 I'm really honored to be part of this panel and to come after this really fascinating presentation. And I'm really glad to have this opportunity to present an early version of this work examining the connection between types of wartime local governance and the stability of post-war governments. So of course quite a lot of work has already been done on the determinant of civil war recurrence and the obstacles that may prevent post-war government from achieving stability. And so this existing argument emphasized the importance of the way in which a conflict or at least its first phase has ended. They suggest that durable peace is more likely after one side in the war achieves a clear victory because having a military dominant victor details further challenge. And in contrast settlement leads to a more fragile peace because even if the parties agree to the terms that have been negotiated they have no guarantee that these terms will be enforced. And so these are really important mechanisms and we do have evidence that civil war recurrence is less likely after conflict that ended with a victory of one side. But at the same time there are important patterns of variation that this mechanism live and explain. So not all victory leads to a stable peace and not all settlement fail. And there is value in understanding this unexplained pattern. More specifically it's important to better identify the threats that post-war government face even when they were formed by victors. And it's also important to better identify some of the resources that have allowed some of the governments that emerged out of settlement to achieve stability. So maybe one way to make progress in this area is to look more closely at the specific processes through which the post-war order can unravel. And so I focus on three mechanisms of processes. So the first is a process of fragmentation. So it can be the fragmentation of the ruling coalition or some of its constitutive entities. But it can also be the fragmentation of the insurgent movement with whom the government is negotiating. The second mechanism that can threaten post-war stability is incomplete or failed demobilization. So demobilization program or efforts whether they are structured and supporting by international agencies or not are notoriously hard to implement and flawed implementation often means that demobilization is not thorough enough to allow the government to provide security. And a third mechanism that can undermine stabilization is post-war government's reliance and dependence on local power broker. So the task that post-war government face is that they must rebuild and stabilize the country but often without the benefit of strong institution. And so in this situation they often form alliances with different type of local power broker who have local capacity for social control. But these alliances are fragile and costly and therefore they limit the possibilities for genuine recovery and for durable peace. So I'll talk more about the mechanism when I get to the case study. But I want to get to my core argument which is that the vulnerability of post-conflict government to these three mechanisms. So mechanism fragmentation, failed demobilization and dependence on local power broker depends at least in part on how dominant arm group in the conflict have approached local power structure during the war. So what I mean by local power structure are the various form of community level organization that shape people's everyday behavior and interaction as well as local political processes. And so we can make a broad distinction between some arm group that have established themselves by co-opting local power structure. So examples of this would be the Tuareg rebellion in Niger and Mali which mobilized in part around existing confederation. And another example would be the Shan State Army in Myanmar which capitalized on traditional local ruling dynasty. The SPLM in South Sudan which had plans to establish a whole different new administrative structure but in practice ended up often co-opting existing traditional chiefs. And in the other category you have other arm group that have mostly displaced and transformed local power structure. Including various group from the Viet Minh that set up people's revolutionary community in villages throughout northern and central Vietnam or in Afghanistan, the Taliban who started to set up shadow governments in the early 2000s. So of course this is an oversimplification and there's a lot of really important work that has been done that shows variation in how rebel group approach local power structure. Both across the territory is a control and also over time throughout the course of the conflict. So yeah, so this doesn't capture this really important within case variation but still we can say that there are differences in rebel groups, dominant and aspirational strategy. And so my main point is that it is harder for post-war government to achieve stability after conflict in which dominant actors adopted strategies of cooperation rather than transformation. Why? Because there are more severe and more misaligned incentive within organization that have adopted strategies of co-optation. So when rebel group have built themselves and grown by co-opting local power structure, they remain quite dependent on the local elite who control this structure. And so the top leadership of the group cannot easily, in case of internal conflict, the top of leadership cannot easily replace this local elite. If, let's say like I'm going to talk about the case of Somalia, like a clan leader wants to leave the armed group, he will take his community and his faction with him. So this local elites maintain this kind of exit option, which means that there are greater risk of fragmentation both during and after the war. Second, to preserve the status, these local elites are likely to resist genuine demobilization. So for an armed group to disarm, it's not enough for its top leadership to agree to demobilization. An important condition of successful demobilization is the presence of an institutionalized integrated chain of command that is committed to the effort and that is recognized as legitimated by the fighters. And so in organization that grew through a co-optation, this condition is less likely or unlikely to be present. Finally, when armed group adopt strategies of co-optation, it's then difficult for post-war government to consolidate power and rule autonomously. This is because the elites that have been co-opted become strong men that can be effective rival to the post-war government. And when on the contrary, armed group have used strategies of transformation, they've already