Environmental change and forced migration


“So if we really want to deal with the root causes of forced migration, the first step is to stop Northern practices that make things worse in the poor countries of the South.


A much more important – but also more difficult – second step is a reform of global rules on trade, investment and intellectual property to give countries in the early stages of development and industrialisation better economic and social opportunities.


Eliminating forced migration is thus a long-term project, closely bound up with global power relations and the North-South division.”

Scilla Elworthy

Stephen Castles

6th December 2001

"Fellow of Green College, Director, Refugee Studies Centre, University of Oxford" 'Stephen Castles is Research Professor of Sociology, University of Sydney, Australia and Research Associate Director at the International Migration Institute (IMI), University of Oxford, UK. He is a sociologist and political economist, and currently works on migration and development, effects of migration on both origin and destination countries, transnationalism, and social transformation

and human mobility at a global level. From 2001-2006, he was Director of the Refugee Studies Centre at Oxford University, and from 2006 to 2009 founding Director of IMI."


source: Age of Migration


It is a great pleasure and an honour to present this lecture to you this evening, and I am grateful to Green College, and especially the Warden, Sir John Hanson, for giving me this opportunity. In fact, in a sense Sir John is responsible for this lecture – or perhaps I should say he is to blame – for as a sociologist working on migration and refugee issues, I had not been closely concerned with environmental matters until he asked me to speak on ‘forced migration and the environment’ a few months ago. My perspective then, is that of a social scientist and not of an ecologist, although working on this theme has certainly made me realise the need to bridge this disciplinary divide.


I must say that I was initially somewhat daunted by the task. However, this was tempered by the discovery that two important sources of expertise were close at hand. Dr Norman Myers, who has written extensively on environmental change and population displacement is a Visiting Fellow of Green College. His work has been highly influential, especially a report entitled Environmental Exodus: An Emergent Crisis in the Global Arena, written with Jennifer Kent, and published by the Climate Institute of Washington DC in 1995 (Myers and Kent, 1995). My other source was Dr Richard Black, Senior Lecturer in Geography at Sussex University, and co-editor of the Journal of Refugee Studies (the main academic publication of the Refugee Studies Centre). Black has done a great deal of research on refugee issues in Africa and elsewhere, and has written a book entitled Refugees, Environment and Development (Black, 1998). All I needed to do was to study these two works, and possibly consult the authors, and I would understand the issues.


Unfortunately, this was not the case. I rapidly discovered that my two eminent colleagues disagreed totally on many key points. Myers highlights the plight of what he calls ‘environmental refugees’:


There are fast-growing numbers of people who can no longer gain a secure livelihood in their homelands because of drought, soil erosion, desertification, deforestation and other environmental problems. In their desperation, these ‘environmental refugees’ ...feel they have no alternative but to seek sanctuary elsewhere, however hazardous the attempt.


Black questions the value of the very notion of ‘environmental refugees’.


A UNHCR Working Paper by him is entitled "Environmental Refugees: Myth or Reality?" (Black, 2001). He leaves the reader in no doubt that he sees the concept as a myth – and a misleading, highly politicised and potentially damaging one at that.


Myers claims that there were at least 25 million environmental refugees in the mid-1990s, and that this unrecognised category exceeded the then 22 million refugees as officially defined. He thought the number of environmental refugees might well double by the year 2010, and could rise even more quickly as a result of global warming. As many as 200 million people could eventually be at risk of displacement (Myers, 1997).


Black argues that there are no environmental refugees as such. While environmental factors do play a part in forced migration, they are always closely linked to a range of other political and economic factors, so that focussing on the environmental factors in isolation does not help in understanding specific situations of population displacement.


Myers believes that environmental changes and the natural and man-made disasters associated with them are forcing millions of people to flee their homes. This does not imply that environmental factors always lead directly to displacement. Rather, environmental pressure leads to land competition, impoverishment, encroachment on ecologically fragile areas and impoverishment. These events in turn cause political and ethnic conflicts which may precipitate violence and war – often the immediate cause of flight.


