Aid, Conflict & Peace Building in Sri Lanka

Foreign Aid & Sri Lanka's Military Expenditure

Aid, Conflict & Peace Building in Sri Lanka
2000 - 2005 [full text in PDF]

Jonathan Goodhand and Bart Klem with
Dilrukshi Fonseka, S.I. Keethaponcalan, and Shonali Sardesai

"There is scope to think more creatively about the interfaces between diplomatic, development, humanitarian, and human rights actors, so that the distinctive approaches of each reinforce and complement (rather than undercut) one another. The same also applies to complementarity between countries - for instance the "good cop" roles of the European countries, versus the "bad cop" roles of India and the U.S..."

Foreword

This is a seminal study on the relationship between aid, conflict, and peacebuilding in Sri Lanka. It builds upon a previous strategic conflict assessment conducted in 2000 and commissioned by the United Kingdom's Department for International Development. Since then, development and diplomatic partners in Sri Lanka have tried to pursue a more coherent "joined up" approach to addressing the challenges of peacebuilding. This study, by example, represents a joint initiative by five partners - the governments of the Netherlands, Sweden, and the U.K., the World Bank, and The Asia Foundation - to analyze political events since 2000 and prospects for a political solution to Sri Lanka's "national question."

Support for democratization, peacebuilding, and conflict resolution has become an important component of international development assistance programs in countries affected by civil conflict, none more than in Sri Lanka where nearly a quarter century of civil war has taken a high toll in terms of lives and economic opportunity particularly among the poorest and most disadvantaged. Sri Lanka's economy has managed to grow at a respectable rate despite years of civil war but this resilience merely underscores the much higher growth potential the country has, if peace is restored. This would significantly reduce poverty and propel Sri Lanka into the upper ranks of middle income countries within a generation.

For this reason, international donors have increasingly realized that they can no longer work "around" conflict. Particularly since the beginning of the peace process, many donors have calibrated their assistance according to the broader objective of building peace. Two of the chief lessons from this experience have been first, that assistance programs must be based upon a sound and thorough understanding of this particular conflict and second, that more strategic and coordinated approaches from the donor (and wider international) community are needed. Strategic conflict analyses can be useful tools for both shared understanding and coordinated responses, especially here.

Indeed one of the study's useful insights is that the "international community" is not monolithic, and different members have different assumptions and priorities and will play different, although hopefully supportive, roles in the Sri Lankan peace process. So instead of pursuing unanimity, which has been a cumbersome and elusive process in the case of Sri Lanka, the authors argue that donors should strive for "strategic complementarity" that emphasizes and builds on areas of convergence in the aims of key members of the international community.

We are happy to have supported a study of this quality that promises to have a shelf life of years not months despite the fluidity of Sri Lankan politics. We hope that it will contribute to the international community's efforts to support a sustainable peace in Sri Lanka and serve as a source of inspiration for future joint initiatives in the area of conflict analysis in Sri Lanka and beyond.

Douglas Bereuter President,The Asia Foundation
Suma Chakrabarti Permanent Secretary International Development Cooperation, Government of the United Kingdom of Great Britain and Northern Ireland
Maria Norrfalk Director General Department for Swedish International Development On behalf of the (Sida) of Foreign Affairs
Ruud Treffers Director General Development Cooperation Agency Netherlands Ministry
Shengman Zhang Managing Director, The World Bank

Executive Summary

1. Introduction

This Strategic Conflict Assessment (SCA) follows and builds upon a previous assessment conducted for the U.K. Department for International Development (DFID) in 2000 (Goodhand, 2001). Like the previous study, it aims to do three things: First, to provide an analysis of the structures and dynamics of conflict and peace in Sri Lanka since 2000. Second, to examine how international engagement has interacted with conflict and peace dynamics, with a particular focus on aid donors during this time period. Third, to identify how the strategies and approaches of international donors can best engage with and help strengthen domestic peacebuilding efforts. The primary end users of this report are expected to be aid donors, but it is hoped that it will be of interest to a wider audience inside and outside Sri Lanka. This volume is the first in a six-part series that includes five supplementary studies that are part of the SCA.

