Despite the continued use of the term in media reporting about human mobility in the context of disasters and climate change, it has long been established in academic circles that ‘climate refugees’ do not exist. The definition of a refugee in the 1951 Refugee Convention does not identify environmental factors. Consequently, most scholarship, as well as policy initiatives, have focused on alternative mechanisms for preparing for and addressing the cross-border movement of people in the context of disasters and climate change.
The Protection Agenda
The most comprehensive response to this issue can be found in the 2015 Agenda for the Protection of Cross-Border Displaced Persons in the Context of Disasters and Climate Change (‘the Protection Agenda’), which was produced as the synthesis of a series of regional and sub-regional consultations with state, civil society, academic and other actors. Although the Protection Agenda recognizes the duty on states to consider whether individuals who enter into their jurisdiction and seek international protection qualify as refugees under Article 1A(2) of the 1951 Convention, the main focus of the document is on the elaboration of ‘effective practices’, including temporary protection and stay arrangements, that were documented during the consultative processes.
The response reflected above is pragmatic, as it sets out to develop and promote a ‘toolkit’ of practices that states can use as and when they are faced with the (attempted) arrival and stay of people who are displaced across borders in the context of disasters and climate change.
Reviewing national case law
However, the rejection of the notion of the climate refugee has resulted in an under-examination of the scope of the Refugee Convention in some cases. Identifying this gap as part of doctoral research I conducted from 2012 until 2018, and which is summarised in my book Climate change, disasters and the Refugee Convention (CUP, 2020), I set out to examine instances where individuals had actually sought recognition of refugee status and who had identified adverse environmental conditions as being of relevance to that claim. Using UNHCR’s RefWorld database, I searched cases for terms such as ‘cyclone’, ‘drought’, ‘flood’, and so forth, and identified a cohort of approximately 200 relevant cases, mostly from Anglophone jurisdictions. Reviewing these cases enabled me to see the kinds of circumstances that people felt gave rise to a need for international protection, and how judges interpreted and applied the law in relation to such claims.
One finding of the research was that judges were far more prepared to recognize individuals as refugees in situations where a human actor of persecution was identifiable, such as in cases where the threat of violence emerged in a disaster setting. For instance, in Refugee Appeal No. 76374, the New Zealand Immigration and Protection Tribunal recognized as a refugee a woman who was at risk of being imprisoned for an extensive period for her role in delivering humanitarian assistance in the aftermath of Cyclone Nargis in 2008. The disaster in this case was a mere backdrop to familiar patterns of persecution. Similar cases were identified in the United Kingdom and Australia.
Of perhaps more interest than the reasoning in cases that were successful, however, was the way judges reasoned around some of the cases that were not successful. Examining the multiple reasons why judges considered the Refugee Convention did not apply revealed assumptions about the refugee definition itself, and a significant part of my research was devoted to the scrutiny of this material.
The cases that revealed the most interesting reasoning in this context were those relating to claims for international protection in situations where the harm the person feared being exposed to on return concerned failures of state protection, rather than the direct and intentional infliction of serious harm. Failures might relate, for instance, to landslides that could have been prevented, or the adversity that accompanied rising sea levels. What came through clearly in this analysis were assumptions about both the temporal and the personal scope of the refugee definition.
In relation to the temporal scope, there was a strong tendency for judges to see ‘persecution’ as an event, with examples such as killing or torturing reflecting such conduct. With notable exceptions from the New Zealand jurisprudence, it was rare indeed for any judge to interpret the experience of ‘being persecuted’ as a condition of existence that extends temporally beyond the moment the person actually suffers an ‘act of persecution.’
In relation to the personal scope, the tendency was to see any person who risked being exposed to serious harm as facing a risk of ‘persecution’. Thus, even persons whose exposure to serious harm was not related to their race, religion, nationality, membership of a particular social group or political opinion, could nevertheless, on this interpretation, be accurately described as having a fear of ‘persecution’.
Re-defining a ‘refugee’
Using the methodology of treaty interpretation set out at Articles 31-33 of the 1969 Vienna Convention on the Law of Treaties, I argue that dominant interpretations of Article 1A(2) inaccurately interpret the experience of being persecuted as being essentially synonymous with being the victim of a human rights violation, with the consequence that the distinguishing aspect of discrimination that is inherent in the predicament of being persecuted is lost, with implications for how refugee status is determined. Retaining the best parts of existing interpretations, I define being persecuted as a condition of existence in which discrimination is a contributory cause of (a real chance of being exposed to) serious denials of human rights demonstrative of a failure of state protection.
I argue that this interpretation invites decision-makers to ask more nuanced questions in cases where individuals express a fear of being exposed to disaster- and climate change-related harm if forcibly returned home. The interpretation draws attention to the broader risk landscape, which for certain people is infused with structural violence that from time to time also manifests in exposure to serious harm, whether directly and intentionally inflicted, or as a consequence of vulnerability and exposure to disasters, disease or economic shocks.
I conclude that this approach does very little to expand the availability of refugee status to a wider group of people. Most people displaced across international borders in this context will still not qualify, which is why initiatives such as those reflected in the Protection Agenda remain of critical importance. However, adopting the maxim that claims for international protection deserve to be examined with ‘anxious scrutiny’, it is important that the task is undertaken with the sharpest possible tools.
It is with great pleasure that I am now able to continue this line of research in a collaboration with the Ludwig Boltzmann Institute, where we are examining decisions in claims for international protection brought in Austria and Sweden.
 RN (Returnees) Zimbabwe CG
 RRT Case No. 0903555