Protecting States, Not People

Like all “landmark” pieces of international jurisprudence, the closer one reads the International Court of Justice (ICJ)’s advisory opinion on the Obligations of States in respect of Climate Change of July 2025, the more one finds things the Court might have done better. Among the areas that have attracted dissatisfaction is the brevity with which the Court treated two closely interconnected questions: the continuity of statehood in the face of territorial loss and the obligations of states towards people on the move because of climate change. The movement of people amid climate change, usually referred to as “climate mobility”, includes both internal and international migration, and spans from forced displacement to voluntary relocation or “managed retreat”. As this post explains, the ICJ concentrated on the principle of non-refoulement, which only applies to those forcibly displaced across international borders. This minimalism has attracted some criticism, as some were hoping for a more substantial treatment of climate mobility. In this post, I look at what the states intervening in the proceedings had to say about state obligations towards those on the move because of climate change, to see if there is a difference between what we think the Court should have said, what the Court actually could have said, and why it would or wouldn’t do so.

What the ICJ Said and What It Could Have Said

In a recent post on this blog, Alex Green discussed how General Assembly Resolution 80/263 expands upon what the ICJ said about state continuity in the face of the existential threat posed by rising sea levels. The ICJ had used deliberately cautious, possibilistic language, stating that “once a State is established, the disappearance of one of its constituent elements would not necessarily entail the loss of its statehood” (para 363). The General Assembly went further, expressly “affirm[ing] the continuity of statehood in the face of sea level rise” (para 7). This is significant, as states appeared willing to stand behind fellow states whose territories face the consequences of climate change, and to affirm that the disappearance or transformation of territory does not put an end to their international legal personality. Yet they didn’t demonstrate the same willingness when it came to the displacement of those states’ populations. In relation to climate mobility, the General Assembly limited itself to reproducing verbatim (para 8) the ICJ’s stance that:

“States have obligations under the principle of non-refoulement where there are substantial grounds for believing that there is a real risk of irreparable harm to the right to life in breach of Article 6 of the ICCPR if individuals are returned to their country of origin” (para 378.)

Jane McAdam has argued that, despite its limitations, the Court’s engagement with the topic nevertheless provides a legal foundation on the link between climate impacts, threats to the right to life, and non-refoulement that can be operationalized in international protection claims. Others were more critical and saw the ICJ’s single paragraph as a missed opportunity to expand more on state obligations regarding climate mobility.

The ICJ’s one-sentence formulation is striking when compared to the Inter-American Court of Human Rights (IACtHR) which, only weeks earlier, addressed the human rights obligations of states on climate mobility in considerably greater detail in its Advisory Opinion OC-32/25. On the one hand, the IACtHR was specifically requested to address this aspect, among others, while the General Assembly did not explicitly ask the same of the ICJ. On the other hand, the World Court’s advisory jurisdiction is not confined to the literal wording of the questions it receives (Western Sahara, para 19; Kosovo, paras 49–51), so it could have addressed climate mobility more extensively, especially since the request’s preamble expressly identified displacement among the consequences of climate change (point 8). In particular, Judge Aurescu described the treatment of non-refoulement as incomplete in his separate opinion and argued that the Court should have gone beyond the negative obligation not to return individuals to situations of serious risk. In his view, international human rights law could also entail positive obligations, including individualized risk assessments, admission, temporary residence permits, and measures protecting individuals against circumstances that could result in indirect refoulement.

Yet, the task of the ICJ in an advisory opinion is to ascertain and clarify existing international law, not to create new legal obligations, and the Court had limited evidence to work with. On the one hand, there is some evidence that a practice of applying non-refoulement to people displaced by climate change is developing. The most prominent point of reference is the 2019 Teitiota v New Zealand decision, in which Human Rights Committee accepted in principle that environmental degradation and climate change could, in sufficiently serious circumstances, trigger non-refoulement obligations where return would expose an individual to a real risk of irreparable harm to the right to life. The ICJ expressly relied on Teitiota in formulating its own conclusion. Some states’ courts (but not governments) have also started to recognize climate and environmental factors in their jurisprudence on international protection and non-refoulement. The IACtHR’s advisory opinion has clearly derived several state obligations regarding climate mobility, including the prohibition of refoulement, from the Pact of San José, and there are substantial expectations that the African Court of Human and Peoples’ Rights will expand on climate displacement obligations in its own forthcoming advisory opinion.