The environmental refugees may end up in urban slums, or camps for internally-displaced persons within the country of origin. Millions, however, leave their country. They may seek refuge in neighbouring countries of the South, where they may cause further environmental problems and conflicts. But many, according to Myers’ gloomy prognosis will try to obtain asylum in the developed countries of Western Europe and North America. The issue of environmental refugees thus ‘promises to rank as one of the foremost human crises of our times’ (Myers, 1997, 175). The rich countries are closing the door, but it will be impossible ‘to hold back the rising flood’ of refugees. Refugee camps and shantytowns will become ‘breeding grounds for civil disorder, social upheaval and even violence’. There may be ‘substantial outlays to counter pandemic diseases and deficits of food, water and energy’. The result could be threats to social cohesion and national identity, leading to ethnic tension and civil disorder (Myers and Kent, 1995, 151-3).


Black rejects this apocalyptic vision, and considers it a neo-Malthusian approach based on dubious assumptions. Moreover it constructs refugees and migrants as a threat to security. He claims that there is no evidence that environmental change leads directly to mass refugee flows, especially flows to developed countries. He sees the emphasis on environmental refugees as a distraction from central issues of development and conflict resolution.


Finally, it is important to note a difference in methodology. The Myers and Kent study refers to about 1000 sources, but takes its main impetus from broad-ranging global prognoses on such themes as population growth, climate change and resource constraints, for instance from the UN’s Inter-Governmental Panel on Climate Change (IPCC). Black by contrast uses mainly national or local studies (including his own fieldwork) on the links between the environment and forced migration in specific situations.


Why do distinguished colleagues disagree so profoundly? Is the difference a disciplinary one between ecologists and geographers, or is the issue the varied perspectives of an environmental expert and a refugee specialist? Is the disagreement methodological or essentially political?


Myers and Black represent opposing positions in an academic and political debate that started in the mid-1980s with a paper on environmental refugees published by the United Nations Environment Programme (El-Hinnawi, 1985). Since then, much research has been done, and several special conferences have been held by such bodies as UNHCR and the International Organisation for Migration (IOM). How can non-specialists make sense of this debate, and weigh up the merits of such divergent viewpoints? It is important to do so, for publicly-accepted views on the linkages between environment and forced migration has far-reaching political consequences, as I will discuss later on.


I will now share with you my personal efforts to gain some clarity. I will discuss some of the research evidence on environmental change as a direct cause of forced migration; then look at claims that environmental change causes violent conflicts which in turn cause mass refugee flows. I will look at transmigration in Indonesia as an example of the complex relationships between migration, environmental change, ethnic conflict and displacement. Then I will return to the concept of environmental refugees and the politics of definitions, before concluding with some observations on the need for fundamental changes in North-South relationships if we really wish to eliminate involuntary migration. I should note that for reasons of time I am going to omit a closely related issue: whether refugee inflows and camps cause environmental problems in areas of settlement.


Environmental change as a direct cause of forced migration

What is the evidence that environmental change cause large flows of refugees or migrants? Myers sees the causes of environmental displacement in such factors as desertification, deforestation, lack of water, salinisation of irrigated lands, and bio-diversity depletion. All of these are linked to rapid growth of population in less-developed countries as well as to global climate change. These macro-level changes lead to pressure on land and other resources. They also exacerbate the effects of extreme weather events, natural disasters and man-made disasters (like Bhopal or Chernobyl) (Myers and Kent, 1995). Based on these observations, Myers provides lists of the millions of people at risk of displacement from desertification, deforestation, rising water levels and so on. However, he does not provide figures on people who have actually been displaced by such problems. Rather, the linkage appears simply as ‘common sense’ – if water levels rise, or forests disappear, it seems obvious that people will have to move.


But is it? Black, after reviewing a wide range of studies on these phenomena, claims that there is no convincing evidence that they lead to large-scale displacement. In some cases, he argues that the precipitating causes themselves are doubtful. Desertification, he claims, has been shown by new techniques like satellite imaging to be mainly a cyclical phenomenon linked to rainfall patterns. Migration in the Sahel zone and similar regions is a coping strategy used by the people for centuries, and is also cyclical rather than permanent. Migration is thus ‘an essential part of the economic and social structure of the region, rather than a response to environmental decline’ (Black, 1998, 28) By contrast, Black sees problems of rising sea levels, declining water supplies and so on as very real. But he finds little evidence of actual permanent large-scale displacements caused by these factors.


