectiveness of individual communities in addressing these threats’ (Dynes, 2006, p. 1). Aligned with this viewpoint, Daniel Aldrich (2012) demonstrated a strong correlation between high levels of social capital and rapid recovery. Using the 1923 earthquake in Kanto, Japan, as a case study, Aldrich verified a strong correlation between high levels of social capital and rapid recovery. In neighbourhoods of Tokyo, residents built informal and formal networks through repeated meetings, activities, and protests, establishing expectations of reciprocity and assistance. Over time, they worked together to rebuild their community (Aldrich, 2012, p. 411). A similar pattern occurred in the aftermath of Hurricane Katrina in 2005, when the members of Village de l'Est in New Orleans, Louisiana, US, ‘did everything they could do to maintain their connections’, in contrast to other affected neighbourhoods (Aldrich, 2012, pp. 412–413).

Overall, a strong body of evidence has concluded that social capital provides financial (such as loans and gifts for property repair) and non‐financial (such as search and rescue, debris removal, childcare during recovery, emotional support, sheltering, and information) resources (Aldrich and Meyer, 2015, p. 259), which enhance the capacities of communities when facing catastrophes. Exploring the social fabric that underpins the disaster context can help us to see beyond the destruction of physical capital (Dynes, 2005, p. 23).

3. SOCIAL CAPITAL'S CONTRIBUTION TO WILDFIRE RESILIENCE

Some disaster scholars have assessed how social capital shapes community resilience to wildfires. Notably, Yoko Akama, Susan Chaplin, and Peter Fairbrother (2014) studied four Australian fire‐prone communities, examining how the three types of social capital, bonding, bridging, and linking, influence collective action to build adaptive capacity.

Bonding social capital refers to the ties established between individuals who have similar socio‐demographic characteristics (or social identity), such as age, ethnic group, and class (Putnam, 2000, p. 30; Szreter and Woolcock, 2004, pp. 654–655); bridging social capital is the connections between individuals with different backgrounds and characteristics; and linking social capital, only advanced in recent years, describes relationships with people in positions of authority or economic power, that is, representatives of public and private institutions. Bridging is horizontal, joining actors with similar social positions, while linking social capital is vertical, connecting individuals based on power differences (Grootaert et al., 2004, p. 4).

As noted by Akama, Chaplin, and Fairbrother (2014, p. 4), wildfire research overlooks the elderly and newly arrived residents who experience self‐isolation in fire risk communities. Social ties and networks can tackle self‐isolation and facilitate information exchange (Akama, Chaplin, and Fairbrother, 2014, p. 4). The range of formal and informal community ties can foster information flow, serve as a local knowledge repository, and create interdependent connections within and beyond the community, providing access to advice, services, support, and resources (Akama, Chaplin, and Fairbrother, 2014, p. 9).

Adaptation and mitigation strategies are sustained, however, when there is a ‘balance between bridging, bonding and linking social capital in a social system’ (Pelling and High, 2005, p. 310). To illustrate, a recent study conducted in Greece (Wardropper, Sparks, and Hovardas, 2025) revealed that bonding social capital alone cannot build community resilience to fires, and that bridging and linking social capital are crucial for long‐term adaptation strategies. The same has been found in collective initiatives for forest landscape management in rural Italy (Paletto, Ferretti, and De Meo, 2012). Moreover, factors such as depopulation and migration can weaken collective action, highlighting the need for external connections.

Therefore, the common belief that the utility of social capital lies in network density or closure is questionable. Mark Granovetter's (1973) work on ‘weak ties’ highlights their importance in accessing information, ideas, and influences that would otherwise be unavailable. Those linkages are ‘bridges’ in an individual's network once they ‘create more, and shorter, paths’ (Granovetter, 1973, pp. 1361–1364).

Tara McGee and Stefanie Russel (2003) demonstrated the role of social networks in wildfire preparedness in a high fire risk rural community in the state of Victoria, Australia. In the context of the US, Menka Bihari and Robert Ryan (2012) found that residents who perceived higher social capital, such as social cohesion, engaged in proactive planning and collaborated with neighbours and agencies to mitigate fire risk.

In addition, what is common across these different social and geographical settings is the impact of place attachment (or topophilia) ¹ on community cohesion and action. Once it entails ‘great personal significance’, individuals foster close relations with neighbours and are more involved in local associations and activities (Jakes et al., 2007, p. 194; Bihari and Ryan, 2012, p. 259). Thus, understanding the meanings attached to sense of place helps us to understand individual and collective inertia or transformative change (Masterson et al., 2017).

Disaster literature recognises the importance of social networks during the recovery phase, especially when government and official responses are inadequate to meet people's needs and address recovery losses (Meyer, 2018, p. 270). For instance, it has been found that some volunteer fire groups in Catalonia (Spain), Greece, and Portugal emerged in response to severe wildfires and the lack of official support. The shared experience triggered grassroots initiatives (forest fire suppression, preparedness activities, and awareness actions) to prevent future events; however, the developed activities were limited, as much as the community's access to resources (Górriz‐Mifsud, Burns, and Marini Govigli, 2019).

