Conflict Potential in a World of Climate Change

Address by Gareth Evans, President, International Crisis Group, to Bucerius Summer School on Global Governance 2008, Berlin, 29 August 2008.

It is a real pleasure to return to the Bucerius Summer School, bringing together as it does policymakers, academics and business leaders from a variety of fields, but still early in their careers, to discuss some of the most difficult and pressing questions facing the world today. It’s not often that I get a chance to speak to a group of the world's best and brightest professionals whose contributions are still largely ahead of them rather than behind them!

This year's program covers, as always, many of the world's most important and urgent challenges, both longstanding and recurrent problems like regional insecurity, inequality and the need to make our international institutions more responsive and capable, but also new sources of anxiety, instability and potential human misery which are different in nature or scale, or both, from anything which policymakers have previously had to confront. And climate change is the mother of all such issues.

Climate change research has developed rapidly over the past decade, with findings resting on a much more robust and comprehensive set of data than ever before. There is now broad agreement that human activities are increasing global surface temperatures at a significant rate and that temperatures will continue to rise through the 21st century and beyond, even in the event of concerted action to stabilise emissions. The past year – with its big debates at the UN Security Council and in statements from EU and G8 states, all following on from the groundbreaking reports of the Intergovernmental Panel on Climate Change – has underscored the central position that climate change now occupies in international policy and strategic thinking.

A recurring theme in the debate – and certainly a staple in all the rhetoric associated with it – has been the potential impact of climate change as a cause of deadly conflict. It has to be said, however, that this dimension of the debate has not always been as nuanced as it might, and that some of the contributions to it might be more persuasive if they were a little more cautiously expressed.

What I'd like to do today is take you through what I believe can, and cannot, be said about the relationship between climate change and conflict, making three basic points:

First, there is unquestionably a general causal connection between the two, at least in the sense that climate change is a "threat multiplier".

Second, that there are, nonetheless, real problems in trying to assess the future impact of climate change in any particular country or region, which has implications for the kind of policy prescriptions that can usefully be offered by governments, or research and advocacy organisations like my own.

Third, that said, there are still some useful general policy positions that can be taken, and that we should be working now on implementing.

General causal connections

The Intergovernmental Panel on Climate Change's 2007 report, and the wealth of publications, events and statements that followed, tell us that many societies may already be suffering the early effects of climate change, and have predicted that in a number of broadly defined regions there will be over time major drops in food production, with shifts in rainfall patterns, accelerating desertification rendering land infertile, or sea-level rise inundating farmlands and furthering the spread of disease.

There are significant variations in the extent to which societies are dependent on climate sensitive resources, and in their likely capacity to withstand the socio-economic impact of climate-change induced shortages. Nicholas Stern’s 2006 report on the Economics of Climate Change for the UK Government foreshadowed some of the findings of the IPCC. He argued that developing countries are particularly vulnerable – due to their dependence on agriculture, high population growth, weak infrastructure and reduced capacity to adapt to climate pressures. That the developing world has contributed least to current rates of global warming makes these findings all the more disturbing.

It is easy enough in all of this to identify highly credible causal connections between climate-change and new security problems, with climate impacts generating wholly new tensions, or intensifying existing societal fault lines and operating as a "threat multiplier" in already fragile regions. There are four connections in particular which seem highly plausible:

• Diminishing access to water, land, or returns on the use of land could increase competition for resources and in turn lead to violence.
• The same declining access to resources could cause people to move in mass numbers – "environmental refugees" – potentially destabilising neighbouring areas.
• Increased climate variability – in the form of drought, flooding, cyclones – can produce economic shocks, reducing employment opportunities and increasing recruitment to armed groups, in turn increasing the capacity of those groups to wage war.
• Environmental migration not just to neighbouring states but to the global North, and divides between responsible and affected states, could strain already fragile relations between North and South – in turn compromising efforts to strengthen dialogue on many issues that demand a genuinely global response, including security issues like responding to terrorism and mass atrocity crimes.

