From disaster to devastation: drought as war in northern Uganda

This paper proposes a shift from the concept of disaster to one of devastation when dealing with the destructive consequences of climate change. It argues that today, a discourse of climate-change disaster has become dominant, in which present disasters are seen as harbingers of a future of widespread climate disaster, products of a global nature in upheaval. The paper contends that one needs to go beyond the series of dichotomies that the climate-change disaster discourse relies upon: future/past, global/local, natural/social. To frame climate disaster as a product of global climate change, and conflict the product of those climate disasters, is to occlude the forms of environmental violence and experience of climate change among disaster-affected communities. Through an exploration of the drought in Uganda, the paper asserts that disaster should be understood as embedded within ongoing, longstanding, multiscalar processes of devastation produced by histories of human engagement with the environment, including that of war.

precipitation and flooding.The evidence is overwhelming'.Concerning drought in East Africa specifically, Oxfam (2018) adds: 'On the back of 18 months [of ] drought caused by El Niño and higher temperatures linked to climate change, the Horn of Africa region is now going through a further drought, caused by a mixture of influences from La Niña and the Indian Ocean weather pattern.Millions of people are facing acute food and water shortages'.
Even if the definitive attribution of any particular climatic disruption-and the resulting disaster-to global climate change still remains technically problematic at the moment, the association between natural disasters and climate change has been largely cemented, at least in much of public discourse.Indeed, it is rare to find an account of a drought, flood, or hurricane in the media that does not raise the possibility that it is a consequence of global climate change: as of this writing, the website of The New York Times (2017) features a story with the headline 'the relationship between hurricanes and climate change', with reference to Hurricane Harvey in August 2017.Whatever the difficulties of determining the etiology of specific weather events, each climate-related disaster is seen to bear with it the ominous message that such disasters are going to increase dramatically in the future.Phenomena such as the East African drought are mobilised as evidence of the reality of global warming and of the need to mitigate climate change and to promote adaptation and resilience in preparation for inevitable future disasters (Hulme, 2009).Contemporary disasters, in this climate-change disaster discourse, are therefore largely about the future; they are portents of what is to come, manifestations of the new global reality of a planetary nature out of joint.
This paper draws attention to the impacts stemming from this incorporation of disaster into the global climate-change discourse, both with regard to how disasters are comprehended as well as the politics and ethics of disaster-oriented practice informed by that understanding.Conceptually, it asserts that incorporating disaster into global climate change establishes a series of dichotomies through which such events are viewed: future versus past; global versus local; and natural versus social.The first term of each of these pairs is privileged when disaster is integrated into climate change, so that the climate-change disaster discourse places its emphasis on future disasters, on the global etiology of those disasters, and on their ontological footing within the realm of nature, even if that nature has been shaped by human activity.Heightened conflict, in this model, is among the future ramifications of climate disaster.
This paper contends, however, that it is these dichotomies, and the overall climatechange disaster discourse, that prevent understanding of how climate change and disaster are actually experienced in the very parts of the world declared as the most vulnerable.Instead, analysis needs to start with the experience of climate change and disaster in different locations and begin to reconstruct an understanding from those locations.This paper seeks to go beyond these dichotomies and to rethink the very concept of climate disaster by starting not from the hazard/vulnerability framework common in disaster studies, but from the lived experiences and the histories of climate change and disaster in specific parts of the world.Drawing on postcolonial environmental studies, it maintains that understanding climate change and disaster as they are faced in different locations requires that they be situated within multiscalar, plural histories of environmental violence, encompassing local, national, and regional dynamics along with the broadest set of global dynamics.War can be an integral part of these histories, and so one needs to incorporate conflict into understanding of climate change in new ways.
To go beyond the limitations imposed by the dichotomies of future/past, global/ local, and natural/social on understanding of climate change and disaster, this study proposes the concept of devastation.Devastation can comprehend better the phenomena that typically are seen as comprising climate change and climate disaster.The objective is to reframe climate disasters as comprised of specific aspects of longstanding forms of violence that exist within a complex matrix spanning past and future, global and local, natural and social.A turn to devastation can open paths towards an understanding and practice that is epistemologically, politically, and ethically adequate with respect to divergent global experiences of phenomena associated with climate change and disaster.This concept of devastation is explored using the case study of drought in 'post-conflict' northern Uganda.The recent drought is situated in the context of broader processes of environmental devastation and violence, revealing how political violence is bound up with destructive environmental change in ways that demonstrate the limitations of the dichotomies of future/past, global/local, and natural/social.

Climate disaster between hazard and vulnerability
The incorporation of natural disaster into climate change can be widely found in current policy documents.Representative of these is Uganda's National Climate Change Policy, adopted in 2015.'Most natural disasters that occur in Uganda are related to extreme weather events' (GoU, 2015, p. 10), it explains, and so 'Disaster Risk Management' is 'a frontline defence for adapting to the impacts of climate change' (GoU, 2015, p. 27).The report speaks of climate change as 'one of the greatest challenges facing humanity in the century', and declares that it 'is likely to disrupt the Earth's ecological systems and have serious negative consequences on agricultural production and productivity, forests, water supply, health systems and overall human development' (GoU, 2015, p. v).Indeed, Uganda is already seeing some of these effects, the document reports: 'the average temperature in semi-arid areas in Uganda is rising', as is 'the frequency of hot days', while rainfall is becoming less predictable and extreme weather events are intensifying (GoU, 2015, p. vi).As the policy attests, 'Uganda has committed to the adoption and implementation of policies and measures designed to mitigate climate change and adapt to its impacts', which requires 'preparedness . . . in all sectors to ensure that the country is resilient to the adverse impact of climate change' (GoU, 2015, p. v).