Effective land management strategies and collaborative planning must be developed, therefore, under trusting relationships between governmental agencies and communities in the wake of a wildfire (Sharp et al., 2013, p. 11). The communication processes—particularly how individuals assess trustworthiness based on beliefs in the trustee's ability (such as knowledge, skills, and competencies), benevolence (the trustor's belief that the trustee will act in their best interest), and integrity (the trustor's perception of attitudes aligning with shared values and norms)—as well as the timing of recovery efforts (when people might feel overwhelmed and exhausted) all influence the outcomes of the post‐wildfire phase (Sharp et al., 2013, p. 18).

4. METHODOLOGY

The two research questions that guided our study were as follows:

• Do collective dynamics in Portuguese communities contribute to understanding wildfire resilience?

• What are the necessary conditions under which social capital shapes wildfire preparedness and recovery in Portuguese rural communities?

To answer them, we adopted Nan Lin's (2001b) social capital theory as a conceptual framework, which is grounded in social network theory. The advantage of the latter lies in its ability to connect micro and macro levels of analysis, enabling the operationalisation and measurement of social capital, and reducing its controversial aspects (Adam and Rončević, 2003, pp. 167–168; Portugal, 2007, p. 20). Network analysis theory avoids tautological assumptions and simplifies the transition between macro and micro perspectives by clarifying the processes through which social capital is constructed and mobilised (Portugal, 2007, p. 20).

In Lin (2001a, p. 29), social capital has three components: resources; social structures; and actions. The actions involved are investment, access and mobilisation, and returns. Investment depends on structural factors and individual position in the network, which influence social capital investment. Access and mobilisation relate to a person's location within the network and the use of contacts. Returns include instrumental benefits such as wealth and power, and expressive benefits such as life satisfaction and health (Lin, 2001b, pp. 243–249).

According to Brian MacGillivray's (2018, p. 117) critical viewpoint, social capital can potentially undermine resilience. The aspects of the ‘dark side’ of social capital, such as the exclusionary nature of bonding social capital, the nature of reciprocity, unresponsive linking capital (when political rebuilding aims are disconnected from the conditions and needs of local communities), and the conservative character of social capital, can impact communities' resilience to hazards (MacGillivray, 2018, p. 119). Therefore, not only the structure counts, but also the content of networks offers greater value in theorising and analysing the social dimensions of resilience.

Among the network's embedded resources (network node attributes) that shape resilience are social norms (views and values), mental models of risk, which include shared beliefs (related to hazards), experiences, collective memories, and shared expectations of action. Even if the community accumulates resources within social networks, they still might remain dormant. Consequently, context matters and social capital is rarely used for collective action unless there are shared beliefs and resources within a community that pertain to its ability to achieve a specific goal (MacGillivray, 2018, p. 122).

Moreover, networks fulfil different functions within a community. Both generic networks (such as information sharing) and specific ones (such as land sharing) contribute to resilience. Contrary to what is often alleged, social networks are not task‐independent structures with ‘a stock of generic sources to be deployed’ (MacGillivray, 2018, p. 122). Rather, each one requires tasks, resources, and forms of collective action. Hence, the arrangement depicted in Figure 1 was constructed.

FIGURE 1.

Modelling social capital. Source: authors, adapted from Lin (2001b).

The first dimension shows the preconditions for social capital, including factors that enable or hinder investment in it. The second indicates social capital elements, while the third lists possible outcomes and returns (Lin, 2001b, p. 245). The transition between the first and the second dimensions accounts for the inequalities of social capital formation. The social structure, or the ‘geography of social capital’, and the content of social networks influence opportunities to build and sustain social capital (MacGillivray, 2018, p. 123). This helps us to understand ‘patterns of differential distributions for social resources that are embedded, accessed, or mobilised’ (Lin, 2001b, p. 246).

The second dimension, mobilisation, concerns access to social capital and its utilisation. This enables consideration of action choices, taking structural factors into account and the resulting impacts. The greater the level of access to embedded resources, the higher the chances of social capital being mobilised for intentional actions.

Lastly, the second dimension, together with the third, relate to how social capital is transformed into returns and benefits. In the context of wildfire resilience, the specific gains can be linked to an individual's access to key resources, such as shelter, food assistance, help with land clearing tasks, and emotional support. Through their connections, they will be better prepared and more likely to recover from wildfires.

