The data consists of qualitative interview notes in English for 246 interviews: 111 interviews across four locations in Sri Lanka and 135 interviews across four locations in Nepal. The locations are comprised of the respective districts (three per country) in which the study was undertaken, plus the capital cities of the two countries (for national data-mapping). 13 of the Sri Lanka interviews and 20 of the Nepal interviews are with the brokers who were the focus of the study. Political economy critiques of the mainstream literature on statebuilding and post-conflict reconstruction, highlight its ahistorical, functionalist and technocratic orientation (Cramer, 2006; Migdal et al; Schlichte, 2005; Roxborough, 2012; Pugh et al, 2008). This critique emphasises the historically-divergent and contested trajectories of state formation/building, and the importance of studying the state as it actually exists rather than as an ideal type. The corollary to this is the need to disaggregate the state through coalitional analysis, to appreciate the role that coercion and the distribution of rents play in shaping political (dis)order and the critical importance of informal networks, brokers and power relations that underpin formal state structures and institutions (North et al 2012, De Waal 2009, Mac Ginty 2010). However, important though these political economy insights are, they rarely deal explicitly with questions related to the territorialisation of power, or the spatial dimensions of scarcity, abundance and dependence (Le Billon, 2012). The implicit spatial assumption is that post-war statebuilding and development involves the creation or rebuilding of institutions at the centre, followed by the diffusion or radiation of power outwards to the margins of the state. This research challenges this narrative, drawing upon insights from political geography, political ecology and border studies which examine the interactions between territory, space, scales, resources and political processes (Agnew, 2008; Jessop et al, 2008; Korf et al, 2009; Nugent, 2003; Paasi, 2011; Watts, 2004). The research focuses on the specific histories of conflict and post-war statebuilding, the networks, brokers and institutions which link centre and periphery, and the evolving geographies of war to peace transitions. From this perspective, the margins are not merely reflective of power relations at the centre, but may actually be constitutive of those power relations (Scott, 2009; Goodhand, 2008, 2013). The research aims to open up the black box of subnational governance and study how different kinds of peripheries and differing brokering relationships may define institutional arrangements and political processes at the national level. The margins may be crucial arenas which determine how peace is built in the aftermath of war. The research consists of a structured, focused comparison of the spatial dynamics of war to peace transitions in the borderlands/frontier regions of two conflict-affected states in South Asia - Sri Lanka and Nepal. The comparison between two different types of peripheries - borderlands, which span an international border as in the case of Nepal, and internal frontiers which lie on the margins of an island state as in Sri Lanka - will be particularly fruitful in revealing differing dynamics of conflict, brokerage and post-war consolidation. These cases also represent two contrasting post-war settlements: while Sri Lanka's war ended with a military victory leading to a victor's peace; Nepal's war concluded with a negotiated settlement and a subsequent period of protracted bargaining between the centre and borderland regions. These contrasting characteristics will help us to develop an understanding of how different contextual features shape the role of borderlands and brokers in post-war transitions, and broaden the applicability of these findings to a wider set of cases in South Asia and beyond. The research will have three strands (national mapping study, district-based studies, and programme/broker studies), which will allow us to capture different levels of analysis and explore the connections and pathways linking structures, institutions and agents. In order to shed light on the complex international/national/local interface, the research deploys a multi-sited research design that mixes methods, including interviews, surveys, life histories and historically-informed contextual analysis.

The research consisted of three strands (national mapping study, district-based studies, and programme/broker studies) which allowed the study to capture different levels of analysis and explore the connections and pathways linking structures, institutions and agents. In order to shed light on the complex international/national/local interface, the research deployed a multi-sited research design that mixed methods, including interviews, surveys, life histories and historically-informed contextual analysis. In Sri Lanka, interviews were carried out with participants in the capital (Colombo) as well with participants in three districts. In Nepal, interviews were carried out in the capital (Kathmandu) as well as with participants in four districts. Data for each country is bundled into five zip files: broker interviews, 3 x district-level interview files, national interview files. Therefore, the data for the whole project is contained in ten zip files. The data is accompanied by an interview list 'map' that has anonymised metadata for the whole collection. Some interview notes consist of full interview transcripts; others are in bullet point or summary form; others are detailed notes taken during the interview by the investigator/interviewer.