1. Introduction

This paper aims to identify the actors and dynamics that influence how globally circulating frames of urban nature are translated onto the ground in the African context, and with what implications for whose voices are or are not included. This is of interest because frames, understood here as discursive definitions of a topic, shape global agendas, but also matter locally in terms of socio-material implications of programmes and projects that seek to implement them on the ground.

Studying how ideas and knowledge from one geographical context influence, or translate to, other contexts has a long tradition, from political scientists researching the transfer of policies as best practices between nation states (Dolowitz & Marsh 1996) to geographers focusing on globally circulating urban policy models and policy knowledge arriving in cities as part of politicised processes in mutated forms (Peck & Theodore 2010; McCann & Ward 2013). Relationality plays a central role in both tracing policy ideas and analysing how local policies are arrived at from a myriad of elsewheres (Lane 2022; Robinson 2015). This paper draws from literature that emphasises the role of transnational actors as ‘agents of policy transfer’ (Stone 2004) who:

interpret, frame, package, and represent information about best policy practices, successful cities, and cutting-edge ideas. (McCann 2011: 114)

For example, Laeni et al. (2021) show how the Dutch water management approach was framed as a global water solution for urbanised deltas and, through international cooperation, ‘translated’ to inform the development of Vietnam’s Mekong Delta Plan of 2013. Goh (2020) illustrates that this happens within global-urban networks, through which capital, knowledge and influence flow. She calls for more research on how and with what consequences globally circulating ideas in the climate context are ‘deformed and refracted’ (p. 2234) in response to local political-economic and spatial conditions. This paper seeks to engage with such questions by investigating how and by whom global frames are deployed, negotiated and configured locally, in the global-to-local translation of frames.

The case of urban nature is chosen to explore this. This is timely as the role of nature is of growing importance in global discourses and agendas that seek to find ways for more climate-resilient urban development pathways (Goodwin et al. 2023; Li et al. 2021). In global environmental governance, researchers have identified various frames of urban nature currently circulating in the climate context, each of which emphasises particular challenges and solutions (Tozer et al. 2022). Framing lays the conceptual groundwork for possible future courses of action by selecting and making sense of policy-relevant issues (Van Hulst & Yanow 2016). This also means that by drawing on frames actors find different justifications and entry points towards working with urban nature, with consequences for and by whom solutions are configured (Tozer et al. 2022). It is important to recognise that generations of researchers have grappled with the question of how alternate frames of nature have been mobilised to advance various development ideologies (e.g. Dryzek 1997). In relation to justice concerns, for example, political ecology provides insights into the social injustices resulting from commodified nature (Harvey 1993) and the unequal distribution of resources and contestation over power and resources at the urban scale (Heynen et al. 2006). More recently, scholars have emphasised the risk of social exclusion and displacement through green gentrification and the financialisation of urban nature (Anguelovski et al. 2018; Anguelovski & Corbera 2022). Besides, as Williams (1972) was already emphasising decades ago, people interpret and define nature based on their cultural and historical contexts, and studies have revealed how for example local indigenous understandings of nature remain unaccommodated within Eurocentric constructs of urban nature in South African towns influenced by colonial history (Cocks et al. 2020). Research has also found that many greening interventions are attached to top-down visions for ‘acceptable’ nature in the city (Sekulova et al. 2021), or are not catering enough to world views and rights of traditional and indigenous communities (Melanidis & Hagerman 2022). Hence, scholars urge for co-production that accounts for diverse forms of knowledge and expertise to create diverse, participative, accessible forms of nature (Mabon et al. 2022; Sekulova et al. 2021; Tozer et al. 2020).

Resonating with concerns raised on top-down visions of nature, the literature on climate urbanism in the global south provides ample evidence of transnational actors deploying top-down and technocratic approaches, at the expense of local actors being included in the process (Diep et al. 2023; Falzon 2021). For example, the Rockefeller Foundation’s 100 Resilient Cities programme was critiqued as imposing a global model of urban resilience from a distance, upon which a market of technocratic resilience tools, instruments and practices can be built – an approach that the City of Durban in South Africa contested (Webber et al. 2020; Roberts et al. 2020). More broadly, critical scholars describe global climate finance as reproducing neoliberal environmental governance at the expense of civil society voices (Bracking 2021). The observed deployment of external knowledge and expertise in the design of interventions is problematic as it effectively undermines African agency, whereby the complexity of urban adaptation and resilience further exacerbates this phenomenon (Spaliviero et al. 2020; Pelling et al. 2018; Castán Broto 2014). A host of scholars problematise maladaptation outcomes of interventions by transnational actors in the global south, and criticise, for example, technocratic and externally driven approaches for risking reproducing vulnerability as a result of a lack of in-depth contextual understanding (e.g. Eriksen et al. 2021).

