Community engagement in the Transition movement: views and practices in Portuguese initiatives

ABSTRACT As the need to address climate change is ever more urgent, many have emphasised the importance of community-level responses. The Transition movement has advanced community-based action to increase resilience for over a decade and has expanded significantly. Thus, it is a critical setting for examining community engagement towards climate change in practice. Our study is based on 39 interviews with facilitators of Transition initiatives in Portugal, coupled with observational data, and is guided by two main research questions: how do Transition initiatives promote community engagement at the local level? What are the factors constraining or facilitating community engagement within Portuguese Transition initiatives? We identify several aspects of Transition’s constructions of community resilience and engagement that indicate ambivalence towards, or avoidance of, certain issues. They relate do agency, structure, power and inclusion, as well as to the modes of engagement and the communication practices of Transition initiatives. We argue that strategies for community engagement should be specific to social contexts rather than internationally uniform and be based on participatory approaches. Drawing on an extensive empirical analysis, the article contributes to theory building on the Transition movement beyond the Anglo-Saxon context and to the wider field of community-based environment initiatives.


Introduction
Collective action strategies at the community level have been proposed as one of the most appropriate scalar responses to global climate change (Hopkins 2008, North and Longhurst 2013, Aiken 2014) and there is a growing interest in the many ways communities are organising themselves to mitigate, resist and adapt locally Collura 2012, Aiken 2015). The Transition movement, which appeared in 2006 in the United Kingdom, has been identified as one of the most promising (environmental) social movements to emerge in the last few decades precisely due to its notable focus on the community level (Bridge 2010, Barr and Devine-Wright 2012, Bay 2013, Aiken 2014, Alloun and Alexander 2014, Power 2016. The movement posits that what it views as the largest challenges of our times, namely climate change, a peak of oil supply and the economic crisis, can all be effectively confronted through community-led responses (Hopkins 2013). The idea of "community" is thus at the heart of the Transition movement (Aiken 2012, Neal 2013, Dilley 2017 with initiatives all over the world being encouraged to engage local communities in transition processes (Transition Network 2017a).
As claimed by Naomi Klein (2014), climate change "changes everything", and demands transformative approaches to energy production and consumption, transports, food, housing, i.e. to the ways economies and societies are organised and run. Transition builds on the notion that by taking back control over meeting our basic needs at the local level, it is possible to stimulate "new enterprises, while also reducing our oil dependency and carbon emissions" (Hopkins 2013, p. 27). For Feola and Him (2016), the Transition movement represents the potential of a social movement to "create spaces of possibility for alternatives to mainstream, neoliberal economies" (p. 1). However, critical voices have stressed that in order to constitute an effective response to current challenges, the movement needs to overcome its apolitical/post-political nature (e.g. Trapese Collective 2008, Mason andWhitehead 2012, Kenis andMathijs 2014), as well as its passive approach to diversity and inclusivity (e.g. Connors andMcDonald 2011, Grossmann andCreamer 2016).
There is some evidence that the process of engaging local communities is influenced by context (Connors andMcDonald 2011, Seyfang andHaxeltine 2012), and that social contexts can hamper or constrain community engagement in the Transition movement (Connors and McDonald 2011). However, the vast majority of studies have been conducted in the U.K. (e.g. Haxeltine and Seyfang 2009, Bailey et al. 2010, Aiken 2012, Mason and Whitehead 2012, Felicetti 2013, Neal 2013) and/or have been focused on a single case study (e.g. Biddau et al. 2016, Dilley 2017, not allowing to ascertain how the movement is diffusing, transforming, migrating and embedding beyond the U.K. context. In addition, while previous studies have touched uponsometimes only incidentallythe issues of community engagement in Transition initiatives (Feola and Nunes 2013), there has been little detailed examination of how initiatives enact community engagement, which is likely to be key to their success.
This study extends the literature on the Transition movement by exploring views and practices of community engagement in local Transition initiatives in Portugal. It responds to calls for researchers to consider the relevance of local contextual factors in community-based initiatives (Botes and van Rensburg 2000, Feola and Nunes 2013, Collura and Christens 2015 and the need to geographically expand the analysis of the Transition movement beyond the Anglo-Saxon context (Mason andWhitehead 2012, Staggenborg andOgrodnik 2015). Two key questions underpin this study: how do Transition initiatives promote community engagement at the local level? What are the factors constraining or facilitating community engagement within Portuguese Transition initiatives? We posit that the practice of engagement of community members needs to be examined in detail and that analysing how Transition initiatives discursively construct and facilitate community engagement can reveal important insights into the strengths and weaknesses of the movement.
Our findings suggest that the Transition movement needs to develop a thorough reflection on and critical analysis of processes of community engagement. In line with extant literature, we argue that there is a need for recognition of the complexity inherent to the notion of community and engagementwhich includes contestation, conflict and exclusion (e.g. Smith 2013, Brodsky 2017)and that community engagement ought to be sensitive to power relations and structures. Moreover, we argue that strategies for community engagement should be specific to social contexts rather than internationally uniform and be based on participatory approaches.
The paper is structured as follows. After briefly introducing the Transition movement and the Portuguese context, we discuss the meanings of community and of community engagement, followed by a review of the literature on power and diversity issues in the Transition movement. The paper then moves to the empirical study by detailing its methodological aspects. The presentation and discussion of findings are organised into a set of themes that are critical to community engagement in the Transition movement.