Disaster preparation and risk reduction, in this framework, are thus about mitigating and minimising hazards while also addressing vulnerability.This reflects the dominant understanding of what constitutes a disaster: a natural hazard occurring in the context of economic, political, and social vulnerability (Wisner, Gaillard, and Kelman, 2011).Different approaches to disasters within the academic and policy literature may emphasise one or the other of these aspects (O'Brien and Eriksen, 2007;Methmann and Oels, 2014): highly hazard-focused approaches may usually stress technical fixes such as improved seed varieties or securing critical infrastructure, while socially and politically informed treatments concentrate on ameliorating vulnerability (Blaikie et al., 1994).Indeed, vulnerability studies provide an important corrective to purely hazard-centred approaches to disasters, pointing out how the spotlight on natural hazards can ignore the way in which vulnerability to climate disaster is produced 'on the ground', using Jesse Ribot's (2010) phrase.That is, without a history of economic and political deprivation and marginalisation, the hazards caused by climate change would not present the dramatic risk that they do to certain communities.The implication is that policy interventions that concentrate on adapting to natural hazards risk leaving unaddressed the economic, political, and social conditions that make communities vulnerable in the first place (Ribot, 2011).Hazards and vulnerability are found now in most treatments of climate disasters, including in Uganda's National Climate Change Policy (GoU, 2015).
The distinction between natural hazard and social vulnerability becomes more polarised, however, as disasters become entwined with ideas of global climate change.With climate-change disasters, a focus on globally-produced climate hazards, and on global climate change as a kind of ultimate, universal natural hazard, endows the natural hazards side of the equation with new primacy (Ribot, 2014, p. 671).A clear line is drawn between the global and the local in the climate-change disaster discourse, with weight shifting to the global as the location from where inevitable, overwhelming hazards originate and disasters are driven.The global is constructed in opposition to the local, with the latter conceived as the site where particular, delimited forms of vulnerability are found.Accordingly, bringing disasters within a climate-change framework produces and relies upon a distinction between the global and the local, bound up with the distinction between the natural and the social: global climate change is seen to produce natural disruptions in climatic and meteorological patterns that, when extreme enough, intervene in contexts of local social vulnerability to create disasters.
This implies a globally differentiated epistemology following North-South lines.The global is the realm of objective universal truth, an object knowable through technical forms of natural scientific knowledge being produced in Northern laboratories mostly via computer models, in which the role of the rest of the world is to provide data (Edwards, 2010).The producers of this knowledge are also constructed as those who are themselves largely exempt from the consequences of climate-change disaster, those whose resilience is high, who are above the rising water line.The local is the realm of social particularity, comprising those parts of the world that are said to be at risk of suffering the most due to future climate disasters.These 'local' areas in the global South are deemed to be excluded from producing or accessing universal knowledge of global climate change and become locations defined by varying degrees of socially-produced vulnerability to these global forces.The local is constructed as divergent from the developed North through its vulnerability, defined by its lack of the North's resilience.It is constructed as knowable through humanities and social sciences, but those methods are only useful to discern local specificity, only the particularity of local vulnerability; they have nothing to say about universal global climate change or about transformations in nature.
The dichotomy between global, natural hazard and local, social vulnerability in the climate-change disaster discourse entails another dichotomy, that between future and past.A distinction is drawn between, on the one hand, disasters linked to increasing global climate change, and, on the other, disasters and broader environmental crises that are not directly connected to global climate change.The focus on climate change thus infuses an ethos of urgency into those disasters seen to be caused by climate-change-related hazards, as they rise to primary importance in domestic and international policymaking.The consequence is that past or present crises within 'agricultural production and productivity, forests, water supply, health systems and overall human development', as enumerated by the Ugandan strategy (GoU, 2015, p. v)-that is, instances of locally significant environmental destruction whatever their origin-are rendered invisible or less important by concentration on global warming and intensive preparation for global-warming-induced climate disasters.Non-climate-change-related environmental violence or disaster can be incorporated into the climate-change disaster framework only in a secondary role as contributors to vulnerability to climate-change-related disasters.
In short, even as the vulnerability approach makes clear the need to take the past, the local, and the social into account, the climate-change disaster discourse inevitably privileges the future, the global, and the natural.More fundamentally, however, it is the dichotomies themselves that may represent the greatest obstacle to understanding climate disaster today.Attention to the place of conflict-or, more broadly, violence-in the climate-change disaster discourse can illustrate the need to go beyond these dichotomies.With the incorporation of disaster into climate change, an urgency comes to characterise interventions and policies seeking to deal pre-emptively with the heightened conflict that is expected to arise from global climate change, from 'climate refugees' to 'resource wars'.This can be clearly seen in the Ugandan strategy in the connection it draws between disaster and conflict.The report invokes looming conflict as a result of climate disaster: 'Disaster risk management is also a key aspect of addressing socio-environmental conflicts and human security concerns, both locally and regionally, in respect to environmental refugees and management of transboundary resources' (GoU, 2015, p. 27).Disaster preparation is about preventing future conflict and insecurity caused by climate change.Despite important academic work questioning the linkage, the dominant understanding of the relation between climate change and conflict remains one of cause and effect: climate change is proposed as the cause of conflict though various mechanisms (Busby et al., 2013), 1 most predominantly that climate change or heightened climatic variability increases resource pressures or introduces shocks to already vulnerable communities, members of which respond by turning to violence.The assumed existence of a stress-response model can reinforce the inability to 'move beyond linear connections' between conflict and disaster and to embrace a 'multi-dimensional, context specific and historically relevant understanding of climatic disaster[s] and their impact on local politics' (Siddiqi, 2014, p. 886).
