Researching Local Subjectivities in Contested Contexts: Using Intersecting Methodologies to Understand Large Green-Energy Projects in Kenya

Abstract

This article critically reflects on the use of an ‘intersecting methodologies’ approach to generate understanding of the diverse experiences of people living near contested large green energy projects in Kenya. Local subjectivities are essential to understanding the contestations engendered by these projects, especially in areas that are distant from decision-making and global capital. However, researching residents’ experiences in these settings is challenging due to historical divisions and recent conflict triggered by the projects, and the unequal power relations that exist between stakeholders. We discuss the opportunities and challenges of using intersecting methodologies for research in contested settings, which involved weaving together qualitative, ethnographic, community-based participatory research and participatory video. Challenges included the difficulties of supporting peer researchers in remote places to conduct complex research processes from afar, negotiating the legacies of prior research practices and the knowledge economy, and balancing research sense-making with video production expectations. We argue that interlacing methodologies from different epistemological traditions helps to uncover the various ways that differently positioned local stakeholders ‘see’ large-resource-based investments. It also brought learning from the different epistemologies and viewpoints into conversation, and this created productive tensions to transform understanding of conflict and open pathways to more peaceful relations.

Introduction

Large-scale resource-based investments are inherently contested, even when transitioning to green energy sources, where the rationale for lowering emissions while expanding energy production seems compelling. Such projects can be highly sensitive, resulting in significant disruption for those living near sites where infrastructure is being laid by developers to harness resources. Additionally, these places are often at the rural margins, and therefore distant from capital cities where decisions are reached about which energy sources to invest in, where and for whose benefit. Issues of difference, division, and inequality are raised—compounded in many contexts by corruption and unaccountable forms of politics. This makes it important to think carefully about appropriate methodological approaches when studying the realities and affects for communities living near these projects.

This article critically reflects on the use of an innovative ‘intersecting methodologies’ approach during the interdisciplinary study ‘Seeing Conflict at the Margins’ (SCM), ¹ which explored different local stakeholder perspectives of large-scale resource-based projects in rural Kenya. The ‘intersecting methodologies’ approach (see illustrative video ² ) was developed to understand the experiences and views of local residents who live adjacent to an extensive new windfarm near Lake Turkana and expanding geothermal power-plants in Ol Karia—respectively the largest wind power and geothermal developments not only in Kenya but in all of sub-Saharan Africa. This approach incorporated methodologies from different epistemological traditions, including community-based participatory research (CBPR), participatory video, and qualitative, ethnographic and historical methods. We use the term intersecting rather than combining or integrating, as the methods were not blended to create changed forms but maintained their coherence whilst interlaced. This meant they sometimes bisected or converged within the same research exercise, and sometimes occurred at separate locations or with different participants in parallel, as detailed below. Specifically, in this paper we ask how methodologies from different epistemological traditions can be woven together to reveal the various local subjectivities and competing perspectives in places characterized by contestation, conflict and struggle?

The ‘Seeing Conflict at the Margins’ project sought to understand how people living nearest to large scale green-energy projects view and respond to the investments. Rather than assume that common interpretations can be established uniformly across experiences and contexts, a guiding research design assumption was that it is more constructive to lay bare the various subjective explanations and perspectives that exist. Centring the priorities of residents in these ‘communities of place, identity and interest’ (Banks et al., 2013), we thus aimed to ‘open up’ the multiplicity of possible interpretations, influenced by differences such as gender, age, ethnicity, employment and social status.

In settings of contestation and conflict, beyond the familiar constraints on area access due to security and administrative controls, there are challenges in researching local subjectivities—or the perspectives of those living within a situation (Shotter, 2006). The first of these arises from unequal social relations and divisions between different local stakeholders. A related second challenge is the acute vulnerability of many to the influence of powerful local actors, who may intimidate or threaten violence. This can disrupt or distort information flows owing to fears of reprisals and/or losing assistance. Individuals may thus express experiences in ways that ensure their personal security. Suspicion and uncertainty about how research outputs will be used is the third challenge, especially given histories of racism, colonialism and misogyny and the dubious use of ‘findings’ to maintain systemic oppression. This can negatively affect relationships with stakeholders, and unintentionally prioritise dominant actors’ interpretations, however incorrect. A fourth challenge is the risk of research processes being manipulated by different stakeholder groups as part of wider political manoeuvring. Misinformation is rife, so the data gathered can be unreliable, or at least difficult to interpret. Some groups may make claims or construct counter-narratives backing their own agendas, which complicates analysis of the competing viewpoints or obscures alternative readings. Finally, in researching subjectivities, a fifth challenge is the risk of inadvertently opening old wounds and reigniting rivalries as different actors attempt to influence the ongoing dynamics of struggle and control (Hume, 2007).

There is already considerable innovation in designing research for contested contexts through mixing quantitative and qualitative methods (Acar et al., 2020; Khan Mohmand et al., 2017; Thaler, 2017). However, in settings characterised by conflict and division, fewer researchers have used qualitative approaches to generate situated knowledge alongside longer-term participatory action research (PAR) to build trust, address power dynamics, and drive iterative learning - despite the potential (see Hyslop, 2021; Santamaria, 2021). This is especially in places where new large investment projects generate tension and divisions.

The ‘intersecting methodologies’ approach was developed collaboratively between UK-based researchers ³ and practitioners affiliated with the Pastoralist Development Network of Kenya (PDNK) and Friends of Lake Turkana (FoLT), organisations that work respectively with communities in Ol Karia and the Lake Turkana basin. Collaboration between team members predated the project by many years. Colleagues from FoLT and PDNK helped to identify Kenyan focal sites, based on their knowledge and existing relationships with communities grappling with tensions generated by new resource projects. Residents were informed about the project before inception, and some peer researchers identified. As UK-based researchers, our role in research on the ground was circumscribed, and largely centred on key moments (providing training on methodologies, accompanying peer research teams to pilot and contextualise the various research methods in the focal sites, and convening discussion and feedback meetings during extended research processes that were driven by peer researcher teams).