2. Background

The period under study can broadly be divided into four phases:

1. Run up to the ceasefire: in a context of an enduring military stalemate and declining economic conditions, the United National Front (UNF) wins elections in December 2001.

2. Ceasefire and peace talks: a ceasefire agreement (CFA) is signed within a month and the UNF government and the Liberation Tigers of Tamil Eelam (LTTE) embark on six rounds of peace talks.

3. The breakdown of talks and political instability: Talks become deadlocked; the LTTE suspends its participation and subsequently submits a proposal for an Interim Self Governing Authority (ISGA). This sparks off a political crisis in the South, with the President taking over three key ministries, then proroguing parliament and declaring new elections for April 2004. A coalition of the Sri Lanka Freedom Party (SLFP) and the Janatha Vimukthi Peramuna (JVP) or People�s Liberation Front wins the election. The combination of political instability in the South and an LTTE split prior to the elections that leads to growing violence in the East, means that the prospects for resuming peace talks appear to be remote.

4. The post-tsunami response: Negotiations between the government and LTTE about a post-tsunami response mirror the political dynamics of the peace process. It takes almost half a year to reach an agreement on a Post-Tsunami Operational Management Structure (P-TOMS), thus boosting the hopes for peace, despite the lack of legal clarity and the turmoil generated among both Sinhalese and Muslim constituencies.

3. Conflict Structures

In spite of the ceasefire agreement and peace negotiations, the structural dimensions of the conflict within Sri Lanka have remained relatively stable. There has been no "seismic shift" in the "tectonic plates" underpinning conflict in Sri Lanka. The constellation of factors that contributed to the outbreak and sustenance of violent conflict - including the nature of the state, its political culture, the institutional framework of policy, uneven development patterns and competing nationalisms - remains largely unaffected by the peace process. In many respects the "peace" that followed the signing of the CFA has had the effect of freezing the structural impediments to conflict resolution.

On the other hand, there has been a significant change in the external context at both the regional and international levels. The global "war on terror," growing international engagement in "post conflict" contexts, and Sri Lanka's integration into a dynamic and increasingly assertive wider Asian region have together created new (and sometimes competing) incentives for domestic actors. Though these changes in the external context may have helped create the preconditions for peace talks, they have not yet led to a radical reordering of political forces inside the country.

4. Conflict Dynamics

By 2001, the conflict had reached a "hurting stalemate." For a range of external and domestic reasons, neither side felt able to further their political goals purely through military means. The UNF-LTTE peace negotiations followed a phased approach that involved ending the violence, creating a peace dividend, and dealing with the core political issues. International actors were central to this strategy by providing security guarantees and reconstruction assistance, and facilitating peace negotiations. Although this strategy was a success in the sense that the ceasefire has outlasted the peace talks, which is unprecedented in Sri Lanka, it failed to deliver a lasting or even interim settlement.

First, the CFA froze rather than transformed security dynamics. Both parties continued to re-arm and strengthen their military capabilities. Although "no-war, no-peace" has meant an end to large-scale militarized conflict, there have been high levels of political violence, including over 3,000 ceasefire violations. Insecurity has grown in the East since the emergence of the Karuna break-away faction of the LTTE.

Second, although there was a peace dividend of sorts, it has been unevenly distributed and its impacts attenuated. Reconstruction funding was caught up in the politics of the peace process, thus limiting the peace dividend in the North-East. In the South, macro economic reforms introduced by the UNF undermined the economic dividend and led to the perception that the government was unconcerned with the plight of the poor.