Another key concept is the social network, which Pierre Mercklé (2004, p. 4) defines as ‘a set of social units and of the relationships that these social units maintain with one another, directly or indirectly, through relational chains and paths of varying length’. Connections between social units encompass a range of interactions, including monetary transactions, information sharing, and the exchange of goods and services. These can be face to face or remote, ongoing or occasional (Mercklé, 2004, p. 4). Social networks can be assessed by querying: Who? What? How? (see Figure 2). First, who are the individuals involved? Second, what type of content or resources are exchanged? Third, which regulatory norms enable their functioning (Portugal, 2006, p. 140). The nodes representing network elements and their relationships with the ego are identified. The ties among these nodes, which describe their relationships, include a range of characteristics (Portugal, 2007, p. 24).

FIGURE 2.

Operationalisation of a social network. Source: authors, adapted from Portugal (2007).

Accessed resources can be divided into two main types: instrumental resources, which include support, shelter, increased risk perception, advice, and warnings; and expressive resources, which cover emotional support, affection, and sociability.

As for ‘how’, this concerns the social governing networks. Understanding how they function and their nature (that is, reciprocity, expectations, obligation, memories, beliefs, authority, etcetera) is important because it enables the creation and maintenance of ties (Portugal, 2006, p. 145).

Stephen P. Borgatti, Candace Jones, and Martin G. Everett (1998) proposed standard ego‐network measures (such as size/degree, density, heterogeneity, composition quality, effective size, closeness, constraint, and betweenness) and their relation to social capital (positive or negative) to comprehend when a community's wildfire resilience is improved.

It was important, therefore, to specify the characteristics/morphology of social networks (ego‐networks) that represent positive social capital. Positive social capital involves ‘knows or likes’, whereas negative relations are ‘hates or is not speaking to’ (Borgatti, Jones, and Everett, 1998, p. 30). Table 1 demonstrates how we applied social network characteristics in relation to social capital.

TABLE 1.

Standard ego‐network measures.

Name	Description	Relation to social capital
Size/degree	‘Number of alters that an ego is directly connected to, possibly weighted by strength of tie’ (Borgatti, Jones, and Everett, 1998, p. 30).	Positive. More connections mean a high chance of getting the needed resources, in order to be more prepared or to recover faster from wildfires.
Density	‘The proportion of pairs of alters that are connected’ (Borgatti, Jones, and Everett, 1998, p. 30).	Negative. If the alters are tied to each other, the richness of accessible resources is rather limited. Dense networks are more constraining (Burt, 1997).
Heterogeneity	Diversity of alters regarding, inter alia, sex, age, race, occupation, and talents.	Positive (if only there is no conflict with compositional quality).
Compositional quality	‘The number of alters with high levels of needed characteristics (e.g., total wealth or power or expertise or generosity of alters)’ (Borgatti, Jones, and Everett, 1998, p. 30).	Positive. The more useful contacts that the ego can access, the more social capital.
Effective size	‘The number of alters, weighted by strength of tie, that an ego is directly connected to, minus a “redundancy” factor’ (Borgatti, Jones, and Everett, 1998, p. 31).	Positive. If the individual has ties with different network positions, the potential information and control benefits are higher.
Constraint	Amount of ego investment in a single node/alter.	Negative. The more constrained the actor, the fewer opportunities to be wildfire‐resilient.
Closeness	The distance between the ego and all others in a network.	Negative. ‘Cohesive contacts – contacts strongly connected to each other – are likely to have similar information and therefore provide redundant information benefits (access, timing, and referrals)’ (Burt, 1997, p. 340).
Betweenness	The number of times that the ego creates a bridge between otherwise disconnected actors. Or the extent of an individual's control and coordination of the network.	Positive. If an individual has high betweenness, they link actors who were unconnected, paving the way to information and control benefits.

Source: authors, adapted from Borgatti, Jones, and Everett (1998, pp. 30–31).

Frane Adam and Borut Rončević (2003) argued that to keep the concept scientifically relevant, research should focus on social capital as a resource within social relations in egocentric networks, which is the most common measurement procedure in social network analysis (Portugal, 2006, pp. 153–154). The method can be defined as a ‘collection of individuals who know and interact with a particular target’ (individual or couple) (Milardo, 1988, p. 20). This technique elicits a list of ties from the ego, and the relationships between them are identified (Lin, 2001a, p. 16). The advantages consist of pinpointing content areas, as naming items, and ‘the mapping of ego‐network locations and characteristics as well as social resources embedded in the ego‐network’ (Lin, 2001a, p. 15).

To implement this methodological approach, semi‐structured interviews were chosen as the main data collection method, since the topic has not been explored in Portugal and qualitative studies on wildfire resilience and social capital are still underrepresented (Blanchet, 1985, p. 86; Lemieux and Ouimet, 2012, pp. 45–48; Sousa et al., 2022, p. 1115; Zhao et al., 2025). Interviews are a common tool in the social sciences, offering participants greater freedom to express their ideas and beliefs in detail, in contrast to surveys. In addition, the semi‐structured interview provides a shortcut to the norms and values systems within social and cultural studied contexts, frames of reference, and interpretations of situations and experiences, thereby capturing the meaning that individuals give to their emotions, behaviour, and lived experiences (Portugal, 2006, pp. 160–161; Atkinson, 2017, pp. 70–74).

