Migration, whether permanent or temporary, has always been a traditional response or survival strategy of people confronting the prospect, impact or aftermath of disasters (Hugo 1996). However, today, more than ever, the complex nature of disasters such as Hurricane Katrina in the city of New Orleans, bring with them an enormous potential for the uprooting of large numbers of people. The increasing complexity of disasters is rooted in the interplay of social and economic factors in the environment, exacerbating the vulnerability of people and environments and intensifying their impacts when they occur. Hurricane Katrina’s impact on New Orleans was compounded by an excessive dependence on technology and half a century’s assault on the natural defenses of the environment of southern Louisiana, leaving the city tragically vulnerable. Virtually the entire population of the city has been displaced, a large proportion of it perhaps permanently, presenting a host of profound economic, social and psychological challenges to individuals, communities and all levels of government.

Despite technological and scientific advances in prediction and mitigation, we have seen a serious increase in both mortality and economic losses from disasters since 1960, particularly in the developing world. Disasters are, in fact, increasing in impact and scope through the combined effects of economic, social, demographic, ideological and technological factors. Greater numbers of people are more vulnerable to natural and other hazards than ever before, due in part to increases in population, but more so to their location in dangerous areas. In fact, disaster risk and losses have dramatically increased, but unevenly so according to region (International Federation of Red Cross and Red Crescent Societies 2003). However, regardless of region, some form of displacement of individuals and communities frequently results from the threat or impact of a disaster.

Both disaster and forced migration are terms that are used to describe a wide variety of environmental and social processes. Perhaps because the term is so widely and loosely used, disasters are quite difficult to define. Although the term “disaster” actually refers to a process, essentially the disruption of social functions, it is generally employed to characterize an event or agent such as Hurricane Mitch or the Northridge Earthquake.

Despite a popular construction of disasters as “acts of God” or “fate” in which nobody is really responsible, there has been a general reconsideration in the scientific community of this event/agent focused perspective. Hewitt (1983) posited that most natural disasters are more explainable in terms of the “normal” order of things, that is, the conditions of inequality and subordination in the society rather than the accidental geophysical features of a place. This perspective shifted the focus away from the disaster event and towards the “on-going societal and man-environment relations that prefigure [disaster]”(Hewitt 1983: 24-27).

Thus, disasters occur at the interface of society, technology and environment and are fundamentally the outcomes of the intersection of these features. To adequately analyze disasters the barrier between human activity and eco-system activity must be collapsed, transforming a relationship of difference into a relationship of mutuality. In very graphic ways, disasters serve as indices of the success or failure of a society to adapt, for whatever reasons, to certain features of its natural and socially constructed environment in a sustained fashion (Oliver-Smith 2002).

The natural or technological agents, however, cannot be relegated to a secondary role. The broad array of “objective” natural and technological phenomena that produce or trigger disasters can create a wide variety of physical and social impacts, according to the context in which they occur. In Hewitt’s view, disaster agents include natural hazards (atmospheric, hydrological, geological, and biological), technological hazards (dangerous materials, destructive processes, mechanical, and productive), and social hazards (war, terrorism, civil conflict and the use of hazardous materials, processes, and technologies) (1997: 26).

The concept of vulnerability links the relationship that people have with their environment to social forces and institutions and the cultural values that sustain or contest them. Vulnerability refers to the totality of relationships in a given social situation producing the formation of a condition that, in combination with environmental forces, produces a disaster. Disaster risks and outcomes are socially produced at the intersection of a complex and dynamic range of hazard and vulnerability patterns, associated with underlying social, economic, territorial and political processes operating in specific locales. The concept of vulnerability links general political economic conditions to very particular environmental forces to understand how basic conditions such as poverty or racism produce susceptibilities to very specific environmental hazards. Vulnerability, thus, integrates not only political economic, but environmental forces, defined in terms of both biophysical and socially constructed risk. The working definition provided by Blaikie et al. is currently among the most utilized:

By vulnerability we mean the characteristics of a person or group in terms of their capacity to anticipate, cope with, resist, and recover from the impact of a natural hazard. It involves a combination of factors that determine the degree to which someone’s life and livelihood is put at risk by a discrete and identifiable event in nature or in society (1994:9).

Vulnerability in their approach is generated through a causal chain of root causes embedded in ideological, social and economic systems, the dynamic pressures of a demographic, socio-economic or ecological nature and specific sets of unsafe conditions which, when combined with a natural hazard, produce a disaster. This more complex understanding of vulnerability enables researchers to conceptualize how social systems generate the conditions that place different kinds of people, often differentiated along axes of class, race, ethnicity, gender, or age, at different levels of risk from the same hazard and suffering from the same event. Therefore, a single disaster can fragment into different and conflicting sets of circumstances and interpretations according to the experience and identity of those affected, thus motivating in many instances different responses. Depending on their experience, some groups may migrate because of the disaster and other groups will not.