The lack of a clear communication strategy about either the peace process or the reform agenda accentuated this view. Third, the step-by-step approach was based on the assumption that a limited peace could ultimately lead to a transformative peace. With hindsight, however, there could never be complete "normalization" until the core political issues were addressed. It proved impossible to circumnavigate or deal indirectly with the pivotal core of the conflict, this being the question of power sharing and LTTE hegemony in the North East. Without a clear road map for peace talks, the nature of the end goal was always unclear, which created anxieties among external and internal stakeholders. The peace process acted as a "lightening rod" for wider political and societal tensions, exposing the multi-polar and multi-dimensional nature of conflict in Sri Lanka. The bilateral government-LTTE relationship could not be addressed in isolation from other key inter and intra group relationships, which are briefly outlined below.

Sinhalese Politics

The southern polity holds the key to peace in Sri Lanka. Some of the preconditions for sustainable peace include: a level of stability in the politics of the South including a bipartisan approach to peace negotiations, a strategy for limiting the effects of, or co-opting conflict spoilers, and a significant and stable constituency for peace.

There have been some positive trends in recent years partly due to the restraining influence of the proportional representation (PR) system on the nationalist politics of the South. It has acted as a brake on the historic processes of ethnic outbidding and contributed to both mainstream parties' greater willingness to explore a negotiated settlement when in power. However, as this latest round of peace talks shows, it has not produced the stability and longevity of government required to move from a cessation of hostilities to a peace settlement. Perversely, PR has produced more moderate mainstream parties, while encouraging more communalist minority parties.

Furthermore, a political settlement requires state reform and thus constitutional change, but achieving the necessary two thirds majority in parliament is less likely under the PR system. Perhaps the most significant change in the political landscape of the South has been the emergence of the JVP as a political force. The JVP have picked up the baton of Sinhala nationalism, dropped at least temporarily by the SLFP and UNP. Both the JVP and Jathika Hela Urumaya (JHU) benefited from, and mobilized around, anxieties in the South generated by the peace process and UNF government policies - namely unpopular macro economic reforms, concerns about LTTE appeasement, and a perception that the peace process had become "over internationalized." That the JVP have entered mainstream politics is a positive development, but their role in relation to peace negotiations (and P-TOMS) has been to check movement toward federalism.

Tamil Politics

The peace process exposed a complex set of tri-polar dynamics in the North-East (LTTE, Karuna faction, the Muslims) and the South (UNP, SLFP, JVP). It also brought out in sharp relief the LTTE's Janus-headed character and the tensions between its military and political 'faces'. In parallel with brutal repression of internal dissent, continued re-armament and repeated ceasefire violations, there has been a new 'offensive' in pursuit of international and domestic legitimacy. 'No war-no peace' has enabled the LTTE to extend its control. But it has also brought new challenges to its hegemony, namely the re-emergence of eastern regionalism, the growing radicalization of Muslims and the demands that it conform to international norms on human rights and democracy, associated with the internationalization of the peace process. Given the current level of instability in the North-East and South at the time of writing, it appears unlikely that in the short term the LTTE will either come back to the negotiation table or resort to full-scale hostilities. While the immediate impact of the upcountry Tamils on the peace process is limited, their growing sense of grievance and radicalization is a notable development since SCA1, and in the longer term may constitute a new basis for conflict in Sri Lanka.

Muslim Politics

A bipolar model of conflict resolution marginalized the Muslims, which contributed to growing tension, and sometimes open violence, between the Muslim and Tamil communities in the East. It also exposed divisions within the Muslim polity, hardening fault lines between Muslims in the southeast (who form a relative majority), the North-East (who form a fragile minority) and areas less affected by the war (central hill country, south coast, Colombo). A further set of tensions has grown between the political leadership and an increasingly radicalized constituency of societal leaders and Muslim youth in the southeast. There is a striking parallel between the growth of Tamil nationalism in the 1970s and present day Muslim radicalization.

The Tsunami

The tsunami accentuated rather than ameliorated the conflict dynamics described above. In spite of initial hopes that the tsunami response would provide a space to re-energize peace negotiations, it had the opposite effect, deepening political fault lines. Protracted negotiations about the institutional arrangements for delivering tsunami assistance to the North-East mirrored earlier peace talks and exposed the deep underlying problems of flawed governance, entrenched positions, and patronage politics. Though ultimately the P-TOMS agreement was signed, the process itself further undermined trust between the two sides. As for its potential to catalyse the peace process, much depends now upon whether and how it is implemented in practice.

