• This field is for validation purposes and should be left unchanged.
  • Earth.Org Newsletters

    Get focused newsletters especially designed to be concise and easy to digest

  • This field is for validation purposes and should be left unchanged.

Should Climate Refugees Be Recognised as Such?

CRISIS - Viability of Life on Earth by Matt Giles Global Commons Mar 23rd 202313 mins
Should Climate Refugees Be Recognised as Such?

The number of people fleeing climate change is rising exponentially. Yet, most of these people are not recognised as climate refugees under the terms of the 1951 Refugee Convention, meaning that states have no legal obligation to grant them entry. Should this be the case or do states have a legal and moral duty to admit those displaced by climate-related disasters? 

Climate change is fuelling environmental degradation on an unprecedented scale. The degradation takes many forms, from floods and desertification of arable land to the gradual disappearance of entire small island nations as a result of rising sea levels. This, in turn, produces “overcrowded cities, loss of jobs, poverty, famine and wars,” as the people affected compete for territory and resources which have now become scarce. 

Many of the afflicted seek refuge in neighbouring countries, constituting around one-third of the record high 103 million people forcibly displaced worldwide. It is estimated that by 2050, the number of people forcibly displaced by climate-related phenomena alone will rise to 200 million.

These are the climate refugees: people who have fled their home country because environmental degradation, whether fast or slow onset, has “jeopardised their existence or seriously affected the quality of their lives”.

You might also like: 5 Facts About Climate Migrants You Should Know About

The problem of the climate refugee, therefore, is one of the foremost human crises of our times. Yet, as noted by British environmentalist Norman Myers as far back as 1997, it has so far been treated as “a peripheral concern, a kind of aberration from the normal order of things”. This attitude might explain why there remains, to this day, “no well-established legal basis upon which States are obliged to assist people displaced by climate change”. 

Indeed, the term “climate refugee” itself has no legal meaning: people displaced across borders by climate-related phenomena do not generally qualify for refugee status under the 1951 Refugee Convention because they are not considered victims of discriminatory persecution. The states to which they flee as asylum seekers are therefore under no legal obligation to grant them entry, and many are forced to return to their country of origin, where they carry out a precarious existence in poor conditions under the constant threat of further disaster. 

Many argue that this exclusion of climate refugees from refugee status is unjustified. Their argument is twofold. Firstly, they assert that states adopt too superficial a view of climate refugees’ circumstances. A more holistic investigation would, in many cases, reveal that climate refugees are in fact victims of discriminatory persecution and thus legally entitled to refugee status under the Convention. Their second argument is that the Convention definition itself is morally untenable. In identifying discriminatory persecution as the key characteristic upon which a state’s duty of protection to a displaced individual ought to hinge, the Convention defines the scope of refugeehood too narrowly. A broader definition, based on the degree of the individual’s need rather than its cause, would better satisfy the moral duty that states owe to displaced people. 

The Dominant View

149 countries are signatories to the 1951 Refugee Convention or its 1967 Protocol.

The Convention defines the refugee as “a person who, owing to a well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country”.

The dominant view is that this definition excludes people displaced by climate phenomena. Dawson J in A and Another, an asylum case decided by the High Court of Australia, put this in unequivocal terms: “No matter how devastating may be the epidemic, natural disaster or famine, a person fleeing them is not a refugee within the terms of the Convention”. 

Part of the reason for this lies in the Convention’s requirement that the asylum seeker must fear “being persecuted”. Persecution implies a persecutor; the idea that the predicament of the asylum seeker is brought about by human agency. As noted by Jane McAdam in ‘Climate Change, Forced Migration and International Law’, this “persecutor” is hard to identify in the context of climate refugees. Natural disasters are assumed to be the result of “the uncontrollable forces of nature” from which the “role of human agents is entirely absent”. 

In the context of climate change, human agency is more relevant. Molly Conisbee and Andrew Simms, authors of “Environmental Refugees”, suggest that policies that are pursued in full knowledge of the harm they cause to people (such as those followed by the major emitting states since the impacts of climate change were established) should be classed as a form of “environmental persecution”. McAdam, however, counters that “persecution is not enough” to establish refugee status under the Convention; that persecution must also be discriminatory. 