The assumptions behind this model are familiar from much of the writing on violent conflict in Africa: rebels are supposed to be rational actors who turn to violence as a way of ensuring access to resources for survival or profit, while rural populations face resource scarcity and are ready to turn to ethnic violence when that scarcity becomes too great (Verhoeven, 2014).It follows that interventions to promote adaptation and resilience to climate change are assumed to be needed as a mechanism of conflict prevention.Uganda is thought to be at special risk as a highly vulnerable African country; within Uganda, the north is often presented as the area with the most significant chance of climate change leading to conflict (Oxfam, 2008, p. 2;MER, 2015, p. 3).Past or ongoing conflict is included only insofar as it produces and exacerbates people's vulnerability to looming climate hazards (Wisner, 2011).A vicious circle is thus established: conflict social vulnerability ( + natural climatic hazard) conflict . . .What the paper seeks to show with the case of northern Uganda, however, is that political violence is not something that can be relegated to a future consequence of climate disaster or to a contributor to vulnerability.Instead, political violence is something that is already bound up with transformations in the climate, locally, nationally, regionally, and globally.This paper argues that a consideration of drought in northern Uganda illuminates how it can be understood as embedded within the legacies of war, as a form of violence that has continued into the post-conflict period.Political violence has caused climate change and continues to cause climate change; or, perhaps more precisely, climate change is political violence, which becomes clear when one is able to get away from the dichotomies structuring the dominant understanding of climate disaster.Consequently, a focus on the relation between political violence and climate disaster can reveal the need to rethink the dichotomies and perhaps to go beyond the very concept of disaster.

From disaster to devastation
The irony of the dichotomies embedded in the climate-change disaster framework is that now, in the age of the Anthropocene, global climate change itself is understood to have rendered no longer viable the distinction between the natural and the social within climatic or environmental processes (Chakrabarty, 2009;Latour, 2014;Ghosh, 2016).The anthropogenic character of climatological transformations, as well as of broader disruptions in the biosphere, hydrosphere, lithosphere, and so on, means that a 'natural' hazard can no longer be taken to be natural in any essential way that fundamentally sets it off from the social processes that give rise to vulnerabilities.Various ways of comprehending the global co-constitution of the natural and the social have been developed, such as Jason Moore's (2015) idea of a 'web of life', which entails broad histories of human-nature interaction (see also Parenti, 2011).That fundamental distinctions between the natural and the social need to be rethought is not a novel argument today, given the rapidly expanding literature sparked by the debates around the Anthropocene.Nevertheless, it may feel somewhat out of place in the more operationally-oriented disasters literature, in which meditations on the unstable ontology of earthquakes might appear an indulgence when faced with unstable buildings.
Productive paths towards new understandings of climate change and disaster have been charted by those working in the 'postcolonial environmental humanities' and specifically within what Anthony Carrigan (2015) calls 'postcolonial disaster studies'.For him, research and practice should '[t]reat postcolonial studies as a form of disaster studies and vice versa' (Carrigan, 2015, p. 117), for it is 'impossible in many cases to disentangle catastrophic experiences from colonial and neocolonial power dynamics' (Carrigan, 2014, p. 4).Michael Niblett (2014, p. 109), drawing, like Carrigan, on the work of the Barbadian poet and literary scholar Kamau Brathwaite, explains that a tsunami, for instance, should not be understood as 'an external force that impacts upon capitalist civilization, but as something constitutively implicated in its unfolding'.Both slavery and Hurricane Katrina, therefore, are 'constitutive moments of a single unfolding spiral of catastrophe', according to Niblett (2014, p. 109).
For these critics, any particular disaster can be dissolved back into much longer histories and broader contexts of colonialism and capitalism, in which catastrophe is the dominant narrative.This postcolonial perspective reveals that large-scale environmental disasters are not something that are yet to come from nature in a future of global climate change.Rather, for most of the world, disasters and dislocations are the reality of the present and have comprised centuries of the past.Hence, research needs to take account of 'how environmental change is entwined with the narratives, histories, and material practices of colonialism and globalization.Postcolonial approaches emphasise how experiences of environmental violence, rupture, and displacement are central ecological challenges across the global south' (DeLoughrey et al., 2015, p. 2).
I agree that the processes that tend to be distinguished as climate disasters can be productively reframed within these longer histories of environmental violence, rupture, and displacement.However, I would look less to (global) colonialism and capitalism as constituting a single dominant narrative of catastrophe, within which specific events or processes need to be framed.Instead, I would maintain that understanding the environmental violence confronted in different locations around the globe requires that violence be situated within histories stretching across many registers, that confound the dichotomy between global and local by encompassing local, national, and regional processes along with sets of global dynamics (Peluso and Watts, 2001;Peet, Robbins, and Watts, 2011).