Sri Lanka's current situation may best be characterized as a "pause in conflict" rather than "post conflict." Though one should not interpret all political and societal changes through the lens of the peace process - since many factors predate negotiations and have their own dynamics - it has raised the political stakes as different groups jockey for a position at the table. At the time of writing, the dominant players in this process appear to have stronger incentives for the status quo than structural change. In other words, there has been a shift from a "hurting stalemate" to a "plain stalemate." A negative equilibrium has developed in which it is about managing the ceasefire rather than advancing the peace process.

5. International Engagement

One of the most salient changes in the political landscape since SCA1 has been the "internationalization" of peacebuilding. Although the policies and practices of different international actors varied significantly, two broad trends can be identified. First, there has been a more robust and multi-faceted international response to conflict and peace dynamics than has historically been the case. This has included security guarantees, ceasefire monitoring, facilitation of peace negotiations (Tracks One and Two), and humanitarian/development aid provision (Track Three). Second, there have been changes in the division of roles between various policy instruments and actors. Reflecting contemporary trends in "liberal peacebuilding," there has been a blurring of the traditional distinction between the conflict resolution and the economic aspects of peacebuilding.

Ceasefire Monitoring

The formation of the Sri Lankan Monitoring Mission (SLMM), a Norwegian-headed body tasked with monitoring the ceasefire and addressing truce violations, was one of the provisions of the CFA. To the extent that there has been a ceasefire for more than three years, the SLMM has been successful. Its limited mandate and problem-solving and consensual approach have helped defuse incidents and maintain the commitment of the key protagonists to the ceasefire. However, under the guise of a ceasefire, the permissive conditions have been created for pervasive human rights abuses and criminality. The emergence of the Karuna faction has complicated the situation, since he was not a signatory to the CFA and therefore not bound by the agreement. The credibility of the CFA and its monitors has become increasingly tenuous, as the number and intensity of the violations increase. Arguably, a broader mandate and greater operational capacities could enable the SLMM to play a more effective role in monitoring and maintaining the ceasefire.

Peace Negotiations

The entire architecture of the peace process has been built around international engagement. The UNF government put its faith in the "international security net" in order to bail them out if things went wrong. Norway was seen as an acceptable and non-threatening facilitator by the main protagonists. Although the LTTE pulled out of negotiations in April 2003, Norway continues to facilitate communication between both sides, and Track Two initiatives are ongoing. Therefore, though peace talks have stalled there is still a peace process. Some of the lessons to be drawn from Norway's role in this process are as follows: first, negotiations were based on a bilateral model of the conflict and sought to forge an elite pact between the main protagonists. Arguably, the exclusion of key stakeholders provoked spoiler behavior.

Second, there was a constant tension between the imperatives of conflict management and human rights concerns. The perception that the international community was prepared to soft pedal on human rights issues, particularly in relation to the LTTE, played a role in undermining the credibility of the UNF government in the eyes of India and the southern electorate. Third, there was a growing perception that the peace process changed from being internationally supported to internationally driven, shaped by the priorities and timeframes of external rather than domestic actors. Yet even with an "international tailwind," it proved impossible to "bring peace," showing that international actors cannot simply engineer peace and complex socio political processes are not amenable to external micro management. Finally, the importance of Track Two initiatives should be highlighted. Backdoor talks helped initiate the peace process and have played a vital role in maintaining communication since the suspension of negotiations. A critical challenge appears to be one of building a more robust architecture for the peace process that strengthens the interface and synergies between Tracks One, Two, and Three.

Development Assistance

SCA1 donors have increasingly calibrated their policies and programs according to conflict and peace dynamics within Sri Lanka. Their attempts to do this can be divided into three areas of engagement. First, applying peace conditionalities to reconstruction and development aid. Second, dealing with the consequences of conflict. Third, addressing the underlying causes of conflict.

