Refugee Law Initiative Declaration on International Protection in the context of Disasters and Climate Change (2024)

1. The protection risks arising in the context of disasters and climate change contribute to driving internal and cross-border movements by people. These risks are also increasingly being raised before national governments and courts in claims for international protection by persons who are outside their own country. Thus, there is a need for clear legal guidance on determining refugee status and applying human rights non-refoulement rules in relation to such claims.

2. The Refugee Law Initiative is an academic institution that brings together specialised refugee law researchers and practitioners from across the world. In this Declaration, we set out guidance on how the international protection rules expressed by refugee law and human rights law apply to claims relating to disaster contexts and climate change in the country of origin.1 This guidance is directed towards refugee status decision-makers, judges, lawyers, policymakers, scholars and others.

SITUATING INTERNATIONAL PROTECTION AMONG WIDER LEGAL/POLICY RESPONSES

• 3. International protection frameworks in refugee and human rights law are but one of a number of legal and policy tools that can facilitate admission and/or stay in another country for people affected by a disaster and/or the negative impacts of climate change in their country of origin. International protection law is likely to be pertinent in only a relatively limited set of circ*mstances.

• 4. Migration frameworks at the national and international levels – including bilateral, sub-regional, regional and other international arrangements – are often the primary tools for admission and stay of non-nationals affected by a disaster in their home country. Both regular and exceptional migration categories can offer potential pathways; and some provisions of migration law and policy expressly address the situation of disaster-affected persons or allow special treatment in such cases.

• 5. In practice, then, international protection frameworks are likely to come into play only where a person facing serious risks arising in the context of disasters or the negative impacts of climate change in their home country is unable to access such migration pathways or where the migration frameworks do not adequately respond to protection needs in the individual case.

CONCEPTUALISING INTERNATIONAL PROTECTION CLAIMS INVOLVING DISASTER ELEMENTS

• 6. For determining international protection claims, decision-makers must engage with the particular nuances of how risks of harm can arise in disaster and/or climate change contexts. Five key considerations for conceptualising this risk of harm can be derived from the research:

  • (a) In disaster contexts, a risk of harm can arise from the effects of a hazard, the ensuing disaster conditions, and for other disaster-related reasons (e.g. State persecution of people who criticise the disaster response), regardless of whether the situation meets formal definitions for a “disaster” or whether it is designated as such by the authorities.2

  • (b) In disaster contexts, the risk of harm generated by a hazard, even a natural one, is shaped not only by its frequency and intensity but also by human factors that determine the extent to which people are exposed and vulnerable to the hazard and can manage the risks.

  • (c) In any disaster context, different kinds of hazards – natural, human-induced and socionatural – can exist. They may also interact to produce risks of harm at different time scales. Sudden-onset disasters are triggered quickly or unexpectedly, whilst slow-onset disasters emerge gradually over time; although sudden- and slow-onset events often interact.

  • (d) Risks of harm in disaster contexts often overlap with risks of harm deriving from other societal processes such as armed conflict and inter-communal tensions. A holistic approach must consider the full spectrum of all the risks posed by these overlapping sources of harm, in addition to the different kinds of hazards that may be present.

  • (e) Climate change can act as a threat-multiplier. It has the potential to shape the ways in which existing weather- and climate-related hazards in any locality manifest themselves over time and to produce new ones. It can also negatively affect the capacity of societies to cope with disasters and manage conflicts.

ASSESSING INTERNATIONAL PROTECTION CLAIMS INVOLVING DISASTER ELEMENTS

• 7. Where the international protection regime is applied, the legal approach to claims relating to disasters or climate change in the country of origin is rooted in two fundamental principles:

  • (a) Claims constituted against the factual backdrop of disasters are capable of engaging the rules of international protection. The disaster context does not, in itself, make such claims ineligible for protection; nor does it automatically lead to the conclusion that there is a need for international protection. As with any other case, each claim must be assessed on its own facts against the international protection rules ordinarily applicable in that jurisdiction.