In this way, the insights of Anthropocene thinking and postcolonial eco-criticism can allow a move away from the dichotomies of natural-social as well as of globallocal and future-past that are being entrenched within the predominant climatechange disaster approach, and in particular away from the privileged value placed on the first term in each pair.Instead, one can start, as eco-criticism suggests, from the experience of climate disaster where and as it occurs, using that as a window on to the dense local, national, regional, and global forces, both natural and social, that congeal at particular points to cause widespread destruction or suffering through the environment.For a subsistence farmer whose crops are not growing, who faces a drought or a flood, it would not seem to matter much whether or not scientists can attribute that particular hazard to global climate change.By taking the farmer's experience and understanding of the environment as the starting point, instead of rising global greenhouse gas levels or deviations from scientifically-determined climatological norms, one is prompted to reconceive climate change and climate disaster.Droughts and floods, whatever their origin, as well as the disasters they give rise to, are climate change in the sense that they represent the most important alterations to people's lived environments that affect their opportunity to survive and flourish.Furthermore, the environmental devastation that proceeds from deforestation, displacement owing to infrastructure development, oil extraction, rapid urbanisation, rural proletarianisation, or war are all wrenching, immediate, devastating forms of climate change for those who experience them, however attenuated their relationship may be to global warming.The present and the past are thus not ignored in favour of the future; the local is given equal standing with the global; and the social is seen as co-constituted with the natural all the way down.Starting with these narratives is fundamental because what the environment is and how people relate to the non-human world will be subject to many different visions, so determining what exactly the 'climate' is that is changing is crucial (Taylor, 2015).This signals the need to dissolve yet another dichotomy, that between climate change and disaster itself.
In short, climate disaster can be seen not necessarily as a result of global warming caused by rising greenhouse gas emissions, as global and future-oriented.Rather, climate disaster is represented by local, concrete, and present forms of devastation, interlaced with national, regional, and global processes, which matter to people, and which are transforming their environments and climates today.These forms of devastation can render environments uninhabitable in ways that are imagined only by the most dramatic of future-oriented global climate-change scenarios.The worst projected disasters of the future are already present for many of the world's poor, and have been so for decades or even centuries (Davis, 2002).Clearly, global warming will play a part in these forms of climate change and climate disaster, but this role can be understood only once a different orientation for the etiology of climate disaster is employed-one that does not look to rising global greenhouse gas levels and their future impacts, but rather looks to the past, to specific histories of environmental and social devastation that produce droughts or floods now.An attention to how natural and social processes are co-constituting becomes increasingly important (Ingold, 2011;Taylor, 2015), and disaster can be de-exceptionalised (Hilhorst, 2013).
In place of the concept of disaster, Carrigan (2015) and Niblett (2014) seem to suggest catastrophe to capture this long-term process of the natural and social onslaught of capitalism and colonialism.Catastrophe, however, may entail too much of a sudden, exceptional character; I would propose instead the concept of devastation, which places the emphasis on long-ranging and lasting ramifications and can bring together a wide set of forms of violence within it.With a temporal dimension more akin to Rob Nixon's (2011) idea of 'slow violence', but able to include periods of overwhelming, direct, atrocious violence as well, devastation entails both violence against people and against the land, plants, and air; it is the outcome of broad violence against ecologies in which human and non-human are entwined.
By beginning from the ways in which climatological disruptions are experienced and understood within environmental, political, and social histories, the category of devastation can avoid becoming simply an undifferentiated sweep of violence.Analytically, narratives of climate change and disaster among those experiencing them can reveal the economic, political, and social processes that produce devastation (Quarantelli, 1998;Oliver-Smith and Hoffman, 2002, p. 18).The researcher can work to integrate those narratives into other narratives of climate disaster, which would include regional or global meteorological transformations and regional or global histories of capitalism, colonialism, and racism (Hage, 2017).The narratives created by researchers can be of use both to those immediately affected by a disaster as well as those working in the disaster prevention and response industry and the professional scientific community.

Drought as war
The East African drought, therefore, is not best understood as a globally-induced natural hazard affecting local vulnerability.Instead, it should be seen as embedded within ongoing, longstanding, multiscalar processes of environmental devastation and generated by equally multiscalar histories of human engagement with the material processes of the planet, which can be framed within numerous, but often overlapping, narratives of devastation.In northern Uganda, war provides the dominant narrative of devastation within which climate disaster and drought can be comprehended.Civil war raged in the region from 1986-2006, pitting the rebel Lord's Resistance Army against the Government of Uganda (Branch, 2011).The insurgency was infamous for its atrocities against civilians, but the government's counterinsurgency also was devastating.From 1996 in some locations, and from the early 2000s in others, the Government of Uganda forcibly displaced and interned the entire rural population of the Acholi sub-region, comprising more than one million people, in wretched camps (Dolan, 2009;Branch, 2011).The forced internment created a humanitarian disaster, and led to widespread cultural, economic, and social destruction.Movement out of the camps and back to rural villages and homesteads began haltingly in 2007, following the conclusion of the war; in some places, it is still an ongoing process.
War is thus part of the past, but its legacies shape the present and certain forms of war-time violence continue into the present.Hence, the violence of the war, past and present, constitutes the dominant narrative for understanding the devastation that people face today, including in relation to the environment.As argued below, even drought, which the climate-change disaster framework would associate with the natural and global, with its relevance for conflict residing in the future, can be seen as part of this history of war as devastation.Indeed, drought is commonly viewed as tied up with the violence of the war.As explained below, the state and the structures of violence that it supports, in the context of the devastation wreaked by the counterinsurgency policy of long-term mass forced internment, are seen as responsible for the drought through one key route: the mass extraction of trees.Climate change can be understood as the set of wrenching transformations that have resulted from the broad legacies of the war as well as the continued forms of environmental violence confronted by the community at present.
Instead of the vicious circle of conflict social vulnerability + natural hazard conflict . . ., a different relation between disaster and conflict can be posed.War is the primary framework for comprehending a devastation that incorporates what is understood by the climate-change disaster discourse as social vulnerability plus natural hazard.This opens the way for a more complex appreciation of the relation between conflict and disaster, in which they are intertwined within multiscalar ecological, economic, and political processes.