Rather than looking at global forecasts, he argues, it is important to examine the strategies adopted by communities and governments in specific cases. These may be prevention strategies (eg, building dykes to stop flooding or replanting trees to stop land erosion) or coping strategies, such as providing relief and rehousing people after disasters. Often natural disasters (such as volcanoes and earthquakes) lead to temporary displacements. After the Kobe earthquake in Japan 300,000 people were displaced, but within 3 months the figure had fallen to 50,000. In the Philippines, by contrast, many of the people displaced by the Mount Pinatubo eruption were still living in temporary camps or squatter settlements several years later.


This points to the important role of the state: a strong, efficient state can deal with environmental problems much better than a weak and possibly corrupt state. The key problem then is perhaps not environmental change itself but the ability of different communities and countries to cope with it. This in turn is closely linked to problems of underdevelopment and North-South relationships.

William B. Wood, the official Geographer of the US Department of State has stated that:


       "Anti-immigrant rhetoric and apocalyptic forecasts of environmental disaster ..may also be

       obfuscating a rational policy discussion... Indeed, focussing attention primarily on such a

       long-term and worldwide phenomenon could mask the more immediate reality of many

       dispersed, and localised ecological crises and the fact that there is usually no simple

       relationship between environmental causes and societal effects."

       (Wood, 2001)


So it seems that general forecasts and common sense linkages do little to further understanding. It is essential to look at specific cases. Most researchers working in this field do just that. Myers and Kent discuss, among other examples, Ethiopia and Eritrea, Somalia, Sudan, Rwanda, Nigeria, Bangladesh and China. Without going into detail, their argumentation appears largely deductive: country X has environmental problems and also has large numbers of emigrants and refugees. Therefore there must be a causal linkage. The method seems particularly problematic when applied to future prognoses: Nigeria, they show has rapid population growth and is likely to have future problems of desertification, soil erosion and water pollution. Therefore mass exoduses are likely to take place. Yet Nigeria is not a significant country of emigration – in fact it attracted many immigrants in the oil boom years of the 1970s. It is rash to make predictions on such tenuous evidence.


The issue is above all one of causality. Where Myers and Kent go into detail, they find a wealth of contributory factors: ethnic tensions, ineffective and mistaken government responses, economic problems and so on. On what basis are environmental factors assigned primacy in complex situations? This never becomes clear. The Rwanda disaster is often portrayed as a classical case of population growth putting pressure on scarce land, and thus precipitating ethnic conflict between the Tutsis and Hutu. Yet it could just as well be seen as a political struggle for power in which both ethnicity and natural resources played a major part. The legacy of Belgium colonial practices of divide- and-rule also played a part. Were the millions of refugees in 1994 fleeing environmental pressure or genocide and civil war?


Other scholars have sought to disentangle the complex causes of forced migration. Korean researcher Shin-wha Lee explores the ‘environment-security nexus’ and puts forward a ‘model of the causes of environmental refugees’ (Lee, 2001). She also looks in detail at a number of cases, including Bangladesh, Sudan and North Korea. In fact her model shows the complex interaction between ecological factors, human-induced diasters, governmental factors (such as inaction, incapacity and corruption, as well as harmful policies), and international factors. In the same way, all her cases studies show multiple causes of forced migration.


Bangladesh, with its extremely dense population and its exposure to cyclones and flooding, appears as the quintessential example of environmental displacement. Yet even here Lee finds complex causes for impoverishment and flight, including land ownership patterns, ethnic divisions, economic development projects such as dams, and political conflicts. The action – or more often the inaction – of the Bangladeshi government is a major factor causing forced migration. Even the Indian government plays a part, since the Farakka dam on the Ganges upstream from Bangladesh did much to reduce water supply and endanger agricultural production in the Ganges delta (Lee, 2001, 73-83).


Lee’s study of the famine in North Korea which has claimed 2-3 million lives comes to similar findings: the country was hit by unprecedented flooding and drought in the mid-1990s, but the real blame for starvation lies with the country’s military-first policies and inefficient command economy. Moreover, international food aid has become a political foot-ball: the regime seeks to use it as a bargaining counter in international relations, while diverting food for military purposes; donor countries try to use food aid as a lever to achieve political objectives concerned with stopping the development of nuclear weapons and bringing about talks with the South. In the meantime, North Koreans starve, or seek to flee to China, where they get a frosty reception.


