The most fundamental limitation of PSA is the fact that it takes the nation state as its sole frame of reference. The underlying conceptual framework is one of elites bargaining at a national level within a territorially defined state. This reinforces the statist approach adopted by development agencies, and is reflected in the way that the development industry organises itself, including the division of the world into country teams, national planning and budgeting processes, statistics aggregated at the national level, and the location of country offices in capital cities – all of which limit understandings of borderland dynamics.

This nation-state framework of analysis is problematic for three reasons:

First, it underplays international and regional dimensions of political settlements. Domestic elites’ strategies to secure their interests are often oriented outwards, particularly in regional conflict systems in which violence, networks and flows (of weapons, goods and people) operate across borders. The political survival of national elites in such circumstances depends upon capturing transnational resources, building alliances with external patrons and mobilising cross-border political or religious networks. Domestic political settlements are thus often heavily shaped by neighbouring states. Myanmar’s political settlement, for example, is deeply influenced by the country’s relationship with China, while in East Africa the domestic political settlements within Kenya, Somalia and Ethiopia are all interconnected. Power dynamics do not fit within the ‘container’ of the state but are intimately shaped by transnational power structures, networks and flows.

Second, PSA underestimates the importance of subnational bargaining processes. These often differ significantly from those at the centre but play a key role in shaping national-level political settlements. Greater understanding is needed of the specific challenges that borderlands pose to ruling elites, including: histories of weak state control and contested legitimacy of state authority; and the challenges of co-opting borderland elites into national coalitions when the availability of cross-border sources of weapons, revenue and support give borderland elites significant power and disruptive potential. In this sense, border regions can be understood as ‘special political zones’ that frequently occupy a disproportionate amount of the attention of ruling elites and where repeated challenges to the overarching political settlement are likely to arise. PSA does not provide clear entry points for engaging with conflicts where relatively stable political settlements at the national level co-exist with high levels of subnational borderland violence.

Third, PSA is rarely attuned to the significance of shifting centre–periphery relations in post-war transitions. It assumes that once order is established at the centre, this will provide the foundations for peace throughout a country’s territory.

In other words, political order radiates outwards from the centre into unruly peripheries. However, many conflicts emerging from the state’s margins are driven by contestation for control of borderland regions and longstanding grievances against central state authority (often linked to ethnic, religious and linguistic differences). Post-war bargaining therefore revolves around questions of political representation and inclusion/exclusion, distribution of resources, and access to services and government positions. In many post-war countries there may be a level of stability and settlement at the centre alongside ongoing conflict and ‘unsettlement’ in borderland regions. For instance, in Nepal in 2006, as discussed elsewhere in this publication (see p.48), there appeared to be a broad and inclusive settlement forged in Kathmandu, but this was not accepted in parts of the Tarai – the southern plains region bordering India, where two-thirds of the population identify as ‘Madhesi’ – leading to violent contestation.

There is a need to bring analytical frameworks that focus on power, institutions and resources more explicitly into conversation with approaches that deal with space, place and territory. Taking the state margins as the starting point from which to understand processes of state contestation, fragility and development addresses a number of key weaknesses in how PSA is being used in peacebuilding policy and development interventions, by:

1. clarifying the drivers and dynamics of borderland violence

2. providing tools to analyse the agents and dynamics of change in borderland regions

3. emphasising the importance of the ideas and beliefs of borderland communities to the dynamics of war-to-peace transitions

We go into greater detail of how these operate below.

PSA adopts a reductionist view of violence, viewing it as a tool used instrumentally by actors (invariably elites) to re-shape or protect political settlements. This framework does not capture the varied causes and functions of violence in conflict-affect countries, how violence is mobilised and constrained by traditions, beliefs, norms and ideologies, or the ways in which it can remain central to the post-war order even after a political settlement has stabilised. As outlined above, a state-centric PSA framework assumes that elite agreement at the centre creates the foundations for re-establishing order in unruly borderlands – overlooking why borderlands can become important sites of contestation, and the specific challenges they pose for stabilising violent conflict.

In order to better conceptualise the varied causes and functions of borderland violence, we draw on research by Cheng, Goodhand and Meehan to distinguish between three broad types of violence: competitive, embedded and permissive violence.³

• Competitive violence occurs between warring elites to contest the distribution of power in society, and is the kind of violence prioritised in PSA.
• Embedded violence is entrenched in how a political settlement works. The privileges elites gain by committing to a political settlement are not only economic (e.g. control over certain resources, import licences) or political (government positions) but also include the ‘right’ to use violence. These ‘violence rights’ determine who has the ‘right’ to enact violence, upon whom, for what reasons, and with what level of impunity. Embedded violence can have distinctly gendered dynamics and in many conflicts – notably El Salvador and Guatemala – wartime strategies of gender and sexual violence became embedded in the peacetime tactics deployed by security forces to enforce deeply inequitable forms of post-war order.
• Permissive violence relates to activities, such as forms of criminal violence, that occur in areas where the state lacks a monopoly of violence, but which neither challenge the political settlement nor become embedded in how it works.