In this paper we explain and explore the intersecting methodologies approach. Following this introductory section, in section 2, we locate the contestation and division generated by large green-energy developments in Kenya in reference to the specific fieldwork sites. This provides orientation to the research purpose and context, and the methodological challenges. Section 3 then discusses the research approaches and methods, including the rationale for their inclusion as components of our overall approach. In section 3, we also outline the extended phases of the ‘Seeing Conflict at the Margins’ research processes. In section 4, we explain the intersecting methodologies approach, through focusing on the interlacing of participatory video methodologies with other qualitative and participatory methods. We also describe the methodological contributions and tensions in this section before turning to the challenges we faced (in section 5). Methodological lessons from the project are synthesised in section 6. Overall, we argue that it was the methodological interlacing, which purposively drew on the distinct contributions of each approach, that enabled diverse subjective viewpoints to be expressed and then discussed within and across contexts. It also brought learning from the different epistemologies (e.g. interactive, observational, visual and action-reflection ways of knowing) into conversation, and this deepened understanding of the tensions underpinning conflict.

Contestation and Division Around Large Green-Energy Developments in Kenya

Investment in large-scale renewable-energy projects has increased substantially in recent years as governments capitalise on the turn to ‘green’ energy. The prevailing perspective is that renewable-energy investments are broadly positive—generating public ‘good’ by expanding national energy production from green sources while contributing to economic growth. Yet, benefits for communities adjacent to project sites are more ambiguous, with inadequate attention to the impacts on social cohesion or the rights of local peoples (Avila-Calero, 2017; Cormack & Kurewa, 2018; Dunlap, 2018; Hughes & Rogei, 2020). Civil society, including global and national human rights organisations, and watchdog and community-based groups, have challenged renewable energy developments, asserting that they cause significant environmental and social harm (Danwatch, 2016; Renkens, 2019). In practice, these projects can both generate new tensions between those who benefit and those who do not and exacerbate localised conflicts that are rooted in longer histories of social and political division.

Kenya is at the forefront of green energy development, ranking first in sub-Saharan Africa in investment in renewable resources. The Kenyan government is banking on continued large investment in green-energy production and transmission to help it achieve middle-income, industrialising status under its Vision 2030 plans (Republic of Kenya, 2007). Already, the country hosts the largest geothermal and wind power developments in Africa. The Ol Karia geothermal complex has expanded substantially in recent years, with construction underway of pipelines, wells and plants estimated to generate 1000 MW of energy. In Marsabit County, the Lake Turkana Wind Power (LTWP) site—consisting of 365 turbines and a high voltage substation—was connected to the national grid in 2019.

Ol Karia has a population of about 20,000, living in scattered villages across a vast, rugged rangeland several hours’ drive from Kenya’s capital Nairobi. Residents are mainly Maasai, but members of other ethnic groups live here too; including Turkana, Samburu and Kikuyu, who have long intermarried with Maasai. Many people practise pastoralism (keeping cows, sheep and goats), but livelihoods have diversified in recent years. People now also work, for example, as traders, labourers, charcoal burners, blacksmiths, and tour guides as well as for geothermal companies. Jobs with these companies are often low-paid and casual; young people complain that the best jobs go to non-Maasai. Most people do not have land titles, a major grievance making them vulnerable to land-grabs and forced eviction to make way for geothermal plants. There are few health clinics, and the nearest public hospital is in Naivasha (more than 40 kms away). Transport is infrequent and expensive, forcing people to walk 15 kms or more to other services and work.

The LTWP windfarm sprawls over 160 square kilometres to the east of Lake Turkana in Kenya’s far north. Notable as the largest single private investment in Kenya’s history, the US$800 million development was constructed in rangelands inhabited by interacting groups of Turkana, Samburu, Rendille and El Molo peoples. Pastoralism and, nearer the lake, fishing are the main livelihoods. The area is remote and was very inaccessible when construction commenced in 2014. To transport turbines and other large equipment to the site, developers constructed a 208-km road from Laisamis, a small town on the main road to the south. While this has improved accessibility, the road stops abruptly at the windfarm. Lake Turkana and Loiyangalani, the largest town (5000 people) and regional administrative centre, are reached by a 40-km rocky track, which makes transport and services even more distant and costlier than Ol Karia.

Emphasising the national benefits of largescale resource investments like the Ol Karia geothermal plants and the LTWP windfarm, means that local experiences are often invisible or unintelligible to project planners and therefore overlooked. Existing research shows that the state’s vision of green transformation has not translated into significant livelihood improvements for many people living in the project areas (Drew, 2020; Hughes & Rogei, 2020). Life continues to be deeply insecure for most, and residents of Ol Karia and the small settlements near the LTWP windfarm talk about the projects in ways that differ, sometimes significantly, from ways that governments, developers, advocacy groups, and researchers do.

Given the widening push towards clean energy transitions, which is intensified by the COVID-19 pandemic and the call to ‘build back better’ (OECD, 2020), it is crucial to critically interpret the way people living near to largescale green-energy developments ‘see’ the situation (Lind et al., 2020a). This is reflected in growing literature on how populations at these geographic and political margins experience conflict and insecurity—or ‘security in the vernacular’ (Lind & Luckham, 2017; Vaughan-Williams & Stevens, 2016). Yet research is particularly challenging in settings like this, where the contestation dynamics are inextricably linked with the resource projects that divide local opinion. The following section discusses the methodological approach developed to navigate mistrust and local disagreements, and make sense of competing subjective perspectives, in the Ol Karia geothermal and LTWP settings.