The Transition movement and local initiatives
The international Transition Network defines its aims as "to inspire, encourage, connect, support and train communities world-wide as they self-organise around the Transition model" (Transition Network 2017a). The movement offers an example of steps that can be taken towards the creation of resilient communities (Barr and Devine-Wright 2012). It works on what may be called preventive resilience, as it gives communities the possibility to anticipate change and avoid poor outcomes by developing coping strategies (Hopkins 2011, Wilson 2012. The movement's goal is to ensure that communities do not collapse at the first sight of oil or food shortages and that they become able to adapt locally to climate change challenges (Hopkins 2008).
The Transition movement has grown rapidly over the last few years to over 1100 "Transition Initiatives" in 43 countries (Bridge 2010, Feola andHim 2016). In Portugal, implementation commenced in 2010; in July 2017, there were 18 initiatives registered in the Transition Network (Transition network 2017b) and a significant number of non-registered ones, including some highly active ones (e.g. Aveiro in Transition). A national platform was inaugurated in 2014 with the objective of representing members as well as bridging local initiatives with the Transition Network (Transição Portugal 2017). The Portuguese context is of interest for at least three reasons: the country's vulnerability to climate change; the traditional low level of engagement with environmental issues; and the recent economic crisis, which enhanced prior socio-economic differences. First, Portugal's geographical situation and climates make the country particularly vulnerable to climate change, with an expected increase in heat waves, droughts and other extreme weather occurrences (European Environmental Agency [EEA] 2016). Secondly, despite some changes in the last three decades (Kousis 1999, Matos 2012a, levels of citizen engagement with and support for environmental movements are generally low (Rodrigues 1995, Garcia et al. 2000. The Transition movement, however, has achieved a significant success in terms of the number of local initiatives (Transition Network 2017b, Transição Portugal 2017. Finally, the 2008 financial and economic crisis launched the country into a deep recession, which had impacts on critical sectors, such as employment (Yeh-Yun Lin et al. 2013). Besides, in comparison to the U.K. context, the Portuguese society has a long history of lower income and education levels (Eurostart 2017) and bigger socio-economic differences. Income and education inequalities are often negative influences on levels of citizen engagement (Putnam 2000, Oxendine 2004).
The Transition movement emerged at the peak of the economic crisis, at a time when community resilience would be critical, postulating the need to train communities to cope with various criseseconomic and environmental. A recent study has shown an increase in the number of alternative solidarity-based projects in Portugal in the last years, with transnational diffusion and the involvement of foreign nationals being important factors (Baumgarten 2017). However, there is a lack of empirical studies focusing on the specificities of the Transition movement in Portugal (Matos 2012b), namely concerning the practices and challenges of local Transition initiatives in promoting community engagement.

Meanings of community and community engagement
Whereas "transition" and "resilience" are key conceptual underpinnings of the Transition movement's approach, several scholars have noted that it is "community" that is its main driver (Aiken 2012, Bay 2013, Dilley 2017. Within Transition, communities are considered key agents of change (Coenen et al. 2012, Feola 2014. In its stated objectives, the movement shows a strong commitment to community approaches and invites people across the world to act as organised and engaged communities through Transition initiatives (Hopkins 2013). Hence, community engagement is crucial to "transition" and to the success of the movement (Hopkins 2013, Transition Network 2017a. "Community" is a non-univocal concept with a plurality of meanings. On the one hand, different meanings of "community" are likely to coexist for different people and even for the same person (Brodsky 2017). On the other hand, the term "'community' has both positive and negative connotations to all who consider themselves a part of one community or another" (Azzopardi 2011, p. 180). Smith (2013) argues that the notion of community encompasses a wide range of concepts such as trust, mutuality, commitment, and solidarity, as well as contestation, conflict, and exclusion. Others argue that "being a community" implies that community members undertake action to improve their lives and to solve common problems (Mannarini and Fedi 2009). In turn, Walker (2011) has looked at uses of the term "community" in carbon governance-related matters and identified six meanings: community as actor, community as scale, community as place, community as network, community as process, and community as identity. Thus, the concept of community can have multiple geographic, cultural, political, and/or symbolic definitions.
The Transition Network defines community as place, putting a strong focus on the idea of working together as a group based at a physical location (Aiken 2012). Transition initiatives (local chapters) are presented as the places "where a community-led process helps town/village/city/neighborhood become stronger and happier" (ECOLISE 2017).
Likewise, the term "community engagement" is theoretically and empirically contested (Dempsey 2010. Academic literature tends to be vague about what community engagement looks like in reality and the concept is often viewed differently by different players: for some, it is ethical practice and principles that underpin the way community engagement is enacted that are most important ( … ) for others it is the process, such as acting in a socially just and equitable manner in the way decisions are made; and for others it is about the impact or outcome achieved through community engagement. (Smith et al. 2017, p. 32) Theoretically, community engagement has often been defined as a mechanism or a process that allows for the reduction of social gaps and helps communities to efficiently obtain and manage resources (Daley andMarsiglia 2000, Azzopardi 2011). Besides, it entails the creation of opportunities for meaningful engagement (Campbell 2014) and is intrinsically linked to the concept of empowerment (Montero 2009). Community empowerment requires community members' engagement towards the development of skills and resources to control their life situations and to act in a compromised, aware and critical manner to address their needs and aspirations (Montero 2009). In these community processes, there is a need for greater attention to the role of context and culture in framing how people experience and respond to threats to their well-being and recognise their agency, power and resilience (Campbell and Burgess 2012).
The Transition movement encourages participation and collaboration across economic, political, social and cultural divides (Connors and McDonald 2011). However, Grossmann and Creamer (2016) found that the movement appears to be more effective in bonding activities, which tend to "reinforce exclusive identities and homogeneous groups", whereas it finds more barriers in bridging activities, which "encompass people across diverse social cleavages" (Putnam 2000, p. 22). Therefore, despite the Transition movement's intentions to actively engage local members (Feola 2014), studies show that local members' engagement in Transition initiatives tends to be low (Bhattacharya 2012, Feola 2014, Grossmann and Creamer 2016. 1 These studies suggest that there is a need for better understanding the factors that enable and limit Transition initiatives in terms of their ability to work with local communities and promote community engagement.

