Climate Refugees: The Notion Remains Largely Taboo
“Migration in response to environmental changes has been a constant throughout human history and represents a rational adaptation strategy,” UN Secretary-General Antonio Guterres once said when he was still UN High Commissioner for Refugees. Nevertheless, at the COP17 UN Conference on combatting desertification in Mongolia this summer, the development of a protection framework for those commonly referred to as ‘climate refugees’ was once again poorly addressed by the COP’s agenda. The only item listed was a high-level meeting devoted to ‘security and migration’, while, in a striking contrast, pastoralism was being celebrated and promoted as well1 – a form of nomadism that tends to disregard the administrative divisions imposed by national borders.
This comes as no surprise. However appalling the IPCC’s assessments of global warming’s impacts may be – and however devastating the disasters punctuating this summer (heatwaves, droughts, mega-fires, floods, tornadoes, the landslide in the Himalayas); however alarming the projections concerning the number of people who may be forced to migrate as a result of environmental degradation linked to climate change – and however untenable the climate conditions in the very regions where refugee camps are often then located; finally, however frequent the declarations of intent made at major international conferences… the issue of ‘climate refugees’ has so far failed to materialise in specific and concerted international mechanisms, let alone in appropriate legal protections.
Individual states, protective of their sovereign functions and reluctant to open their borders, balk at making further concessions to international law, which they perceive as a form of dispossession. In addition, the joint development of international mechanisms to assist and protect climate-displaced people – particularly within frameworks like the COPs – would also stumble over the persistent question of cost-sharing between nations: the most vulnerable countries, invoking ‘climate justice’, argue that the poorest nations are the most affected despite having historically contributed the least to global warming; developed countries first insist that China take its share; and China, in turn, invokes its status as the world’s manufacturing hub to justify its position among the developing countries. The dust of the roads into exile appears to be swiftly swept from the diplomatic carpets. Meanwhile, the shrinking of habitable land as the oceans rise offers an enduring image of the slow closure of the field of possible action. National leaders behave like landowners bound by rigid patterns. There has been little visible progress.
A Politicised Issue
Although the current period is increasingly tense and divided, both nationally and internationally, concrete legal protection specific to climate refugees ought, in principle, to command consensus – both in the name of solidarity and human dignity in the face of a climate crisis that is planetary in scale and systemic in nature, and by virtue of the normal and indispensable adaptation of the law to a cause of displacement that existing instruments were never designed to address. One thinks here of the large international consensus around the protection that was granted to refugees through the 1951 Geneva Convention and its 1967 New York Protocol, which States adopted en masse after the horrors of the Second World War and despite the Cold War.
In France in 2024, during the campaign for the snap legislative elections held in the wake of the European Parliament elections, the Nouveau Front Populaire coalition (NFP, uniting the French left, including La France Insoumise (LFI) – the Communists, the Socialists, and the Greens) promoted the creation of a specific status for ‘climate-displaced persons’ in its manifesto – a first in France. A month earlier, Manon Aubry, co-president of the Left group in the European Parliament, made a similar demand during the European campaign, using the term ‘climate refugees’ instead. Other Green and left-wing politicians voiced similar proposals in the European Parliament even earlier.
This 2024 electoral proposal by the united French left may well have then drawn more comment from the right than from the left. “Some observers fear an acceleration of migratory flows,” ran a headline in the conservative daily Le Figaro. On the far right, the magazine Valeurs actuelles denounced a proposal it deemed ‘arbitrary’, devoid of ‘scientific foundation’ and amounting to ‘banal environmentalist activism backed by an increasingly aggressive legalism’. Some contours of the broader debate already emerge here: uncertainty around numerical projections as well as reception issues; the still-missing definition of the very concept of ‘climate-displaced person’; and the divisive tensions around how legal developments and collective political will interfere. The list, however, has to be extended further: What should be the legal form of a new status, and which branches of law would be relevant? How should this issue be articulated with the international ‘climate diplomacy’ conducted through the COPs? And, lastly, should the response be national, international, or anchored in collective mechanisms at the regional level in the first place?