None of these forms of violence are unique to borderlands, although they may be distinct in such regions. Addressing forms of competitive violence can be especially challenging in borderlands since these regions are often central to processes of statebuilding and economic development and zones where state authority is heavily contested. Borderlands are often also valuable sites of cross-border trade, especially where different systems of regulation and commodity valuation heighten the exchange value of goods on different sides of the border and make cross-border trade especially profitable. This can increase competition for control over cross-border networks and flows, especially in countries such as Myanmar where border regions are themselves the site of lucrative resources.

Borderland regions also provide practical advantages for those challenging state authority. Cross-border spaces can offer protection from government attacks, either through arrangement with neighbouring governments or with other armed groups operating in the margins of neighbouring states. The supposed inviolability of international borders limits state authorities’ efforts to curb activities beyond their boundaries. Attempts by governments to co-opt borderland elites may be particularly difficult in contexts where borderland elites are able to access cross-border support systems, strengthening their autonomy and negotiating power. For example, in Afghanistan and Tajikistan, central governments have struggled to co-opt borderland elites whose power is derived from their control over illicit cross-border economies.

Forms of embedded violence can be particularly pronounced in contested borderlands where government attempts to gain control are reliant on coercion and violence, and lead to suspending rather than extending the rule of law. Such responses often create forms of protracted cyclical violence. In Colombia and Myanmar, attempts by central governments to extend authority rely on alliances with paramilitary-style organisations. In these contexts, stability is less about bringing an end to violence than franchising out violence to secure control over contested territories. In some cases, stability at the centre may rest on agreements that tolerate or even exacerbate violence in borderland regions. In Myanmar, the rampant extraction of and exploitation of resources – especially timber, jade, drugs and land – in the country’s borderlands underpins the political settlement that has emerged. Bargains between military elites, national and transnational business elites, and in some cases leaders of non-state armed groups, have stabilised armed conflict in many areas but have subjected borderland populations to violent dispossession of land, environmental destruction and the negative consequences of illegal drugs.

Disaggregating forms of borderland violence points to the need for a systemic reappraisal of current conflict resolution and peacebuilding policies. In particular, it emphasises the need to understand how violence can become an important component of post-war state consolidation and economic development, rather than being caused by the absence of the state or the economic marginalisation of borderland regions. This warns against the assumption that peacebuilding, economic development and the expansion of state authority are necessarily mutually reinforcing and emphasises the need to understand the trade-offs that often surround these policy goals.

PSA tends to be based on a structuralist understanding of political change and development, and struggles to make sense of the dynamics of change and sources of agency within political settlements. Political brokerage provides a lens to understand the shifting dynamics of political settlements. Borderland brokers are the go-betweens, gatekeepers or representatives that span spatial divides between competing elite coalitions or connect political elites to their constituencies. They seek to occupy and monopolise a ‘deal space’ – a point of friction and an interface – which links the centre to periphery, the (trans)national to the local. The dynamics of brokerage and the nature of the deal space are shaped by three key factors: timing, space and scale.

First, the ‘deal space’ is usually extremely time-sensitive. As noted by Bell and Pospisil (2017), periods of post-war transition are often characterised by periods of protracted ‘unsettlement’ and include moments of rupture when new rules of the game are renegotiated. These periods of flux create an opening and a demand for actors that can mediate between different levels, spaces, and social and institutional boundaries. For example, in post-war Nepal, Madhesi elites in the Tarai borderland mobilised against the new constitution, making demands for more substantive federalism. A new deal space was opened up by a violent movement in the Tarai, which was followed by an economic blockade that Madhesi brokers were integral to. Conversely, over time, the space for brokerage may close down as new power-sharing agreements are forged or the central state establishes a stronger foothold in previously ‘unruly’ borderland regions.

Second, brokerage is shaped by the distinct characteristics of each borderland space, including the degree and form of institutional and social hybridity. In the post-war period, the (re)negotiation of centre–periphery relations shapes formal debates related to constitutional change, transitional justice and economic development, as well as the informal bargaining linked to the distribution of rents and political positions. This is an uneven and ‘ragged’ process – some borderlands are more salient to the central state than others, which means that some brokers have greater or lesser significance. Political brokerage therefore differs according to the spaces and ‘synapses’ that brokers occupy.