As suggested by grey literature, transnational actors are increasingly promoting nature-based solutions (NBSs) and developing programmes and projects in cities as a means to address climate risks and build resilience. Yet research on urban NBS in Africa, which remains in its infancy, has not attended to this (Nyika & Dinka 2022; Chausson et al. 2020). Against this background, it is important to explore how global frames of urban nature are translated to the African context. It becomes relevant to ask whether the translation process is characterised by a replication of top-down approaches driven by transnational actors observed in research on climate urbanism or whether (and which) other actors and dynamics influence how globally circulating frames are translated onto the ground in African cities. To that end, the paper is guided by the following research questions:

• How and by whom are global frames deployed, negotiated and configured, when transnational actors are involved?
• What implications does this have for whose voices are or are not included?

Empirically, this paper focuses on a resilience-building intervention targeting informal settlements in Lilongwe, Malawi, in which various actors at the transnational, national and local levels were involved between 2019 and 2022. The case therefore lends itself to gathering rich material for analysis and drawing empirically grounded conclusions.

The paper is structured as follows. The next section introduces the case and explains the research design. The results of the analysis are presented and discussed in Sections 3 and 4. The final section concludes by outlining the broader implications and suggesting further research.

2. Research design

An in-depth case study design was chosen to obtain data for answering the research questions. The paper focuses on the case of Lilongwe, Malawi. Between 2019 and 2022, actors at various levels were involved in logically connected efforts to increase urban resilience in informal settlements. The process commenced with a participatory planning exercise through the United Nations Human Settlement’s (UN-Habitat) City Resilience Action Planning (CityRAP) tool, in which urban nature was prioritised as a strategy for nature to provide recreation, social inclusion and safety through green public spaces in the informal settlement of Mtandire. Building on the planning effort, eventually a pilot NBS project was implemented that focused on flood risk reduction through afforestation and solid waste management around river banks in the informal settlements of Kawale and Kayileka. Curious about the different configurations of strategies and practices related to urban nature, as well as the shift in location over time, the authors designed data collection and analysis for tracing back the trajectory, to identify actors and dynamics influencing the overall global-to-local translation process.

For this paper, the general components of a global-to-local translation process are conceptualised as follows. A global frame is a discursive definition of a topic and cognitive construct that transnational actors deploy through funding allocation for implementation of programmes and projects (Laeni et al. 2021; Tozer et al. 2022). Through those, the frame (e.g. nature for resilience) eventually becomes instantiated in space and time and embedded in the local context in the form of a particular strategy and associated practice (e.g. flood risk reduction through increased vegetative cover). The latter is negotiated and configured by the actors who are involved in developing and implementing the programmes and projects. The work of Tozer et al. (2022) is drawn upon instrumentally to identify frames of urban nature in the case. Accordingly, in global environmental governance, four frames of urban nature circulate, each emphasising particular challenges and natural solutions in the climate context: ‘nature for mitigation’, ‘nature for resilience’, ‘integrated benefits of nature’ and ‘nature first’. While through the first two urban nature is argued to be able to mitigate climate change and contribute to adaptation and disaster risk reduction, the third highlights the potential of nature to address multiple intertwined sustainability challenges simultaneously (including biodiversity loss, climate change, energy sustainability, as well as health and wellbeing) and the fourth centres on halting biodiversity loss.

A mixed-method approach was employed to capture the perspectives of various actors. Key information stems from the first author’s participant observation during the CityRAP process in Lilongwe between May and August 2019, which involved two weeks of stakeholder workshops with a total of 105 meeting participants (27 female and 78 male), and remote support to the local government until validation of the plan, while working as a consultant for UN-Habitat. In this role she also participated in the development of a concept note for the ‘Risk Award 2021’. While this positionality may have impacted the data collection as some of the interviewees were known to the researcher, it provided advantages in gaining access to meetings for participant observation and for establishing rapport during interviews. Fifteen semi-structured interviews with representatives from transnational actors, national and local governments and non-governmental organisations (NGOs) were undertaken in 2023 (see supplemental data online). The interviewees comprised representatives from organisations applying for funding in the phases presented in this paper, and hence represent an adequate evidence base for the perspective of partners applying for funding. Informants were selected based on their knowledge and involvement throughout the trajectory of the resilience-building effort and by employing the snowballing sample technique. The interview guide was designed to understand the most important steps in the translation process; actors along the journey; how and why NBSs became part of it; and aspects of negotiation/contestation and compromises. Interviews were audio-recorded and selectively transcribed. This information was complemented by transect walks in the project area and an audio-recorded focus group discussion (FGD) with community representatives in April 2023. The FGD was facilitated by a local NGO staff member in Chichewa, and subsequently transcribed and translated by a research assistant. Further, project-related documents were analysed (see list of documents in the supplemental data online).