The 2024 NFP initiative has direct antecedents in Europe, including texts from the European Parliament – notably a 2016 resolution on “human rights and migration in third countries” adopted by 339 votes to 333, which ended solemnly with a request that the European Union (EU) ‘participate actively in the debate on the term “climate refugee.” In fact, the original draft resolution, drawn up by the Parliament’s Committee on Foreign Affairs, was more developed and more concrete, calling for the EU to participate “in the possible formulation of a definition in international law.” The removal of this passage from the resolution as finally adopted, together with the narrow final vote, exposes the limits of consensus: agreement can be reached on the principle, but its practical implementation remains taboo.
This was confirmed five years later in 2021, when the European Parliament held a highly instructive debate on the report “on the impacts of climate change on vulnerable populations in developing countries,” which included, in particular, a proposal for ‘climate visas’. The report, this time presented by the Committee on Development, advocated the following: 1) a stronger orientation of EU external policy around explicit principles of climate justice; 2) an enhanced external humanitarian policy directed toward ‘geographic vulnerabilities’ and at-risk areas; and 3), regarding climate refugees, it “call[ed] on the Commission and the Member States to consider climate change-induced livelihood destruction as an eligibility criterion for humanitarian protection, call[ed] for the EU and the Member States to offer protection to persons forced to flee from a country that in part or in its entirety is becoming, or has become, uninhabitable due to climate change, notably through issuing humanitarian visas and granting temporary or long-term admission and measures to avoid statelessness; invite[d] the Commission and the Member States to discuss such proposals in international forums, in parallel to other EU initiatives.”2
However, the debate that followed in the European Parliament’s plenary session on 17 May 2021 clearly revealed both the points of consensus and division, as well as the respective political positions. The Greens/EFA group (Caroline Roose), aligned with the report, also highlighted the need to support smallholder farming, pastoralism, and agroecology everywhere, and called for a thorough reform of free trade agreements. Similarly, The Left group (Miguel Urbán Crespo) insisted ‘on the ecocidal policy of large multinationals’, likewise demanding a radical change of model. The centrist liberal Renew group agreed on the necessity of legal evolution and announced it would vote in favour of the report. The conservative EPP group, while pushing for its own amendment proposal, demanded that the Members of the European Parliament (MEPs) focus on scaling up external aid but drop the issue of migrants – an MEP from the group also framed the vision of aid within a principle of ‘climate justice’ and ‘historical and moral responsibility’.
As for the remarks of MEPs from the radical right, they ranged from a polemical, climate-sceptic, anti-immigrant, and anti-woke tone – including a complaint about birth rates in Africa – to a reminder of the multiple causes of migration, and the denunciation of state failures in at-risk countries. Finally, the European Commission’s representative closed the debate by focusing solely on EU external aid, without a single comment on the question of visas or the protection of climate refugees. Two days later, the final parliamentary vote buried the report.
Figures and Fears
Public opinion appears stunned; yet one thing is certain: people know the term. France’s “first climate refugees”: this phrase was used in July by Cécile Duflot, Executive Director of Oxfam France (former National Secretary of the French Greens and ex-Minister for Housing), while speaking to journalists about people forced to evacuate their homes because of forest fires (around 200,000 people evacuated around Bordeaux that summer). While Duflot was seizing the opportunity to call for heightened awareness at the crossroads of environmental and humanitarian emergencies, the media likewise seized the opportunity – to echo a well-known narrative dating back to the 1990s, namely the fear of a sudden influx of refugees fleeing entire regions of the world rendered uninhabitable.