  • (b) In assessing the international protection claim, no new or special rules apply. Legal principles and evidential rules ordinarily applicable in the jurisdiction apply unchanged to these claims. In particular, there is no need to determine whether the situation meets formal definitions for a “disaster”. That it has been classified as such by the authorities may be indicative of the risks posed to an individual claimant but is never determinative.

• 8. There are varied ways in which disaster contexts can generate risks of harm potentially relevant to claims for international protection. Illustrative factual scenarios include:

  • Scenario 1: where disputes or controversies around disasters, climate change or related environmental issues place inhabitants at risk of harm at the hands of other people.

  • Scenario 2: where disasters contribute to a breakdown in public order or aggravate dynamics of conflict, violence or exploitation, thereby producing or exacerbating for inhabitants a risk of harm at the hands of other people.

  • Scenario 3: where State or non-State actors induce a disaster by severely degrading the natural environment, whether deliberately or as an accidental consequence of other dangerous activities, placing inhabitants at risk of harm.

  • Scenario 4: where the nature of the response by the State or other authorities to disasters, or their disaster risk management (DRM) and/or climate change adaptation (CCA) action, produces or exacerbates the risk of harm for inhabitants.

  • Scenario 5: where the manifestation of natural hazards, and/or the disaster conditions that they trigger, presents a risk of harm for persons exposed to them.

The existence of one or more of these (or other similar) common factual scenarios is indicative, not determinative, of the potential for harm to a claimant. Whether their circ*mstances legally engage the rules of international protection will depend entirely on the particular facts of the case.

• 9. In assessing claims arising from a disaster context, decision-makers will need to be alert to the important ways in which human or societal factors (political disputes, ethnic discrimination, gender considerations, etc.) often shape the risks of harm arising in these scenarios. Moreover, decision-makers are likely to need to supplement familiar sources of country of origin evidence with evidence from sources more directly concerned with disasters, climate change and the environment.

INTERPRETING REFUGEE DEFINITIONS IN RELATION TO DISASTER-RELATED CLAIMS

• 10. Established refugee law principles and standards apply unchanged to the assessment of claims arising from a disaster context. Decision-makers should not characterise a person or claim by using non-legal or confusing terms such as “climate refugees” or “disaster refugees”. Instead, the existing refugee definitions ordinarily applicable in the jurisdiction must be applied to such claims.

• 11. How the elements of international refugee definitions should be interpreted in disaster-related claims is set out below for: inclusion as a refugee under Article 1A(2) of the Refugee Convention (paras 12–24) and under complementary refugee definitions in regional instruments (paras 25–26); and for cessation of refugee status under Article 1C(5) of the Refugee Convention (para 27).3

Being persecuted

• 12. In claims for refugee status arising from disaster contexts, it may be tempting to assume the dangers are “natural” and thus not capable of constituting “persecution”. But many of the risks of harm that arise in the five identified disaster Scenarios (para 8 above) are at least partly constituted by human acts or omissions. In cases where intentional human actions underpin the anticipated harms arising in these Scenarios, they have the potential to be persecutory in nature.

• 13. Disaster contexts can generate direct forms of persecution, for example, where certain people are targeted for ill-treatment due to their stance on disaster-related or wider environmental issues, or where the breakdown in public order following a disaster exposes certain people to a heightened risk of ill-treatment, or where a community’s water sources or other essential natural resources are intentionally destroyed or “weaponised” against them as a means to harm them (Scenarios 1–3).

• 14. In tandem, persecutory harm in disaster contexts can result also from broader discriminatory practices that expose certain people to additional forms of danger in the specific context of disasters, for example, by forcing certain inhabitants to live in areas particularly exposed to hazards, excluding them from risk reduction programmes, leaving them in harm’s way when others are evacuated, or denying them access to life-saving relief in the aftermath of a disaster (Scenarios 4–5).

• 15. All of these diverse forms of persecutory harm arising from the disaster context, as well as any similar persecutory harms faced by the claimant from sources existing outside the disaster context, must be taken into account “cumulatively” in determining whether the “severity” threshold implicit in the “being persecuted” concept is met by the anticipated harms in the particular case.