The complexity of disasters today is demonstrated by the processes in which they can combine with and compound each other. For example, in 1998 Hurricane Mitch (a natural agent) produced floods in Honduras (a socio-natural phenomenon) that inundated warehouses full of pesticides and fertilizers (a technological hazard), producing what might be called a compound or complex disaster (Jansen 2003). Recently in the fall of 2004, Hurricane Ivan threatened New Orleans with just the same conflation of dangers. Hurricane Katrina has just fully realized the nightmare of Ivan. There is no question that environmental changes, particularly in the form of degradation, have increased the severity of socio-natural disasters. Moreover, disasters, singly or in combination, can further be compounded by the incidence of political upheaval, such as war, ethnic cleansing, or terrorism, or social factors such as racism, exclusion or religious persecution. And disasters can contribute to political instability that can lead to conflict with the potential to displace people.

To the degree that disasters force people to relocate either temporarily or permanently, disaster victims have been seen as a subset of the category of environmental refugees, a term that has generated a considerable amount of recent debate. As associated with disasters, the phenomenon of forced migration is also complex. Unless explicitly limited to referring to permanent, involuntary transfer to distant locations, the concept of forced migration refers to a variety of demographic movements, such as: flight, evacuation, displacement, resettlement, as well as forced migration.

  • Flight – escape
  • Evacuation – removal of people from harm’s way
  • Displacement – the uprooting of people from a home ground
  • Resettlement – relocation of people to new homes
  • Forced migration – people must move to a new and usually distant place

If the threat of disaster is immediate, flight or escape to the closest safe location is a frequent response. An impending threat may result in an evacuation that resembles flight or may be more organized, administrated by internal or external agents. Displacement similarly can occur as the result of flight or be more planned in the sense that people are organized and obliged to move from one residence site to another either temporarily or permanently. If the movement is thought to be permanent, resettlement in the form of the creation of a new residence site may actually be the outcome. Finally, as mentioned earlier, forced migration involves permanent, longer distance moves generally into completely different environments. Some of the forms of demographic movement may lead to others—flight or evacuation, for example, may lead to displacement and resettlement or eventually to forced migration.

Each form of demographic movement may vary along a number of scales or continua associated with certain characteristics that refer largely to the social and environmental relations expressed in the particular context.

  • Proactive – reactive (Richmond 1993)
  • Voluntary – forced
  • Temporary – permanent
  • Physical danger – economic danger
  • Administrated – non-administrated

These five pairs are best viewed as poles on a series of continua rather than closed or opposing categories. These concepts have to be treated with a certain flexibility, and should not be taken as hard and fast categories because the reality of particular occurrences of forced migration tends to be too complex to nail down within rigid categories. Looking at each kind of demographic movement along the various continua presented reveals the wide variability that each can display. Flight, for example, generally tends to be proactive and forced, but not administered: it can be temporary or permanent, tends to be associated with physical danger and may result in permanent displacement, resettlement or forced migration. Evacuation, usually a response to physical danger, can have similar outcomes: it can be proactive or reactive but tends to be administered to a greater or lesser degree. Displacement can be an administered process of moving a population only a short distance. It can equally take place by people moving themselves out of harm’s way, but remaining in the same area or environment. Displacement can be temporary or permanent, voluntary or involuntary, and may be a response to both physical and economic harm. Forced migration involves moving further away, to different environments and for longer periods of time, if not permanently. Except in extreme cases, the coercive power or push factors in disaster induced forced migration will vary and may be balanced to some degree by pull factors or positive inducements to move.

There has been a recent debate over this issue with claims of millions of environmental refugees being produced versus counterclaims that the evidence is uneven, unconvincing, and counterproductive. There is little question that some disasters, but not all, force people to migrate. The nuclear accident at Chernobyl forced many thousands of Ukrainians to migrate out of the contaminated zone. Had the disaster not occurred, there probably would not have been any significant out-migration. Such cases as Chernobyl, with widespread and long-lasting physical and economic danger, seem to be occurring more frequently. The South Asian tsunami clearly displaced millions and now Hurricane Katrina has uprooted more than a hundred thousand people, the majority perhaps permanently. Many disasters actually also trigger a wide variety of factors both prior to and after onset that displace people.

Since the 1980s researchers have linked the issue of environmental change with human migration, explicitly designating as “environmental refugees” people who are forced to leave their homes, temporarily or permanently, due to the threat, impact or effects of a hazard or environmental change (El-Hinnawi 1985). Although environmental studies have traditionally focused on the natural world, the impacts of pollution, deforestation, soil erosion, degradation, desertification and other environmental processes on human beings have also been a source of both interest and concern to ecological and social scientists. Indeed, the impacts of many of these processes have often been framed as “disastrous” because they create stress, disrupt normal social processes, and force people to adapt by making temporary adjustments or permanent changes in how, where and when they do things in life. Myers has asserted that recent human-induced environmental change, such as desertification, deforestation, or soil erosion, compounded by natural and man-made disasters, could force as many as 50 million people to migrate from their homes by 2010 (1997). He sees environmental change and disasters as triggers or detonators that lead to land competition, resource degradation, occupation of fragile regions and impoverishment that eventually force people to migrate.