This further maligns the claims of environmental refugees because, as McAdam notes, the impacts of climate phenomena are seen as “largely indiscriminate”; fires and floods do not differentiate between their victims based on “background or beliefs”. Thus, Conisbee and Simms’ concept of environmental persecution, though persuasive in principle, is not persecution in the Convention sense of the word; the damage done by major emitting states affects so vast a multitude of different races, religions, nationalities and political/social groups worldwide that none of them can plausibly claim to have been singled out.  

This perceived absence of discriminatory persecution in the context of climate refugees means that the principle of surrogacy, in which international refugee law is rooted, is not engaged. According to Andrew Shacknove, author of the influential 1985 paper ‘Who is a Refugee?’, a “bond of trust, loyalty, protection, and assistance” exists between citizen and state, the severance of which obliges the international community of states to provide “surrogate protection” to the individual via asylum. 

Shacknove suggests that climate events are deemed insufficient to break this bond because they are seen as “sources of vulnerability beyond social control”; their unpredictable, unstoppable, and overwhelming nature means that they impose “no obligation on a government to secure a remedy”. In contrast to the state which actively turns against its citizens, hence breaking the “bond of trust”, it is assumed that the state hit by a natural disaster is doing and will continue to “do its best” for its people, who therefore do not require the surrogate protection of another state.

The Holistic Approach

The dominant view arguably rests upon flawed assumptions. The very idea of the “natural disaster” elides, both linguistically and conceptually, two notions that should be recognised as distinct; the “natural” hazard event on one hand, and the human “disaster” which results on the other. 

Floods, hurricanes and earthquakes are not, of themselves, disasters; they are merely “natural hazards”, the result of “extremes in geophysical processes.” They can engender “disaster” only to the extent permitted by the social conditions prevailing in the area which they strike. 

For example, a hurricane that hits a well-fortified city equipped with accurate predictive technology, efficient evacuation procedures, and well-funded emergency services will have a much less serious human impact than an equally forceful hurricane that hits a settlement without such capacities. Only the latter is likely to be called a “natural disaster”. 

Similarly, the same climate event will affect different sectors of the same general area differently. In the case of a flood, people living in temporary settlements along river banks will be more vulnerable than people living in secure housing further away. This is evidenced by the Pakistan floods of 2022. This refutes the dominant view’s assumption that the impacts of natural disasters are indiscriminate. Instead, climate-fuelled disasters tend to disproportionately affect the poor, who often live on land cheapened by its susceptibility to environmental hazards, and whose livelihoods “are not resilient in the face of shocks”.

It might be countered that this simply reflects the everyday economic injustices of any given society, rather than the active discriminatory persecution required by the Convention. Matthew Scott, however, in ‘Climate Change, Disasters and the Refugee Convention’, argues that it is overly simplistic to suppose that people “simply end up living in places that are exposed to natural hazards”. In his view, it is no coincidence that the people living in such areas are often members of “marginalised groups” such as “ethnic minorities and political dissidents”. The primary reason for their exposure and lack of resilience to natural disasters is their poverty. But, as Ben Wisner argues in ‘At Risk: Natural Hazards, People’s Vulnerability and Disasters, their poverty is a result of the “specific relations of exploitation, unequal bargaining and discrimination’ which they suffer ‘within the political economy.”

This echoes Johan Galtung’s theory of structural violence: the idea that discrimination against marginalised groups is “built into the structure” of society, presenting itself as “unequal power and consequently as unequal life chances”. Members of such groups have limited access to education, healthcare, adequate shelter, work, property, food, information, and participation in public life, all of which conspire to render them more vulnerable than members of majority groups to “a range of shocks”, including natural disasters. 

You might also like: ​​How Marginalised Groups are Disproportionately Affected by Climate Change

There are multiple examples of this. 