Apex brokers are from borderlands with high salience. They constitute the spine of a political system, linking the centre to core coalitions and constituencies. They have privileged access to key figures in the central state, to major on-budget or off-budget resources and to crucial sources of information and intelligence. They may have a major role in the use of ‘competitive violence’ to enforce or renegotiate the terms of the political settlement. Nangarhar province in the eastern borderlands of Afghanistan, for example, with its powerful tribal structures and strategic location on the Pakistan border, has always been home to apex brokers who could make and unmake national political settlements. On the other hand, tertiary brokers are located either in less salient borderlands or they broker relationships within borderland regions rather than directly with the central state. They facilitate the circulation of power, ideas and resources in spaces that do not determine the overall stability or otherwise of the national political settlement; here, ‘permissive violence’ can feature without being a significant concern to ruling elites.

Third, as well as connecting different spaces, brokers operate across and frequently jump between different scales. For example, those involved in illicit economies circumvent the central state and directly ‘plug in’ to regional and international markets. Gaining access to international aid and connections may open up the deal space of borderland brokers vis-à-vis the state. These brokers are therefore not constrained by the ‘national order of things’ and, though they may by locally embedded, they operate in a regional and international environment.

A further criticism of PSA is its reductionist analysis of elite interests and incentives. Elites are assumed to be driven by the pursuit of wealth and power, and political settlements are presumed to be the result of conflict and negotiation over material resources. This understates the foundational role of traditions, ideologies, beliefs, cultural norms and notions of legitimacy. Although none of these issues are unique to borderland regions, they do have important spatial dimensions.

Cultural norms and belief systems are inseparable from notions of place, space and territory, and frequently border regions are at the nexus of clashing or incommensurate world views and belief systems. On the one hand, national identities and state-based discourses about citizenship may not resonate with or may be weakest in borderland areas, where the history of state authority has been limited and contested. For example, the ethnic minorities who occupy borderland regions in Myanmar, or the Madhesi population in Nepal’s Tarai region, have alternative histories, visions and cultural practices that clash with central elites’ efforts to forge exclusivist national identities and ideologies of rule. Conversely, forms of extreme nationalism and ethnic chauvinism may flourish in the periphery, as for example Sinhala nationalism in Sri Lanka, which emerged from the state’s southern periphery and was mirrored in turn by the emergence of Tamil nationalism from the north-east. The nationalist political imagination has depended on the constant invocation of border threats and dangers to the territorial integrity of the nation.

Borders, as well as being containers of nationalism, are also conduits of transnationalism. People, ideas and commodities cross and challenge the border, and ideologies and visions of security and development do not stop at international borders. For example, diaspora communities may pursue alternative visions of post-war reconstruction that challenge state-based narratives. On a much larger scale, China’s vision of security and development extends well beyond national borders into the borderlands of Myanmar, Laos and beyond.
In the post-war moment, when the political settlement is being renegotiated and people’s understandings of the world around them are in a state of flux, there are heightened opportunities for brokers to mediate across competing narratives and fields of meaning – the deal space is not only about negotiating rents, but also about ‘translation’ and sense making. For example, when the Taliban came to power in Afghanistan in 1994, it drew upon local clerics, religious education, shari’a law and rural village culture to mobilise support and draw key constituencies in the eastern borderlands into the new political settlement. This is illustrative of the fact that mobilising legitimacy (as well as capital and coercion) – particularly in strategically important peripheral regions – may be key to establishing stable political settlements.

Similarly, the power of borderlands elites is linked not only to their access to resources and the means of coercion, but to their ability to represent and vocalise the demands and beliefs of borderland populations. This is especially important for understanding inclusion and exclusion at the margins of states. PSA concentrates mostly on the importance of ‘horizontal’ inclusion – i.e. inclusion of competing elites – to stabilising violent conflict, but it provides little scope for analysing the dynamics of ‘vertical inclusion’ – i.e. inclusion of the interests of non-elites.

Exploring the role of ideas, ideologies and beliefs also draws attention to factors that mitigate or exacerbate exclusion at the margins of the state, such as whether borderland elites can act independently of the populations they claim to represent, or whether their power is conditional upon delivering certain promises or services. For example, political negotiation, brokerage and ideologies can become hyper-masculinised in conflict-affected borderlands in ways that marginalise women from decision-making and ensure that their interests are not an important factor in shaping how elites mobilise support and compete for power. However, the need for borderland elites to offer credible alternatives to state authority can also heighten the importance of service delivery – such as health, education and justice – as a source of legitimacy for borderland elites. This suggests that external peacebuilders need to better understand the vernacular of local politics and in particular local understandings of legitimacy.

