killmicrosoft
29-02-2008, 11:13 PM
just found this look at the bit talking about new world order
http://www3.hants.gov.uk/forced_migration.pdf
NEW ISSUES IN REFUGEE RESEARCH
Working Paper No. 70
Environmental change and forced migration:
making sense of the debate
Stephen Castles
Refugees Studies Centre
University of Oxford
E-mail : stephen.castles@queen-elizabeth-house.oxford.ac.uk
October 2002
Evaluation and Policy Analysis Unit
Evaluation and Policy Analysis Unit
United Nations High Commissioner for Refugees
CP 2500, 1211 Geneva 2
Switzerland
E-mail: hqep00@unhcr.ch
Web Site: www.unhcr.ch
These working papers provide a means for UNHCR staff, consultants, interns and associates
to publish the preliminary results of their research on refugee-related issues. The papers do
not represent the official views of UNHCR. They are also available online under
‘publications’ at <www.unhcr.ch>.
ISSN 1020-7473
Introduction
This paper seeks to find a way into current discussions about the links between
environmental change and forced migration for non-specialists. The perspective is
that of a social scientist working on issues of refugee movements, asylum and
migration, and not of an that of an ecologist, although working on this theme certainly
makes one realise the need to bridge this disciplinary divide.1
Links between environment and forced migration certainly appear to be an important
theme, which has generated a large volume of literature, as well as some heated
public debates. . My starting point in trying to assess the issues was to look at the
work of two authors who have published significant recent works on the theme. The
first was Norman Myers, who has written extensively on environmental change and
population displacement. 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. 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). I assumed that all I needed to do was to study these
two works, and I would understand the issues.
Unfortunately, this was not the case. I rapidly discovered that the two authors
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 United
Nations High Commissioner for Refugees (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
1
This article is based on the Brian Walker Lecture on Environment and Development presented at
Green College, Oxford, in December 2001
1
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 will be discussed later on.
2
In an attempt to gain some clarity in this debate, this paper will discuss a very small
selection of the voluminous research literature on environmental change as a direct
cause of forced migration. It will then look at claims that environmental change
causes violent conflicts which in turn cause mass refugee flows. The article will
briefly discuss 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.
Environmental change as a direct cause of forced migration
What is the evidence that environmental change causes 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.
3
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 disasters, 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.
4
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.
A reasonable 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)
5
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
6
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: example of 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. This can be illustrated with an example from Indonesia. 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
7
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 us 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
8
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 circumstances, 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.
9
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 reality is that there is no consensus for extending the refugee
regime. Most receiving states want to restrict it further rather than improve it. The
United Kingdom’s 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 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 one
can 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-
10
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.
It is important to 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.
In mid-2001, 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.
11
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.
12
REFERENCES
Black, R. (1998). Refugees, Environment and Development. London: Longman.
Black R. (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.
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.
Tirtosudarmo, R. (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.
13
14
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.
http://www3.hants.gov.uk/forced_migration.pdf
NEW ISSUES IN REFUGEE RESEARCH
Working Paper No. 70
Environmental change and forced migration:
making sense of the debate
Stephen Castles
Refugees Studies Centre
University of Oxford
E-mail : stephen.castles@queen-elizabeth-house.oxford.ac.uk
October 2002
Evaluation and Policy Analysis Unit
Evaluation and Policy Analysis Unit
United Nations High Commissioner for Refugees
CP 2500, 1211 Geneva 2
Switzerland
E-mail: hqep00@unhcr.ch
Web Site: www.unhcr.ch
These working papers provide a means for UNHCR staff, consultants, interns and associates
to publish the preliminary results of their research on refugee-related issues. The papers do
not represent the official views of UNHCR. They are also available online under
‘publications’ at <www.unhcr.ch>.
ISSN 1020-7473
Introduction
This paper seeks to find a way into current discussions about the links between
environmental change and forced migration for non-specialists. The perspective is
that of a social scientist working on issues of refugee movements, asylum and
migration, and not of an that of an ecologist, although working on this theme certainly
makes one realise the need to bridge this disciplinary divide.1
Links between environment and forced migration certainly appear to be an important
theme, which has generated a large volume of literature, as well as some heated
public debates. . My starting point in trying to assess the issues was to look at the
work of two authors who have published significant recent works on the theme. The
first was Norman Myers, who has written extensively on environmental change and
population displacement. 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. 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). I assumed that all I needed to do was to study these
two works, and I would understand the issues.
Unfortunately, this was not the case. I rapidly discovered that the two authors
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 United
Nations High Commissioner for Refugees (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
1
This article is based on the Brian Walker Lecture on Environment and Development presented at
Green College, Oxford, in December 2001
1
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 will be discussed later on.
2
In an attempt to gain some clarity in this debate, this paper will discuss a very small
selection of the voluminous research literature on environmental change as a direct
cause of forced migration. It will then look at claims that environmental change
causes violent conflicts which in turn cause mass refugee flows. The article will
briefly discuss 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.
Environmental change as a direct cause of forced migration
What is the evidence that environmental change causes 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.
3
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 disasters, 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.
4
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.
A reasonable 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)
5
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
6
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: example of 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. This can be illustrated with an example from Indonesia. 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
7
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 us 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
8
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 circumstances, 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.
9
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 reality is that there is no consensus for extending the refugee
regime. Most receiving states want to restrict it further rather than improve it. The
United Kingdom’s 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 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 one
can 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-
10
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.
It is important to 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.
In mid-2001, 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.
11
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.
12
REFERENCES
Black, R. (1998). Refugees, Environment and Development. London: Longman.
Black R. (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.
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.
Tirtosudarmo, R. (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.
13
14
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.