No wonder Lee comes to the conclusion that both Bangladesh and North Korea illustrate Amartya Sen’s principle that the roots of famine lie not in lack of aggregate food supply, but in the failure of individuals’ entitlements to food. The problem is primarily political and social – not environmental.


My conclusion from this research literature is therefore that the notion of the ‘environmental refugee’ is misleading and does little to help us understand the complex processes at work in specific situations of impoverishment, conflict and displacement. This does not mean, however, that environmental factors are unimportant in such situations. Rather they are part of complex patterns of multiple causality, in which natural and environmental factors are closely linked to economic, social and political ones. This is where we need much more research and better understanding, if we are to address the root causes of forced migration.


Forced migration and conflict

If environmental change does not lead directly to forced migration, is it rather a major cause of violent conflict, which in turn lead to flows of internally displaced persons and refugees? If this case can be made, then environmental change would indeed be a major global security issue. Observers have noted attempts to ‘securitise environmental issues’ (Goldstone, 2001, 38), drawing attention to the Myers and Kent Report, as well as a major project designed to examine the links between environmental scarcity and violent conflict headed by Thomas Homer-Dixon of the University of Toronto (Homer-Dixon and Percival, 1996). Homer-Dixon claims that we ‘are on the threshold’ of an era in which traditional security concerns such as armed conflicts will come frequently, if not primarily, as a result of environmental change (Homer- Dixon, 1991)


Such claims are an important issue in the growing sub-disciplines of ‘conflict studies’ and ‘political demography’. Forced movements of population are increasingly perceived as a major factor in generating conflict between states and the use of force. Many of the international military interventions of recent years have had the prevention of refugee flows as one of their main objectives. The list includes the establishment of a ‘safe haven’ for Kurds in Northern Iraq after the Gulf War, the US intervention in Haiti in 1994, and the NATO intervention in Kosovo in 1999 (Castles and Miller, 1998, 122, 293; Weiner and Russell, 2001, 5). Fear of refugee flows has also played a major part in Western strategies in the current Afghanistan conflict. If environmental factors lead to refugee flows this would be a powerful reason for the ‘international community’ to take pre-emptive action.


So what is the evidence? Black reviews 11 major refugee- producing conflicts of the 1990s and states that several of them, far from being in resource-poor areas, are precisely about control of valuable resources, especially oil, for instance the Gulf War, Sudan. Azerbaijan and Kazakhstan.


In other cases, like the Great Lakes Region, Somalia and Sierra Leone, environmental factors do play a part, but seem far less important than ethnic conflicts and political power struggles (Black, 1998, 31-5).


A more comprehensive study of conflicts between 1980 and 1992 by Hauge and Ellingsen (Hauge and Ellingsen, 1998) found a positive correlation between land degradation, deforestation and water scarcity and civil war. However, the magnitude of the effect was very small, raising the probability of conflict by only around 1 per cent. The causal effects of other risk factors such as poverty, regime type and current and prior political instability was far greater. A number of other major studies summarised by Goldstone come to the same results: long-term environmental change factors are not of themselves major causes of violence (Goldstone, 2001, 41-2).


This does not mean that environmental factors do not cause conflict, but rather that it is usually not violent conflict, especially civil or international war. This is because disputes on such matters as water rights on a river that crosses international boundaries cannot be resolved through military force. The costs, as well as the environmental and human damage caused by war will almost always exceed any potential gains. Such disputes are, according to Goldstone, a new type of non-violent environmental and demographic security issue , that has to be resolved through negotiation or arbitration. He gives examples of negotiations on water rights between India and Bangladesh, Israel and Jordan, and Hungary and Slovakia (Goldstone, 2001, 43).


On the other hand, short-term disasters such as hurricanes, droughts, floods, earthquakes and industrial accidents may well have major political repercussions. This is not due to the event itself but to responses by the government concerned. In Nicaragua for instance, the Somoza regime used international aid after the 1972 earthquake as a source of self-enrichment. The resulting protests led to the Sandinista uprising. The response of the Pakistan Government – dominated by West Pakistani elites – to the 1970 cyclone in the then East Pakistan was one of indifference. This helped precipitate the conflict which led to the birth of Bangladesh. In China, by contrast, the Government responded to the 1998 floods (which were partly a result of Government-supported logging), with a massive relief effort led by the army. This helped gain support for the regime. Thus natural diasters do not in themselves lead to rebellion or civil war, but only where inadequate responses show the failure or corruption of the regime (Goldstone, 2001, 45-6).