Borderlands, particularly those affected by conflict, are sites of contestation but also constant negotiation. Maintaining, containing and securing borderlands may benefit some constituencies to the detriment of others, and produce violence, some forms of which may be less visible or more disguised than others. A gendered borderlands lens focuses attention on specific identities (masculinities, femininities, sexuality, race and class), and how these intersect to shape processes of inclusion and exclusion. It is also a tool for understanding how borders and associated structures and institutions in borderlands are used to maintain power and gendered inequalities.

Therefore, ‘gendering borderlands’ helps increase understanding of the distinct relationships that diverse social groups have to the border, as well as the kinds of movement the border enables – or disables. This raises important questions about who has control over the border – who has access to movement, and who is contained and imprisoned by it. It also looks at how and why certain groups – such as rural women and men, ethnic minorities and young people – and power relations are left out of national development and peacebuilding processes in borderlands. A gender perspective highlights the fact that the ‘international borderline’ is only one component of a network of different types of boundary. Gendering borderlands reveals the multiple physical, virtual, legal, personal and political boundaries that diverse groups and individuals experience and negotiate as part of their lives on the margins.

A gendered borderlands lens also highlights resistance – how those excluded at the margins resist agents and systems of domination, and the spaces that open up to transform structures that perpetuate the exclusion of particular groups.

* With contributions by Dr Mandy Sadan (Reader in the History of South East Asia, SOAS)

PSA has helped develop a more rigorous political economy analysis of the drivers of violent conflict and the trajectories of post-war transitions. However, the nation-state spatial framework that underpins PSA has limited the insights it can provide on borderland violence and post-war transitions. Addressing this analytical gap involves thinking about the interconnections between power, space and time – which do not generate a simple set of policy prescriptions. To some extent it reinforces what is already known to be good practice: taking context and history seriously, and understanding power relations. And perhaps its chief value to policymakers is to provide another analytical lens – along with several others, including gender, conflict and the environment – that can be deployed in contexts where borderland dynamics are a significant factor. This should lead to more targeted, contextually attuned policies, which are cognisant of processes on both sides of the border.

A borderland perspective leads to a set of questions with valuable implications for international peacebuilding practice:

To what extent do formal structures and institutional arrangements align with existing configurations of power?A borderland perspective focuses explicitly on the spatialisation of power and how political settlements have subnational and transnational dimensions. Although international actors have neither the capacity nor the legitimacy to micro-manage political settlements or empower borderland elites, they do need to better appreciate underlying power relations and their spatial dynamics and the vernacular and idioms of local politics. Interventions can perhaps create the conditions for more productive ‘conversations’ between states and borderlands – or at the very least not create disincentives for such conversations to take place.

How do brokers influence the relationship between centres and borderlands? How can brokerage arrangements promote security and, in the long-term, support more progressive and inclusive post-war orders? Engaging with these questions does not mean fixating on finding ‘good’ brokers to support while avoiding ‘bad’ brokers. Rather, it should be based on an understanding of the environments brokers work in. This provides a starting point to explore how interventions can influence the incentive structures of brokers to reduce the use of violence as a negotiating tool, and how service delivery – including to marginalised groups – can become a more important foundation for power and legitimacy.

What are the trade-offs between different sets of policy goals and interventions? A borderland perspective calls into question several mainstream assumptions, including that: statebuilding and peacebuilding are synonymous with each other; extending the state footprint into borderlands will bring peace and stability; economic integration will reduce insecurity and poverty in border regions; and promoting good governance will help stabilise borderlands. There is therefore a need for more conscious deliberation on the trade-offs between different goals, and who bears the costs of various interventions.

Is borderland insecurity generated by policy regimes and decision-making in metropolitan centres? A borderland perspective exposes the links between insecurity and poverty in borderland regions, and stability and prosperity in metropolitan centres. Therefore the ‘pathologies’of the margins are generated by – and need to be addressed by – policy regimes and initiatives emanating from the centre.

¹ See for example: Khan, Mushtaq. Political Settlements and the Governance of Growth-Enhancing Institutions (Unpublished, 2010), available at http://eprints.soas.ac.uk/9968/; Di John, Jonathan and James Putzel. Political Settlements: Issues Paper. (Birmingham, GSDCR, 2009).
² Political settlements analysis has been at the forefront of DFID’s work on ‘Building peaceful states and societies’, as well as a key focus of major DFID-funded research projects, notably the Political Settlements Research Programme based at Edinburgh: www.politicalsettlements.org/
³ This typology of violence is based on research covering a wide range of cases as part of the recent ‘Elite Bargains and Political Deals’ project completed by the UK Government Stabilisation Unit. See: Cheng, Christine, Jonathan Goodhand and Patrick Meehan. Securing and Sustaining Elite Bargains that Reduce Violent Conflict (London: Stabilisation Unit, 2018).