Power and diversity in the Transition movement
Literature on the Transition movement appears to converge in two interlinked critiques: the apolitical nature of the movement and its passive approach to inclusivity. A standard critique of the Transition movement has concerned its apolitical or post-political nature (Trapese Collective 2008, Aiken 2012, Neal 2013, Kenis and Mathijs 2014. This is based on the movement's philosophy of being for change and "not against" something (Bay 2013), maintaining a non-adversarial and non-confrontational position (Felicetti 2013). Avoiding direct confrontation of powerful interests (Connors and McDonald 2011, Aiken 2012, Kenis and Mathijs 2014), the Transition movement explicitly rejects the "us against them" discourse found in traditional environmental movements (Kenis and Mathijs 2014). This lack of politicisation has been attributed to an unrealistic analysis of power (Kenis and Mathijs 2014) and to difficulties in developing oppositional forms of political resistance (Mason and Whitehead 2012).
Whereas power and conflict issues are strongly avoided, communities are viewed in the Transition movement as having the ability to harmoniously design strategies leading to a full reconfiguration of habits, practices, and infrastructures towards transformative change (Aiken 2015). 2 The apolitical (or post-political) nature of the movement assumes relevance in the debate about the capability of the movement to contribute to social change, with some studies concluding that being more political would make the movement more effective (Trapese Collective 2008, Kenis and Mathijs 2014). 3 A number of scholars have argued that the movement suffers from an inability to expand much beyond the usual middle-and upper-middle-class, middle-aged, highly educated participants, post-materialist progressive white people, who have the time and privilege to engage in environmental activism and who are already environmentally conscious (Connors and McDonald 2011, Aiken 2012, Seyfang and Haxeltine 2012, Alloun and Alexander 2014, Kenis and Mathijs 2014, Grossmann and Creamer 2016. These findings are unsettling considering that promoting inclusivity and social justice is a principle guiding Transition's action (Transition Network, 2017d). Grossmann and Creamer (2016) maintained that, although diversity and inclusivity are fundamental to Transition philosophy, the movement takes a passive approach rather than an active one towards inclusivity and diversity. Besides, the movement's lack of ability to include diverse social groups has been explained by the lack of a critical approach to the role of community (Walker 2011, Aiken 2012) and by a strong tendency to idealise the notion of community, ignoring dimensions such as power, conflict and exclusion (Kenis and Mathijs 2014). In this respect, some have suggested that the movement needs to attenuate the spirit of localism (Mason and Whitehead 2012); pay more attention to power and social justice issues (Kenis and Mathijs 2014); and take an active approach to diversity and inclusivity (Grossmann and Creamer 2016).
The critiques enunciated above lead to the question of whether the Transition movement can promote community engagement without actively considering communities as internally complex and the relations between communities and other social and political entities/institutions as potentially problematic (Walker 2011. Additionally, questions arise on the role of contextspecific aspects of communities, their nature and relations.

Method
We conducted a qualitative study involving interviews, observation, and analysis of online websites and Facebook pages. In this paper, we focus especially on the semi-structured interviews with "Transitioners", coupled with observational data. A qualitative approach allows for the studying of a social phenomenon as a constructed entity, revealing the situational context of the Transition movement, its actors, theirs views, and their strategies to surpass difficulties.

Participants
All the initiatives registered in the international Transition Network in March 2013 were contacted and invited to participate in this study. In a universe of 19 initiatives, 16 accepted to participate (two of those were not included in this paper, as they are implemented in university contexts, rather than local communities). Transition initiatives in Portugal are spread throughout the country, and thus some are in rural contexts and others in cities.
In total, we conducted 39 interviews with Transitioners in 14 initiatives (2-3 per initiative). The term Transitioners is used here to refer to the founders and/or current facilitators of Transition initiatives. Participating Transitioners were aged approximately 25-50, 22 male, and 17 female. Interviews were conducted between September 2013 and January 2014.