Indeed, the figures put forward over the past three decades, and regularly disputed by researchers, refer to tens or hundreds of millions of migrants3 driven into exile by extreme environmental degradation – figures that vary according to who is speaking, which definitions are used, and which scenarios are chosen. If climate mitigation (in line with the Paris Agreement) fails, and in the absence of swift and ambitious national adaptation, populations are or will be trapped by rising sea levels and extreme aridity (gradual phenomena, also known as slow-onset events), or by disasters such as cyclones and floods (fast-onset events, also called extreme events). Therefore, they may be forced to migrate toward regions perceived as safer, including, provided they have the means to travel, toward countries offering sufficient employment opportunities, the Global North among them. Hollywood, meanwhile, created a successful new ‘climate’ sub-genre of disaster film out of this real-world scenario, in which solidarity clashes with selfishness, heroism with cowardice, or fatalism with willpower, as so many fictional, supposedly cathartic devices.
Governments, for their part, duly commissioned their own reports, essentially from an administrative, security-driven, and geopolitical perspective. In these reports, they notably highlighted the growing fragility of the worst-affected states to the point that these states could no longer guarantee human security – food, health, basic services, a functioning economy, fair access to justice, and so forth – and, therefore, could no longer contain civil conflicts nor prevent the rise of transnational criminal networks, thereby rendering the entire world more unstable in turn.
As for the peaceful coexistence and integration, even if only temporary, in host countries of those who fled, this represents a concrete challenge persistently raised by international organisations, which nevertheless struggle to be heard at a time when migration policies on the ground are increasingly eroding human rights, while both ballot-box results and outbursts of xenophobic violence – in Northern Ireland, South Africa, and elsewhere – point to various degrees of hostility toward incomers among parts of the population. Although “primary responsibility for the protection and well being of affected populations rests with the States concerned,” as Guterres warned on behalf of UNHCR in 2012, the traditional approach of responding to disasters and displacement “in emergency mode, establishing camps” must be reconsidered, he insisted. “The billions of dollars spent on relief in recent decades,” Guterres explained, “have evidently not led to the sustainable strengthening of national and local capacities. (…) A development-oriented and area-wide approach is thus required, ensuring that the most vulnerable and marginalized members of society are able to enjoy the human security and human rights to which they are entitled.” Against this background, Guterres evoked temporary protection solutions, and advocated for specific ‘planned relocation’ for those whose living areas have become permanently uninhabitable. Guterres was referring here chiefly to internally and regionally displaced persons affected by climate change, but his observation holds more broadly: if we fail to anticipate, both as a community of nations and as individual states, we will end up just building more camps. We need to look up, not look away.
But what are the ‘real’ figures? Alarmist or sceptical – with these terms, researcher François Gemenne, in his article entitled ‘How they became the human face of climate change’, describes how the subject has, in fact, been constructed since the 1980s around two poles: first a humanitarian and governmental pole, associated with alarmist data and geared toward early warning, anticipation, and operationality; then another pole, which Gemenne calls ‘sceptical’ and which is mainly composed of researchers specialising in migration (and whose findings, in fact, are also used by the first pole). Those researchers question the alarmist figures, stress the multicausality of migratory movements, as well as insist on the fact that those who are internally displaced (toward a neighbouring area or through rural exodus followed by precarious urbanisation) are far more numerous. They also detail the trap of extreme poverty, which climate change further amplifies to the point that households are sometimes simply condemned to stay. Finally, these researchers, just like international organisations and the at-risk populations themselves, remind us repeatedly of the deeply rooted desire of the people concerned to remain on their inherited land. As for the future effects of ongoing adaptation and mitigation policies, they must be factored into the equation as well… In other words, no numerical projection is reliable: this is something everyone at least agrees on.
Naming the Displaced
Because of this intrinsic multicausality of migration, the multiplicity of fields of study and actors, and also the absence of any international definition – the very designation of ‘climate refugees’ fluctuates: ‘migrants’, ‘refugees’ or ‘displaced persons’, with either ‘environmental’ or ‘climate’ as a qualifying adjective. We also find expressions such as ‘(im)mobilities under constraint’ which recall that the worst victims of climate change may be unable to migrate at all. The UNHCR, for its part, favours an explanatory formulation: “forced displacement in the context of the adverse effects of climate change and conflict.”4
The term ‘displaced person’ has an undeniable advantage over the other competing terms: it encompasses internal displacement (displacement occurring within national borders) – whereas the term ‘refugee’ implies transnational movement by definition. The term ‘displaced’ additionally conveys more clearly the idea of involuntary movement – like ‘refugee’.