The material gathered was analysed and reiteratively interpreted. Qualitative coding paid attention to the following themes: frames, deployment mechanisms, actor configurations, included/excluded actors, aspects of negotiation/contestation, influential dynamics, and translation outcomes in terms of local instantiation of the frame (strategy and associated practice). Frames were identified by thematic analysis of objectives and issues that were meant to be addressed through working with nature. Here, referring to the four frames identified by Tozer et al. (2022) helped to identify shifts over time. Reiterative analysis of all data sources led to the interpretation of results presented in the next section. These are structured around three main phases identified in the case study, preceded by an explanation of the geographical context.

3. Global frames (re-)configured for Lilongwe’s informal settlements

Malawi’s capital, Lilongwe, has an official population of 989,318, of which an estimated 76% reside in informal areas (National Statistics Office 2019). The city has an ambivalent colonial inheritance. Designed as a new post-independence capital in the late 1960s, counterintuitively it bears the strong influence of British and South African town planning, seen in sharp zoning of land-use types and the ‘garden city’ concept that allowed areas of unbuilt-upon land to separate and segregate residential areas from one another (Mwathunga & Donaldson 2022; Englund 2002). Today, the spatial differentiation between poor and wealthy areas remains but much of the ‘garden’ of the city has perished. Heavy reliance on fuelwood and charcoal for domestic energy, which recurrent bans on charcoal alone are unlikely to solve, has impacted the country’s natural resources and deforestation did not spare the rapidly growing city (Zulu 2010; Bone et al. 2017). From 1986 to 2021, the city lost 44% of its green coverage (Nazombe & Nambazo 2023). Construction, sand mining, brick moulding and agriculture-related activities contribute to the loss of wetlands, riparian vegetation and urban greeneries in Lilongwe and exacerbate vulnerability to flood risks (Afionis et al. 2020; Mkulama et al. 2020). Most vulnerable are informal settlements located along main streams, such as Mtandire, Kawale and Kayileka (Makuwira 2022). Meanwhile, since the country’s independence, various transnational actors have engaged in urban development in Lilongwe, through developing plans and strategies and funding projects (Croese et al. 2023). Today, much of this effort is directed at reducing risks and building resilience given the city’s vulnerability to recurrent climate-related hazards (GoM 2019). This involves working with nature and implementing NBSs, as exemplified by this case study.

Three phases of a translation process can be observed, understood as successive and logically connected efforts by actors at various levels at increasing urban resilience. First, Lilongwe City Council (LCC) and UN-Habitat, through its CityRAP tool, generated a Resilience Framework for Action (RFA), involving community representatives and various stakeholders. LCC decided to focus on Mtandire, and in future potentially replicate the process in other vulnerable areas. With this goal in mind, community-based organisations (CBOs) from Kawale and Kayileka were invited to participate and undertook risk mapping for their localities. For Mtandire, eventually, green public space was identified as one of six priority areas to address urban resilience (see Figure 1). After validating the plan, there was no funding for its implementation in place and both LCC and UN-Habitat aimed at raising funds. Second, following a dormant period of about a year, jointly with other stakeholders, a concept note was developed in response to a call for proposals for the ‘Risk Award for Eco- and Nature-Based Solutions for Disaster Risk Reduction’, issued by the Munich Re Foundation and the United Nations Office for Disaster Risk Reduction (UNDRR). The concept note did not win funding. However, third, it laid the foundation for an NBS pilot that was eventually implemented about a year later in Kawale and Kayileka by a local NGO through donor funding as part of the Swedish International Development Cooperation Agency’s (Sida) programme contribution to UN-Habitat. Each phase of the process was characterised by a different mechanism of deployment, which led to the frame’s instantiation and local embeddedness as strategies and associated practices of working with nature. As the analysis reveals, those strategies and associated practices related to urban nature were altered over time. Driven by the dynamics of funding opportunities, they were negotiated, reconfigured and relabelled, at the same time influencing who was involved and excluded (Figure 1). The results are presented in detail below for each phase.