Procedures
The interview script included questions related to five major themes: initiatives' emergence and development; relations with local communities; communication practices; partnerships with local, national and international actors; and political representations. Questions were formulated at a general level and phrased in an open-ended format to allow participants to express their opinions and feelings in an unconstrained manner. All interviews were tape-recorded, fully transcribed, and subjected to a thematic analysis inspired by grounded theory (Boeije 2002). The Nvivo software was used to support the thematic analysis. The process of data analysis began with open coding and was followed by selective coding guided by research questions. The analysis was discussed among all team members several times.
All interviews were conducted in Portuguese, the native language of the researchers and most of the participants (only one participant was not a native Portuguese speaker, although fluent in Portuguese and having lived in Portugal for over a decade). Interview excerpts were translated from Portuguese to English after data codification in Nvivo software. All authors participated in the process of translation by translating or revising the translated text. Researchers were attentive to cultural meanings and interpretations; by assuming both roles, the researchers/translators could clarify terms and concepts that would otherwise remain ambiguous or decontextualised (Tarozzi 2013). Researchers started by translating the text literally, word-by-word. Small adaptationsmainly in terms of grammatical and syntactical structureswere made later to improve readability (Birbilli 2000, Filep 2009).
In this paper, the main data source were the interviews, which were complemented with field notes from meetings and events organised by the movement. The latter included a national meeting for all the initiatives, courses on how to implement a Transition initiative, several thematic workshops, regular meetings of local initiatives, debates, and community shared-meals. Participation in these events allowed for a much closer observation of the movement, which fostered a broader analysis. Moreover, extended observation was carried out in two localities, one urban (Coimbra) and one rural (Aldeia das Amoreiras) by spending time with Transitioners and community members, and by attending multiple activities. Based on these various forms of (participant) observation, we wrote down our impressions as notes during and after ending field contact, and regularly discussed interpretations of what we had observed among team members. In this paper, those notes were used as complementary materials to interpret findings from the interviews. The study also included a large number of interviews of community members, but due to space limitations, we will only refer to a few key general aspects here.

Findings and discussion
Six complementary and interlinked themes will be discussed in this section. Theme one, "understandings of community", refers to the ways participants define community and engagement. Theme two, "diversity and inclusion", outlines diversity and inclusion-related challenges in Portuguese initiatives. Theme three, "agency, power and socio-political structures", explores participants' discourses about (individual and community) agency, power and social-political structures. Theme four, "messaging", shows how the avoidance of communication about climate change and the focus on positive messages can weaken the impact of movement. Theme five, "modes of engagement", discusses the modes of community engagement put in place at local initiatives and the potential of participatory approaches. Finally, theme six, "perceived constraints in community engagement", analyses obstacles to the process of community engagement.

Understandings of community
In all the interviews that we conducted, we identified an intense focus on the idea of community as a starting point. Hence, it seems pertinent to explore how Transitioners (re)construct the meanings of "community" and "community engagement" in the Portuguese context. As found by Aiken (2012), participants in our study perceived community as both the purpose of transition as well the means by which to get there. In other words, community engagement and community building were viewed as both paths to resilience and as goals per se. In both rural and urban initiatives, Transitioners valued those aspects highly, based on the idea that only through social networks within the community will resilience be possible: "My work is a communitarian work, which seeks to change some of the ways we organise ourselves in society, so that we, as a community, may be more resilient" (Gabriela). Multiple Transitioners referred to community organising as the main reason for being involved in the movement. The greatest goal of many participants was to promote community engagement, and they viewed themselves mainly as promoters of such engagement: "It is what moves me ( … ) I am interested in encouraging people to collectively do practical things" (Francisco). That perspective was also shared by Sandra, who stressed that the focus of Transition initiatives was not on environmental matters, but on organising a community: "I don't see Transition movements like that. I see Transition movements more as initiators or facilitators of the organisation of communities ( … ) towards meeting their own needs." Similarly, the majority of participants in our study considered environmental issues to be opportunities for constructive and positive change in the community.
In short, interviewed Transitioners place community engagement at the core of their personal and collective mission. They present it as the main driver of their action and as a central purpose to be fulfilled. The engagement of community members is viewed as a key element in the path to stronger, more resilient communities, where people have more control over their lives. Community engagement was also associated with attaching importance to social and cultural values that have been eroded or lost. These views seem to be very much aligned with the discourse of key figures in the Transition movement (e.g. Hopkins 2013) and reproduce closely the discourse of the Transition Network, which presents community engagement as a fundamental principle of the movement (Transition Network 2017c).