However, the term ‘climate’ attached in the expression itself raises categorical, semantic, and political difficulties. Attributing forced displacement (whether temporary, long-term, or even permanent with no possibility of return) to the sole cause of global warming’s consequences is, indeed, problematic in several respects. Does ‘climate’ refer only to sudden-onset disasters such as floods and cyclones – or does it also encompass slow-onset processes such as rising sea levels, progressive desertification, and the increasing erosion of land – which then raises the problem of defining a ‘threshold’ for those slowly developing events? And are all these events not caused as much by the unpreparedness of public authorities – in short, political ‘climate inaction’ – as by climate change itself? In contrast, might the term ‘climate’ become a deterrent to legal progress, given that this could conflict with climate-sceptic states and judges? As for other environmental disasters (for instance, extreme industrial pollution accidents), should they also be considered, in which case should we speak of ‘environmental refugees’ instead? Likewise, does an earthquake, a not-climate-related disaster, fall within this framework?
Two of these questions – namely the perimeter and the ‘threshold’ (the ‘tipping point’) – have in fact been partly addressed by international judges. In Teitiota v. New Zealand (2020), the UN Human Rights Committee clarified that climate-related harm to be taken into account for individuals seeking asylum may not only result from sudden-onset but also from slow-onset events (see UN OHCHR’s analysis). Its own concrete interpretation of the threshold was nevertheless minimalistic in the submitted case, and the judges rejected the displacement protection request. As for the possibility of addressing ‘disaster’ victims independently of any reference to climate, this appears to be an avenue already being explored – by the multilateral Platform on Disaster Displacement, which emerged from the European-born Nansen Initiative, as well as through the decade-long elaboration of a binding multilateral UN convention on the ‘Protection of persons in the event of disasters’, which has been planned for adoption at a UN General Assembly session in 2027.
Finally, the term ‘(climate) refugee’ itself – although it is likely here to stay in everyday language – shifts the debate too quickly onto constrained normative ground: the legal status of ‘refugee’ is based almost exclusively on the status established by the 1951 Geneva Convention, which refers to victims of persecution – a status further completed and marginally, yet significantly, extended by regional agreements5 and sometimes by parallel national provisions (such as the French ‘constitutional refugee’ status for ‘freedom fighters’, actually rarely applied). Furthermore, the national administrative practices and regulations that, in turn, govern the 1951 status apply the granting of this status with varying degrees of restrictiveness or flexibility: interpretation and political orientation both matter. In short, were a climate refugee status to be enforced – assuming agreement could be reached on its definition – this would be first addressed at one of these levels: the original Geneva Convention, which might then need to be renegotiated; regional agreements with the consequent need to have them transcribed into national policy and case law; or national protection schemes of various kinds, which yet proves to be difficult, if we consider the experience of some Nordic European countries.6
The original strict Geneva definition, which entails persecution, makes this technically difficult, though not impossible: the Tribunal of the UN Human Rights Committee in the ‘Teitiota’ case ‘determined’ that no presumptions of non-applicability exist (View 2.8). That said, other already existing forms of (subsidiary or temporary) protection may be considered as an alternative. The EU Temporary Protection status – created in 2001 “in the event of a mass influx of displaced persons from third countries who are unable to return to their country of origin” (Directive 2001/55/CE, art. 1), with the former wars in ex-Yugoslavia in mind, was applied in 2022 for the first time for Ukrainian refugees, but is in theory one of them. As for the mere possibility of allowing non-refoulement on grounds of danger, this too represents a challenge in its own right, and currently constitutes a scrutinised stake in national courts just as what precedes. This, in any case, explains why legal scholars remain opposed to the use of the term ‘refugee’ in this debate, despite its symbolic force.