There is ready evidence that any of these pressures can exacerbate humanitarian and security strains with dire effects. Food and fuel shortages in parts of North and West Africa, Haiti and elsewhere, for example, while essentially political and economic in origin, resulted in a series of violent protests earlier this year, in some cases prompting a brutal government response. In Mali, competition for territory and access to natural resources has driven a deadly, decades long conflict between Tuareg rebels and government forces, in a region already marked by political instability and widespread poverty. And there is ample experience in many contexts of large-scale movements of people, across borders or within conflict affected states, generating humanitarian and security risks.

The need for policy caution

But what does all this mean for policymakers here and now? How does it translate into an actionable policy agenda? Here it is important to recognise the very real limitations on what we can sensibly say about the climate-conflict connection, as we move from broad generalisations to particular policy prescriptions.

The first big problem is simply predicting where, precisely, the impact of climate change will fall. Despite considerable and continuing advances in climate change research, we are not yet able to determine where and when changes will occur with any kind of precision. Climate science is not so precise as to enable the assessment of impacts to be country-specific or even wider region-specific, let alone at the sub-national level.

There are for a start unresolved issues which affect prediction generally, like the acknowledged difficulty in assessing the "feedback" effects of global warming. Will melting ice caps mean that with more open water and land exposed to solar radiation, absorbing more of it because they are less reflective than ice, there will be more warming and even more melting? Or will increased temperatures cause particular cloud cover changes, meaning that incoming solar radiation will be reduced and future warming more limited?

There are also more geographically specific problems. Gaps in data, particularly in regions affected by conflict, often in developing countries (and particularly in those affected by conflict), impede systematic and analysis of climate trends, both generally and geography-specifically. As the IPCC’s recent report states, the short time scales of specific climate studies and their limited spatial coverage limit constitute further barriers to precise regional mapping.

And in assessing impacts sub-national and localised environmental factors become relevant. In a country as large as Kazakhstan, for example, environmental conditions vary considerably – with pollution levels, past radiation exposure and farming practices continually altering water access and supply.

All these difficulties have at times combined to produce radically divergent accounts of political and security risks in certain regions. One recent paper, for example, exploring the relationship between climate change and the onset of civil conflict in Sub-Saharan Africa, used a climate model to analyse likely rainfall patterns in sub-Saharan Africa from 2006 through 2059. The authors found that overall levels of rainfall were actually expected to increase, while at the same time rainfall variability would remain relatively stable, over the next five decades. This finding was in some tension with global climate change projections of scarcity and drought, and increased variability, and led the authorts to suggest that, "the cataclysmic predictions linking climate change and human security" may not apply to sub-Saharan Africa.1

But then again they may. The point is we don't quite yet know: the climate impact predictive models are just not yet good enough., More work is being done to make climate projections more specific and enhance our ability to assess country or regional impacts with confidence. But we are certainly nowhere near, now, the level of confidence that policymakers would like to have.

Even if we could be confident about how future climate change will physically impact on particular societies, the second big problem is judging how that will actually affect human societies, economies, patterns of cooperation and confrontation – and in turn how all this will affect the incidence of violent conflict. And here again we need to be very cautious indeed about jumping to conclusions about particular cases.

We certainly know from experience that natural resources can play a decisive role in conflict situations. Political radicalisation, internal violence and inter-state tension are some of the visible outcomes of the "resource curse" – where energy and minerals-rich countries either lose the wider benefit of their incomes through exchange rate effects, or waste them outright through corruption or misallocation, failing to diversify their economies, educate their people and develop effective and accountable institutions. Conversely, frustration stemming from chronic resource shortage can serve as an important impetus to take up arms.

But in looking to the reality of today's conflicts, and tomorrow's likely ones, identifiable environmental factors invariably interact with multiple other variables – the all too familiar issues of poor government, failures in leadership, ethnic tension and inequitable systems for distributing resources that together drive some of today’s most violent and intractable wars – making it difficult to judge how environment will affect a particular situation.