Diversity and inclusion
Given the centrality of the idea of community to the Transition movement, and the key aim of engaging communities, it is important to delve further into the ways community is defined and circumscribed. In the discourse of the Transition movement and of the participants in this study, community is defined in terms of locality (e.g. city, village, and neighbourhood). First of all, the names attributed to local initiatives reveal a perception of community based on the locale (e.g. Linda-a-Velha in Transition, São Brás in Transition). Second, in all the conducted interviews, we have found several references to the "locale" and the "neighbourhood", with some explicitly considering that "Transition should be something done from the apartment building, the neighbourhood, [thus] very local" (David). The community was considered the most effective level of action: "It makes much sense to act from here [the local level] as a starting point. From ourselves. From our house, our neighbourhood, our city. The global materialises at the local level, so it is a matter of scale" (Gabriela).
Looking at a community as a locality does not tell us much about how the make-up of the said community and the differences between their members are considered. Our study suggests that Transitioners' views on community (and, therefore, on the people targeted in community engagement) build on a naïve, consensual imaginary that overlooks difference and conflict (including their productive dimensions). Thinking the concept of community through calls for consideration of both the interests and the perspectives of the various groups within communities, class struggles within those spaces, gender issues, and so on. However, those issues tend to be absent in the Portuguese initiatives as in other contexts (Kenis and Mathijs 2014, Grossmann and Creamer 2016).
Although one Transition initiative included in this study has explicitly defined itself as oriented to working with disadvantaged groups (Coimbra in Transition), overall findings point to a lack of diversity and of representativeness of the variety of community members in Transition initiatives. For instance, the level of schooling among Transitioners was much higher than the Portuguese population's average (all had university degrees, including several with master and PhD degrees). Besides, the overwhelming majority of the 39 Transition participants that we interviewed described a strong commitment to environmental issues, several past experiences of engagement in social movements (e.g. "Indignados" and "Zeitgeist") and in environmental organisations (e.g. GAIA). As a member of a Transition initiative put it, engagement in some of those projects was replaced by involvement in the Transition movement. Another revealing aspect is that many participants reported professional and/or academic experiences involving a strong engagement with environmental issues. On the one hand, it seems clear that academic and professional choices are themselves a consequence of previous interest in environmental issues. On the other hand, the academic milieu appears to have further stimulated the interest in environmental issues. For example, one Transitioner highlighted her experience in a student union as the setting to the emergence of her interest in such issues. Hugo, from an urban initiative in a suburb of Lisbon, also mentioned his educational experience as stimulating to his interest in the environment. Some interviewees found their community as the ideal context for mobilisation and engagement. Their past community engagement experiences in the neighbourhood were stressed, and for them the creation of a Transition initiative seems to represent a continuation of their active engagement with the community: "My involvement with the Transition initiative came as a result of the involvement with earlier projects in this neighbourhood" (Bruno).
In sum, we found that Portuguese Transitioners are a mostly homogeneous social group with several mutually reinforcing traits and experiences. It is vital that Transition initiatives look further into the engagement of non-engaged citizens, as well to the inclusion of different social groups, especially underprivileged ones. If we bear in mind that those most affected by climate change will be the already disadvantaged groups (Hossay 2006, Riemer andReich 2011), it becomes apparent that this is particularly critical in Portugal, a country that is highly vulnerable to climate impacts and where significant sectors of the population have low levels of income and are lacking other crucial resources (moreover, the recent economic crisis worsened the condition of those social groups).

Agency, power and socio-political structures
Transitioners' discourse is ambiguous regarding agency in enacting change. Whereas there is a strong focus on community action and community-led responses, individuals are often viewed as the core "locus" of transition and of responsibility towards change. On the one hand, Transition initiatives worldwide focus their carbon-reduction strategies on communities (around three main areas: local food initiatives, reskilling workshops, and projects that reduce consumption of goods or fossil fuels) (Bay 2013); on the other hand, the Transition movement assumes that transition to a lowcarbon world needs to bring together and integrate inner and outer dimensions (Power 2016). What seems worth highlighting here is that participants in our study defined inner transition as the fundamental condition for achieving collective change. Leonor, for example, clearly defended the importance of individual change, arguing: "you can only do it [transition] if you change your own behaviour, if you carry out an inner-transition (… ) this entails a whole change of behaviour, a new vision of life, and of your role in it". Joana claimed that "Transition is a process, which is, first, individual, and then must happen in a group, in a community. Yet, it must always come from an individual level of change". Therefore, each individual is considered an actor of change, and is responsible for both an individual and a broader transition by propagating Transition's ideals: inner transition is then to fuel collective change.
Although agency is typically spread through a variety of actors at different levels, focusing on individual responsibility atomises action and backgrounds structural power-related issues: many aspects are not in individuals' power to change (e.g. access to public transportation) and different individuals, like different social groups, have different degrees of power to address problems (e.g. money to invest on home insulation).
In spite of the focus on individual-led change, most Transitioners also postulate a vision of community development that would arguably require citizens' involvement withand having a voice incollective decision-making processes. For example, Abílio stressed that "real" development cannot take place if people are not the protagonists: "I believe that real development must arise from within the community ( … ) because a key aspect of development is that the community perceives itself to be the protagonist of its own development, otherwise there is no real development." Gabriela phrased the idea explicitly in terms of power, although she conflated the different notions of "power to communities" and "power of community": Change must be something collective, at a level where people feel they can make a difference ( … ). I believe [in Transition] because it can bring power to communities. Bring back the power of community. People getting together and doing things.
Several participants in the study voiced similar constructions of collective agency and empowerment stressing the importance of giving power to communities as a way of promoting social change and moving towards a more sustainable society. However, Transition avoids dealing with the politics of power distribution at community level and larger scales. Rather than struggling for (political) power or confronting authorities, corporations, or even the capitalist system, the Transition movement seems to be focused on the adaptation of daily activities and practices of community members (Bay 2013). As in other national contexts (Trapese Collective 2008, Mason and Whitehead 2012, Kenis and Mathijs 2014, Biddau et al. 2016, Transition initiatives in Portugal tend to prefer a "collaborative", "positive" approach: We are not against anything, instead we are in favour of many things. ( … ) it was a day of protest in Lisbon, and I remember that we were all day planting trees, doing what I believe should be done, through a positive approach. (Jorge)