For reasons recognised by the Convention

• 16. In disaster contexts, persecution in all these forms can occur “for reasons of” any one or more of the Convention grounds. The nexus with a Convention ground can derive from either the risk of harm or the failure of State protection, as including both acts and omissions by the State (para 22 below). In Scenarios 1–3, provided that the infliction of harm is on Convention grounds, the fact that such persecution takes place in a disaster context is irrelevant to establishing this nexus. Disasters, climate change and wider environmental issues are “political” in many societies.

• 17. Likewise, in Scenarios 4 and 5, the person’s exposure and vulnerability to disaster-related harms must be linked to one or more Convention grounds. That link must also be sufficiently proximate to satisfy the “for reasons of” nexus test as applied in the decision-maker’s national jurisdiction. For instance, this may not be the case where the person’s exposure or vulnerability to disaster risk results merely from structural patterns of disadvantage in society rather than from specific practices of discrimination towards particular profiles of persons by State or non-State actors.

Well-founded fear

• 18. In claims for refugee status that concern disaster contexts, the “well-founded fear” test applies unchanged. Given the forward-looking nature of this test, common sense dictates that persecutory harm must be “reasonably foreseeable”. However, assessment of the risk of such harm eventuating in the future depends not on broad statements about its “foreseeability”, but on a careful analysis of the evidence as to the individual and contextual features said to give rise to the feared scenario.

• 19. There is no legal requirement that such a risk must be temporally “imminent” in order for it to be “well-founded”. Time instead operates as a sliding scale as to the strength of evidence required to show that a fear is “well-founded”. The further into the future that the feared scenario lies, the more compelling the evidence will need to be of the likelihood of its occurrence and the risk to the individual claimant and the absence of potential mitigating factors.

• 20. Particularly for non-recurrent hazards, past events may not be determinative of future risk where circ*mstances have changed in the interim. But, as many natural hazards are recurrent, seasonal or worsening in the context of climate change, decision-makers should be hesitant to find that the fear of such a hazard recurring is not “well-founded”. Moreover, even if a sudden-onset disaster has passed, decision-makers must consider whether any discriminatory practices underpinning a claimant’s exposure and vulnerability to hazards or lack of access to support remain in place and may produce a future risk of other discriminatory harms.

National protection

• 21. Given the range of persecutory harms that can arise in disaster contexts (paras 13–14 above), the decision-maker’s assessment of national protection in the country of origin must take account of any risk-reducing activities by the State. The form of protection must be appropriate for addressing the particular kinds of harm(s) feared. It may include not only national protection mechanisms for mitigating the risk of harm from human actors (police, judiciary etc.) but also DRM, CCA and sustainable development mechanisms for addressing disaster risk.

• 22. Where a State’s unwillingness to protect is for Convention reasons, this may make the resulting exposure of the claimant to serious harm (including from natural sources) persecutory. Where a State is willing but unable to protect, the serious harm feared may be persecutory (a) where the harm is inflicted by non-State actors (i.e. Scenarios 1–4) for Convention reasons, or (b) where the claimant’s exposure or vulnerability to a hazard (i.e. Scenario 5) results from a sufficiently close causal link with discriminatory practices by non-State actors for Convention reasons.

• 23. In Scenario 5, the prospect of “planned relocation”, as an organised measure used in the context of disasters and climate change to move people out of harm’s way, may also need to be considered as part of the national protection assessment. However, decision-makers should be cautious in determining that relocations provide adequate protection for current risks, unless a programme already exists and has a strong track record of effective implementation.

Internal protection alternative

• 24. All disaster-related and other risks of harm or suffering in a potential internal protection alternative (IPA) site need to be cumulatively factored into assessing whether, for the particular individual, (a) there is a risk of persecution there and, if not, (b) its accessibility and suitability as an IPA. Disaster-related risks in these sites are always pertinent to the IPA assessment, regardless of whether the persecution in the claimant’s home area is for disaster-related reasons or not. In Scenario 5, any planned relocation measures may also need to be considered (para 23 above).