To take one such example, a number of suggestions have been made over the last two or three years that the "real roots" of the conflict in Darfur lie in long periods of drought in the 1970s and 80s – pushing nomadic communities southwards and leading to confrontations with sedentary Fur and Masalit tributes – and that this is the "first climate change war". But this is an extreme simplification of a very complex situation in Sudan, the main driving dynamic of which has been the determination of the political centre – the NCP regime in Sudan – not to allow the transfer of any real power to the country's long marginalised peripheries. Darfur is simply one of a number of overlapping circles of conflict – there's the five-year-old war between the Darfur rebel movements and the government; the proxy war that Chad and Sudan are fighting by hosting and supporting the other's rebel groups; and the longstanding conflict between North and South, with the negotiated peace there highly fragile as recent events in Abyei have vividly demonstrated.

There have been recurring land tensions between sedentary and nomadic tribes in Darfur and elsewhere, but, as recently underlined by the UN Environment Program for Sudan, increasing rural poverty and displacement to cities have been a more measurable outcome of environmental degradation than conflict as such. More generally, it is interesting to note that the United Nations Human Development Report of 2006 concluded that while reduced access to water constituted a significant threat to the realisation of development goals, it had not proved a major security risk. Any detailed analysis of the evolution of the crisis in Darfur or anywhere else, makes clear the sheer number and complexity of factors which affect when and where violence will emerge, and whether it will continue.

Further complicating any attempt to leap into confident predictions about the impact of climate in generating conflict, is the growing body of work stressing the potential climate change may actually have for generating intra- and inter-state collaboration – in other words, conflict prevention. Water is an important example. While its distribution has certainly often generated tension between states – as it is now doing for example in Central Asia – historically, water scarcity has more often worked to favour cooperation between them. Interstate dialogue prompted by diminishing water supplies, particularly, can build trust and institutionalise cooperation on a broader range of range of issues. Pakistan and India is a current example, with the need to manage water distribution being an incentive to conflict negotiations, with one of the six committees established to manage tensions in 2004 being explicitly devoted to water management.

A much more localised example comes from Uganda. In two regions of this country long affected by a brutal civil war, international organisations have been working with local NGOs, local service providers and affected communities to develop water provision schemes to service the local area. All parties have been brought together to identify how project implementation might best be used to reduce existing community divisions and promote dialogue between local governments and the localities they serve, dialogue that has continued and consolidated since. The value of the project has not only been to reduce current and latent source of tension. It also worked to institute accountable democratic structures and, critically, strengthen capacity to manage conflict without recourse to violence, thereby helping to reduce the potential for its future emergence.2

All this makes for a need for caution in any talk about future "resource wars". Environmental stress can form an important backdrop to future violence, reduce opportunities and avenues for conflict resolution and fuel long-term patterns of instability. But it is rarely sufficient in itself to explain large-scale violence, and it may even lead to cooperative outcomes where we least expect it. The crisis in Darfur, violence in the oil-rich Niger Delta, ongoing tensions over access to water in Central Asia all have a clear environmental dimension – but we must not lose sight of the specific and political causes of violence that fuel instability and sadly will likely continue to in the future. The bottom line is that every conflict has its own dynamic, and there is no substitute for understanding all the factors at work.

Bottom lines for policymakers

For all the need for caution that I have been stressing, none of this means that there is nothing useful that can be said or done in policy terms about the climate-conflict nexus. One can take action by way of mitigation, and adaptation.

For all the problems that exist in making the connection in particular cases, the connection in general terms is compelling enough for this to be an important reinforcing argument – supporting all the others out there – for effective global action, now, to mitigate climate change, through a coordinated effort from all states, backed up by the necessary resources, to reduce carbon emissions in order to slow and hopefully eventually stop global warming. This is a critical time to be taking stock, as states meet for the third time this year, in Accra, to discuss how to broker a new post-Kyoto agreement that can drastically curb emissions, while assisting developing countries meet targets without undermining work to reduce poverty and promote economic activity. Without binding commitment to stem the pace of climate change our efforts to manage its impacts will always be working against the tide.