replaced strong men by strong organizations that can be then key levels that allow them to rule autonomously. So that's the theory of the mechanism that the degree to which post-war government are vulnerable to fragmentation, stole demobilization or lack autonomy depends in part on the extent to which armed group, during the war, have co-opted rather than transformed local power structure. So how can we look at this empirically? So one way to do this, and this is still a work in progress. Sorry, I didn't manage the... Oh, sorry, I was wrong. I have this data set that has a range of viable, capturing ways in which armed group have approached local power structure. And so I've documented both practice in which they have transformed this structure by creating new institutions, by redistributing land and assets, by empowering disempowered group, and practices through which they have co-opted. And so the plan is to look at whether we see some kind of connection between these wartime practices and the likelihood of civil war recurrence. But for today, I want to present evidence from a comparison between two cases that ended similarly with a victory of the insurgents and yet the post-war government had extremely different trajectory. So in both Ethiopia and Somalia, insurgent groups overthrew the incumbent regime in 1991. And so in Ethiopia, a coalition of rebel groups led by the Tigray People Liberation Front overthrew the regime of the DERG. In Somalia, different rebel groups, including the United Somali Congress and the Somali national movements, but there were more groups overthrew the regime of the DERG. So in Ethiopia, the rebel coalition that came to power stayed in power from 1991 until 2018. And it's been a period of severely constrained political right for the country, but also a period in which the government was able to enact successful reform, leading to lower poverty and better access to basic services. And throughout this period, there were localized conflict and mass protests, but none of this conflict became an existential threat to the government. In Somalia, the level of violence did not really decline after the fall of the previous regime, even after the UN sent a mission to enforce a ceasefire, and the transition government was not formed until 2000. So it was too divided to agree on the path forward. It was replaced by a new transition federal government, which was not really more successful at restoring stability. And if we zoom in at this period that immediately follows the war, we see the role of the three processes I mentioned earlier, so fragmentation, incomplete demobilization, and dependence on local power brokers. So in Ethiopia, the group that the TPLF, the group that dominated the ruling coalition, continued to experience severe internal crisis after the war, just how did that during the war? But this severe internal crisis did not lead to a splintering, in part because of its rules and procedures, but most importantly because no member of the top leadership had a personal following, he was able to withstand the crisis without splittering. And we see something completely different in Somalia. The two main groups, so the USC and SNM, both disintegrated. Somalia went through a long period of fractional war. So we can take the example of the United Somali Congress. There were two factions that fought until 1998, and even when they started to negotiate that triggered a new splintering. I'm probably not going to have time to go through all the mechanisms, but I wanted to say something about demobilization. So the TPLF was able to demobilize a lot of the insurgent troops, and those troops that were not demobilized were integrated into the National Army, and they even arrived at a situation where the insurgent troops agreed to fight under the leadership of National Army officers, meaning of the officers of the army they had just been fighting a few years earlier. In Somalia, the insurgents were unable to demobilize their fighter. There are many examples, but even in Somalia, which is now viewed as part of Somalia that achieved order or stability, there was so little control of the insurgent faction that they ended up abusing their own clan or the clan they were coming from. So two very different paths, and the question is, do we have evidence of a connection or continuity to wartime mode of local governance? And largely, we do. So I want to start by saying a few words about the TPLF strategy of local engagement. So the TPLF established its support base by dismantling the existing village-level political structure, replacing them with new popular councils, and then it used this council to enact a very radical land reform programme, like redistributing land, which is one of the most consequential and radical reforms you can enact in an agrarian society. And so this reform helped the TPLF build a robust support base. It helped the organisation recruit during the war, but it also helped with demobilisation because then there was plots or lands that the fighters could go back to after the war. And in Somalia, the mobilisation process was very different. Initially, there were efforts to establish each of these organisations as multi-clan organisations that would be based on a shared political platform rather than on clan identity, but these attempts mostly failed and they failed for a range of reasons. Some of them was because the government strategy was to arm a clan faction against each other, but also it failed because once they transitioned from mobilisation to actually taking arm against the government, it was more efficient to build combat units around the clan lines and to try and recombine the units they had. So those were the main points I wanted to touch on and to conclude, I want to say briefly a few words about some of the theoretical and practical implications of the distinction I was making. So one theoretical question that it might be relevant for is always this question of whether new wars are different from previous wars in that they contribute to the erosion of state institutions rather than state building. Maybe the distinction can help us look differently at the question of under what conditions do wars make the state and one of these conditions maybe that the dominant actors have initiated a political process that undermines the rival of the state. And finally, there is also this argument that has been made sometimes that foreign intervention prevents a return to stability by preventing the emergence of a clear vector, but as we've seen, insurgent can defeat the adversary without having the capabilities of achieving a secure peace. And so, yeah, I'll stop here. Thank you very much.