Messaging
Interviewees acknowledged that Transition's "messages" are not readily comprehensible for citizens. One participant spoke of a "message that is hard to understand. It is difficult ( … ) because transition requires an explanation: transition from what to what? It is complicated to explain this" (Ricardo).
Climate change is viewed as a problem that people "fear" and "prefer to avoid facing", according to Tiago, who also said that the Transition initiative he facilitates had struggled with reaching people via this issue and those at its basis. "If you start talking about climate change ( … ) about oil, it's over ( … ) [Oil] is still a sacred cow". Alternatively, he argued "if we speak about positive things, like biking and walking more, having fun with the kids, then we can get to people".
There is then a tendency to avoid (talking about) problems and to concentrate on responses: "The strategy is to focus on the solution, rather than the problem. The problem is irrelevant" (Nélio). Some of the solutions praised by the movement were referred to as already existing in the communities, and as being accessible through the revitalisation of popular knowledge and cultural practices. As one participant stressed, there is a need to value people's knowledge and sustainable cultural practices: "Valuing what people do, and appreciating what people do, establishes bridges of communication" (Tomás). The development and revitalisation of skills have been referred to as necessary for building a resilient community, which is able to transform its physical and social environment in order to adapt to and mitigate climate change (Archer and Dodman 2015).
This focus on (individual) skills eschews the structural issues mentioned above (social, economic, and political). Moreover, this strategy is used with no further explanation or (preferentially) debating around the reasons why people need to change their lifestyles. If people do not understand the need for change, alternative lifestyles are more likely to alienate than inspire (Portwood-Stacer 2013).
We came across interesting and valuable attempts at framing the experience of "transition" in "positive" terms. Whereas the message of the movement could be perceived as being one of reduction (e.g. of consumption) and sacrifice, some Transitioners frame it as being "experiencing abundance"abundance in vegetable gardens, and in gardens in generalturning localities "more beautiful" and making people "happier" (Sara Rocha, facilitator of the Transition initiative in Coimbra, interview to Rádio Universidade de Coimbra, 23 January 2014). Some also speak of collective definitions of the "good life" (Annelieke van der Sluijs, facilitator of the Transition initiative in Coimbra, interview to Rádio Universidade de Coimbra, 23 January 2014). However, the meaning of "good life" or "happier life" seems to be previously fixed by Transitioners. Paths to the "common good" should arise from critical discussions involving a plurality of perspectives (Barry 1999).

Modes of engagement
Based on interviews, observation and analysis of Facebook pages, we found that the capacity of Portuguese Transition initiatives to mobilise and involve local people has been quite limited. Our interviews suggest that Transitioners view broad engagement and participation by local community members as goals that are difficult to attain: "The greatest challenge was the mobilisation of the community" (Tomás). Ana reinforced this idea arguing that "with one exception, in Portugal none of the initiatives is working with the community the way it should ( … ). In the last [national] meeting, I noticed that all [national initiatives] have similar difficulties in getting the community involved". Whereas there is agreement about the importance of a wide participation, most interviewees do not seem to believe in its materialisation. It is, therefore, crucial to understand how Transition initiatives approach local communities and promote such engagement.
Transition initiatives appear to mostly develop what can be termed weak modes of engagement. Their activities are used to disseminate information about Transition and to let people know about the principles and goals of the movement. Examples include a film cycle, an open debate, a workshop, or a course, or, more rarely, a thematic party or dinner (the hands-on, practical nature of some of these is particularly praised by Transitioners). Those activities are said to simultaneously aim at promoting community engagement, deepening relations, and developing exchanges among people within the community. In most cases, however, Transition initiatives have limited itself to the organisation of this kind of one-off activities. For those community members who have come in contact with Transition, the face of the movement has been such isolated events, which only occasionally achieve significant attendance, and in general stay short of fulfilling the movement's message, or of bringing forth the transformational promise that it makes.
In addition to the modes of engagement described above, it was consensual amongst interviewees that Transition should attempt to create spaces for citizen participation through the exploration of their needs, aspirations, and wishes, and by using a participatory approach. Susana's words expressed the opinion of the majority of Transitioners in our study: "I believe that this has to be done over time, by being with people, and by talkingby talking about what is important in life, about people's conditions, and their dreams." Our fieldwork and interviews with local residents suggest that participatory strategies to involve an extended community in the Transition project have been relatively successful in the rare cases where they were implemented. The best example is an extensive exercise called "Dream Village" carried out in a rural initiative (Aldeia das Amoreiras): "We put everyone together in groups of interest, which resulted from grouping people that shared the same dream [for the community], and we facilitated meetings with these groups" (Susana). This led to the joint construction of a series of plans for the locality and, for a while, it achieved a high degree of adherence and dedication from the local population. However, commitment and interest waned with time and it is hard to pinpoint whether lasting changes resulted from this project.