Addressing the Methodological Challenges of Researching Local Subjectivities in Contested Contexts

Local subjectivities are well-recognised as crucial to understanding contextual nuances (Paerregaard, 1998; Reason & Bradbury, 2006), and avoiding the unintended negative consequences of largescale projects (Lind et al., 2020a). This is because people’s thoughts, feelings and beliefs influence how they interpret their social worlds, and, in turn, are the root of their attitudes and actions. Communities are not homogenous (Cooke & Kothari, 2001; Gujit & Kaul Shah, 1998), and each person will be differently positioned in relation to resource investments, with some benefiting and others not. The most suitable combination of methods depends on the specific circumstances and dynamics of the conflict that is being studied. The suite of different methodologies included in our ‘intersecting methodologies’ approach therefore purposely employed the distinctive contributions of their different epistemological foundations, as explained below.

Qualitative methodologies respond to the ontological assumption that there are multiple realities that are socially determined, and that epistemologically what we can know depends on our way of knowing (Yilmaz, 2013). Qualitative methodologies are suitable for generating in-depth understanding about people’s experiences from their perspectives (e.g. Hammett et al., 2015, p. 130). These subjective meanings grounded within a situation (Shotter, 2006) are crucial to explaining why people act as they do. In this research project, the qualitative methods of data collection included semi-structured interviews and focus group discussion of key issues with diverse and differently positioned informants, as well as ethnographic observation methods and oral-history interviews. Connected with the interpretative research tradition these methods are underpinned epistemologically by a close relationship between knower and known, or utilising researcher subjectivity in meaning-making (Lincoln and Denzin 2003). Interview methods involve deep listening and adapting questions and the direction of group discussions reflexively in response to what people say, which makes them suitable for understanding various grounded perspectives and responding to participant’s concerns in the conflict context. The participant-observation and historical interviews methods were important to generate reflections on everyday realities including what was not said, and to historically contextualise people’s experiences and viewpoints. They also provided the means for local research teams to develop interpretive capacities through comparing insight across methodologies to deepen understanding of the conflict dynamics.

Participatory methodologies are also forms of qualitative research, but they have a different epistemological focus. They respond to the challenges to researcher authority from feminist, post-colonial, and critical scholars who questioned whether people’s lived realities can be legitimately and objectively interpreted and represented by external researchers (Lewin & Shaw, 2021). The use of participatory and action-orientated methodologies has greatly increased as ways of conducting collaborative research that prioritises the agendas, knowledge and viewpoints of marginalised and previously unheard groups (Lincoln & Denzin, 2003).

In our research, we first interlaced the selected qualitative methods with accessible participatory activities as appropriate to learning from people in situ about their everyday realities (Bergold & Thomas, 2012). The particular participatory visual, interactive and action-learning methods we chose enabled participants to surface and then discuss their knowledge about local resources, agencies and actors. Historical timelines and livelihood trend line analysis (Irwin et al., 2015) provided insight on their perspectives of past and recent changes; and collaborative network mapping (Schiffer & Hauck, 2010) identified the connections they perceived between key stakeholders and their influence over resource project planning and implementation.

However, to avoid being extractive and to address suspicion (Khan Mohmand et al., 2017), these qualitative and participatory methods were incorporated within the meta-methodology of community-based participatory research (CBPR) (Minkler & Wallerstein, 2008) for the following reasons. As a type of participatory action research (PAR), CBPR intends to democratise knowledge production and build participants’ influence (Kindon et al., 2007), through involving them in interpreting their own situations. However, CBPR goes further than other qualitative participatory research approaches because the research methods are conducted, and participatory activities facilitated, by local people. Trust can be more easily fostered between these peer researchers and other community members than with ‘outsider’ researchers (Wallerstein et al., 2021), and, in turn, this can develop the honesty required for reliable insight (Reason & Bradbury, 2006; Wicks & Reason, 2009). As well as building local capacities, this process can strengthen local agendas, and, crucially, increase the likelihood that marginalised participants can mobilise their own knowledge and sustain action on an issue.

Participatory video (PV) was adopted as an umbrella approach to drive and mediate the CBPR processes, which means participatory video-led processes in turn encompassed the various methodologies discussed above. This is due to PV’s potential to generate trusting and enabling research contexts. Participatory video is a facilitated process, and itself incorporates other visual and performative methods such as storyboarding and participatory drama (Abah, 2021) to prompt dialogue and group sense-making (Shaw, 2021). The initial group-forming stage of participatory video centres on short video recording and playback ⁴ (viewing) exercises (Shaw & Robertson, 1997) to build participants’ confidence and capacities. Importantly, given the need to generate inclusive research contexts, video-structured turn-taking generates democratic space in which all group members can participate actively (Shaw, 2021). This backs up the facilitation approach that is necessary in participatory processes to counter inherent power imbalances and prevent dominance by the most influential stakeholders (e.g. Ledwith & Springett, 2010). This early participatory video stage also addresses the challenges of including the most marginalised viewpoints and fosters local ownership of the research processes—both aspects that are considered vital to successfully transforming trust, power dynamics and understanding when using PAR in conflict settings (Hyssop, 2021, p. 340). This important overarching methodological contribution is the reason we focus on the participatory video processes more than the other methods in this article.

At the same time there are practical tensions to navigate in using participatory video effectively and ethically, e.g. between the opportunity to bring neglected perspectives to public attention and the risk of negative reactions (Wheeler, 2012; Shaw, 2015, 2020). Thus, in order to ‘do no harm’ (Hyssop, 2021, p. 341), it is only later in our participatory video processes that participants created video stories or messages for external audiences (Shaw, 2021). At this later stage, participatory video makes another methodological contribution in prioritising and elevating the perspectives of those who are usually less ‘heard,’ which can productively change existing power relations. This is because marginalised local people are positioned more influentially in debates happening within the wider community through these video-mediated communication processes, and because screening of videos in which participants communicate compellingly about their experiences at events with duty-bearers ⁵ increases the likelihood of their views being listened to and valued.

This discussion of extended participatory video processes highlights the longer-term and iterative nature of our research engagement. However, as participatory video is mechanism for CBPR, as well as a methodological umbrella incorporating the other approaches, the discussion also clarifies how we sought to address the research challenges in these contested contexts. Peer researchers were considered more able to build trust and allay suspicions, the exclusion of vulnerable and marginalised local people was addressed through taking time to build trust and capacities, and allowing research communication processes to unfold for deeper understanding. Participatory video was used to enable those participating to drive their own agendas and shape the discourses about the green energy projects.