Again, in Uganda, this war is not the only history, and not the only narrative that matters.Indeed, global meteorological processes, themselves socially embedded, are part of the devastation.However, this narrative of war should assume primary position in understanding drought and framing responses to it instead of abstracting certain elements from that history needed to comprise the categories of 'natural hazard' and 'social vulnerability'.With this revised interpretation of climate disaster, no longer are climate shocks seen as producing resource scarcity and consequently leading to conflict and insecurity.Instead of climate disaster causing local conflict within a global climate change framework, conflict might be viewed as causing climate change, or perhaps both the violence of disaster and the violence of conflict are elements of broader, longer-term histories of violence.This also helps to qualify some of the more nuanced, empirically-grounded studies of conflict and disaster: Siddiqi (2014, pp. 888-889) shows that these studies frequently present a natural disaster as opening up a political space into which either the state can reassert itself through aid provision and reconstruction assistance, or else the political opposition can establish its legitimacy by filling that gap.Climate disaster, in this sense, can spawn a change, therefore, from the legitimacy of one political regime to another (Pelling and Dill, 2010).In northern Uganda, however, the dynamic is different: it is not the state's response, or lack thereof, to climate disaster that has political meaning; rather, the drought itself is seen as political, as a continuation of state violence.By being part of longer processes of social-natural devastation, climate disaster may not comprise a disruption, but rather an intensification of, and continuity with, these existing political arrangements and structures.
Northern Uganda demonstrates that the effect of disaster is not necessarily to reorient people politically, but to confirm their political subjugation by an essentially predatory and antagonistic state and its agents, a subjugation that increasingly threatens the very basis of their livelihoods.People's desperation and insecurity in the face of state-violence-driven climate disaster thus does not lead to conflicts among rural people, but to a confirmation of their oppression by the state.One should not view potential future violence as resource conflict in response to climate disaster; instead, it should be seen as a continuation of the violence of the war, perhaps even involving new armed opposition to the state.
Coming to terms with the shortcomings of the dominant climate disaster/conflict paradigm is important not only as a conceptual exercise.As climate change is increasingly perceived as the primary threat to development and security in Africa, climate change adaptation and resilience are coming to dominate the international governance regime (USAID, 2013).In Uganda, as in many other African countries, this policy shift towards climate-change adaptation and resilience is still in its initial stages.Nevertheless, the dominant climate-change-centred understanding of disaster and conflict looks likely to have an increasing impact on the economy, political structures, politics, and people's lives.

'Climate change' in Acholi
Disruptions to and changes in the environment were the major subjects of interviews and discussions conducted among rural communities in northern Uganda's Acholi sub-region in 2017. 2 Ongoing drought and broader problems with rain were raised consistently: less rain, especially during the planting seasons, was noted, but so too was the fact that when rain did come, frequently it was very powerful.Along with reduced rainfall there were periods of brighter sun that ruined crops.In addition, increased winds destroyed crops or homesteads, as did unusual weather events such as hailstorms.As an older woman explained in a focus group discussion: 'In the past, rain would always come gently and smooth and the food would grow well; but now, you hear it has rained but the rain has destroyed gardens, and when the wind comes it is very strong and ends up blowing and destroying everything.Now the ground is just drying up.For me, those are the changes I see, and indeed there are many changes'.
These changes to the weather, though, are not seen as isolated developments or as fundamentally different from other environmental changes that may not be related to global climate change.Indeed, the environmental transformations that people often explained as being the most important-what 'climate change' means in the context-seem to have begun as people returned home from government-run internment camps.In most discussions, the division marking the beginning of the difficult present time is the departure from the camps and people's embarkation on an effort to re-establish lives and livelihoods in an ecological, political, and social environment that had been torn asunder by two decades of brutal civil war and up to a decade of forced confinement.Contemporary conditions were almost always contrasted to the time before the phase spent in the camps, with the latter blamed for many current problems.
As people returned from the camps in the late 2000s and early 2010s and sought to establish livelihoods in a radically changed setting, the challenges they faced in doing so became clear.First, the landscape itself was unfamiliar to the returnees after years away.Fields were overgrown, homesteads ruined, water sources destroyed, and many large trees, essential for boundary demarcation or orientation, as well as having cultural and spiritual importance, had been felled.New 'security roads' had been opened up by the Uganda People's Defence Force (UPDF), while old footpaths had been swallowed by the undergrowth (HURIFO, 2007).Landmines and unexploded ordinance were scattered, posing dangers to those clearing their fields or to children playing in the bush; the many people killed during the war had left certain areas considered dangerous and possibly haunted by spirits.Perhaps most difficult of all, people were returning to their land without cattle, which had been the primary form of wealth but had been looted during the early years of the war by cattle rustlers and the Ugandan military.