Perceived constraints in community engagement
Whereas community engagement is considered crucial for Transition, engagement practices often fail to result in the involvement of most members of local communities, as mentioned above. Hence, it is necessary to examine the kinds of constraints that may be inhibiting engagement. In this study, interviewed Transitioners identified cultural and contextual factors as the main obstacles. One of the most frequently reported obstacles was a general lack of disposition of communities to engage with and to commit to civic issues. This appeared in association with references to a culture of weak civic participation, which some participants view as specific to Portugal: "It is very hard. It is hard because we have a culture of low participation, and when it [participation] occurs, people have many ready-made ideas which are well-defined at the outset" (Gabriela). Although some recent research has pointed to a changing pattern of civic participation in Portugal, as new movements have emerged and forms of protest have diversified (Kousis 1999, Matos 2012a, in our study, participants associated the lack of proactivity and commitment with a culturally based non-demanding, non-protesting attitude (even when people are unhappy and unsatisfied about something). Historical legacy was also pointed out as hindering civic engagement, as in this statement about rural Southern Portugal: "Our mentality about land … the tradition of working for the landlords is, I believe, still embodied in our roots" (Abílio).
Several Transitioners suggested that people are distrustful and insecure about what is different from what they know or are used to: "People assume that either you have a personal interest in it, or you are representing an NGO ( … ), and it is assumed that you don't like to do anything in life, and that you just criticise what others do" (Leonor). This lack of trust is also considered to be the result of a lack of belief in environmental problems, which makes people suspicious about actions to solve them: "Regarding [environmental] problems, people are non-believers, they are suspicious by nature" (Sérgio).
For other participants, the low level of community engagement was explained by lack of time, which is a consequence of the pace of contemporary life. Some see the economic context in Portugal over the last few years as particularly unfavourable for citizens' involvement: When compared with four years ago [2009], people are more tired, have less time, have more financial problems, and they are increasingly nomadic, dealing with job uncertainty, etc. Therefore, if there is no physical lack of availability there is a mental and emotional lack of availability. (Susana) It should be highlighted, however, that few Transitioners analysed the challenges involved in engaging communities in connection with the actual practices of Transition initiatives, and, in general, explanations for the lack of engagement were attributed to cultural, contextual, or people's characteristics. For instance, some interviewees placed the problem at the "receiver's" end. Carolina said: "[Transition's message] is probably is not well understood, but in order to understand anything, you have to be open to that. It is very hard to deliver a message when the receiver is not open to it." In Bruno's opinion, "the message was understood by those that were open to that type of messages".
Nonetheless, it is likely that other factors are at play in the relation between Transition facilitators and other community members. Illustrating the diversity challenges discussed above, we observed some signs of "symbolic boundaries" between the two groups in both rural and urban Transition initiatives. Some practices, such as "community lunches", appear to help interaction between the two groups of people; however, we sensed a certain separation between them. Several local people suggested that a social distance exists between Transitioners and themselves, and that their relationship is not fully horizontal. In one initiative, the fact that Transitioners had moved to the village only a few years ago, bringing with them a whole set of unfamiliar ideas and practices (e.g. vegetarianism), and that they were socio-demographically distinct from the local population (e.g. highly educated, with several single or divorced mothers), appears to have created some identity-related resistance or scepticism from the local population. Still, in the localities where we carried the fieldwork, Transitioners maintain close contact and seemingly good personal relationships with many community members. The majority of the facilitators who were interviewed highlighted the importance of strengthening interpersonal relationships with the local community in order to promote trust, cooperation, and a sense of community.
At a national meeting for Transition initiatives held in September 2013, the main priority identified was working to further community engagement. Improving communication strategies was also discussed in tandem with this purpose. Inviting community members as "specialists" to participate in Transition initiatives and to get involved with local groups and organisations were some of the means proposed for helping connections and for engagement with communities and the promotion of community resilience. This is likely to be a slow and long-term process; yet, it is encouraging that Transition initiatives are thinking creatively about ways to tackle it.