To sum up: there does not appear to be a convincing case that environmental factors cause major violent conflicts which in turn lead to massive flows of forced migrants. Other factors, such as political divisions, ethnic rivalries and economic interests seem far more important in causing violence and war. Again, we should not neglect environmental issues, but rather understand them as part of much broader processes of societal change.


Indonesian transmigrasi as an example of complex linkages between environment, conflict and displacement

This means that scholars of forced migration (and indeed of migration in general) need to sharpen their perception for environmental factors. I suspect that if we do this, we may find environmental factors in many migratory situations where we did not suspect them in the past. I would like to illustrate this with an example from Indonesia, where the Refugee Studies Centre is currently establishing cooperation with the Institute of Cultural and Social Studies at LIPI, the Indonesian Institute of the Sciences. The aim is to set up a research unit on internally displaced persons (IDPs).


For many years, migration specialists have been very interested in Indonesia’s transmigrasi (or transmigration) programme, designed to relieve population pressure in densely populated Java, Bali and Madura by encouraging families to resettle in the sparsely populated outer islands, especially Sumatra, Kalimantan and West Papua. Transmigrasi was actually started by the Dutch colonial power in the early 20th century, but continued and expanded by both the Sukarno and Suharto regimes. It can be seen as the world’s biggest ‘demographic engineering’ project. Indonesia’s five-year plans aimed to move very large numbers: 250,000 families from 1974-79; 500,000 families from 1979-83 and 750,000 families from 1983-88. These targets were not fully achieved, due to lack of institutional capacity, but very large numbers were moved. In addition there were many ‘spontaneous transmigrants’ who took advantage of the infrastructure and economic opportunities opened up by the official programme (Tirtosudarmo, 2001a).


Most scholars have been interested in transmigrasi because of its economic and demographic consequences. However, a new aspect arose in the years of instability around the collapse of Suharto’s ‘New Order’ regime in 1998: violent conflicts broke out between indigenous groups and settlers from the inner islands. These were particularly intense in Kalimantan, where Dayak tribesman attacked the predominantly Madurese transmigrants. Hundreds were killed and thousands of Madurese families fled their villages and had to be evacuated by the Government. Such actions were widely portrayed as the result of long-standing ‘ethnic hatreds’ which came to the surface once the lid of military control was removed,. No doubt cultural and religious differences did play a part, but it soon became apparent that there were other factors at work, including environmental ones.


Transmigrasi was not just a demographic policy but also part of a strategy for economic modernisation and introduction of market crops into areas previously given over to subsistence agriculture and hunting. The Dayaks practised traditional swidden agriculture – that is slash-and-burn cultivation with long fallow periods to regenerate the fragile soils. The Madurese transmigrants cleared the land, and tried to set up the sort of permanent agriculture they were used to, with little knowledge of local conditions (Tirtosudarmo, 2001b). They also cleared native forest to plant cash tree-crops like rubber and coconuts. Many also worked as wage-labourers for logging companies. Transmigrants in Kalimantan and elsewhere, became a significant factor in deforestation (Sunderlin and Resosudarmo, 2001). The effects for the Dayaks were devastating:


A close observation of the lives of the indigenous Dayak people under the rapid external pressures induced by the central government’s development strategy basically confirmed that the very resources upon which they have depended for centuries – the land, forests, and rivers – will no longer be able to sustain them. The ongoing exploitation of natural resources in Kalimantan, in the last two decades, has forcefully transformed the local people into marginal peasants, estate workers and urban wage labourers. (Tirtosudarmo, 2001b, 18)


This is the environmental background to these ethnic conflicts. But even here, it is clear that focussing on one cause alone is misleading. As already pointed out, transmigrasi was a both a demographic and an economic policy. But it was also concerned with central control of the outer islands, and with strengthening the role of the Javanese military establishment. The Suharto regime saw transmigrasi as an instrument of nation-building and modernisation, through which indigenous cultures and economies would be transformed and integrated into the wider nation (Elmhirst, 1999; Tirtosudarmo, 2001a). Add to this the poor management of the programme and the lack of preparation of transmigrants for the very different environmental and cultural conditions, and we have a recipe for disaster.