4. Understanding the dynamics and actors in the translation of frames

Transnational actors played a key role in how global frames were translated to Lilongwe’s informal settlements. The findings resonate with conceptualisations of transnational actors as agents of policy transfer. The case suggests that they take on various roles in this logic. They can provide finance for projects that Meijerink and Huitema (2010) conceptualised as ‘frontline practices’, which constitute an essential strategy by transnational actors for demonstrating how solutions that are part of a frame can be implemented and locally embedded. Another role is the interpretation and further normative development of global frames, such as in phase 3, where a strong emphasis emerged on the ‘blind spot of the nexus between informality, resilience and NBS’ (UN interviewee A), highlighting the need for more efforts on NBSs in informal settlements. At the same time, transnational actors are entangled in the machinery of climate and development finance as implementing or executing entities who are part of the competition for resources and in the process enter into partnerships with local governments and NGOs. The hunt for erratic funding was a key dynamic influencing how and with what consequences globally circulating frames were ‘deformed and refracted’ (Goh 2020) in the translation to local projects in Lilongwe. Funding opportunities shaped a reconfiguration of the strategy and associated practice (i.e. from green public spaces for social inclusion, safety and recreation to afforestation and SWM for flood risk reduction), and later relabelling or associating it with biodiversity and other integrated benefits of nature, to suit global frames of urban nature deployed. This reemphasises that in the process of translating international policy from the global to the local space, various tactics are used by practitioners to gain access to funding for implementation (Mannell 2014).

Turning to wider implications, three groups of issues emerge. First, the donor impetus caused a significant thematic shift in how nature was locally embedded as a strategy, which resulted in nature taking a more central value as flood risk reduction measure vis-à-vis grey infrastructure (roads and drainage) previously identified in the RFA. This reemphasises the powerful position of transnational actors in deploying frames of urban nature. Ultimately, the shift in how nature became strategised meant that the value of nature as providing for social inclusion, safety and recreation was neglected. The question arises as to why the actors seeking funding for the RFA did not opt for building on the identified priorities of green public space by using them as flood retention zones, thereby aiming to realise multiple benefits. The empirical material does not allow for a conclusion. It could have been that, as the topography and risk profile in the settlements make areas close to the river more susceptible to floods and soil erosion, it was deemed a sensible approach to focus on these areas. Creating new public space could also have been complex to implement as it may have necessitated land reallocation. However, flood risk reduction through public green space in flood hotspots was found an appropriate measure in Kibera, Nairobi (Diep et al. 2022), thus it would be valuable to research this further to propose contextualised and evidence-based solutions for Lilongwe.

Second, it would be an oversimplification to characterise the translation as a top-down imposition of global frames. The local government took several key decisions, many of which were pragmatic. Shortly after the finalisation of the RFA, LCC negotiated funding from the UNDP and DoDMA for grey infrastructure flood risk interventions in Mtandire that were closely aligned to the priorities identified in the RFA. When the funding opportunity for NBSs emerged in phase 2, LCC suggested targeting other flood-prone settlements, to spread thin external resources. As such, various external funders and projects were coordinated under the common goal of building resilience to floods in informal settlements. This suggests that subnational governments can bring in their priorities by strategically manoeuvring funding opportunities in the global-to-local translation of frames. This resonates with studies that challenge the dominant interpretation of interventions of transnational actors in African cities representing the imposition of external ideas. Croese et al. (2023) found, related to master planning efforts supported by transnational actors, that local planning actors often carefully, pragmatically and opportunistically mediate external support.

Third, which local actors were included in partnerships to develop the project and thereby had a voice in the translation process was determined by strategic concerns (e.g. in phase 2 the choice of WESM, to provide expertise concerning environmental issues) or more pragmatic ones (e.g. the choice for CCODE as an actor with existing networks in informal settlements), and largely driven by the partners’ desire to access funding. Potentially most problematic was the lack of inclusion of communities in decisions taken in phases 2 and 3. This contrasts with the first phase, when communities were heavily involved in identifying and prioritising resilience-building actions. Research compares the CityRAP process in Mtandire positively with a described general lack of participation activities conducted by LCC, and attributes the involvement of communities to UN-Habitat’s influence (Zimba 2022). Indeed, the aim of CityRAP is to bring together local governments and communities to assess and prioritise action in data-scarce contexts based on local knowledge and perceptions with minimal external interference (Spaliviero et al. 2020). However, in the subsequent phases, in the competition for funding, decisions were taken by the local government and partners that resulted in thematic and geographical shifts, without wider consultation of the beneficiary communities. As various assessments had already been done, and because local NGOs were closely involved in collaboratively developing the concept note with LCC, it might have seemed an adequate justification for the proposed project. Nevertheless, procedural and distributive justice aims may have been better served if the community had fully participated in the development of these projects. There is ample evidence that working with local-level stakeholders does not automatically guarantee that actions will address the needs of local vulnerable groups, and lack of community inclusion may ultimately hamper the sustainability of the interventions and potentially cause maladaptation outcomes (Eriksen et al. 2021; Omukuti 2020).