The interview guide included open‐ended questions related to: sociodemographic aspects; place attachment; type of involvement in the community; past experiences of wildfires; mitigation plans and strategies; and perspectives on local and national fire risk management institutions. There were also directed questions to map an egocentric network in a wildfire scenario.

As a complementary method to in‐depth interviewing, the participant observation technique was also employed to understand and contextualise local interactions and social rituals. As Randal Collins (2004, 2020) has shown, social life is observable through the ordinary, trivial, and mundane aspects of everyday life and interactions.

5. CASE STUDY COMMUNITIES

Drawing on Charles Ragin and Howard Becker's (1992) theory of case study research, we aimed to comprehend and compare the impact of social capital on a community's preparedness for and recovery from wildfires. For this, two regions in central Portugal (União de Freguesias de Figueiró dos Vinhos e Bairradas and Ferraria de São João) were selected based on predefined criteria, such as susceptibility to wildfires, having experienced the 2017 wildfires, and participation in prevention and mitigation governmental programmes such as ‘Safe Village, People Safe’, initially created in 2006–07 but improved in 2018 in response to the 2017 mega‐fires (Associação Safe Communities Portugal, 2018).

After a few initial communications and some negative answers, we mobilised the personal networks of the participants, which helped us to establish contact with potential gatekeepers in fire risk communities. Additionally, geographical proximity to the researchers and the fact that the central region of the country is one of the areas most severely affected by fires (Molina‐Terrén et al., 2019) influenced the selection of the chosen case study communities.

We conducted 10 interviews in União de Freguesias de Figueiró dos Vinhos e Bairradas with the assistance of a gatekeeper who holds a municipal position and is well acquainted with the region and its inhabitants and knowledgeable about fire‐related impacts among the population. He first contacted the potential interviewees, and then we called them to schedule an interview. The gatekeeper played a crucial role in building rapport and gaining trust from the very first moment. In Ferraria de São João, access to the field was not fully facilitated by the local gatekeeper, a village resident. Recruitment of interviewees proved particularly challenging, especially of male participants. The majority of the village's residents are elderly and socially isolated, often expressing a degree of distrust towards unfamiliar people. Ultimately, eight people opted to participate in the study, but they demonstrated a willingness to engage with the research, and trust was gradually established over the course of the researchers' stay. Despite the recruitment issue, we ensured that the sample had a gender balance, age diversity, varied types of community involvement and positions, and representation of vulnerable groups (such as the elderly and persons with physical/mental health issues).

In total, 18 interviews were conducted in Figueiró dos Vinhos and Ferraria de São João village in the parish of Cumeeira (see Figure 3). Snowball sampling was employed owing to the exploratory nature of the research in Portugal. The fieldwork took two months, starting in January and ending in February of 2023.

FIGURE 3.

Location of the case study communities. Source: authors.

After the fieldwork, the interviews were transcribed and analysed using a coding system within MAXQDA software. The coding system was constructed deductively, following the conceptual framework, research questions, hypotheses, and key variables that emerged during analysis.

6. FINDINGS

In recent decades, the older generation in Portugal has outnumbered the younger population (Nazareth, 2009, p. 29). Ageing has persisted, but with geographical differences: the coastal zones are more populated by younger people (aged 14 or younger), whereas in inland regions, the population has been declining and ageing since the 1970s (Nazareth, 2009, p. 39; Rodrigues, 2018, p. 12). Penela and Figueiró dos Vinhos municipalities are among the areas with the most elderly residents (Nazareth, 2009, p. 50), which helps to explain why the average age of the sample is 56.8 years, with ages ranging from 28‐74 years. Nearly 40 per cent of the people in the sample are retired, but previous occupations were also considered. The number of women working as housekeepers and manufacturing labourers exceeds that of men. This pattern is not new, as men generally hold higher‐status jobs than women, whose careers are often considered to be ‘feminine’, reflecting sexual labour segregation (Coelho and Ferreira, 2018). Women also frequently hold two jobs in care‐related fields (farming, domestic labour, eldercare, etc.) even after retirement.

Ten of the interviewees had lived in their communities for more than 20 years, whereas six had been present for between 10 and 20 years. The length of residence affects an individual's social capital. It has been shown that newcomers' networks are relatively limited as compared to those of people who have lived in an area for at least 10 years. The newcomers, or neo‐rural people as the literature refers to them, justified their migration to rural areas on the basis of family property, networks of relatives, and/or a search for an idyll—these motivations are also discussed by Elisabete Figueiredo, Maria Partalidou, and Stavriani Koutsou (2020, p. 134).