What can we learn from this example? Obviously it illustrates yet again the complexity of the causes of conflict and forced migration, but it also shows the significance of environmental factors in a conflict which appears at first sight to have other causes.


The growth of forced migration and the politics of definitions

Let me return to terminology and definitions. By now it should be clear that the term ‘environmental refugee’ is simplistic, one-sided and misleading. It implies a mono-causality which very rarely exists in practice. However, it is problematic for other reasons as well, The term ‘refugee’ has a precise meaning in international law. A refugee is defined by the 1951 UN Convention Relating to the Status of Refugees as a person outside his or her country of nationality who is unable to return because of a ‘well-founded fear of persecution on account of race, religion, nationality, membership in a particular social group, or political opinion’. Clearly, someone who flees due to environmental problems does not fall under this definition. Nobody gets asylum just because of environmental degradation.


The term ‘environmental refugee’ could only have a legal meaning in the narrow sense of people forced to flee when repressive forces use environmental destruction, such as defoliation or polluting water, as an instrument of war against a specific group. Cases include US use of Agent Orange during the Vietnam War and actions of the Iraqi Government against the Marsh Arabs. However, refugee claims of such groups would be based on persecution itself rather than the form of it, making the term environmental refugee redundant.


US State Department Geographer Wood therefore suggests using the term ecomigrant as a broader concept to include anyone whose need to migrate is influenced by environmental factors. Indeed he points to the useful ambiguity of the ‘eco’ part of the term, which can refer to both ecological and economic factors. He argues that migration very frequently has an element of both, and a clear separation between the two is impossible. This idea is useful to highlight the fact that ‘environmental factors influence migrations and migrants alter environments’ and that this has always been part of the human condition (Wood, 2001).


However, we cannot get around the definitional issue this easily, for definitions are crucial in guiding the policies of governments and international agencies towards mobile people.


Definitions reflect and reproduce power, and none more so than the refugee definition I just referred to. It makes a big difference whether people are perceived as refugees, other types of forced migrants or voluntary migrants.


The 1951 Convention is the main yardstick for deciding whether people fleeing persecution will get asylum – in extreme cases, the definition can literally decide over life or death.. Since the end of the Cold War, receiving states have become more and more restrictive in their interpretations of the refugee definition. Discourses about ‘bogus asylum seekers’ and ‘economic refugees’ have paved the way for increasingly restrictive asylum and immigration rules. Many refugees cannot lay claim for the status because visa rules and sanctions against airlines imposed by Northern governments make it impossible for them ever to get out of their countries. The ‘non-departure’ regime of the Cold War has been replaced by the ‘non-arrival regime’ of the New World Order. More people now die trying to cross the Straits of Gibralter or the Rio Grande than were ever shot on the Berlin Wall.


In these circumstance, many refugee advocates and non-governmental organisations have pointed to the inadequacies of the 1951 Convention definition. It is Eurocentric in its origins and ignores the reality of mass displacement through war and generalised conflict in countries of the South. The majority of persons in need of protection and assistance do not count as refugees.


At the end of 2000, there were 12.1 million refugees recognised by the UNHCR in the world, the great majority in poor countries of the South (UNHCR, 2000). The number of refugees has fallen considerably from its peak in the mid-1990s – not because less people have to flee, but because it is harder and harder from them to gain admission and recognition. Nonetheless, refugees have a powerful institutional protector with a mandate supported by some 140 countries – the UNHCR.


The number of people displaced internally by conflict (IDPs) is not accurately known, but is estimated at around 25 million. IDPs are often worse off than refugees because they remain in the state where they are or have been oppressed. The government is unable or unwilling to protect them, but generally rejects attempts to do so by international agencies. There is no legal or institutional regime specifically designed to protect IDPs.


In addition there are millions of people displaced by development projects like dams, airports, urban development and industry. The World Bank estimates that 10 million people are displaced in this way every year – or 200 million in the 1980s and 1990s (Cernea, 2000). Again there is no specific protection regime.