In this hostile environment, people often described their surroundings, including the weather, as a force in opposition to them.The word that is typically used in the Acholi language to denote weather conditions is piny, as in piny lyet-it is hotor piny ngic-it is cold.Piny can also refer to conditions of darkness or daylight or to one's surroundings more broadly, touching on its environmental, social, and even spiritual dimensions, given the deep meanings with which the landscape is invested (Finnström, 2008).A consequence of the war was that, as one person put it, 'now the environment has become unfriendly'.Another expanded on these interlinked factors: 'In the past when trees were not cut down, we used to carry out farming peacefully because our crops would grow well and there was no army.But now we came back from the camps, we lack water, our crops don't grow well, and we don't have money to pay fees for our children'.The aspect of this bad environment of most concern was the perceived falling productivity of land and the inability to survive from farming alone.As one person noted: 'during that time before we ran off to camp, everything was okay, you would dig and your food would grow nicely.You would dig one acre [and] you would even get four bags, but now . . .even though you dig, there is no food that comes out'.Very concrete legacies of the war were identified as causing the fall in productivity.During the conflict, granaries had been destroyed and seed stocks lost, as had farming implements and household supplies.Many young people, having grown up in the camps, lacked basic knowledge of building homesteads or planting crops.Others identified toxic legacies of the war: 'In our area here, there were very many bombs, so the chemicals from the bombs, sometimes we think, is what has brought trouble in this place'.People also blamed the chemicals reportedly being used by the expanding agricultural plantations, as well as the dust created by new quarrying works.The cause most commonly cited for declining productivity, however, was the destructive changes in weather patterns attributed to the massive loss of trees.
There was some debate over the exact way in which cutting trees was leading to decreased rainfall and to drought.Some claimed the chainsaws were keeping the rain away; others pointed to the inability to perform the rituals that had brought rain now that the trees, the site of these practices, were gone.Yet others blamed the foreigners doing the cutting, contending that they were performing their own rituals to keep the rain away.Almost everyone to whom the research team spoke, though, agreed that the most important factor was the forests' role in attracting rain and retaining water.Without forests, rain will not come, and, when it does fall, it will fail to soak into the soil and will drain away.As one woman put it: 'The way I see it, piny is not as it used to be in the past.Back then, if the rain decides to fall, it shows that real rainfall is coming, even the air changes.The wind moves like this and then the trees could shield excess wind.But these days, they have cut the trees, the wind moves anyhow and now even the rain periods come and pass without anything happening.From January all the way to March, then it would start raining and go all the way like to June.Then two months after that, people start planting food in their gardens again.But now, you can go up to August like bang! . . .The sun is shining so bright and hot and there is no rain'.A group of women in a village in Amuru District added: 'Yes, [tree cutting] is there.They cut almost everything, they started way from up there to all the way down.Now there's nothing; the rain has also gone silent and now, for us, we are dying of hunger because the ground has dried up.We are going to die of hunger, there is nothing to eat-you see this, they come and cut everything'.In a nearby parish, a similar account was provided: treecutting 'is ruining the forest.There isn't any forest anymore.You know that the forest also brings rain and prevents the wind.Now the forest isn't here to stop the wind and bring the rain'.One interviewee drew attention to differences with the past: 'People dig by themselves for two to three years with nothing to gain from it.It is not that we don't dig, we do, but then it is the little rainfall we receive that is a problem and that is because of the trees'.
This fall in productivity and decrease in rain has dangerous implications for the future.One person underscored that it is the 'children who will suffer': 'They will not have any trees left.It is the war that caused all these problems-in the past, people respected each other but now people do not do that anymore because people just cut down any tree on any hill they come across'.
These negative consequences were not limited to declining agricultural productivity.Deforestation has harmed people's broader livelihoods, as the commons that all depended on for building materials, firewood, grazing land, and medicinal plants were destroyed.A group of young people in a nearby sub-county stated that the loss of trees was also destroying the foundation of communal solidarity by erasing the physical reference points for traditional authority within the community.A spiritual devastation followed, as the homes of spirits and trees with significant historical resonance were cut and shipped off.One person predicted that, in five years, 'this place will turn into desert'.
These narratives of tree cutting are not to be understood as a 'cultural' interpretation of an 'objective' phenomenon, a 'local narrative' of a 'global' process of climate change (Daniels and Endfield, 2009;Taylor, 2015).Rather, the tree cutting itself, together with the destructive ecological, social, and spiritual changes caused by it, including drought, are the disaster; they are climate change.Yet, these destructive transformations in people's surroundings were themselves not isolated; they were connected to longer histories of political violence.Together, they form parts of a narrative of devastation focused on the violence of the war, past and present.And so, to understand the drought and its consequences, as well as climate change, one needs to explore the history of these cultural, ecological, political, and social developments to grasp how drought is constituted as a social-natural phenomenon, part of a history of devastation.Central to comprehending these histories today is charcoal. 3

The war against trees
While hardwood timber extraction, the spread of commercial farming, and the expansion of grazing lands were all cited as factors causing the loss of forests, one culprit was dominant: large-scale charcoal production.The recent and still expanding industrial production of charcoal was widely blamed for the mass extraction of trees and the clear-cutting of forested land.As one woman explained: 'For me, I see that the time that charcoal burning was so much was in the year that just ended.There was a lot of charcoal burning, it was really too much; it was almost taking place every day'.Another emphasised that 'because of charcoal burning, hunger has come in.Charcoal has thrown hunger our way'.These sentiments were repeated in discussions across the region.
Grasping the origins of charcoal production requires exploring the demand for charcoal, the modes through which rural communities have been incorporated into it, and the larger trade networks that enable its generation and marketing.Charcoal production and trade are immediately related to the war and its legacies, whereas the demand for charcoal is determined largely by national and international political economies.The drought is, in this sense, spawned by forms of violence that emerged and were consolidated during the war and that are part of a longer history of devastation, while also being integrated into global political economies of energy.