Other researchers dispute the accuracy of the term “environmental refugee,” finding it misleading. They attribute the displacement of people to a complex pattern of factors including political, social, economic as well as environmental forces (Wood 2001, Black 2001, Castles 2002). Natural disasters are seen to cause temporary displacement, but not authentic, i.e., permanent, migration. Indeed, if permanent migration does occur as the result of a disaster, it is seen as more the result of deficient responses of weak or corrupt states rather than the environment as expressed in the form of a natural hazard impact. Black’s critique that focusing on environmental factors as causes of migration often obscures the role of political and economic factors is well-taken, and echoes the position held by most disaster researchers today that focusing solely on agents reveals little about the political or economic forces that together with agents produce disasters or, for that matter, any forced migration that might ensue.

In some sense, the objections of Black, Wood, Castles and other scholars to the term “environmental refugee” are based on the construction of human-environment relations as a duality, in which each domain is separate and capable of causing things to happen in the other. Seeking single agent causality in the environment tends to elide the fact that the environment, and its resources as well as its hazards, is always channeled for people through social, economic and political factors, even in the best of times.

It is important to remember here that a disaster is also not defined in terms of the event itself, but in terms of both the processes that set it in motion and the post-event processes of adaptation and adjustment in recovery and reconstruction. Forced migration can be part of the process prior to the event or after, but it is not inevitable. Disasters can act as triggers or accelerators of imminent change or change already underway. For example, migration of elites from highland Peru to coastal cities had been going on for more than half a century in 1970. The Peruvian earthquake of 1970 unquestionably accelerated that process. By the same token, migration of rural peoples to highland towns and cities was also very intense during the same time period, but was made more so by the fact that disaster aid distribution in the form of housing was primarily located in urban areas (Oliver-Smith 1992).

Disasters are brought about by the interrelatedness of a range of factors of different orders: cultural, social, environmental, economic, institutional and political—all of which are taking place in the context of imposed space change and of local level responses and initiatives. Moreover, these changes are taking place simultaneously in an interlinked and mutually influencing process of transformation. Forced migration associated with disasters, therefore, is commonly the result of the interactions that both bring about the disaster and are then accentuated by the event itself.

Most local displacement by disaster tends to be temporary, but may become permanent, particularly if the disaster permanently alters or destroys a local economic base. However, that outcome is usually not entirely the result of the agent alone, but rather government response. The government decision not to reconstruct Homestead Air Force Base, a major regional employer, after Hurricane Andrew led to a permanent alteration of the economy of south Dade county and may account for some of the permanent migration that ensued

The Great Flood of 1927 in the lower Mississippi Valley displaced nearly 700,000 people, approximately 330,000 of whom were African Americans who were subsequently interned in 154 relief “concentration camps” where they were forced to work (www.mvd.usafe.army.mil/MRC-History-Center/gallery/flood/flood2.html). Although there were many reasons for African Americans to leave the South, the flood and its consequences were the final motivation for thousands to migrate (Barry 1997: 417). When Hurricane Andrew hit Miami-Dade County in 1992, it inflicted nearly 30 billion dollars worth of damage and displaced roughly 353,000 people temporarily. Forty thousand of the displaced (roughly 11%) permanently migrated. But, of the 40,000 that migrated about half (20,000) moved only about a half an hour’s drive north, sparking a population boom in Plantation and other Broward County communities. The net loss of 20,000 was soon outstripped by people whose desire to move to Florida was not deterred by the hurricane (Gainesville Sun 2004).

Many of the emerging disaster trends and characteristics that have been noted will very likely increase the number and scale of forced migrations in the relatively near future. The combination of increasing population, population density, increasing poverty, and occupation of hazardous sites has accentuated vulnerability to both natural and technological hazards and increases the probability of forced migrations. Technology has also vastly increased the numbers of hazards to which populations are exposed. When socio-natural disasters trigger technological disasters, the resulting complex processes may force people to migrate because the disaster impacts, in combination with local environmental contamination, make the environment uninhabitable. While many of the changes associated with increasing state and market integration have established more resilient infrastructures in some regions of the world, they have also frequently undermined traditional adaptations of rural populations to natural hazards. In addition, the effects of global climate changes, including increased risks of flooding, storms, deforestation, desertification, soil erosion, and sea level rise increase the probability of disasters contributing to internal and international forced migration. The physical and social processes recently triggered by Hurricane Katrina underscore that emerging reality.