6.1. Being bodily present: involvement in the community

Regarding local and cultural norms, residents often assist their neighbours out of genuine goodwill and general trust. This kind of social solidarity is frequently seen in rural communities (Bernard, 2019; Dal Bello, Sacramento, and Marques, 2021, p. 27; Uyttewaal et al., 2023).

Helping and supporting vulnerable people, such as the elderly, is indeed one of the main features observed and heard. For instance, an older woman in Figueiró dos Vinhos actively engages with the community by helping her elderly neighbours, accompanying them, and being a close friend. The same is true in the village of Ferraria de São João, where Vitória Guedes, around 60 years old, recently widowed, and living alone, receives strong support from neighbours such as Teresa Domingues (71 years). Teresa, who acts as a social companion and helper, stated: ‘I have been helping her with everything she asks me to’. Another case in point was shared by a younger resident:

Those people who still live here end up helping and supporting the more vulnerable others. For example, the woman who had recently become a widow [her husband committed suicide], as well as all the neighbours and the people who stay here during the day.

Additionally, when neighbours have not seen each other for a while, they promptly try to reach out. Rosa Costa, 74, a retired smallholder, shared that her neighbours are very attentive to her and her husband: ‘They rarely relax; when they don't see me, they call immediately. If there's no answer in the morning, they try again in the afternoon’. The couple is elderly and faces health and financial challenges: Rosa's husband suffers from Parkinson's disease, mobility issues, and chronic asthmatic bronchitis. The close‐knit nature of rural life, which also applies to newcomers, is also common in other geographical locations such as areas in Northwestern Europe (Uyttewaal et al., 2023).

Two women and one man have links to trans‐local associations, namely in Vila Facaia, in the municipality of Pedrógão Grande. Anouk Halls has connections with the former (woman) leader of the Association of Victims of the Pedrógão Grande Fire, as well as with Miguel Marques (see below), and Alice Rodrigues is the president of a cultural group in Vila Facaia. Anouk Halls is also a member of the Friends' Association of Vale do Rio, whose aim is to improve older people's well‐being in the region.

Local projects and activities were commented on by Lia Silva (including cheese and bread workshops, a village's book, and cinema sessions) and Miguel Marques, whom we consider a hub in terms of his social capital, in Ferraria de São João. His connections facilitated access to the financial resources needed to spur these collective initiatives to revitalise the village. Some scholars argue that new residents, with their knowledge and experience, have contributed to making rural communities more resilient to fires (Reis, 2014; Dal Bello, Sacramento, and Marques, 2021, pp. 32–33).

Social integration indeed takes place through social networks. Rosa Costa whose house is in a hidden and risky location (surrounded by forest), has received help from her neighbours, preventing her from being in a life‐threatening situation, especially during a wildfire.

Lastly, village cohesion is partly defined by relationships with temporary residents—those who visit on weekends and holidays. Some interviewees described second‐home owners as emigrants who inherit houses or urban residents who buy and rebuild properties. Community members with diverse local ties may be more likely to socialise and form friendships with temporary residents.

6.2. Facing the inevitable threat: communities' perceptions and experiences of wildfires

Even six years after the major fires of 2017, people were still grappling with lingering trauma. A long‐term resident mentioned that when the next wildfire occurs, he plans to leave the community to avoid hearing the desperate screams that he recalls to this day. In the case of Dalila Antunes (70 years old), another long‐term resident, who developed post‐traumatic stress disorder after the mega‐fires of 2017, she reported that she is still unable to sleep well, as a ray of light triggers her memories of the wildfire. Following that intense event that almost took her life, Dalila sought psychiatric help offered by the local government (municipality).

Interviewees demonstrated a strong awareness of the danger of wildfire, often considering their own preventive measures and preparedness plans, whether individual or collective, such as land clearing and acquiring fire protection equipment like sprinklers, fire extinguishers and hoses, water storage containers, and power generators in case of an outage. After the 2017 fires, risk management agencies and organisations have been especially criticised for their lack of action so far. In an interview, Penélope Alves, 51 years old, remarked:

At this moment? If another wildfire occurs, it will likely be worse than the 2017 one.

Why do you say that?

There is nobody who cleans or cares. It is good when the government talks about precautions and cleaning the roadsides, the required distances, but afterwards, you see nothing done.

Another participant, Anouk Hals, 56 years old, added:

Wildfires are the greatest threat we face living here… I believe that because of politics and money, the danger will never disappear. Since the 2017 wildfires, nothing has changed, and that is serious. I think it is a very serious situation.

Why do you think so?

Many people have died, and there are more eucalyptuses [trees] than in 2017. There are new rules, but nobody does anything. Nothing has changed. On paper and in words, yes. But the reality, living here, watching the area, the village, our valley where I live every day, nothing has changed.