Aboriginal people were especially vulnerable to the 1985 Alice Springs floods because they lived in flimsy accommodation on low-lying land. In the 2007 Indian floods, Dalits, who occupy the lowest stratum of the caste system, represented the overwhelming majority of fatalities because their homes were more prone to damage. This is no coincidence. The exposure of both groups to these natural hazards is a result of the unequal life chances that the discrimination built into the social structure has, over generations, afforded them. These long-term causes of exposure can be compounded by discriminatory state practices immediately before, during, and after the natural hazard event occurs. The Aboriginal people were not alerted to the oncoming floods in 1985 because flood warnings were only broadcast on radio channels customarily used by the white population. Whether deliberate or not, this oversight is a clear breach of the state’s duty to protect all of its citizens equally. The same is true for the 2007 floods; relief efforts were coordinated by “majority population representatives”, meaning Dalits were “often the last to receive assistance”. 

In cases such as these, where discrimination is obvious, refugee status will likely succeed. The Dalit situation closely accords with the example given by Hathaway and Foster, authors of ‘The Law of Refugee Status’, in which a government limits its relief efforts to “victims who are members of the majority race”, in turn “forcing a minority group to flee to another country in order to avoid starvation or exposure”. In this case, they argue, the members of the minority group are clearly refugees within the Convention definition, because the state has failed to protect them from serious harm. This is required by Lord Hoffman’s description of persecution in Islam, a UK asylum case. Since that failure is on grounds of race (or, in the case of the Dalits, “social group”), the “requisite element of civil and political differentiation is present”. 

The failure to protect, therefore, though lacking the active and intentional agency implied by the term “persecution”, can qualify as such for Convention purposes. This is because the state has a duty to afford its citizens protection on an equal basis; failure to do so breaches the “bond of trust” between citizen and state. 

Yet cases are rarely so clear-cut; government policies are rarely so openly discriminatory, and rarely lead so directly to the harm suffered. Scott argues that in most cases, discrimination is hidden in “apparently neutral measures or omissions” which, by reinforcing “dominant norms”, generate “systemic inequality for people outside the dominant norm.” It is therefore “subtle and even invisible,” perceived as “normal” by dominant and marginalised populations alike.

For this reason, the term “being persecuted” is preferable to “persecution”. Not only is this more faithful to the text of the Convention, but it also removes the common tendency to think of persecution as consisting only of individual acts of direct persecution. This tendency, Scott argues, “narrows the temporal scope of the refugee definition to the moment the harm is experienced, and thus detracts attention from the wider social context in which the risk of exposure to such acts arises”. “Being persecuted”, conversely, implies a “condition of existence entailing a real chance of being exposed to serious denials of human rights”. 

Therefore, even if the 2007 floods had never occurred, the Dalits would still have been persecuted, since being persecuted lies in the exposure to the risk of serious harm. By the same logic, once a disaster event is over, the risk to those who were vulnerable to it is not; if returned, they will reenter the same social system which caused their extreme vulnerability to that event, and will likely be equally vulnerable to the next one. Even if that future event never occurs – which, given the context of climate change, is increasingly unlikely – they will be victims of discriminatory persecution. 

This “slow violence”, therefore, while more diffuse than the torture of a political dissident or even the exclusion of marginalised communities from disaster relief, is no less discriminatory, persecutory, or harmful. Its victims clearly do not enjoy the equal protection of their home state. For this reason, it should be recognised as grounds for refugee status. Yet states continue to satisfy themselves with a surface-level, contemporaneous investigation of the claimant’s situation, failing to delve into the “root causes and dynamic pressures” which underpin it. This blindness may be wilful, since most states are under political and economic pressures to keep their refugee intake low. However, this does not mean it is justified. The common acceptance of natural disasters as “nobody’s fault” is often fallacious, and damagingly so, since it tends to automatically exclude climate refugees from the protection which they desperately need and which, upon a more holistic understanding of their circumstances, they may well be entitled to under the Convention. 

You might also like: Climate Migration Is the Crisis of the Century

Who Is a Refugee?  

So far it has been argued that many climate refugees should be recognised as refugees under the 1951 Convention because a holistic assessment of their circumstances reveals them to be victims of discriminatory persecution. Yet, this is not true of all climate refugees. 