Increasing demand for charcoal is a product of Uganda's rapid urbanisation (United Nations Development Programme, 2013); more than 80 per cent of the population of Kampala, the capital, and other urban centres is reliant on charcoal for meeting energy needs (Ministry of Energy and Mineral Development, 2016;Mukwaya, 2016).For years, most charcoal came from the central region, but with the end of open fighting in the north in 2006-07, a vast new reserve of trees for charcoal production was made available (Ministry of Energy and Mineral Development 2016;Miteva et al., 2017).As people moved back home after the war, having lost their cattle and savings, they were in need of cash to rebuild their lives and to pay school fees and for other essential services.Hence, people turned to charcoal production.At first it was largely small-scale and based on household labour.However, an industrialised mode of production began to take over within a few years.Charcoal dealers started to bring in their own work teams, which they place in camps to clear-cut large swathes of forest using chainsaws.Large trucks ply the remote dirt roads and tracks, ferrying workers and supplies in and innumerable sacks of charcoal out.Hundreds of migrant workers can clear many kilometres of land, extracting even the roots of trees, and leaving behind cratered landscapes of burning pits and the few scrub trees and bushes too small to burn.One parish chief claimed that the damage caused by commercial charcoal producers was so great that when land had been cleared of trees and used as a burning site, planting could not occur again for 10 years.Now, almost 40 per cent of Uganda's supply of charcoal to urban areas is reported to come from the north, whereas a decade ago it was negligible (Ministry of Energy and Mineral Development, 2016, p. 57).Available statistics on the scale of production are dramatic: in 2014, it was reported that Amuru and Gulu Districts' forest cover had been almost halved since 1990, and 4,000-5,000 bags of charcoal are estimated to be removed from just one district every week (Acholi Times, 2016a).Charcoal has come to comprise a 'violent environment' (Peluso and Watts, 2001), centred on a specific 'resource complex' (Watts and Peluso, 2014).
The dominant picture of the charcoal trade painted in the literature tends to posit a highly decentralised, fragmented sector with little vertical integration (Zulu and Richardson, 2013).In northern Uganda, though, the production of and trade in charcoal appears to be directed from the top, driven and made possible by the involvement, sometimes direct, of the state and the UPDF.As one informant underlined: 'The people who bring their workers . . . to produce charcoal here are military commanders'.People often cited the presence of armed men among the burning teams and transporters, as well as the involvement of military personnel without uniforms.Another person commented: 'The government workers give the rich people permission to cut the trees so that even if you complain, nothing will be done for you. . . .Those people have security that we can't stop'.Local government officials spoke about arresting illegal charcoal dealers and impounding their charcoal, only to get a call from military headquarters demanding that the person be released.Human rights activists reported receiving threatening telephone calls when they were seen as interfering with charcoal dealers or traders.Similarly, after the Local Council 5 Chairman of Amuru District, Michael Lakony, imposed a total ban on production, he announced that he had been threatened.He warned that: 'I have names and recorded voices of the big persons in government and the army which I will expose if they continue' (Acholi Times, 2016b).In sum, there is a widespread perception that those with power can grab land or extract trees for charcoal with impunity.
The involvement of the state and the UPDF means that the devastation caused by tree cutting, and the drought to which it is seen as leading, is widely understood as a continuation of the violence of the war and encampment.As one informant pointed out, the community was now experiencing the 'war that has been waged against trees'.The end of the conflict did not mean the end of state violence in the north; instead, it simply took on different forms.Those controlling tree cutting frequently are identified as the very people who were at the forefront of the state's counterinsurgency in Acholiland.Uganda's military has long been accused of looting natural resources, especially in the eastern Democratic Republic of the Congo and southern Sudan, contributing significantly to the consolidation of Uganda's business/political/military elite (Sjögren, 2013).The military seems to have brought these practices back to northern Uganda.The officers who were operating in Acholiland returned with peace to start exploiting its natural resources.There had been a small amount of logging of hardwoods during the war, but this looting escalated dramatically in the post-war period.In some cases, the post-war extraction of trees for charcoal and timber followed exactly the same routes as had the counterinsurgency: many of the security roads that were cut by the UPDF during the war to allow its military vehicles to reach remote locales are now the very roads that are being used to extract charcoal and timber.The militarisation of the north has continued, therefore, into the post-war period, as the same forces that had been involved in the devastation of the war-the government, the outsiders who were its agents, and collaborators within northern society-are now blamed for the deforestation that has produced drought.
Moreover, tree cutting has brought with it other forms of state-driven environmental violence, other forms of climate change as war.Most notably, land grabs for commercial farming are widely reported.Sugar cane plantations have attracted the most notoriety: one case has seen years of struggle between the community and the Madhvani Group, backed up by state security (Martiniello, 2015).But it is not just extreme cases where state violence is involved; as one person noted: 'the local people just wake up and find investors here, ploughing land without even knowing where they came from and for how long they would be staying here because of the connections the landowners have'.
Military force is apparently used to guard some commercial farms belonging to elites, and people assert that some who had opposed giving land to plantations had been threatened or arrested.Land that was formerly part of the commons has been enclosed, and environmental pollution is reported near plantations, including damage to water supplies, thought to be a result of the use of chemical fertilisers, herbicides, and pesticides.Another source of widespread environmental concern in the community is the arrival of large herds of cattle, brought by truck from the south and left to graze on land either leased or otherwise accessed by the cattle owners.These herds are blamed for intensive degradation of the land and for destroying crops and fragile communal areas.Like the charcoal dealers, the cattle keepers are largely considered to be untouchable because they are armed and because of their reported military and government connections.Even relations with wildlife have been incorporated into the matrix of political violence.Elephants have been leaving Murchison Falls National Park and destroying crops and threatening human life.People widely believe that the recent rise in oil exploration and traffic through the park is to blame for the frequent elephant incursions.Some who fight back against the elephants have recounted violent reprisals by park rangers for any suspected harm done to animals.Many communities said that they felt that their livelihoods and lives were being sacrificed by the government for the interests of the oil companies.Game parks, in particular Murchison Falls, are becoming increasingly militarised, and game wardens are seen as the agents of dispossession and state violence (Carmody and Taylor, 2016).