The long formulation favoured by the UNHCR, “forced displacement in the context of the adverse effects of climate change and conflict,“ can be explained in yet another way: outside situations of collective flight in the emergency aftermath of a disaster (floods, typhoons, and so on), migration – as already stated, and as all reports confirm (IOM, UNHCR, the World Bank, NGOs, and others) – is very often multicausal: the current motives behind many migrant stories include the search for better socio-economic conditions following the slow degradation of living environments, flight from growing political instability, and, finally, a lack of trust in the ability of state institutions to manage these ongoing degradation factors. Ultimately, climate thus appears both as one causal factor among others and also as a factor that amplifies other causes: conflicts (including land disputes), impoverishment, and state inability to manage resource scarcity, as well as to meet the basic needs of populations: food, health, education, and other services including security, as well as institutional stability. The unfolding climate crisis is therefore already causing departures, though it is not their sole identifiable cause. That complexity leaves us without reliable indicators – which, as stated by Dina Ionesco (HCR) in her hearing for the French Senate in April, further hampers international decision-making.
Meanwhile, as we see with each COP event, European states are eager to showcase their climate diplomacy in a game they discretely hope will yield geopolitical gains on a rapidly reshaping multipolar stage. But they may well find themselves accused of double standards by the very partners they are courting if they continue to evade the urgent and far-reaching challenge posed by climate displacement, going so far as to refuse even to define the notion and give it the specific and timely attention it requires. Climate refugees are worth everyone’s attention. •
This article first published on the Transform Network website.
Endnotes
- Mongolia hosted COP17 during the UN International Year of Rangelands and Pastoralists and launched three initiatives under its Steppe Action Agenda. Among them, the Rangelands Flagship Initiative attracted the greatest attention and secured the largest share of the funding announced during the Conference.
- The rapporteur Mónica Silvana González (S&D) mentioned in her opening remarks to Parliament: “the protection of existing legal mechanisms to ensure the rights of people who are forced to move internally or across borders, building on the Nansen Initiative for the Protection of Displaced Persons; that the risk to life threatened by climate change be explicitly included among the eligibility criteria for humanitarian protection, following the United Nations Human Rights Committee’s resolution in the case of Teitiota, as stated by UNHCR itself. As a key proposal: a climate visa, a humanitarian visa as a temporary protection measure that ensures legal and safe migration routes for people in need of protection due to being victims of disasters.” (verbatim transcript from the discussion of this report in the European Parliament)
- Over one billion people may be at risk of being displaced by 2050 due to the combination of environmental change, conflict and civil unrest, according to the inaugural Ecological Threat Register (ETR 2020) published by the Institute for Economics & Peace. Source: Geneva Center for Security Policy.
- Hearing of Dina Ionesco (HCR) before the French Senate’s Commission of information on Climate Diplomacy, April 2026.
- Among regional instruments, the legally binding 1969 Convention of the Organisation of African Unity (OAU) adapted the refugee definition to the African context by extending the 1951 Geneva Convention criteria to persons compelled to leave their country “owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of [their] country of origin or nationality.” In Latin America, the 1984 Cartagena Declaration pursued a similar objective by recommending refugee protection for persons who had fled their country because their lives, safety or liberty had been threatened by generalised violence, internal conflicts, massive violations of human rights or other circumstances seriously disturbing public order. The Cartagena Declaration is not, in itself, legally binding; however, its principles may acquire binding force when incorporated into domestic law by individual states. Main source: Morel & Moor, (2012). “Migrations climatiques: Quel rôle pour le droit international?” Cultures & Conflits, (88), Article 88.
- Sweden and Finland once established specific forms of legal protection for people displaced by disasters and climate-related environmental degradation. These measures have since been repealed, and in practice were rarely applied by the courts.