Many natural hazard events are overwhelming enough to decimate minority and majority groups alike, causing members of both to flee across borders. This is especially true of poor countries in which vulnerability levels are high generally. Whilst it may still be the case that members of minority groups are worse off than members of majority groups, the latter are climate refugees nonetheless provided their existence is jeopardised and their quality of life seriously affected. Yet in such situations, even the expansive approach to the Convention advocated above would only recognise the minority group members as refugees, since only their vulnerability would be the result of discriminatory persecution. 

This begs the question of whether discriminatory persecution is the correct characteristic on which to base refugee status.

In “The Ethics of Refugees”, Matthew Gibney argues that the persecution requirement is arbitrary. There is no convincing reason why states should differentiate between people whose lives are equally endangered merely because the threats they face come from different sources. This is akin, he argues, to allocating “access to hospital beds according to how an individual came to be injured”, rather than by the severity of their injury. This is reinforced by Joseph Carens, author of ‘The Ethics of Immigration’, who argues that it is “a mistake” to require refugees to have been “deliberately targeted”; what is morally significant “is the severity of the threat to basic human rights and the degree of risk”, as opposed to the “source or character of the threat”. 

On this view, an individual fleeing the threat of death, injury, disease, and malnutrition as a result of a natural disaster is as deserving of protection as an individual fleeing direct threats of violence by state actors, since the detriment which they flee (namely death or serious physical harm) is the same. Indeed, climate refugees are more deserving than those fleeing forms of persecution which, though direct and targeted, do not involve serious risk of death or bodily harm. The fact that the Convention recognises the latter category of people as refugees whilst refusing the same status to a person facing genuinely life-threatening circumstances as a result of natural disaster reflects a “misplaces set of priorities.” 

The Convention definition, therefore, fails to “adequately capture what makes a displaced individual worthy of moral attention”. This vital characteristic is the degree of that individual’s need, determined by the severity of the threat they flee and the likelihood of that threat coming to pass. The threats of death, injury, disease, and malnutrition are clearly severe, and in the context of climate change, the likelihood of natural disasters is increasingly great. As noted by Wallace-Wells, we will soon see a “planet battered daily by hurricanes and monsoons we used to call ‘natural disasters’ but will soon normalise as simply ‘bad weather’”. This will be the case even if global emissions are immediately lowered. Thus, the idea that the problem of climate refugees can be solved simply by limiting emissions is flawed. Nor can the problem be solved simply by funding projects to improve the resilience of vulnerable countries to natural hazard events (such as the construction of flood defences). 

This is not to say that such endeavours are not valuable; indeed, they are vital to tackling climate change. But they do not help the individuals displaced by climate phenomena in the short term. Such individuals cannot be expected to wait around in precarious conditions for emissions cuts and resilience projects to eventually pay off, because they will most likely die in the meantime. The fact that offering asylum to climate refugees will not solve the “underlying problem that has given rise to their flight” is no reason to refuse it; states have a humanitarian duty to rescue people from life-threatening situations where they possess the capacity to do so. 


There are, therefore, two ways in which people displaced by climate-related disasters can be recognised as refugees. The first is to adopt a more holistic understanding of discriminatory persecution which properly accounts for the structural violence conditioning the vulnerability and exposure of such people to climate events. The second is to do away with the discriminatory persecution requirement altogether, and to assess refugee status on the basis of need. 

Both suggestions are likely to be unpopular. The evident inadequacies of current responses to refugees suggest that states would be reluctant to deal with any more refugees than they already do, let alone the 200 million climate refugees projected by 2050. Yet that does not mean that the suggestions are wrong; climate refugees have a powerful moral claim to admission given the severity of the threats they face, and, as noted by Carens: “The seriousness of the claim is not affected, at the level of principle, by the number of claimants”.      

You might also like: How Ecological Migration in China Could Succeed Without Breaking Rural Traditions

Subscribe to our newsletter

Hand-picked stories once a fortnight. We promise, no spam!

Instagram @earthorg Follow Us