First, at the Tokyo Conference on Reconstruction and Development of Sri Lanka in June 2003, aid donors pledged $4.5 billion as reconstruction and development aid to Sri Lanka, tied explicitly to progress in the peace negotiations. Whether intended by the donors or not, this was interpreted by the parties to the conflict as a form of peace conditionality. In practice, conditionalities or the incentives for increased aid did not have the desired outcome. They were based on an inflated view of aid's importance. There were no mechanisms for ensuring compliance. The common position expressed in Tokyo was subsequently undermined by the reticence of the larger donors to attach political or conflict related conditions to their assistance.

Second, humanitarian and reconstruction assistance for the North-East, in order to address the consequences of war, was a key element of the CFA. Since 2002, there has been a scaling up of assistance to the war-affected areas, but much of it was caught up in the politics of the peace process. A circumscribed peace dividend had the twin effects of undermining confidence in the process and eroding trust between the two sides.

Third, by shifting their priorities to the peace process, some donors arguably strayed too far from their core areas of competence. For these donors, working "on" conflict meant applying peace conditionalities and/or providing a peace dividend. In practice this translated into being sensitive to conflict dynamics in the North-East, while being conflict blind in the South. A focus on short-term conflict management was arguably at the expense of developing conflict sensitive approaches to long-term structural issues such as poverty, governance, and economic development. Therefore, programs aiming to promote peace in the North-East were undermined by policies pursued in the South. For instance, donors encouraged the UNF government to simultaneously force through two major structural changes (negotiating a peace settlement and implementing radical reforms), which created unmanageable tensions within the polity. Although many of the smaller bilateral donors have become more conflict sensitive in terms of how they engage with the peace process and programming in the North-East, development policies in the South, funded principally by the larger bilaterals and multi-laterals continue as they always have done. This has meant very limited adaptation to a conflict setting of programs in the areas of governance, macro economic reform, civil society support, and poverty eradication.

6. Conclusions & Policy Implications

Conclusions

It is difficult to predict how the dynamics described above will play themselves out in the future. Violence (and peace) involves too many unanticipated consequences. It is also important to note that peace processes almost never involve a smooth transition from war to peace. Continued instability and violence, the destabilizing tactics of spoilers, and the emergence of new unanticipated tensions tend to be the norm. To an extent, the fact that the peace procss in Sri Lanka has generated "micro cycles" of conflict is hardly surprising. Whether these conflicts can be contained and transformed is another question. Previous efforts at peacemaking, like the Indo-Lanka Accord, created unmanageable tensions within the southern and northern polities, which ultimately exploded in the form of renewed and intensified hostilities.

The preconditions for a transformative peace are not in place. There is a lack of "settlement stability," necessary for either side to take the risks required to reach a negotiated solution. The most likely medium-term scenario appears to be a continuation of the negative equilibrium, as neither protagonist is yet ready to endure the costs of war. But the potential for a return to war is inherent in the current situation. A possible trigger for this could be the growing "shadow war" in the East.

A bipolar model with a focus on conflict management rather than transformation may have contributed to the current impasse. As, arguably, did the strategy of "economizing" peacebuilding, based on the mistaken assumption that economic incentives could override political imperatives. However, one should not overstate the role of international actors in such circumstances. In the short-term, at least, the traditional tools of diplomacy and the policy instruments of aid donors, have limited traction over domestic state and non-state actors. This is not an argument for international disengagement. As the Norwegians themselves have argued, there is a need for long-term engagement and political commitment. In many respects, it is too early to talk of success or failure.

Implications for Peacemaking

Based on the above analysis, a number of implications can be identified for international actors involved in peacemaking (Tracks One and Two).