The findings have to be understood within the study’s limitations. The study predominantly captures the perspectives of transnational actors, the local government and NGOs involved in the development of the RFA and subsequent fundraising efforts. While this translated into addressing flood risks through afforestation, it would be worth investigating whether and how the initial vision of green public spaces was taken forward by the community or other actors. Further, in focusing on the trajectory from participatory planning to developing the pilot project, the actual outcomes of the implemented project are beyond the scope of this paper and the comprehensive perspectives of communities on the same have to be explored in future research.

5. Conclusions

The paper set out to identify the actors and dynamics that influence how globally circulating frames of urban nature are translated onto the ground in the African context, and the implications for those whose voices are or are not included. The results of tracing back different phases of such a translation process in Lilongwe’s informal settlements suggest that it is not predominantly shaped by transnational actors in a top-down manner. Contrary to evidence from the climate urbanism literature that community groups are often engaged in fact-finding for vulnerability assessments but not for the identification of strategies (Shi et al. 2016), there has been meaningful engagement in the development of the RFA. However, lacking resources for implementing the strategy set in motion fundraising efforts that influenced how local strategies and practices were altered to suit global frames and the resources they generate. In the process, strategic and pragmatic partnerships of actors at different levels were forged, and strategies and practices for implementing (nature-based) solutions were tactically reconfigured and relabelled over time. The case also shows that the local government was able to pragmatically manoeuvre various external funding opportunities for flood resilience, reaffirming findings in literature that planning actors often carefully and opportunistically mediate external support. LCC entered into partnerships with transnational actors and NGOs, where the latter were closely involved in key decisions. Yet, while the overall process commenced in a highly participatory spirit, the community was left out in decisions on the design of the pilot project that was finally implemented. This puts into question whether the outcomes represent their needs and interests and would merit further research to derive lessons for the actors involved in designing the project.

Results from the study suggest several wider implications that speak to critical literature on urban adaptation and resilience. First, international urban and climate finance are key enablers of investments in developing countries. The study suggests that a better understanding of the intricate relations between urban planning and international climate finance is needed. Many local governments engage in meaningful participatory planning processes, often with support from transnational actors. In view of scarce resources, local governments often find themselves reacting to unpredictable funding opportunities to implement strategies arising from such processes. Unless actors at all levels strive to align new interventions closely with existing strategies and promote meaningful engagement with communities on the same, external funding potentially distorts local priorities by replicating top-down visions for adaptation and resilience. Second, NBSs enjoy global recognition and funding support. There is growing evidence that interventions to adapt to climate change and build resilience can only be progressive if goals are associated with the need to improve the prospects of disadvantaged groups (Ziervogel 2019; Vale 2014). To enable transformative change it is therefore vital that efforts to enhance resilience through working with nature are aligned with the priorities of the most vulnerable groups. Third but not least, the findings have consequences for the understanding of frames of urban nature as well as the concept of NBSs and their transformative potential. Global frames of urban nature are cognitive constructs that generate financial resources. As the study has shown, frames are inherently malleable, which means that they can be manipulated towards specific interests. This implies a wider risk that in the pursuit of funding, and manipulation of frames to focus on issues of strategic interests, or by ‘green-coating’ initiatives and using nature as an empty signifier without holistically addressing environmental issues, the multifunctionality of NBSs, which proponents argue to be the strength of the concept (Melanidis & Hagerman 2022), gets lost in translation. To ensure that the potential of NBSs is realised in view of the global climate biodiversity crisis, actors promoting such approaches should be mindful of these risks and put in place measures to address them.

While this study contributed to the understanding of the translation process of global frames of urban nature to African cities, further research should investigate issues of power between actors of various levels in the translation process, and the socio-material implications for the distribution of benefits and burdens of NBSs as outcomes of these processes. Research on NBSs in African informal settlements is nascent and policymakers remain sceptical of the feasibility of working with nature in this context (Diep et al. 2022). The paper has started making important inroads in demonstrating what strategies and associated practices for informal settlements in Africa are deemed relevant by actors at various levels. Building on this, researchers should further investigate how working with nature in urban adaptation and resilience-building interventions can be closely linked to the priorities of disadvantaged groups in African cities, such as communities in informal settlements, to leverage the power of urban nature for more just societies.