Skip to content
International Court of Justice, advisory opinion.

Legality of the Threat or Use of Nuclear Weapons (1996)

Citation: ICJ Reports 1996, p. 226. **Requested by:** the General Assembly. **Instruments:** Statute of the Court, Article 38; General Assembly resolution 1653 (XVI) of 24 November 1961; Declaration of St Petersburg 1868; Geneva Protocol 1925.. Covered in Unit 1 · Nature and Sources of Public International Law.

Say the ratio out loud before you open Reasoning — recalling it unprompted is exactly what the exam pays for.

Why it matters

For the sources paper this is the modern statement of what General Assembly resolutions contribute to the formation of custom. The nuclear question is the vehicle; the doctrine is the cargo. The opinion also shows opinio juris being tested against a divided international community and failing.

The question

The General Assembly asked whether the threat or use of nuclear weapons is in any circumstance permitted under international law.

The custom argument, from both sides

States asserting illegality pointed to a consistent practice of non-utilisation since 1945, and read that practice as expressing opinio juris on the part of the possessors. States asserting legality in certain circumstances invoked the doctrine and practice of deterrence: they had always reserved the right to use such weapons in self-defence against an armed attack threatening their vital security interests, and if the weapons had not been used since 1945 that was not because of an existing or nascent custom but because circumstances justifying use had fortunately not arisen.

Held on custom

No customary rule prohibiting the use of nuclear weapons as such could be found.

Reasoning

Where custom is to be looked for. The Court restated its own formula: the substance of customary law must be "looked for primarily in the actual practice and opinio juris of States". That sentence is worth memorising because it puts practice first and states the two elements as a pair.

The deterrence problem. The Court declined to pronounce on the practice of deterrence. It noted that a number of States adhered to it for the greater part of the Cold War and continue to do so, and that the members of the international community are profoundly divided on whether non-recourse over fifty years expresses opinio juris. In those circumstances "the Court does not consider itself able to find that there is such an opinio juris."

That is a model of how to reason about the subjective element. Where the international community is split on the meaning of an abstention, the abstention cannot be read as law.

Resolutions of the General Assembly. Here is the passage that matters for the sources question. "The Court notes that General Assembly resolutions, even if they are not binding, may sometimes have normative value. They can, in certain circumstances, provide evidence important for establishing the existence of a rule or the emergence of an opinio juris."

And the method for testing a resolution: "To establish whether this is true of a given General Assembly resolution, it is necessary to look at its content and the conditions of its adoption; it is also necessary to see whether an opinio juris exists as to its normative character. Or a series of resolutions may show the gradua1 evolution of the opinio juris required for the establishment of a new rule." (read gradual)

So the test has three elements: content, conditions of adoption, and opinio juris as to normative character; with a fourth possibility, that a series of resolutions shows gradual evolution.

Applying the test. Examined in their totality the resolutions declared that use would be a direct violation of the Charter, and in certain formulations that such use should be prohibited. But several had been adopted with substantial numbers of negative votes and abstentions. They were a clear sign of deep concern, yet "they still fall short of establishing the existence of an opinio juvis on the illegality of the use of such weapons." (read juris)

The argument from resolution 1653 (XVI). The first resolution expressly proclaiming illegality, of 24 November 1961, had referred to earlier declarations and binding agreements from the Declaration of St Petersburg of 1868 to the Geneva Protocol of 1925, and then proceeded to qualify the legal nature of nuclear weapons, determine their effects, and apply general rules of customary law to them. The Court drew a sharp inference from that structure: the very fact that the Assembly had to perform an exercise of legal qualification shows that in its view there was no specific customary rule prohibiting use, for if such a rule had existed the Assembly could simply have referred to it.

The state of the law. The annual adoption by large majorities of resolutions recalling resolution 1653 (XVI), and requesting a convention prohibiting use, reveals the desire of a very large section of the international community to move towards complete nuclear disarmament. But the emergence, as lex lata, of a customary rule specifically prohibiting use "is hampered by the continuing tensions between the nascent opinio juris on the one hand, and the still strong adherence to the practice of deterrence on the other."

Answering points

  • The core proposition: resolutions of the General Assembly are not binding, but may have normative value and may evidence an existing rule or an emerging opinio juris.
  • Give the three-part test, then show the Court applying it and rejecting the resolutions on the conditions of adoption limb because of negative votes and abstentions.
  • The argument from resolution 1653 is the subtlest point in the opinion and distinguishes a good answer. A body that must argue a rule into existence thereby shows the rule did not exist.
  • Distinguish lex lata from lex ferenda: the desire of a large section of the community is not yet law.
  • Contrast with Nicaragua, where resolution 2625 (XXV) did evidence opinio juris on the non-use of force, because there the attitude of States towards the declaration was one of acceptance of the rule's validity. The difference is in the voting record and in the absence of a competing practice.

Related cases in this unit

Parts of the judgment

Precedents cited