Regional refugee definitions

• 25. Regional refugee definitions in Africa and Latin America extend refugee status to people fleeing serious disturbances of public order. The impacts of natural hazards and adverse effects of climate change may amount to disasters that can contribute to creating or exacerbating such disturbances. Serious disturbances of public order often occur in situations where disasters interact with and exacerbate or trigger armed conflict and generalised violence. Depending on the situation, the same may be true where disasters contribute to creating or exacerbating other circ*mstances that seriously disturb public order.

• 26. Under the regional refugee definitions, individual claimants must also face a risk of serious harm as a result of these events or circ*mstances, but this need not be on discriminatory grounds.

Cessation

• 27. Cessation of refugee status on the basis of a change of circ*mstances in the country of origin cannot occur where a well-founded fear of persecution persists (para 20 above). Even where the original basis of the refugee claim made no reference to the risks posed by disaster contexts, decision-makers should consider them at the point of determining whether to cease refugee status if the evidence now suggests that they are relevant factors. They should also pay attention to the extent to which the institutions and infrastructure necessary for protection have been weakened over the short- and long-term by disasters and the impacts of climate change.

APPLYING HUMAN RIGHTS NON-REFOULEMENT RULES TO DISASTER-RELATED CLAIMS

• 28. Non-refoulement obligations in human rights law represent a complementary source of international protection, alongside refugee law. Most of the rights underpinning non-refoulement obligations can be violated not only by harms emanating from human sources but also by those deriving from natural sources, including natural hazards. The assessment of whether harms in the country of origin meet the “minimum level of severity” required to trigger these non-refoulement obligations is concerned with their “cumulative” impact on the individual if returned there and not any comparison between conditions in that country and the host country.

• 29. Disaster contexts can present risks of harm at the hands of State or non-State agents (Scenarios 1–4) and also as a result of hazards and their effects (Scenario 5). The anticipated harms need not be on discriminatory grounds. All of the pertinent dangers arising from the disaster context must be considered cumulatively, in combination with those from other sources (e.g. conflict, political violence, etc.), in determining whether the anticipated harm is serious enough to engage non-refoulement obligations.

• 30. Human rights non-refoulement rules require a forward-looking assessment of risk. The standard is usually expressed as “substantial grounds for believing” that the person faces a “real risk” of harm sufficient to breach the severity threshold. The interpretative approach to the risk test in refugee law (paras 19–20 above) applies equally here, i.e. the more temporally distant the anticipated future harm, the stronger the evidence that will be required to meet this legal test. This assessment of risk must also consider whether measures taken within the country to protect people from these harms are sufficient to displace that “real risk” and whether an internal relocation alternative exists.

Regional approach in Europe

• 31. The European Court of Human Rights suggests that where the “predominant cause” of the harm is poverty or “the State’s lack of resources to deal with a naturally occurring phenomenon”, then a higher severity threshold of “very exceptional” circ*mstances applies than where it is “the direct and indirect actions” of human State or non-State actors. As yet, the Court has not applied this higher threshold to any case involving refoulement to a country where disasters play into conditions of risk on the ground, as compared with its frequent application of this threshold to a line of cases involving challenges to removal on “medical” grounds. This may recognise the distinct nature of disaster contexts in which the risks of harm are so often shaped by human elements (see para 6).

[With thanks to the Refugee Law Institute for providing this document]

1

For a fuller legal analysis and citations, please see the accompanying analytical paper: David Cantor et al, ‘International Protection, Disasters and Climate Change’ (2024).

2

For example, the United Nations Office for Disaster Risk Reduction (UNDRR) has defined a “disaster” as “a serious disruption of the functioning of a community or a society at any scale due to hazardous events interacting with conditions of exposure, vulnerability and capacity, leading to one or more of the following: human, material, economic and environmental losses and impacts” <https://www.undrr.org/terminology/disaster> accessed 30 April 2024.

3

It is assumed that disaster contexts do not throw up any particular issues in relation to exclusion definitions or other limbs of the cessation clauses, which should continue to apply as usual.

Published by Oxford University Press 2024.

Issue Section:

Document

Email alerts

See also

Citing articles via

More from Oxford Academic