What States Did and Did Not Ask the Court to Do

It would be difficult to claim, however, that these steps are enough to prove that a customary rule governing climate mobility or the application of non-refoulement along the lines of what Judge Arescu suggested had already crystallized. And, crucially, nearly no states (and no international organizations) participating in the ICJ proceedings made such a claim. Climate mobility appeared frequently in the written and oral submissions, with one third of submitting states describing displacement as one of the negative impacts they face; but very few went on to elaborate on the obligations that international law imposes towards climate-displaced people, including non-refoulement. States, including in the Global Majority, appeared keener on discussing how they are affected by climate change than taking up new responsibilities.

A prominent exception was the Solomon Islands, whose submissions devoted significant attention to climate mobility and non-refoulement obligations. At the same time, these submissions themselves characterized the relevant protection frameworks as “evolving” norms (para 23). Since states themselves have stopped short of asking the Court to declare a settled customary rule of climate non-refoulement, there was less room for the Court to present a duty not to return – let alone to admit – as existing law. The Court is not bound by states’ explicit submissions in its determination whether a rule constitutes a customary norm, and it may rely on the implicit practice of a “significant majority” of states even when others have explicitly opposed it. But this does not seem to be the case here, as the limited steps described above do not appear sufficient to constitute a consistent practice or opinio juris on non-refoulement, let alone other obligations regarding climate mobility. This helps to explain, at least in part, the Court’s caution in expanding its formulation of the prohibition of refoulement in the context of climate change beyond the position of the Human Rights Committee in Teitiota or elaborating on further obligations.

Climate Mobility, Cooperation, and Responsibility-Sharing

Rather, several submissions emphasized the need for an international cooperative response. This was an attractive position, since climate mobility is not the product of a single event, nor it is attributable to a single state. Its causes, consequences, and the capacity to respond are all distributed unequally. However, state submissions did not specify the form and legal content of such a duty of “cooperation”.  The ICJ ultimately did recognize a general duty to cooperate in addressing climate change, but it did not articulate a distinct duty of international cooperation specifically in relation to climate displacement. Going much further would have required the Court to determine the precise content of obligations – such as admission, temporary residence, procedural safeguards, and responsibility-sharing – which states did not elaborate in their submissions. The General Assembly’s subsequent resolution likewise retained the general language of cooperation rather than developing a specific framework for responsibility-sharing in relation to mobility.

This was also echoed in the recent UN Declaration on Sea Level Rise (focused specifically on the implications of sea-level rise for statehood rather than climate change more generally), in which states again endorsed the continuity of statehood, but made no reference to non-refoulement. The Declaration addresses mobility only by emphasizing “the importance of supporting voluntary choices for communities affected by sea level rise and mobility with dignity”, where international movement is involved, “in agreement with relevant States” and consistently with international law (point 15). In both interventions, states were prepared to articulate a clearer collective position on the continuity of states threatened by sea-level rise but not one on the obligations owed to the people fleeing those states.

The normative foundations for a stronger duty to cooperate in the context of climate mobility are increasingly being explored in scholarship. In her recent article, Samantha Velluti advocates for understanding climate mobility as a “common concern of humankind” requiring cooperation and solidarity-based responses rather than leaving protection entirely to unilateral state discretion. Elizabeth Mavropoulou’s work on responsibility-sharing in refugee law has explored how the principle of “common but differentiated responsibilities” (CBDR), developed within international climate law, might provide a normative basis for distributing migration-management and refugee-protection responsibilities according to states’ different capacities and contributions. I and others have argued elsewhere, building on Mavropoulou’s analysis, that if CBDR can inform responsibility-sharing in migration governance generally, it applies a fortiori to the context of climate mobility.

The ICJ and Its State-Centred World

This is an argument about the direction in which the law could develop, while there remains a substantial, political gap between recognizing climate mobility as a common international problem and agreeing to share the responsibilities that follow from that recognition. The global political environment is one of human rights regression, hardened borders, externalization, anti-migration discourses, xenophobia, and nationalism. Against this background, expecting states to accept a new category of internationally protected migrants, or to assume more substantial responsibility for people displaced from other countries, is apparently politically more difficult than affirming the continued existence of those countries themselves. Seen in this light, the Court’s caution, as disappointing as it may be, is far from surprising. This is not intended to be apologetic or exculpatory of the ICJ’s restraint; rather, it is a realist recognition that international law is made by states, and states and their political leadership are, at best, highly cautious and, at worst, openly opposed to assuming any further responsibility towards migrants and displaced ‘aliens’. The ICJ is the jurisdictional expression of this state-centric international legal infrastructure we live in; and the master’s tool won’t build a house the master does not want to take care of.