As summarised in Table 1, our starting input was to deliver intensive methodological training to the peer researcher teams in Kenya, covering, for example, the specific interviewing, focus group, participatory video, participatory mapping, and ethnographic activities they would use, as well as protocols for ethics and data storage and security. We then spent a week in each focal site accompanying the teams as they ran research activities with community collaborators, which helped embed the training in practices. Over 9 months, these teams then led consecutive 2-week programmes of research activity—the intersecting research processes (IRPs)—with different stakeholder groups.

Table 1.

Research Process in Seeing Conflict at the Margins.

Research phases	Activities	Time period
Team building, training and skills development	• Methodological training	3 weeks
	• Accompany peer researcher teams to Ol Karia and the LTWP project area
Research Phase 1	• Integrated Research Process – led by peer researchers in different villages and with various groups	Ongoing in bursts over 9-month period
	• Local partner support backed up remotely by IDS researchers
Mid-project workshop	• Collective analysis of research materials	2 weeks
	• Accompanied collaborative film-making
Research Phase 2	• Further IRPs in each site with missing stakeholders	Ongoing in bursts over 3-month period
	• Preparation of research materials and production of video outputs with remote IDS support
Research team preparation and presentation of materials	• Preparation for research dissemination video screening events	2 weeks
	• Academic seminar
	• Accompanied video-mediated engagements with Ol Karia communities
	• Research process evaluation
Research phase 3	• Video-mediated engagements within Ol Karia and LTWP project area and with external audiences	Ongoing over 3 months

Twelve IRPs were conducted in total with stakeholder groups including young people, women, elders, businesspeople, and pastoralists; and both those benefiting from the renewable energy projects and those who are excluded.

We anticipated the IRPs would be adapted in each setting by combining different combinations and sequences of research activities. However, we soon realised that interlacing the methodologies flexibly according to group was complex for less experienced facilitators. We worried about instrumentalising a process that should evolve responsively; yet the peer researchers needed more structure. This was the rationale for developing the specific activities in the IRP. The 2-week activity schedule encompassed participatory video exercises such as those focusing on building expressive confidence and capacities, exploring the environment, telling stories through visuals and storyboards; and participatory activities including the construction of historical timelines, resource maps and livelihood trend analysis. It also included interviews and focus groups with community members, and the ethnographic techniques of observation, listening and identifying meanings embedded in daily activities. The way the different methods were interlaced in the IRPs is elaborated in section 4, but Table 2 shows the activities in the first 2 days as an illustration.

Table 2.

Example of IRP Activities.

Day	Activity
Day 1 Basics	• Introductory participatory video exercises—to get to know each other, and build expressive confidence and capacities
	• Introductory participatory video exercises—to generate inclusive dynamic within the group and explore the local area and participants’ concerns
Day 2 Exploring local history using visual methods	• Participatory mapping—Local history timeline
	• Participatory discussion followed by PV edited statements exercise to mediate the participatory synthesis of key insights:
	a) What are the most important changes in people’s lives and livelihoods since geo-thermal/wind power projects began?
	b) What would improve life here and why?
	• Narrative methods—storyboarding and telling stories with visuals to build on earlier interactive research activities
	• Participatory video exercise—Shot-by-shot storyboard (interlacing through incorporating events from local history timeline)

In 2018, following IRPs with a range of groups, peer researchers from each site curated analysis of interview transcripts, video materials and personal reflections, to share at a mid-project workshop. Over 3 days, the wider research team completed various collective sense-making exercises to identify broad notions, deeper meanings and compelling stories that spoke to the bigger picture of contestation in each area. This set the stage for a second phase of research activity and accompanied video-making, including further IRPs with stakeholders missed in the initial work. The final research phase involved a series of screenings and facilitated discussions, both to reflect on findings with residents, and to extend research learning through exchange with wider stakeholders such as local government, company representatives and workers, activists and advocates.

In summary, we used methodologies that not only uncovered diverse subjective realties, but also brought the varying views on the roots and current drivers of conflict into conversation. Different local actors co-produced evidence, and reframed issues and action pathways together. Ensuing exchanges were analysed for insight on the nuanced ways of ‘seeing’ and the diverse claims underpinning the struggles. Dialogue with external audiences was organised not to simply validate the research findings but, instead, to highlight various subjective perceptions at the local level and reflect on how these different viewpoints might be incorporated to encourage both peaceful relations as well as better outcomes for communities from large green-energy projects.

Ways of Seeing—Generating Diverse Subjectivities and Deepening Knowledge Progressively

In the following two sections, we illustrate how participatory video as the umbrella methodology and overarching focus of this paper was interlaced with other methods. To inform this we draw on our research diaries and reflective team discussions, as well as interviews with the peer researchers and key community collaborators about their experience of the research process. In doing so, we highlight the contributions and challenges of the intersecting methodologies approach.

As explained, a key element of our approach was iteration (Wheeler et al., 2018, p. 11) or developing perspectives of conflict through sequenced research activities over time. This involved, first, layering different qualitative and participatory methods in the IRPs. Second, it involved phased collective analysis, knowledge synthesis and video production, as explained in the previous section. Finally, it incorporated video-mediated dialogues as the interactive context for generating deeper understanding between different perspectives, beginning within and across stakeholder communities in Ol Karia and the LTWP area, and then later through dialogue events with external stakeholders. Allowing time for this extended process to unfold was necessary to build trust as well as support the communication capacities of participants, as discussed earlier (Hyssop, 2021, p. 340; Shaw, 2020, p. 19).