Over the past twenty years, colleagues from the field of development-induced displacement, refugee studies and disaster research (Hansen and Oliver-Smith 1982, Cernea 1996, Turton 2003, Oliver-Smith in press) have discovered that the displaced peoples we work with share many similar challenges. Although the places and peoples are geographically and culturally distant and the sociopolitical environments and causes of dislocation dissimilar, there emerge a number of common concerns and processes. Refugees, earthquake victims and displacees experience uprooting and relocation and must cope with the consequent stresses and the need to adapt to new or radically changed environments. All may experience privation, loss of homes, jobs, and the breakup of families and communities. All must mobilize social and cultural resources in their efforts to reestablish viable social groups and communities and to restore adequate levels of material life. These are important similarities that we must recognize and understand both to minimize displacement and to assist in the material reconstruction and the social reconstitution of communities. The catastrophic losses from Hurricane Katrina demonstrate in horrific fashion the urgent need to develop the conceptual, strategic, and material tools to confront the increasing challenges of natural hazards made even more potent and complex by climate change, increasing population densities and environmental degradation in the 21st century.

Anthony Oliver-Smith is professor of anthropology at the University of Florida. He has done anthropological research and consultation on issues relating to disasters and involuntary resettlement in Peru, Honduras, India, Brazil, Jamaica, Mexico, Japan, and the United States since the 1970s. He is a member of the editorial boards of Environmental Disasters and Desastres y Sociedad.


Barry, John M. 1997 Rising Tide: The Great Mississippi Flood of 1927 and How it Changed America, New York: Simon and Schuster.

Black, Richard 2001 “Environmental Refugees: Myth or Reality?” UNHCR Working Papers (34):1-19.

Blaikie, P., T. Cannon, I. Davis, and B. Wisner, 1994 At Risk: Natural Hazards, People’s Vulnerability and Disasters, London: Routledge.

Castles, Stephen 2002 “Environmental Change and Forced Migration: Making Sense of the Debate,” UNHCR Working Papers (70): 1-14.

Cernea, Michael 1996 “Bridging the Research Divide: Studying Refugees and Development Oustees,” in T. Allen (ed.) In Search of Cool Ground: War, Flight and Homecoming in Northeast Africa, London: James Currey/Africa World Press.

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

Gainesville Sun 2004 “After Hurricanes, Sunshine State loses sheen,” September 14: p.3B.

Hansen, Art and Anthony Oliver-Smith 1982 Involuntary Migration and Resettlement: The Problems and Responses of Dislocated Peoples, Boulder: Westview Press.

Hewitt, Kenneth 1997 Regions of Risk: A Geographical Introduction to Disasters, Harlow, Essex: Longman.

Hewitt, Kenneth 1983 Interpretations of Calamity, Winchester, MA: Allen & Unwin, Inc.

Hugo, Graeme 1996 “Environmental Concerns and International Migration,” International Migration Review30:1: 105-131.

International Federation of Red Cross and Red Crescent Societies 2003 World Disasters Report, Bloomfield CT: Kumarian Press.

Myers, Norman 1997 “Environmental Refugees,” Population and Environment 19:2:167-82.

Jansen, Kees 2003 “Crisis Discourses and Technology Regulation in a Weak State: Responses to a Pesticide Disaster in Honduras,” Development and Change 34:1:45-66.

Mississippi River Gallery “The Flood of ‘27” (www.mvd.usafe.army.mil/MRC-History-Center/gallery/flood/flood2.html).

Oliver-Smith, Anthony In press “Communities after Catastrophe: Reconstructing the Material, Reconstituting the Social,” in Stanley Hyland (ed) Community Building in the Twenty-First Century, Santa Fe: School of American Research Press.

Oliver-Smith, Anthony 2002 “Theorizing Disasters: Nature, Culture, Power,” in Susanna M. Hoffman and Anthony Oliver-Smith, (eds.) Culture and Catastrophe: The Anthropology of Disaster, Santa Fe, New Mexico: The School of American Research Press.

Oliver-Smith, Anthony 1992 The Martyred City: Death and Rebirth in the Andes, Prospect Heights, ILL: Waveland Press.

Richmond, A. 1993 “The Environment and Refugees: Theoretical and Policy Issues,” Revised version of a paper presented at the meetings of the International Union for the Scientific Study of Population, Montreal.

Turton, David 2003 “Refugees and ‘Other Forced Migrants’: Towards a Unitary Study of Forced Migration,” Paper presented at the Workshop on Settlement and Resettlement in Ethiopia, January 28-30. Addis Ababa.

Wood, W. B. 2001 “Ecomigration: Linkages between Environmental Change and Migration,” In A.R. Zolber and P.M. Benda (eds) Global Migrants, Global Refugees, New York: Berghahn. Pp 42-61.