It therefore seems appropriate to call for a much wider international protection regime that would embrace all these groups and – why not? – so-called environmental refugees as well. The states party to the 1951 Refugee Convention and its 1967 Protocol are meeting in Geneva next week for the first time ever to discuss this issue. But the reality is that there is no consensus for extending the refugee regime. Most receiving states want to restrict it further rather than improve it. Our former Home Secretary, Jack Straw made such demands earlier this year. Any changes in the Refugee Convention in the current climate are likely to be for the worse.


That is why a notion like ‘environmental refugees’ is not only misleading, but possibly harmful. It can be used by those who want to restrict asylum opportunities for refugees to support claims that those who arrive on our shores are not genuine victims of persecution, but are in fact fleeing environmental degradation and impoverishment. If people making refugee claims are not real refugees in the sense of the 1951 Convention then the case for exclusion is strengthened.


How are we to deal with these unpleasant politics of definitions? It seems to me that in the current climate of hostility to refugees and asylum seekers we need to do our utmost to defend the UN 1951 Convention, while at the same time calling for improved international legal regimes and institutions to protect the other types of forced migrants. This is the approach adopted by Mary Robinson as UN Commissioner for Human Rights. Some progress is being made within the UN to reform protection arrangements for IDPs, for instance through the establishment of the Office for the Coordination of Humanitarian Affairs (OCHA) and the recent introduction of an IDP-unit within it. However, improvements are slow and as usual are bogged down by unwieldy UN bureaucracy and by the reluctance of member countries to provide necessary resources or to make concessions which may affect national sovereignty.


North-South disparities, development and migration

The real issue, though, is not to protect and assist people forced to migrate by environmental and other factors, but rather to adopt policies which will deal with the root causes of all types of forced migration, and make them unnecessary. Here I generally agree with the catalogue of measures put forward by Myers (Myers, 1997), which includes:


  • Promoting sustainable development (defined following the 1987 Brundtland Report as ‘development that meets the needs of the present without compromising the ability of future generations to meet their own needs’).
  • Foreign aid measures designed to alleviate environmental pressures, and to address the needs of the most impoverished groups.
  • Measures for the relief of foreign debt of the poorest nations (although I find Myers’ idea that debt relief should take the form of ‘debt-for-environment swaps’ dubious, since it again seems to define environmental refugees as a threat).
  • Specific initiatives designed to help developing countries confront environmental challenges.


However, I think it is mistaken to focus narrowly on environmental issues in development strategies. It is important to recognise that environmental change has always been part of human development. In today’s developed countries, the natural environment has been profoundly changed and remoulded to fit human needs over many centuries. In the last century we have become increasingly aware of the need to preserve natural resources and bio-diversity, and we are beginning to make some progress in doing so – as the return of fish to British rivers and the increase in forest coverage demonstrates. Are today’s less-developed countries going to be able to make such transitions, or will they be prevented by rapid population growth and climate change? If we impose too much ‘environmental conditionality’ on poor countries, we may condemn them to remain poor.


Let us remember that the rich countries of the North are still responsible for the greatest environmental problems. This applies in a direct sense: the average American produces ten times as much greenhouse gas and global warming as the average Indian or Chinese. But it is also true in a much more pervasive way. Globalisation as a new world order is based on the opening up of all regions of the world to economic activities largely controlled by Northern-based transnational companies and motivated by their profit interests. The global economic institutions – the IMF, the World Bank and the World Trade Organisation – impose conditions of privatisation, free trade and investment and control of intellectual property which protect the dominance of the transnationals. The military might of the only super-power is available to police this world order if it is threatened by those it designates as rogue states, fundamentalists or terrorists.


Underdevelopment is not a natural condition of the South, but a process resulting originally from colonialism and now from the North-South division. The decline in living standards and welfare in Africa and parts of Asia over the last 20 years – at a time of rapid economic growth in the rich countries – illustrates the way whole regions are becoming uncoupled from global development.


A few months ago, a world conference was held to debate ways of controlling the illegal trade in small arms which fuels the local wars, which in turn lead to human-rights violations, misery and flight. The conference failed, because US arms manufacturers, backed by their government, rejected any control even of illegal arms trading as an infringement of their market freedom.