Initially participants may lack expressive confidence and are unlikely to arrive at the project with fully worked-out perspectives. Groupwork phenomena mean people may say what they assume others want to hear and honesty on sensitive issues is unlikely before trust develops (Dentith et al., 2012; Shaw, 2020). Thus, the first step in the IRP was to generate safe interactive spaces and give time for relationships and deeper exchange to evolve. This is crucial to involving less influential community members. Notwithstanding the decision to organise separate groups for this reason, such as for women, men, younger people, or livelihood-related groups, inequitable dynamics often mean participatory processes are dominated by the most confident group members. It is essential to address this to ensure meaningful inclusion (Howard et al., 2018). Participatory video mediated this initial group-building stage as it is well-evidenced for building communication confidence, and establishing inclusive dynamics within groups (Shaw, 2017, 2020). As anticipated, the video opportunities also motivated involvement. Participants became familiar with each other and the research environment, and practised some presentational and videoing skills, through some light-hearted recording and playback exercises. However, facilitation skills more than production expertise is crucial for participatory video to deliver its promise (Shaw, 2017). The first challenge we faced was how to transfer important relational aspects to local peer researchers, alongside the nuts and bolts of the intersecting methodologies approach, given the constrained timescale for training and skills development. This was especially the case as they expected to learn about video production rather than facilitating others in participatory video-mediated interactions, a challenge evident in other participatory video processes (Shaw, 2021). Accompanying them as they practised facilitating research activities with other community collaborators helped. However, the input on interactional process rather than production skills had passed some by, and facilitation came more naturally to some peer researchers than others. Supporting women practitioners in the team, which was deemed vital for the research success, also disrupted prevailing gender dynamics and created tensions with some dominant male team members.

Preliminary generation of subjective knowledge was carried out through participatory exercises and group discussions, which can be part of both participatory video and qualitative methodologies alike. These exercises helped group-members express their tacit knowledge. For example, creating historical timelines and conflict trendlines enabled them to visualise change over time, and make manifest their current understanding. One peer researcher reflected—‘What’s really surprised me about this work is the knowledge that is out there in the local community’ (Mohammed ⁱ -LTWP team ⁱⁱ ).

Using these visual methods then supported a shift to intersubjective sense-making (Lewin & Shaw, 2021, p. 713) because the visible representations established common terms while providing a physical artefact for group-members to refer to as they considered meanings. As another peer researcher commented, ‘in a focus group discussion you get a wide range of views and opinion that carries a kind of weight’ (Mwangi- LTWP team—see endnote ii). However, it is crucial that the group are prompted to reflect critically together (Shaw, 2021, pp. 815–816; Mitchel et al., 2017, pp. 48–70). This is because the outputs produced can be distinct from the dominant understandings that many were socialised to accept unreflexively, or they can reproduce narratives at odds with people’s actual experiences, and the group need to unpack assumptions. It was the iterative process of deliberation, which was mediated by the sequenced and interlaced activities, that generated deeper contextual understanding.

The group discussions that followed activities were audio recorded as research data. Yet, it can be hard to access audio recordings quickly enough to inform the progression of research conversations, due to the sheer volume, transcription time, and need for careful translation. A specific example of interlacing methods to address this was using a participatory video exercise to synthesise a first-level group interpretation after each participatory exercise. Group members reflected on what they had learned and then took turns to record in-camera edited statements ⁱⁱⁱ to capture their key insights. Prompting questions structured this activity. For example, after the historical timeline exercise, each participant answered the questions: ‘What are the most important changes in people’s lives since geo-thermal/wind power was developed?’ and ‘What would improve life here and why?’ After group discussions on public attitudes to largescale resource investments, participants recorded an introductory statement (e.g. ‘We are residents of …. looking at how geothermal/wind power has affected our lives’), followed by two or three statements outlining key impacts/changes and why these had happened, two or three statements on the barriers to benefiting, and a conclusion. Edited statements—a basic participatory video format—created short summaries on video that could be more easily referred to as preliminary research synthesis or prompt for further discussion.

Our perception was the practice of intersecting methodologies, such as the sequencing of mapping methods, participatory discussion, and video-mediated synthesis described above, made logical sense to the local teams. People with no prior experience of either facilitating, or taking part in, group discussion or participatory video were able to move smoothly from a group exercise, into a focussed discussion on what they had learned, or interviewing one-another, and then capturing perspectives on video. The peer researchers as new facilitators also coped broadly well with the processual aspects of running activities, but observations and feedback illustrated organisational and facilitation challenges. They found it easier to work with women and elders, who deliberated extensively with enthusiasm. For example, one peer researcher thought that ‘women in the community - really appreciate the work …. because maybe they are not much involved in issues …. in the society’ (Iris - LTWP team-see endnote ii). However, the teams commented that the youth groups didn’t take things seriously and were reticent to express opinions, which they interpreted as them thinking discussions were not valuable. The peer researchers responded to their difficulties by engaging young people in making creative video dramas (Abah 2021) during and after the IRP processes. ⁶ For example, one drama told of job opportunities at the windfarm leading to women’s sexual exploitation. The dramas humanised some of the complex dynamics, but even with this approach the teams found it difficult to maintain young people’s commitment, which raised the need for more input on relational aspects of practice (see section 5).

Within the 2-week IRP programme of activities, after 4 days of groupwork to start the process (as above), peer researchers spilt up, with some conducting interviews, recording oral-history testimonies or carrying out ethnographic observations in the wider setting. In parallel other peer researchers facilitated the developing participatory video activities with the core groups. Learning from each activity across methodologies informed what happened next. To emphasise—learning during these processes does not have to be videoed to be valuable (Shaw, 2020). Going around with the equipment motivated exploration, but insights can come through unstructured encounters during filming such as when a passer-by asks what is happening. In this way, videoing activities opened possibilities for more ethnographic-oriented research as a form of methodological interlacing - ‘a way of being there, working alongside people, which gives me …. insight into what’s actually going on, with learning coming from the dynamics of the situation not just what people say or record’ (Shaw – researcher discussions ^iv ). Indeed, going around with the camera in a less structured way can be a way to invest time in building trust and the relational conditions for open conversations across the community, which is a way to address the problems of suspicion. For example, a peer researcher who joined the project midway, illustrated how positive community relations across divides can result (here, between elders and youth) not only within project groups, but because a project is visible. He explained how residents initially asked, ‘Why are you bringing that fear machine, do you want to sell our land?’ (Peter - Ol Karia team-see endnote ii). However, he became very trusted and popular as he worked to include everyone’s views. In turn he, as a young man, reported gaining enormously from learning about his own history and culture from the older women and men. Nevertheless, in these contested contexts building trust was inevitably much harder in some cases.

