By recognising COP and CMA decisions as relevant to treaty interpretation, the Court gives institutional decisions a role in shaping the operative content of the climate treaties without formal amendment. This becomes particularly important at paragraph 224, where decisions 1/CMA.3 and 1/CMA.5 are treated as confirming 1.5°C as the parties’ primary temperature goal, even though the Paris Agreement itself refers to keeping warming “well below 2°C” while pursuing efforts towards 1.5°C. Falk’s account of quasi-legislative institutional power is useful here: formally non-binding decisions can acquire normative force through consensus, repetition and judicial recognition. The route through Article 31(3)(a) VCLT is orthodox as stated by Orakhelashvili, but that does not resolve the larger question of how far subsequent agreement can reshape treaty obligations. The concern is sharper for dualist systems such as India, where the legislature consents to treaty obligations on the basis of the text ratified. If that content can later be expanded through conference decisions without renewed legislative approval, the link between international obligation and domestic democratic consent becomes considerably weaker.

The Court’s rejection of lex specialis is convincing at one level, but paragraph 170 raises a real problem about treaty autonomy. The Paris Agreement reflects a carefully negotiated balance, so applying a separate customary duty of prevention at a “stringent” standard risks adding an obligation that States did not expressly bargain for. The Court’s reliance on silence is also uneasy in light of Lotus: it effectively assumes that general international law continues to apply unless States have excluded it, whereas the opposite presumption could also be made. Simma’s criticism in Kosovo points to the same difficulty. At the same time, rejecting lex specialis preserves the customary duty and prevents the Paris Agreement’s “facilitative” and “non-punitive” compliance mechanism from becoming the only meaningful response to breach. Keohane’s emphasis on stable institutional expectations, however, shows why the regime-integrity concern remains important: if external customary rules can supplement a carefully negotiated treaty regime, its boundaries become less predictable.

The Court’s treatment of these principles suggests a distinction between binding rules and interpretive principles: the latter guide the application of law without creating independent obligations. This is defensible, but the Court does not clearly explain the legal status of such principles. Its treatment of equity is particularly unclear: after acknowledging in Continental Shelf (Tunisia/Libya) that equity is a general principle “directly applicable as law,” it nevertheless confines equity to interpretation. The exclusion of polluter pays is also revealing. Although the Court relies on soft-law instruments to support other principles, it rejects polluter pays largely because it lacks specific treaty anchoring. That unstated distinction is significant because polluter pays is the only principle among those considered that directly allocates the costs of environmental harm to its contributors.

The Court’s extension of the customary no-harm principle to climate change is persuasive but leaves an important methodological gap. Corfu Channel and Pulp Mills concerned identifiable, bilateral harm, whereas climate change involves cumulative contributions by virtually all States. Nuclear Weapons establishes that the duty can extend beyond national territory, but that addresses where harm occurs, not whether a State’s conduct can be causally connected to it. The Court therefore moves too quickly from establishing the customary rule to extending its application to a fundamentally different factual context. Barcelona Traction offered a stronger basis for addressing this difficulty through the concept of obligations owed to the international community as a whole. The Court’s reference to climate change as a “quintessentially universal risk” points in this direction, but it does not fully develop the connection.

The Court adopts a functional rather than source-based approach to applicable law, asking what rules are most relevant to the questions before it rather than working through Article 38 source by source. As Higgins suggests, this reflects international law as a process of authoritative decision-making rather than a fixed catalogue of sources. The inclusion of five guiding principles also supports Panezi’s observation that general principles often do more work than the formal taxonomy suggests. The trade-off, however, is predictability: if relevance determines inclusion, the scope of applicable law ultimately depends on judicial assessment. Still, given the integrated nature of climate change, the Court’s approach is arguably preferable to artificially separating treaty, customary and principled obligations.

The Court gives the General Assembly's questions a deliberately broad interpretation. It understands relevant conduct to include all State acts and omissions that adversely affect the climate system through greenhouse-gas emissions, and it rejects any territorial limitation on its inquiry. The Court nevertheless states that it is identifying existing obligations rather than creating new ones. By extending scope to licensing, subsidising and production, the Court relocates the legally relevant conduct from the atmosphere to the domestic regulatory decision. The Assembly asked about emissions; the Court answers about the chain of State decisions that make emissions possible. That is a defensible reading of an ambiguous phrase, but it is a choice, and it determines the reach of every obligation that follows.

The fact-to-norm move in para 137 is unexamined and substantial. Because the risk is "indisputably established" by science, the Court identifies a general standard of conduct in the abstract- two sentences after quoting Bosnia Genocide for the proposition that due diligence "calls for an assessment in concreto" (ICJ Reports 2007 (I), p. 221, para. 430). Scientific certainty about risk is thus the premise licensing a doctrinal generalisation the Court's own precedent resisted. The Court expressly treats the IPCC reports as the "best available science" on the causes, nature and consequences of climate change. The duty to prevent environmental harm cannot be assessed without considering what States knew, or reasonably should have known, about the risks created by their conduct. The precautionary approach reinforces this point: uncertainty cannot be used as a reason for postponing measures where there is a threat of serious or irreversible harm.The difficulty is that the Court does not fully explain the legal status of the scientific material on which it relies. The IPCC is not a law-making institution, and its reports are not themselves sources of international law. They are evidence capable of informing the interpretation and application of existing legal rules. Kelly's argument that custom is increasingly demonstrated identifying the real dange- an obligation of due diligence with no external referent is one whose standard the adjudicator supplies.