Newcomers in Ferraria de São João stated that the village's Protection Zone Project developed in collaboration with the community and supported by national and international partners using municipal and state funds, has somewhat enhanced their sense of security. Yet, they also shared the same views as those above regarding how the status quo has remained unchanged.

The Project created a safe and enclosed area, known as the ‘Village Protection Zone’ (VPZ), wherein hundreds of eucalyptus trees were removed and replaced with native, fire‐resistant species. The VPZ is managed by the Residents' Association, initially organised by Miguel Marques. Community efforts, including identifying landowners and encouraging them to sell their properties to the Association, was crucial to establishing the VPZ. As is the case in other Mediterranean areas, 98 per cent of Portugal's forest land is privately owned, and most of it is mismanaged and abandoned (Camargo and de Castro, 2018, p. 42; Uyttewaal et al., 2023, p. 5). As noted by Fantina Tedim et al. (2013, p. 93), in addition to climate conditions in Portugal, ‘the main driver for mega‐fires occurrence is landscape evolution and land use changes'.

In both case study communities, interviewees described the same extreme fire behaviour, especially the sounds. Eucalyptuses and every possible combustible material were devastated at an extreme rate of speed. A couple had their house destroyed, and a male landowner lost EUR 60,000, as well as his willingness to care for the forests.

In Ferraria de São João, the aftermath of the wildfire fostered a collective effervescence among residents, also driven by the valorisation of the local and well‐preserved Cork Oak trees that prevented the village's destruction. This resulted in group solidarity, and the pioneering national project to prevent future traumatic events. Moreover, the Residents' Association endeavoured to gain formal recognition in order to receive institutional support and funds. This collective effort can be understood as means of actively seeking citizenship or the right to have rights (Somers, 2008) or as a way of achieving ‘worth’ or grandeur in Luc Boltanski and Laurent Thévenot's (2006) terms.

Nevertheless, despite the success of the local initiative, residents of Ferraria de São João are experiencing challenges. Interviewees mentioned conflicts between the older residents and the neo‐rurals, revealing social and cultural divides. This fragmentation exposes the negative aspect of social capital here, as it weakens the social cohesion crucial for the VPZ's future sustainability. A proposed solution was to bring in an external mediator from local institutions.

6.3. The role of social networks in coping with fires

The unprecedented fires caught authorities and people unprepared. Still, they showed solidarity. For example, the residents of Ferraria de São João battled the wildfire together and saved their village. They noted that Cork Oak trees served as a natural barrier, helping to contain the fire. The emergency prompted people to act quickly, almost instinctively. In Lia Silva's words: ‘As a matter of fact, we coordinated very well, really. It was impressive’. It was the mutual support network that saved the village, as two long‐established female residents underlined: ‘This was what saved us… The residents… and especially the oldest’.

In Figueiró dos Vinhos, existing support networks and solidarity norms shaped the community's resilience during and after the 2017 wildfires. A former firefighter chief, now working for the local council, highlighted the community's sense of solidarity. During the fires, local bakeries, grocery stores, social institutions like ‘Santa Casa da Misericórdia’, and scouts collaborated, demonstrating the degree of close connection in the community. For example, Dalila Antunes, an elderly female community member mentioned earlier, received protection thanks to her strong social bonds (see Figure 4): her neighbour and friend, Celestino Sousa, prevented her from entering her burning house. After the fire was extinguished, her brother and sister‐in‐law offered her shelter and emotional support. Dalila's social network is diverse in composition, including both strong and weak ties, which encompass bonding, bridging, and linking connections. This varied social structure enhances her high level of wildfire resilience. The extension and number of bridging ties enable Dalila to reach many people whose social positions differ from hers, thereby increasing social distance (Granovetter, 1973).

FIGURE 4.

Dalila Antunes's social network. Notes: nodes and ties in yellow are family members, pink are friends, orange are neighbours, and green are acquaintances. EDP = Energias de Portugal (an electric utilities company in Portugal). Eng. Vítor = an engineer called Vítor. Source: authors.

Bonding social capital is also equally important in emergencies, as neighbours or family members are often the first responders, providing immediate lifesaving assistance ² .

When the fires broke out, those at risk had not received any official warning. Or, as a female firefighter noted, weather conditions were not taken seriously enough by risk management officials to issue proper warnings. Thus, what helped people to spread information across villages was indeed ties and connections. Interviewees whose networks were less dense or whose linkages were beneficial because of their knowledge of or expertise in wildfires became more aware sooner. We believe this should be a factor in evacuation planning, part of civil protection, given the inconsistencies and limitations in securing evacuees' needs and rights (Oliveira and de Mello, 2025).

From a network structure perspective, Rui Gonçalves, a retired lawyer, has diverse connections with specific attributes, such as forestry experts (see Figure 5). As Robert Milardo (1988) pointed out, some individuals are more socially active because of factors such as education level, occupation, income, and personal traits.

FIGURE 5.