In addition to the in-camera edited summaries, during the second week of the IRP each group made a creative video. Many early videos recorded aspects of daily life, such as a marriage ceremony, livelihood activities such as fishing or collecting water from the village pump or highlighting and explaining the use of locally important vegetation. All of these can be interpreted as traditions that people want to maintain. Other creative videos were issue based: one Ol Karia group focusing on land degradation attributed to geothermal developments. Some early videos reflected social norms uncritically (e.g. women’s roles)—common at this stage (Shaw, 2021; Rogers, 2018). Considering participatory visual methods, Rose (2012) emphasises the importance of the makers’ readings of their outputs, and as with the maps and timelines produced, the intention is that videos materials are used to prompt reflective dialogue. Participants may also ‘speak back’ to their videos (Mitchel et al., 2017) by re-framing understanding or constructing alternative video narratives later, which is another knowledge deepening iteration. This is also an example of the interlacing of action learning and reflective dialoguing methods. Even though both are mediated by videoing processes in this case, they draw on different epistemologies or ways of knowing.

The main mid-project workshop activity was a collective analysis of the IRP learning so far, including the interview data, the ethnographic evidence, and the video content. The full research team—peer researchers from both sites, PDNK and FoLT partners, ourselves, as well as project advisors—were involved. Before the workshop, the peer researcher teams considered all the research generated and selected written materials, maps, photographs and videos to bring. Then to begin the Ol Karia and LTWP teams presented the ‘story’ of the learning from each focal area, alongside a collection of IRP videos to highlight key issues. We observed that the videos assisted peer researchers to present their central findings more systematically to the wider team.

Following this, research team members generated themes and identified meanings from the presentations/screenings, noting each theme on separate post-it notes. The two research teams that led the IRPs documented for the other team and not their own, e.g. the peer researchers for Ol Karia recorded possible themes and meanings on post-it notes for the LTWP site. The post-it notes for each focal area were then gathered and the two teams did an initial sorting, sifting and clustering of these themes, placing the post-it notes on very large mental maps (8 flipchart papers taped together) to create conceptual representations for each focal area. The teams then presented the clusters, which corresponded to larger themes or meanings emerging from the IRPs. For example, land loss and broken promises stemming from corrupt political processes came up in both settings. Positive and negative impacts of development were reflected, but conflict had increased between those assumed to have benefitted and the worst off (linked to ethnic divisions in the LTWP area). Maasai had been displaced to a fenced resettlement village, RAPland (named after the developer’s Resettlement Action Programme), and there was a sense of debasement in shallow cultural performances for tourists to replace traditional livelihoods.

The next step of collective sense-making involved the whole team responding with their own questions and clarifications. Then the teams worked to probe for deeper meanings and connections; and were also asked to think about the discrete issues raised from the perspective of diverse local-level residents, who might ‘see’ the situation very differently. These included school leavers in town, pastoralists, young mothers, elders, duka (kiosk) operators, businesspeople, CLOs (community liaison officers employed by companies), chairmen (in Ol Karia), people with disabilities, local elected officials and other duty-bearers (chiefs, sub-county administrators). They reflected on whose perspectives featured most prominently in the initial phase and which voices (subjectivities) were missing. This informed plans for a second round of IRPs in the next research phase.

This collective sense-making work lasted 2 days, with each team choosing key resonant themes and specific personal stories to illustrate it. To build the peer researchers’ visual story-telling skills, we then accompanied the two teams in video production following a simple model that they could use to generate further videos. For Ol Karia, the team settled on the theme ‘from hope to despair’ that emerged in RAPland. For LTWP, the team settled on telling the story of Sarima, i.e. the fenced resettlement village in the middle of the wind farm. The videos produced during the accompanied filmmaking aimed to reflect contrasting experiences. For example, the video Beyond Despair ^v contains comparative stories about the relocation of Maasai pastoralists to RAPland. One young man had a hard time initially but built a new successful business. However, a women remained despairing after her baby died, which she attributed to giving birth on the hard cold flooring of the new dwelling provided by the developer as part of her resettlement package. This enabled a further step in these local participants driving their own agendas and shaping the discourse.

The accompanied video production process worked well in Ol Karia (Beyond Despair), even though we, as white academic outsiders were present alongside the local peer researchers. It helped that we collaborated during this visit with residents who we had met and worked with during the initial fieldwork, and who had been involved subsequently in IRPs. In contrast, this process stage was far from ideal near the LTWP windfarm.

The area around the windfarm is ethnically diverse. In Loiyangalani, the largest town in the region some 40 km from the wind farm, the team expressed satisfaction that Turkana, El Molo, Rendille, and Samburu women happily exchanged experiences in the same group. However, it was particularly important to run an IRP in Sarima, the settlement situated within the windfarm and thus most directly affected. However, conflict between Turkana and Samburu groups worsened during the research project time period, and this made it impossible to organise this IRP safely. Consequently, the LTWP team asked us UK researchers to accompany their collaborative video-making process in Sarima. Peer researcher Iris was frustrated that Sarima residents mistrusted the research intentions, although it was not surprising given there had been no chance to go through the intended trust-building process. We did actively involve five residents, but the video research here felt extractive because it was not based on the long-term relationship-building that occurred elsewhere in the area, and the tension in this uncomfortable and underserved place was palpable. Despite this, Iris reported that when the video was taken back to the community they understood the process, and ‘after talking they came on board’ (see endnote ii).