Maintaining and strengthening ceasefire arrangements, which ensure the containment of war is vital. Although there are fears that renegotiating the CFA and the SLMM's mandate risks de-stabilizing the current equilibrium, the ceasefire in its current form may not survive, given the level of pressure being placed upon it. There may be a need to consider extending the scope of the CFA to cover the full range of military actors and strengthen its human rights component. In parallel, SLMM's mandate and capacities may need to be revisited with a view to improving its means of investigation, better public diplomacy, and boosted operational capacity, particularly in the East.
More thought can be given to developing a transformative approach to peacemaking and peacebuilding. This would involve raising one's analytical gaze beyond the CFA and the limited peace agenda that the government and LTTE outlined for themselves. Interim processes can help institutionalize political engagement but should not merely freeze the status quo - conditions that show progress toward a transformative agenda can be attached. For instance, questions of human rights, transitional justice, and reconciliation could have been more central to the negotiation process.
A more inclusive approach to conflict resolution could be developed. Arguably the negotiation model was based upon two "killer assumptions." First, that it was a bipolar conflict between two relatively coherent sets of actors. Second, that the leadership would be able to represent a clearly defined constituency and "deliver" a peace deal to them after closed-door negotiations. Bipolarity had perverse impacts in the sense that it generated grievances among those who felt excluded. This was exacerbated by the lack of a communication strategy to reach out to the southern and northern polities. An inclusive approach does not necessarily mean getting everyone around the same table at the same time. But it does mean thinking more carefully about inter and intra group divisions and the vertical linkages between leaders and their constituencies. It is possible to map out a number of areas in which one could expand the scope of the (Track One and Two) negotiations:

o The need to include both mainstream parties in negotiations is a clear lesson from the UNF-led peace process. The two parties command the confidence of 60 percent of the electorate, potentially a formidable constituency for peace. A bipartisan approach is therefore a sine qua non for peacemaking.

o An adequate formula for including Muslim representatives in the peace process needs to be found, which goes beyond merely including a Muslim delegate in the government representation.

o Ways need to be found to engage with the "unlike minded," including nationalist groups such as the JVP and JHU. Ignoring or attempting to exclude such groups has not worked and, arguably, they have some legitimate concerns. Engagement could mean building contacts through Track Two and Track Three processes.

o Strengthening and supporting Track Two activities appears to be critical, particularly at a time when formal negotiations have broken down. In the current context, Track Two constitutes in many respects the backbone of the peace process.

o Although there is a significant peace constituency in Sri Lanka, its impacts are attenuated by its fragmented nature, lack of information, and distance from the levers of power. There is scope to strengthen work in this area through, for instance, more strategic engagement with the media, particularly the vernacular press.

There is a need to rethink the current consensus on harmonization. This is not working in practice, nor does it lead to the most optimal division of labor within the international community. There should be a shift in emphasis away from harmonization toward strategic complementarity. There is scope to think more creatively about the interfaces between diplomatic, development, humanitarian, and human rights actors, so that the distinctive approaches of each reinforce and complement (rather than undercut) one another. The same also applies to complementarity between countries - for instance the "good cop" roles of the European countries, versus the "bad cop" roles of India and the U.S. - and between conflict resolution tracks as the synergies and linkages between Tracks One, Two, and Three could be further strengthened.
International actors should be cognizant of Sri Lanka's regional context. This has a number of implications including listening to what Asian actors have to say about the conflict, incorporating their concerns into emergent strategy and analysis and in so doing "de-Westernizing" international peacebuilding. Given the sensitivities around excessive international involvement and the current level of domestic support for India, now would seem to be an opportune moment for India to consider the role as an additional co-chair, joining the EU, U.S., Norway, and Japan - though it is recognized that India, given past experiences in Sri Lanka, is wary of taking on the role of peacemaker.

Implications for Aid Donors

If donors are to work more effectively "in" or "on" conflict, they must develop a more realistic assessment of their role and impacts. By attempting to stand on the same ground as the diplomats, aid donors have not been playing to their comparative advantages. The implications of our analysis in relation to the "three C's" (conditionalities, consequences, and causes) are as follows:

First, the lesson about peace conditionalities is that applied crudely and without a strong political process to back them up, they have limited or even perverse impacts. Since the tsunami, the aid landscape has changed substantially. The threat of withholding aid in an "over-aided" environment will have very little effect. The debate should now shift toward thinking about positive conditions on aid and gaining influence through engagement. P-TOMs may be one immediate way of doing this. It is extremely important that donors invest the requisite political and financial capital into supporting the practical implementation of this mechanism.

