IN Obligations of States in respect of Climate Change (Advisory Opinion of 23 July 2025) (“the 2025 Opinion”), the International Court of Justice (“ICJ”) unanimously confirmed that states have obligations under several international treaties and under customary international law to protect the climate system and wider environment from anthropogenic greenhouse gas (“GHG”) emissions and that breach of these obligations constitutes an internationally wrongful act entailing international responsibility. Initiated by Vanuatu, the proceedings attracted widespread engagement: there were 91 written statements from states and international organisations, reflecting a wide range of positions. The Court was asked by the General Assembly to address two sets of questions, respectively concerning the international obligations of states to protect the climate system from GHG emissions and the legal consequences for states where they have caused significant harm to the climate system. Although sparse in places, the 2025 Opinion provides much-needed clarity on states’ climate change obligations and shows how what was once considered by some to be a fragmented, somewhat normatively ambiguous area, is in fact relatively integrated, coherent and core to international law.
On treaty obligations, the Court noted that the three core climate change treaties not only complement each other but are also complemented by human rights and law of the sea treaties. UN Framework Convention on Climate Change (“UNFCCC”), Article 2 sets the objective of “stabilis[ing] greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system” and contains basic principles and obligations with respect to climate change, while the Kyoto Protocol and the Paris Agreement “translate these […] into a set of more specific interrelated obligations” (at [120]). States must comply with obligations of conduct under the three climate change treaties, including the obligation of cooperation and assistance set out under the UNFCCC (at [214]–[218]) and concretised in the Kyoto Protocol (at [219]–[220]), and the obligation to exercise due diligence under the Paris Agreement, by employing best efforts to use all available means (at [229]).
States must also comply with obligations of mitigation under the Paris Agreement by preparing, communicating and maintaining successive Nationally Determined Contributions (“NDCs”), which set out plans to reduce GHG emissions. States have limited discretion when it comes to formulating their NDCs and the Court advised that NDCs must, “when taken together, be capable of realizing the objectives of the Agreement”, such as the 1.5-degree temperature goal (at [249]). This is important because it means that individually determined national contributions can be objectively assessed more easily. Of significance is the Court’s recognition that the principle of “common but differentiated responsibilities in respective capabilities” is evolutive (although Judge Xue was very critical of this in her dissenting opinion) (at [226]), which is important for the formulation of NDCs. Nevertheless, important practical questions remain: How are historic emissions to be taken into account and from what date? What precise circumstances of states are relevant in determining the capacity of a state over time? What does the principle mean for the obligation to transition away from fossil fuel dependence?
The ICJ also concluded that the duty to prevent significant harm to the environment applies to the climate system and not just instances of direct cross-border harm and the duty to cooperate to protect the environment reflects customary international law (at [134], [271]ff.). The former obligation is one of due diligence with a “multifactorial and evolutive character”: its fulfilment depends in part on state capacity and resources and, as states’ economic and material capacity increases, what is required for due diligence is heightened (at [292]). The duty to cooperate is tied to the duty to prevent significant harm in that a coordinated response may be crucial for fulfilling that obligation (at [141]). Crucially, the duty to cooperate is founded on the recognition of the interdependence of states, requiring, among other things, “efforts by States to continuously develop, maintain and implement a collective climate policy that is based on an equitable distribution of burdens and in accordance with the principle of common but differentiated responsibilities and respective capabilities” (at [306]).
In considering the relationship between obligations arising from the treaties and from customary international law, the Court recalled that treaties may help in identifying and determining the content of customary rules and customary law principles may develop under the influence of a treaty, such that treaty rules also become customary rules (at [312]). According to the Court, “[t]his is even more likely where the relevant customary and treaty rules ‘flow from a common fundamental principle’” (at [312]). Climate change treaties establish standards “that may enable or facilitate the identification and application of the diligence that is due in specific instances” and may “inform the general customary obligations, just as the general customary obligations provide guidance for the interpretation of the climate change treaties” (at [313]). Compliance with treaty obligations in good faith suggests substantial compliance with customary duties, but the obligations remain independent (at [314]). In a passage of particular import to the US as a major non-party emitter, the Court observed that it is possible that a non-party state cooperating with states that are party to the treaties, in a manner equivalent to that of a state party, may be considered to be discharging its customary international law obligations; but absent cooperation in this way, such a state has “the full burden of demonstrating […] conformity” with customary obligations (at [315]).