The UK and local researchers and group members all have worldviews which are hard to escape, and we aimed to unpack assumptions and advance debate. Through the accompanied collaborative video-making in Sarima, we also confronted what generating subjectivities really means. This settlement in the middle of the windfarm had been relocated 600 m away from the road to protect villagers from the dust from construction vehicles and reduce the danger of accidents. Community members were consulted about the village layout, the fencing and the placement of manyattas, or traditional homes, which were constructed for residents under the like-for-like terms of the resettlement. A community store, toilets and washrooms, an elder’s shelter, a classroom and teacher’s quarters were also built as part an action plan monitored by external third parties. Yet, in the short participatory video engagement, residents complained about being contained within a fenced area after relocation and living in manyattas clustered close together. As outsiders we might well doubt that the developer had promised to build a hospital as was claimed in the Sarima video, as healthcare facilities and services in rural Kenya are a state responsibility (in theory) rather than the responsibility of outside investors. Yet, the purpose of the videoing processes was to surface these contradictions, and the video materials were then used to prompt critical reflection on the inconsistencies between diverse positions and what they mean. When playing back and discussing the video with Sarima residents later in the research process, it was evident that residents’ views in the video conveyed their psychological feeling of containment and of broken promises. These subjective communications resonated with a lived experience that remained deeply precarious, even after the LTWP project was completed and delivering energy to Kenya’s national grid, and revenue for the investor consortium.

It is easier to produce a compelling message on video if the narrative is simple and coherently voiced, rather than complex incorporating a wide range of different views. Although different stakeholder groups were run at each site, this means there was a danger of constructing video narratives that neglected critical nuance and fine-grained difference (Wheeler et al., 2020). Originally we anticipated producing one or two videos summarising the overall research insights, but in response to this risk we decided it was better to try to maintain diversity given the research purpose in revealing the range of local subjectivities. Following Mistry and Shaw (2021), we propose participatory video can evolve social and political dialogue in conflicted contexts, because it contributes to generating agonistic pluralism (Mouffe, 1999). This involves recognising and exposing tensions inherent between opposing points-of-view, rather than unproductively pursuing consensus, which erases differences that are crucial to ‘seeing’ ways to build more peaceful coexistence (Mistry & Shaw, 2021, pp. 199–202). Therefore, a larger range of video outputs for wider audiences were made to maintain agonistic pluralism, by showing the multiplicity of experiences in these contexts. In addition to the accompanied videos (In the Shadow of the Turbines, A Windfall…Who Benefits? and Beyond Despair,^7,8 and the videos produced by participants during the 12 IRPs, there were also a range of ‘insight videos.’ These brought together edited statements from different locations on different themes that the windfarm development was thought to engender, such as compensation, land and ‘social evils’ (prostitution and sexually transmitted disease), and political relations and processes. ⁹ These outputs were screened consecutively at video-mediated events and collected side-by-side on the project website ^vi to generate further thought and discussion.

As a final, iterative step as part of the intersecting methodologies approach, the materials produced were then used to widen the dialogue horizontally and vertically. The teams showed the videos to audiences across each focal area - outwards to other communities facing similar large-scale resource investments, and upwards to duty-bearers, policy makers, investors and academics. Each IRP had generated large amounts of evidence from differently positioned stakeholders in these places, and this informed the progressive research interactions. Bringing the diverse perspectives into conversation both through these video-mediated engagements across the community, and during systematic thematic analysis later, developed understanding of the diversity of subjective perspectives.

Lessons From Navigating the Methodological Challenges in Context

In this section we briefly reflect on the some of the challenges the research team encountered in implementing an intersecting methodologies approach.

The first was the intricacies of the approach. The very aspects that make it well-suited to exploring diverse local subjectivities—that it is interlaced, layered, iteratively progressive and responsive—also point to the challenges of putting it into practice. These were amplified by having to support the peer researcher teams from afar through most of the process. At the outset, we had intended ongoing team mentoring would be provided by partner organisations in Kenya after the initial 2-week training and accompanied practice with us. However, the young peer researchers needed considerable support in navigating the project dynamics alongside the inherent complexities of the intersecting methodologies approach. It became necessary for us as UK research leads to provide sustained backstopping support, and guidance on relational aspects became rather subsumed by technical needs.

Related to this, the second challenge was in balancing the development of video production capacities with the facilitation skills needed to support participants’ sense-making. In projects with a core visual component, there is always a tension between making outputs to communicate research findings and using visual processes to mediate the generation of research knowledge and discussion of insights with diverse stakeholders. In these contexts, where precarity characterises life and people come and go, this was compounded. For example, all those peer researchers initially trained from the LTWP area accepted other work opportunities and left the project immediately afterwards apart from the team leader. He then had to find others and train them himself. He was new to the methodologies, and there was inevitably less emphasis on participatory video’s relational contributions as he was excited about the video-making.

A third challenge was balancing the composition of the peer researcher teams, given that ethnicity, gender and age were amongst the critical social differences shaping different ways of seeing. In the LTWP area, despite efforts to recruit peer researchers from all ethnic groups in the region, the team encountered perceptions that it was biased to the interests of Turkana. Given tensions between the Turkana and Samburu in the region, it was vital to run an IRP in the Samburu village of South Horr. Some peer researchers in both areas also reported that elders did not want to or should not be expected to use equipment. Whilst it is important to respond to the cultural context, Shaw’s previous experience suggests elderly participants in Kenya can greatly enjoy videoing opportunities when appropriately encouraged and supported.

Finally, the fourth challenge was navigating some local resident’s expectations that research participation would bring immediate rewards. Residents in Ol Karia and the LTWP area are accustomed to researchers and consultants visiting and are socialised to an extractive, rather than participatory knowledge economy. IRP participants who committed 2 weeks over full days, often leaving work and childcare to be involved, were compensated for time and travel. Yet, assumptions about further renumeration was an ongoing challenge for the peer researchers. In the LTWP area, Mohammed reported that, on seeing the camera, frequently the first thing people did was ask for money no matter how limited their engagement. He attempted to challenge perceptions by explaining the project’s local importance in including and communicating externally resident’s diverse viewpoints. The same challenge arose in Ol Karia, but Eunice related that over time the wider community took more project ownership, and called for her to come when critical events occurred.