Vulnerability and adaptation to devastation
Drought in northern Uganda cannot be viewed as an isolated phenomenon, as a climate disaster caused by global climate change producing natural hazards that intervene in contexts of local vulnerability.Instead, it has to be seen as part of much broader regimes of devastation in which the dominant history is one of war, regimes that span the social and the natural, the global and the local, and the past and the future.Rather than treating conflict as something that either results from a future of global climate change or that contributes to local vulnerability, it should be seen as spanning both, dissolving the line between global natural hazard and local social vulnerability.The particular hazard at work in this case-drought and changing rainfall patterns-was produced locally by social forces as much as it was produced globally by natural forces (which, in the Anthropocene, were themselves also shaped by social forces).Vulnerability cannot be thought of as something purely local, as there was nothing purely local about the war in Uganda nor about the significant support that the Ugandan government has received from international donors over the past two decades, making possible the conflict and contemporary militarisation.Also stretching far beyond the local are the economic and political interests advancing the extraction of trees.A case in point is the international political economy in which East Africa's urbanisation is being pushed forward relentlessly, but energy for it must come from the massive, destructive dispossession of forests, requiring state violence to be effected.Even global warming can be conceived of as a form of violence, whether the violence of decades of extractive, unequal capitalist development, or, even more directly, the emissions generated directly by the military-industrial complex (Fressoz and Bonneuil, 2016).
Climate change itself does not exist purely in the global sphere, but rather, it is occurring through myriad forces that must be defined, starting from the experience of those living amidst it.And so global climate change also must be reconceived-there are certainly transformations happening to planetary systems that can be discerned using scientific methods, but global climate change is also the aggregate of many forms of climate change taking place in locations across the globe.This means that the firm distinction between future environmental disaster and past vulnerability needs to be dissolved, too.Today's drought, once placed in its ecological, historical, and social context is embedded in the ongoing legacies of the war, but it is also one aspect of a much longer history of devastation, stretching from violence between slave traders and different armed clan groups, to the so-called pacification effort of the British and its forced population displacement, to the violence of the Idi Amin era, neoliberal structural adjustment, and extractivist 'development' now.
Drought, and the broader devastation faced by communities after the war, thus crosses the divides between natural and social, global and local, future and past, but not just along the lines of hazard and vulnerability.Both armed conflict and drought can be seen as parts of a long history of devastation that is punctuated by periods of armed violence and atrocity.To focus attention primarily on future disasters stemming from the natural hazards resulting from global climate change, and on the conflicts to which they might give rise, is to ignore, and do injustice, to this long legacy of violence and war.It is also to disregard the narratives of climate change, drought, and devastation among those who are affected by these disasters, and to pay no heed to what they present as needing to be solved to ensure that there is a future.
Adaptation, resilience, and preparedness as part of the climate-change disaster model thus have the potential to lock in precisely this reality of devastation.Interventions can place existing economic, political, and social conditions beyond question, outside of the realm of what can be addressed by climate-change-focused policies.In fact, these interventions can even entrench the existing economic, political, and social order by endeavouring to increase its resilience and to help it adapt (Barnett and O'Neill, 2010;Brown, 2011;Ribot, 2014).Violence is placed beyond question and prevented from being addressed justly; adaptation and resilience threaten to preserve the unjust order created and enforced through violence.To understand the relation between climate change and conflict as cause and effect may itself reproduce devastation as it informs policy interventions.The challenge is to ensure that the response to climate disasters prepares for the future by building on just engagement with the violence of the past and the present.(2014; 19(4)).See also Bettini (2013), Hsiang and Burke (2013), Theisen, Gleditsch, and Buhaug (2013), and Devlin and Hendrix (2014).
Paska Aber, Phionah Alanyo, Miriam Auma, Anneeth Hundle, Giuliano Martiniello, Laury Ocen, Eric Awich Ochen, Paul Omach, Arthur Owor, as well as Human Rights Focus-Gulu, from July-October 2017.This is one of a series of papers written or co-written by members of the research team.In total, 22 discussions were held in five districts of the Acholi sub-region.They took place in the Acholi language and were then transcribed and translated into English.A dozen individual interviews were held, all in English.The paper also draws on the author's longer-term research and work in the region for the past 15 years.

Correspondence
Dr Adam Branch, Alison Richard Building, 7 West Road, Cambridge CB3 9DT, United Kingdom.E-mail: arb209@cam.ac.ukEndnotes 1 For critical work, see the special issues of Political Geography (2007; 26(6)), Journal of Peace Research (2012; 49(1)), Climatic Change (2013; 123(1)), International Journal of Human Rights (2014; 18(3)), and Geopolitics Research ethics clearance was approved by the Research Ethics Committee of the College of Humanities and Social Sciences at Makerere University, and permission was granted by the Uganda National Council of Science and Technology and the Office of the President.Funding was provided through an Interdisciplinary Innovation Award (grant number AH/P008232/1) funded from allocations to the Arts and Humanities Research Council and the Economic and Social Research Council under the Global Challenges Research Fund.3For a more expansive discussion of the politics of charcoal in contemporary Uganda, upon which this paper draws, seeBranch and Martiniello (forthcoming).