Rui Gonçalves' social graph. Note: the graph shows fewer family members and more neighbourhood and friendship connections. The interviewee regarded the firefighter as a negative acquaintance, so he is positioned further away. Connections on the outer edge are less emotionally close to the ego. Source: authors.

Rui relied on a friend, Fernando Ribeiro (80 years old), who owned a local guesthouse. The interviewee asked him for shelter for his girlfriend, Júlia Alves. Although the guesthouse was completely full of evacuees from the town, Fernando offered his daughter's house as accommodation for Rui's girlfriend.

For Rui, this kind of solidarity is ‘of great beauty’, and it makes him emotional when recalling the experiences. ‘It is one of the things that I learned to like about the people of this land. People who, when things got tough… You could believe that someone will always come up to help you out’, he added.

Another essential node in Rui's network was his lawyer and former university friend, Tomás Martins, 65 years old. He was spending the holidays at Rui's place when the wildfire started, and his knowledge and experience supported Rui in weathering the difficult conditions until 05:00 on 18 June 2017, the day after the fire started.

As for the role of linking social capital (see Figure 6), Miguel Marques's experience highlights its significance. When Miguel's house was entirely engulfed in flames, he called the city's mayor for help. Minutes later, the mayor and the deputy mayor arrived in a fire engine, thereby helping to save the village.

FIGURE 6.

Miguel Marques' social graph. Notes: bonding, bridging, and linking social capital are well balanced. As Marissa King (2020, p. 136) points out, this type of network, expansionist, is linked with popularity, status, and power. ADAI = Association for the Development of Industrial Aerodynamics, linked to the University of Coimbra; AGIF = Agency for the Integrated Management of Rural Fires; MONTIS = Nature Conservation Association. Source: authors.

The response to the disaster in Ferraria de São João is undoubtedly rooted in the community's cohesion and successful inter‐organisational relationships following the wildfire. The consensus among residents about building a resilient community led to the securing of national and European funds, which made the VPZ possible. Miguel Marques's network was crucial in that process, as his bridging and linking social capital enabled access to institutional contacts and expert advice, thereby enhancing the village's adaptive capacities.

The level of preparedness has significantly improved in both communities as compared to the period before the tragedy. There is greater awareness of fire risk among community members, and their communication with neighbours has been strengthened. ‘They are more perceptive than they were… Now we are more prepared’, said Teresa Domingues, a long‐standing resident of Ferraria de São João.

In both communities, prevention is crucial for managing fire risks. However, maintaining and cleaning forests and surrounding areas can be a challenging task when property owners lack the financial resources to do so. This is exactly where the state must intervene, according to the interviewees.

Lastly, in both localities, a fire‐related network exists that facilitates the exchange of information and advice through connections, particularly between permanent residents and second‐home owners.

7. DISCUSSION AND CONCLUSION

This research analysed how people connect and with whom they do so during a community's fire risk coping processes, which affect resilience‐building efforts. The findings suggest that social networks are mobilised for various purposes, including rescue, shelter, protection of belongings, knowledge transfer, and food aid. Both formal and informal ties in small rural communities hold great value for an individual's preparedness and recovery, which benefits the collective where they live and with which they are engaged. Especially when the length of residency is long, connections are broader and more expansive, promoting, as a result, initiatives to mitigate fire risk.

A community whose residents have positive social capital in terms of constraints, closeness, betweenness, effective size, compositional quality, and heterogeneity is more resilient to fire than those villages where homophilous ego networks dominate. Ferraria de São João exemplifies how individuals' social capital, particularly that of neo‐rural residents, can contribute to the collective good, illustrating how the emergency is collectively experienced and how recovery efforts can be built upon afterwards.

However, we have detailed how the structure of social networks and the characteristics of their nodes influence the quality of social capital, whether positive or negative. The social position of the interviewees helps to explain the structure of their networks and the benefits they can receive in a wildfire scenario, including knowledge, advice, fire prevention supplies, mental health support, and land clearing services. And here lies one of the key contributions of our study: social capital is not inherent in individuals and communities; rather, it is a construct that depends on broader social, historical, and economic trajectories, which differ across geographical levels and scales. As applied in other studies, ‘institutions of governance, rights, and power’ are absent from reflections that make social capital perform in the first place (Somers, 2008, pp. 234–235).

We argue that wildfire resilience is enhanced only when a system of values and meanings is firmly embedded, renewed, and maintained through interaction rituals (such as meetups, small talks, favours, and assemblies), mutual support, place attachment, and collective action in a community setting. Culturally normative and contextual practices are not explored enough when discussing background variables that influence disaster resilience.

Consequently, studies focusing on Australia, Canada, New Zealand, and the US would not be able to inform fully a research project and wildfire risk management policies in European countries like Portugal. This is despite some similarities in terms of the Mediterranean‐type climate and the rising trend of individualising responsibilities in wildfire risk governance (Silva et al., 2010; Asiyanbi and Davidsen, 2023).