Conclusions
This paper contributes to understanding how the Transition movement is migrating across countries, and promoting community engagement at local levels. It is based on 39 interviews and observational notes from 14 Transition initiatives in Portugal, which allow us to offer a broad overview about how the Transition movement is embedding in the national context. Overall, we found that the discourse of Transitioners in Portugal reproduces closely the discourse of the Transition Network in various ways, which may be conflicting with the desire of doing (effective) community engagement at the local level. We have also found a great extent of ambivalence about the meaning and roles of community in Portuguese Transition initiatives (community as means and as goal; lack of critical analysis of the idea of community; excessive focus on the role of individual change) and a lack of consideration of contextual factors in promoting community engagement in local initiatives.
As found in other studies (Pyles 2009, Sarkissian et al. 2009, Barr and Devine-Wright 2012, Dilley 2017, the Portuguese Transition initiatives considered community engagement in three ways: a fundamental principle, a process and simultaneously a desired outcome. In our study, community engagement was expressed as the principle goal of Transition initiatives in strong association with the aim of achieving resilience, so much so that community engagement often appeared to become a proxy for resilience. In spite of its centrality, we have identified several aspects of Transition's discursive constructions of community and community engagement that indicate ambivalence towards, or avoidance of, certain issues. They relate to agency, structure, power, and inclusion, suggesting a lack of politicisation of environmental issues among Portuguese Transition initiatives, as in others contexts Mathijs 2014, Kenis andLievens 2014).
In the Transition movement's vision, communities lead change. This is a welcome alternative to disempowering discourses on climate change where perceived lack of agency often deters citizen engagement. However, the Transition movement's construction of the role of community is riddled with various ambiguities and even contradictions. Firstly, Transitioners often speak of individual responsibility for change as if it were independent from the power to act in relation to a number of structures (e.g. related to consumption and mobility). Secondly, community agency is detached from political, economic, and other structures in the discourse of the Transition movement. As resilience arguably depends on the capacity of communities for collective decision-making and action, political power structures at various scales have to be taken into account. In spite of this, the movement displays a systematic avoidance of confrontation with existing political and economic powers (e.g. governments and corporations). Transition's faith in individual and community agency within unchanged structures raises doubts about its capacity to succeed in increasing community resilience in relation to climate change and other sustainability crises.
The movement seems not to actively consider diversity and power issues within communities nor indeed how they affect the possibility of people to engage in participatory spaces in egalitarian forms. These matters are often treated in a relatively unproblematic way, without ascertaining how they are entrenched in existing relationships or exploring the potential for transformative change (Archer and Dodman 2015). As those most affected by climate change are likely to be the already disadvantaged groups, the explicit recognition of climate change as matter of social justice could help address power inequalities in communities. Otherwise, the Transition movement might be promoting the constitution of a "resilience based on power, prestige, position and influence that could ultimately lead to a dystopian future marked out by inward-looking and even 'gated' forms of community" (Barr and Devine-Wright 2012, p. 530). The movement should then explicitly adopt a more critical stance with regard to class dynamics and social inequality (Connors and McDonald 2011) and actively seek to recruit diverse group participants.
Our study suggests that Transitioners tend to emphasise the role of cultural and circumstantial factors in explaining low levels of community engagement. They see those barriers as something that is out of their control: "the Portuguese culture", others lacking time to participate, the economic crisis, a widespread lack of trust, and so on. Although this may correspond (partially, at least) to real constraints, it should not prevent a critical analysis of Transition initiatives' practices. Besides factors already identified in the literature, our research suggests that the communication practices of local initiatives may be constraining community engagement processes. Most Transition initiatives have found communicating about climate change with members of their communities to be difficult and consciously avoid the theme. As they shun a negative narrative about climate change and its impacts, they focus on "solutions" and "positive" messages steering away from the issues that are at the root and heart of the international Transition movement (climate change and fossil fuels). Importantly, climate change is also sidelined in Transitioners' reflections on their individual mission and identity. The avoidance of (communication about) climate changeand the structural issues it is associated withis rather significant in Portugal as efforts towards empowerment for adaptation would be particularly needed among multiple vulnerable communities.
If the problems at the basis of the movement are not clear, the proposed solution, "the transition", will most likely not be well understood either. The kinds of transformations that many analysts consider necessary to address climate change (e.g. Klein 2014) require an extended democratic debate rather than top-down "fixes". While the Transition movement proposes community-based responses, our study shows that there is not enough of an effort to devise fully inclusive and participatory modes of interaction and means of participation of community members, which may (in part) explain the problem of lack of trust. Participatory modes of engagement would allow for multiple imagined goals of "the transition", as well as diverse possible pathways, to become visible. They could empower community members and be a democratising instrument (Montero 2009).
It is likely that factors specific to Portugal, such as relatively low levels of civic engagement and of mobilisation towards the environment, and strong social/economic/educational differences among the population, as well as the period at stake (coinciding with the economic crisis), may have made engagement particularly challenging. So, it is noteworthy that the movement managed to expand significantly the number of local initiatives. Even if the Transition movement often faces barriers to the promotion of participation by most members of the communities, it seems to be strongly appealing to individuals who have a previous interest in environmental issues and an active involvement with their communities, which means that it is likely to last, and to some extent, will continue to be successful. However, in order to fulfil its vision of the future, the movement cannot exist without a broader participation and a greater commitment of local communities, and therefore must continue to devise means and strategies for promoting this kind of engagement. The use of participatory approaches in Transition initiatives appears to have a particularly strong potential, especially if based on pluralism, thoroughly democratic and oriented to deliberation. Growing reflexivity in the movement with enhanced analysis of its practices and greater awareness to communicationrelated challenges suggest that a learning process is ongoing and that some of the problems identified here may be mitigated in the future. However, widespread community engagement is likely to remain hard to achieve if the movement continues to avoidor being ambiguous aboutissues related to power, agency, structure, and inclusion. In addition, strategies for community engagement should be adapted to specific social contexts rather than internationally uniform as the characteristics