In both sites, despite the challenges, the teams generated much qualitative evidence including a remarkable amount of video material reflecting diverse subjective perspectives. They also developed practical appreciation of the processual value of participatory video and video-mediated engagement for generating new dynamics and grounded knowledge, with facilitators emphasising the value of bringing local perspectives to government and developer representatives.

Conclusion

In response to the challenges of researching subjective perspectives in contexts of struggle and contestation, such as mistrust, political manoeuvring and unequal power dynamics, this article has reflected on using an ‘intersecting methodologies’ approach to explore resident’s experiences of large-scale green-energy developments in Kenya. Specifically, our focus was to use methodologies from different epistemological traditions in ways that would uncover and bring into conversation different subjectivities and perspectives. This resonates with the idea that conflict, contestation and violence have diverse embodied and externalised manifestations that it is impossible to wholly comprehend using one methodology (Santamaria, 2021, p. 97)—especially emotional and psychosocial aspects that may be hard to narrate coherently (Hoang, 2021, p. 944). Our work therefore contributes to emerging thinking on how to access these different kinds of knowledge and read across them by using ‘kaleidoscopes’ of methods at different project stages (Santamaria, 2021, p. 959). However, we extended the kaleidoscope approach practically, as we did not apply multiple methods concurrently but separately; neither did we integrate them to create new forms. Instead, we weaved methodologies together so that they were somewhat separated, either within an exercise, group session, or the IRP fieldwork process. We also brought insights from each into conversation throughout the situated engagements in Ol Karia and the LTWP area as the processes evolved. Importantly, CBPR, visual methodologies, and the interpersonal and observational methods involved in traditional qualitative research, are different epistemologies, which offer different processual contributions and knowledge lenses. Interlacing these different ways of knowing did utilise the strengths and offset the weaknesses of each (Khan Mohmand et al., 2017).

Recent literature suggests that PAR processes can transform relationships productively in conflict settings, but that navigating the tensions requires a creative, iterative approach guided by a practical understanding of power dynamics during engagement (Hyssop, 2021, p. 340). Using CBPR to involve peer researchers without NGO or research backgrounds in leading research activities was important in countering extractive research practices and shifting local power dynamics (Wallerstein et al., 2021, p. 661). It also fostered trust with and within the participant groups, despite the challenges, which is crucial to effective engagement in conflict settings (Hyssop, 2021, p. 341). This was aided by the long-term processes - necessary to trust-building in such contexts (Santamaria, 2021, p. 970).

The iterative nature of the learning processes was vital (Hyssop, 2021; Shaw, 2021; Wheeler et al., 2018). For example, creative visual activities within sessions first surfaced embodied and tacit knowledge, and then enabled residents to stand back from their lived experiences and deliberate about complex meanings through participatory dialoguing approaches (Lewin & Shaw, 2021). Insight was developed progressively as participatory evidence was brought into conversation with that generated through qualitative methods (interviews, focus groups, and ethnographic diaries); both informally as peer researchers switched activities throughout the IRPs, and in a more structured way during collective analysis processes. Ultimately this deepened analysis because it prompted critical introspection, as well as situating recent events in longer processes of historical continuity and change. The different epistemologies and traditions not only operated effectively side-by-side. Interlacing the methods so that what was learned from one informed the other generated a constructive tension: both in revealing contrasting subjective experiences, and in enabling the diverse ways of ‘seeing’ the resource projects to speak to each other productively, and thus drive the conversation forward. This is what de Sousa Santos (2010 in Santamaria, 2021, p. 957) terms a ‘dialogue of knowledges.’

Thus, the significance of what we learned in applying an intersecting methodologies approach is that the key in mixing methods is not so much to iron out and reconcile differences between very different methods. Rather, it is to craft and guide processes that accommodate activities and practices emanating from different traditions, often happening separately but still proximate enough to inform the other. This in turn deepens understanding of multiple and sometimes diverging subjective viewpoints.

Our experiences also contribute to wider thinking about how to work in conflict. Participatory methodologies were used primarily to build inclusive communication space and to mediate the unfolding dialogue, but there were also small shifts in the local dynamics as participants confidence and capacities grew. This is a key contribution of PAR processes in contested and insecure contexts (Hyssop, 2021; Shaw, 2017). When marginalised groups generate their own understandings there are inevitable tensions as this disrupts local power relations. Yet, these tensions are not only unavoidable, but a source of learning, with another contribution of the intersecting methodologies approach being making differences visible through the different approaches. This enables intercultural translation towards cross-interest understanding (Santamaria, 2021, p. 957), for instance across different ethnic communities, or outwards to external audiences. Our work to support participants to make video stories and messages aimed at different audiences supported this agonistic pluralism (Mistry & Shaw, 2021).

Overall, the project reenergised local involvement in discussions around resource investments. The video-mediated engagements with external audiences did not progress as far as intended during the project lifetime, but since then local teams have brought grounded interpretations to external spaces using the exhibition and video materials. This changed the dynamics of the conversation, as it has re-positioned marginalised residents more influentially. We attributed these successes to the undisputable local ownership that developed, due to the CBPR approach and particularly the energy engendered by participatory video. As Eunice reflected:

‘They owned it. A big section of the people have embraced the project, have really welcomed us. …. I think they’re seeing this project as a chance or an avenue …… to let the world hear their concerns’ (Ol Karia team – see endnote ii).

In response to Hyssop’s (2021) assertion that conflict transformation most fundamentally requires that the capacities, knowledge and collective agency of local people is built so that they can lead pathways to peaceful coexistence themselves, this is an enduring legacy of the project.