The Court presents treaties and customary international law as independent, dual sources but also in a partial double helix formation: they cannot easily be separated (notwithstanding their formal independence), but they are interlinked through interpretation and state practice, which also form the basis of customary international law and the material for interpreting (and developing) treaty obligations; practice by a state may also discharge one or more obligation, with some but not perfect overlap. Although much of the Court’s view here is not new, its confirmation both generally and in relation to climate change obligations is significant. First, states cannot claim that they do not have obligations to address climate change simply because they are not party to the climate change treaties. This is important given the track record of high-emitting states not signing up to and accepting jurisdictional clauses of climate change treaties. Second, states are prevented from fashioning treaties in a way that avoids customary international law obligations (contrary to the argument made by some states that the treaties displace customary international law obligations). Custom and treaties do not apply to the exclusion of each other. Third, states are encouraged to cooperate in addressing climate change whether parties to the climate change treaties or not. Finally, and crucially, as states have obligations under customary international law, they may be held responsible for conduct that occurred prior to the treaties coming into force, although it remains unclear at what point states acquired such obligations and whether responsibility can only be from the point at which the customary obligations arise. For domestic climate change litigation, the 2025 Opinion should assist parties in making arguments that national law should provide a remedy for breach of these obligations and/or for judicially reviewing acts of governments. In the UK, the Court of Appeal ruled that it can only assess whether the Government has provided a “tenable” view of what the Paris Agreement requires (R. (on the Application of Friends of the Earth Ltd.) v Secretary of State for International Trade [2023] EWCA Civ 14, [2023] 1 W.L.R. 2011, at [50]). The 2025 Opinion may reduce the scope for what may be considered “tenable”.
The second set of questions put to the ICJ concerned the responsibility of states for breach of their climate change obligations. The Court noted that, as obligations to protect the climate system and other parts of the environment from anthropogenic GHG emissions under customary international law are erga omnes in nature, responsibility for their breach may be invoked by any state. This is the first time that the Court has held specific customary obligations to be erga omnes. In addition, obligations under the UNFCCC and the Paris Agreement are erges omnes partes (at [441]). For the Court, “the most significant” primary obligation is the customary law obligation to prevent significant harm to the climate system, but a state “does not incur responsibility simply because the desired result is not achieved; rather, responsibility is incurred if the State fails to take all measures which were within its power to prevent significant harm” (at [409]). The obligation is one of due diligence that must be assessed in concreto. Unlike in Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro), Judgment, I.C.J. Reports 2007, p. 43, at [431], the Court did not assert in the 2025 Opinion that state responsibility for the violation of due diligence only bites if and when the harm occurs. That is the clearer and more logically coherent position.
Crucially, the Court made clear that, while treaties can provide lex specialis on the legal consequences of breaches of climate change obligations, the text, context and object and purpose of the climate change treaties do not support the claim that they represent a derogation from the general rules of state responsibility. In fact, several states entered declarations when becoming parties to the climate change treaties, “affirming the continued application of the customary international law rules on State responsibility” (at [419]). Contrary to the position taken by some states in their oral submissions, the Court emphasised that the internationally wrongful act in question is not the harmful emissions themselves, but the actions or omissions of a state that cause harm to the climate system in breach of their obligations (at [427], [429]) and attribution to states requires a scientific determination. Some states had also argued that the conduct of private actors cannot be attributed to a state for the purposes of state responsibility, but the Court made clear that what is important for the above-mentioned obligations is the state’s own actions or omissions in failing, for example, to exercise due diligence in enacting necessary regulation, making the question of attribution of private actors beside the point (at [428]). Finally, while it may be difficult to establish a causal link between wrongful actions/omissions and the harm arising from climate change on a transnational scale, it is not impossible (at [438]), although the Court leaves open what standard of legal causation may apply in addition to, and alongside, scientific causation. In terms of the consequences of establishing international responsibility, the Court’s observation that restitution may take the form of “reconstructing damaged or destroyed infrastructure, and restoring ecosystems and biodiversity” is noteworthy (at [451]). In confirming that the law of responsibility applies to climate change obligations the Court puts to rest arguments that attribution, causation etc. are too difficult to apply in this context, such that an augmented regime (over which states must negotiate, with all the attendant difficulties and biases this entails) is required.
It is remarkable that the 2025 Opinion exists largely because of the efforts of small-island developing states and many in the Global South. They not only pushed the General Assembly but their submissions largely won the day. The Court also acknowledged that sea level rises will have particularly adverse consequences for small-island and low-lying coastal states, “potentially leading to the forced displacement of populations within their territory or across borders, as well as affecting the territorial integrity of States and their permanent sovereignty over their natural resources” (at [357]). The Court observed that such matters “are closely connected with the right to self-determination” and that “sea level rise is not without consequences for the exercise of this right” (at [357]). This observation is significant, as is the connection between these principles, and so it is somewhat disappointing that the Court did not elaborate upon the implications for the legal obligations of other states, despite the issue having been raised in pleadings (although see the Separate Opinion of Judge Aurescu).
The 2025 Opinion is hugely significant, both for the fight against climate change and for international law more generally, but even the Court acknowledged that the law occupies only a modest place in the face of “an existential problem of planetary proportions that imperils all forms of life and the very health of our planet” (at [456]).