Adaptation means policymakers taking action right now to improve the capacity of societies to adapt to the effects of climate change, whatever they may be and whenever they may be felt. The focus should be on limiting vulnerabilty to potentially damaging socio-economic effects and associated human security risks, for example by:

• developing initiatives to reduce reliance on climate sensitive activities, improve governance and invest in physical infrastructure;
• making efforts to bolster disaster preparedness and early warning, including improving military and civilian rapid response capabilities;
• incorporating forward-looking resource management considerations in peacebuilding and post-conflict reconstruction efforts; and
• accelerating diplomatic efforts to encourage cooperation over resources before environmental stresses increase.

These propositions closely track some of the basic tools that NGOs like mine have long been advocating for conflict prevention generally, and that many governments and intergovernmental organisations like the UN and EU have been working to improve and refine – good governance, early warning, resource management, effective and timely diplomacy. Much of the framework for implementing climate sensitive responses is, as a result, to some extent already in place – the real challenge, as always, is mobilising the political will and the resources to make it happen.

A good example of the role of effective governance and institutional structures in mitigating conflict risk associated with environmental stress – and, by extension, the negative impact of poor governance and a collapse of necessary institutional structures – comes from Central Asia. The erosion in the post-Soviet era of political structures for coordinating access to shared water reserves – centering on two rivers flowing to the Aral Sea – has had a defining role in the escalation of inter-state tensions. The removal of Soviet oversight, rising nationalism and inter-state competition have all played a role in preventing the emergence of viable and accountable systems for negotiating access to supplies. These factors must be set against parallel socio-economic pressures – rising demand for water to service the dominant cotton industry, and falling supplies, driven partly by the widespread use of wasteful and inefficient irrigation practices. Mismanagement and corruption at the state level has in turn prevented translation of cotton revenues into meaningful social and economic development – creating the conditions for economic stagnation, widespread poverty and political repression at home as elites seek to retain the bulk of cotton's spoils. It is a heady and dangerous mix.

Policymakers in most parts of the world have, it needs to be acknowledged, come a long way in recent years in recognising the link between development and security. The fields were for a long time neatly drawn, with little overlap: development specialists dealt with poverty alleviation; diplomats and defence experts focused on security issues. But the idea of "conflict-sensitive development" has increasingly entered the lexicon of international policymaking – on the back of research confirming the myriad costs of violent conflict for human, economic and environmental security and the evident risk that poorly targeted aid poses for exacerbating conflict’s dynamics.

Responding to potential future seeds of instability – the ‘traditional’ development issues of inequality, poverty or resource stress – should form part of a wider agenda of preventing conflict outbreak and recurrence. This includes work to develop accountable political institutions, create equitable systems for managing resources and consolidate the rule of law. Such a forward-looking approach is essential to ensure that societies are able to manage socio-economic stress – whether driven by climate change or otherwise – quickly, effectively and without violence erupting.

In summary, we know that climate change is a hugely important global issue, and we have multiple reasons for tackling it – preventing potential new causes or multipliers of deadly conflict is just one of them. From a conflict analysis perspective, we must remain focused on the intimidating task of trying to understand the multiple pressures and tensions that fuel contemporary violence – of which those induced by climate change are just some. Climate change should not dominate conflict analysis, nor should anxiety about new sources of conflict dominate the climate debate. But there is sufficient connection between them for us to know that in redoubling our efforts at both mitigation and adaptation we certainly won't be wasting our time.

1. Cullen S. Hendrix and Sarah M. Glaser, "Trends and Triggers: Climate Change and Civil Conflict in Sub-Saharan Africa", Political Geography, Vol. 26, Issue 6, August 2007.

2. Saferworld, "Water and Conflict: Making Water Delivery Conflict-Sensitive in Uganda", August 2008 (available at http://www.saferworld.org.uk/publications.php/355/water_and_conflict).