To illustrate the argument, we do not confirm the research hypothesis that a high level of trustworthiness in risk management organisations correlates with better resilience. In fact, the opposite was the case here. Aside from firefighters' work, local and national risk management agencies and organisations have faced criticism for their handling of the fires in 2017. When asked to evaluate risk management efforts in the years since, the most commonly used terms by interviewees were ‘abandonment’ and ‘negligence’. This reflects Portugal's political and economic landscape, including persistent ad hoc policies, a dominant market presence, rural abandonment, and fractured land management planning, all of which have contributed to the political isolation of rural communities (Mateus and Fernandes, 2014; Beighley and Hyde, 2018; Camargo and de Castro, 2018).

Following the unprecedented events of 2017, government and fire agencies were committed to transforming the wildfire management paradigm, aiming to create a more prepared society (Moreira et al., 2020). Nonetheless, after six years, people feel isolated and abandoned. Bottom‐up projects that emerged before formal assistance from and initiatives of the state, such as the ZPA, require sustained institutional and financial support to continue. The community alone cannot address structural issues, such as current laws and bureaucratic systems governing forest ownership and landscape forest management. Furthermore, village conflicts and unresponsive ties exist and could undermine top‐down projects like ‘Safe Village, People Safe’. The social organisation of communities should be understood before developing such programmes (Observatório Técnico Independente et al., 2020).

Margaret Somers (2008, p. 112) highlighted that social capital can be easily instrumentalised by ideological agendas, prompting individuals to depend on their social connections when the state neglects its duty to protect rights (‘marketisation of the social’). As states influenced by market fundamentalism fail to provide essential human services fairly, people are forced to rely on personal ties. The dominance of the paper and pulp industry in the Portuguese economy, which depends on pine and eucalyptus trees for its main raw materials, has been shaping the political culture of forests, agriculture, and wildfires (Mateus and Fernandes, 2014; Camargo and de Castro, 2018; Fernandes, Guiomar, and Rossa, 2019).

We oppose Putnam's (2000) social capital project that sees social relations as having or producing the economic value expected in a market exchange. It is not about commodifying relations but emphasising the non‐contractual arenas of social life (Somers, 2008, pp. 219–225). Therefore, in this research, ties and connections served as ‘buffers’ against the inaction of government and its institutions as well as their unpreparedness to deal with the wildfires of 2017. Thus, social networks emerged as a community's form of resistance to the impacts of market fundamentalism on their right to have rights. The historical and persistent marginalisation of their right to be protected degrades risk communities to second‐class citizenship.

To rethink social capital is to acknowledge the threat of neoliberal governance of risks and disasters. Following the thesis of Karl Polanyi (2001), the noncontractual relations of reciprocity, solidarity, and redistribution are essential for a strong rights‐based civil society, as they are necessary for the survival of democracy (Somers, 2008, p. 117). The arrangement shown in Figure 7 illustrates our argument.

FIGURE 7.

Rethinking social capital: the role of non‐contractual relations in the maintenance of democracy. Source: authors.

Our study has inherent limitations. It is based on case studies and, as such, it does not allow for generalisations. The selected regions are context‐specific owing to resource constraints (time and funding). Hence, future research should be conducted in other regions with distinct features and dynamics, such as the north of Portugal. Given that types of social capital and network configurations are not stable across time, it would be interesting to conduct a longitudinal study that compares and assesses the changing dynamics of social structures that influence resilience. The role of women in social integration and leadership within community networks also warrants further analysis, as fieldwork suggests that they are likely to serve as hubs or catalysts in networks. Additionally, interviewing second‐home owners, trans‐local hubs, and institutional actors could yield greater insights into this topic.

Finally, wildfire policies adopted by Southern European countries over the past decades, including those in Portugal, have neglected structural causes of fires, such as social, cultural, and historical accounts of fire risk landscapes (Tedim, 2018, p. 251; Uyttewaal et al., 2023). A social network analysis lens can provide alternative perspectives, enhance institutional approaches to wildfire management, and pave the way for community engagement in risk governance. For instance, informal warning and information sharing networks could be integrated into state‐led fire management strategies. Consequently, incorporating local community dynamics and needs is not only a crucial step towards a much‐needed paradigm shift in wildfire governance, but the survival of vulnerable territories and their rights‐bearing inhabitants depends on it.

CONFLICT OF INTEREST STATEMENT

The authors have no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

ETHICAL STATEMENT

Our research conduct was guided by informed consent, voluntary participation, confidentiality, no potential for harm, and respect for the participant's right to withdraw from the study without any justification or consequence. The names that appear in this article are fictitious in order to protect interviewees' actual identities. All of the interviewees were presented with and signed a document of informed consent, wherein the research aims and conditions were described.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.