Second, in order to address the consequences of conflict, there is scope (and a need) to substantially scale up assistance to the North-East to build a visible peace dividend. This will help meet immediate humanitarian needs and boost confidence in the peace process. Reconstruction programs may simultaneously contribute to the de-escalation of conflict and address its underlying causes by tackling the problem of chronic poverty in the North-East. This may involve developing pragmatic institutional arrangements in order to deliver such programs and to build capacities at the local level.

Third, there is potential for donors to do more to address the underlying causes of conflict, particularly in the South. A key lesson from Sri Lanka is that peace conditionalities may have limited traction when the broader framework of aid conditionalities remains unchanged - especially when some of these conditions may be inimical to peacebuilding. The larger donors in particular can have a significant impact upon the structural dimensions of conflict by working in a conflict sensitive way on areas like governance, economic reform, and poverty. This however may mean (depending on the donor) a significant reorientation of current strategies and approaches. Becoming more conflict sensitive necessarily means becoming more political, in the sense of being more attuned to the political context and governance structures within Sri Lanka. Some of the implications of this are outlined below.

Governance: In this report, conflict in Sri Lanka is conceptualized as a crisis of the state. In seeking to address this crisis, internationally supported "good governance" programs have often hindered rather than helped. There is a need to develop more conflict sensitive governance programs based upon a careful analysis of "actually existing" politics and the key drivers of change within the country. There is scope to work on governance issues more imaginatively. For example, exploring Asian models of developmental states that may be more applicable to Sri Lanka than Western models; engaging more proactively with political parties in a range of areas including policy dialogue and institutional development; initiating dialogue with a more diverse group of actors - including the JVP - on different options and models of governance; focusing more on governance at the provincial and local levels in order to improve delivery and accountability at the community level.

Civil society: To some extent, donors have engaged with civil society as an antidote or alternative to the state. In practice this has meant avoiding the core governance and peacebuilding challenge of how civil society can engage with and hold the state accountable. Some donors have begun to realize this, but more could still be done to support the political, as well as the service delivery role of civil society actors. There is scope to target actors and organizations at the meso level as they may hold the key to developing stronger links between Tracks Two and Three in order to build a more stable and influential peace constituency. As already mentioned, donors could also reach out more to the "unlike-minded" within civil society, given their significant influence on public opinion. Finally, there may be scope for large-scale and locally managed framework funding for civil society initiatives.

Economic reform: The breakdown of peace talks cannot be attributed to the UNF government's package of macro economic reforms. But the Sri Lankan case does raise serious questions about the scope, sequencing, mix, and speed of reform programs in fragile transition contexts. If peacebuilding is an overriding priority, then there may be a need to rethink models based purely on a calculation of optimum economic "efficiency." More thought could have been given to the political impacts and the distributional effects of economic reforms. There is also scope to draw upon and learn from comparative regional experiences in the area of macro economic reform.

Poverty: Poverty eradication is a declared priority of the Sri Lankan government and donors alike. But the growth of relative poverty and the expansion of pockets of exclusion in the North-East and South have had the effect of undermining faith in the government, the development project, and the peace process. Re-energizing efforts to address poverty and social exclusion would have a wider pay off in relation to the peace process. The JVP are one of the only political groups to put social exclusion at the heart of their agenda. This may be one of the few areas where international donors and the JVP have some common ground to explore.

Tsunami relief and reconstruction: Finally, it is widely recognized that large injections of funding have the potential to adversely affect both short-term conflict dynamics and the long-term causes of conflict. A conflict sensitive approach must involve the accountable and balanced distribution of resources with the participation of affected populations. Support for a coordinated approach through P-TOMS should be prioritized. This is also another area of international support in which regional approaches could be further developed.