So if we really want to deal with the root causes of forced migration, the first step is to stop Northern practices that make things worse in the poor countries of the South. Stopping the arms trade and the trade in alluvial diamonds that has fuelled conflicts in Cambodia, Sierra Leone and the Democratic Republic of Congo is essential. A much more important – but also more difficult – second step is a reform of global rules on trade, investment and intellectual property to give countries in the early stages of development and industrialisation better economic and social opportunities. A package of aid measures, as listed above, is only the third step.


Eliminating forced migration is thus a long-term project, closely bound up with global power relations and the North-South division. Invoking ‘a frightful future of third world ecological ruin threatening first world stability’ is harmful, because it is likely to encourage new immigration restrictions – what State Department Geographer Wood aptly calls ‘Green Walls’ (Wood, 2001, 55-7). We clearly need a much broader approach, which recognises the complex causes and the global dimensions of the crises that force people to flee their homes.


Stephen Castles
Fellow of Green College
Director, Refugee Studies Centre, University of Oxford



Black, R. 1998. Refugees, Environment and Development. London: Longman.

--- 2001. "Environmental Refugees: Myth or Reality?," UNHCR Working Papers(34): 1-19.


Castles, S. and M. J. Miller 1998. The Age of Migration: International Population Movements in the Modern World. London: Macmillan.


Cernea, M. M. 2000. "Risks, Safeguards and Reconstruction: A Model for Population Displacement and Resettlement".


In Risks and Reconstruction: Experiences of Resettlers and Refugees. Eds. M. M. Cernea and C. McDowell. Washington DC: World Bank: 11-55.


El-Hinnawi, E. 1985. Environmental Refugees. Nairobi: United Nations Environment Programme.


Elmhirst, R. 1999. "Space, Identity Politics and Resource Control in Indonesia's Transmigration Programme," Political Geography 18: 813-35.


Goldstone, J. A. 2001. "Demography, Environment and Security: An Overview". In Demography and National Security. Eds. M. Weiner and S. S. Russell. New York and Oxford: Berghahn: 38-61.


Hauge, W. and T. Ellingsen 1998. "Beyond Environmental Scarcity: Causal Pathways to Conflict," Journal of Peace Research 35: 299-317.

20 of 21 06/04/2020 19:44


Environmental change and forced migration http://www.war-no-more.co.uk/items/environmental-change-and-f...

Homer-Dixon, T. 1991. "On the Threshhold: Environmental Changes as Causes of Acute Conflict," International Security 16: 76-116.


Homer-Dixon, T. and V. Percival 1996. Environmental Security and Violent Conflict: Briefing Book. Toronto: University of Toronto and American Association for the Advancment of Science.


Lee, S.-w. 2001. Environment Matters: Conflict, Refugee and International Relations. Seoul and Tokyo: World Human Development Institute Press.

Myers, N. 1997. "Environmental Refugees," Population and Environment 19(2): 167-82.


Myers, N. and J. Kent 1995. Environmental Exodus: An Emergent Crisis in the Global Arean. Washington DC: Climate Institute.


Sunderlin, W. D. and I. A. P. Resosudarmo 2001. Rate and Causes of Deforestation in Indonesia: Towards a Resolution of the Ambiguities, 11 November 2001, www.cifor.org/publications/html/occpaper9/


Tirtosudarmo, R. 2001a. "Demography and Security: Transmigration Policy in Indonesia". In Demography and National Security. Eds. M. Weiner and S. S. Russell. New York and Oxford: Berghahn Books: 199-227.

--- 2001b. Geographic Mobility and the Emergence of Ethnic Politics Jakarta: 1-29, Manuscript


UNHCR 2000. Global Report 2000: Achievements and Impact. Geneva: United Nations High Commissioner for Refugees.


Weiner, M. and S. S. Russell 2001. "Introduction". In Demography and National Security. Eds. M. Weiner and S. S. Russell. New York and Oxford: Berghahn: 1-17.


Wood, W. B. 2001. "Ecomigration: Linkages between Environmental Change and Migration". In Global Migrants, Global Refugees. Eds. A. R. Zolberg and P. M. Benda. New York and Oxford: Berghahn: 42-61.