Provisional Measures before International Courts and Tribunals: Preventing Irreparable Harm Pending Final Judgment

Provisional Measures before International Courts and Tribunals: Preventing Irreparable Harm Pending Final Judgment

I. Introduction

In 2024, the International Court of Justice issued three separate provisional measures orders—on January 26, March 28, and May 24—in the case of South Africa v. Israel . This sequence drew intense global attention because it is exceptionally rare for the Court to issue three distinct emergency orders within a single ongoing dispute.

Provisional measures are temporary orders that protect a party’s rights while a case is still before the Court, stepping in specifically to prevent harm that a final judgment could not later undo.

The Court definitively settled the legal status of these orders in the 2001 LaGrand case , ruling that provisional measures carry mandatory compliance obligations. However, this established bindingness has historically raised persistent questions about actual compliance on the ground by state parties.

The divergence between mandatory legal text and actual state implementation exposes a critical enforcement gap inherent to international judicial oversight, one that becomes clear once LaGrand, Bosnia, The Gambia, and South Africa v. Israel are read together.

II. From Recommendation to Binding Obligation

Article 41 of the ICJ Statute grants the Court the explicit power to indicate provisional measures whenever circumstances require the preservation of rights pending a final judgment. The text entirely lacks any explicit declaration regarding the legal status or mandatory nature of these measures, leaving a structural void regarding their enforceability. Because the drafters chose the permissive term “indicate” rather than an imperative command like “order,” sovereign states routinely interpreted these measures as mere diplomatic recommendations that they could legally disregard. The Court ended decades of scholarly and diplomatic disagreement in its 2001 LaGrand decision by identifying a binding obligation inherent to the treaty’s object and purpose.

The LaGrand case arose after two German nationals, Karl and Walter LaGrand, were sentenced to death in the United States without being informed of their right to consular assistance under the Vienna Convention on Consular Relations. Germany approached the ICJ seeking provisional measures to prevent Walter LaGrand’s execution while the dispute was pending. On 3 March 1999, the Court ordered the United States to take all measures necessary to ensure that the execution did not take place before the Court gave its final decision. Despite this order, Walter LaGrand was executed. In its judgment of 27 June 2001, the ICJ held that provisional measures indicated under Article 41 create legally binding obligations for states. This judgment resolved the long-standing uncertainty surrounding Article 41 and transformed provisional measures from procedural recommendations into binding legal commitments. At the same time, the events of the case also revealed that recognising a legal obligation does not automatically ensure compliance, an issue that remains significant in international adjudication.

The preventive role of provisional measures became clearer in Bosnia and Herzegovina v. Serbia and Montenegro . In 1993, the ICJ issued provisional measures directing the parties to prevent acts capable of causing genocide-related harm while the proceedings continued. When the Court delivered its final judgment in 2007, it held that Serbia had breached its obligation to prevent genocide at Srebrenica. This case extended the doctrine beyond stopping a single, discrete act—here, the Court used provisional measures to address an ongoing, large-scale risk. That preventive logic carried directly into how the Court later approached The Gambia v. Myanmar.

III. The Modern Test: The Gambia v. Myanmar

The Gambia v. Myanmar shows this approach at its most developed. In November 2019, The Gambia brought proceedings against Myanmar under the Genocide Convention, alleging genocide against the Rohingya population, and asked the Court for provisional measures to protect Rohingya rights while the case proceeded. On 23 January 2020, the ICJ ordered Myanmar to take steps to prevent acts prohibited under the Genocide Convention and to preserve evidence relevant to the case . What makes this case matter for the doctrine is not the outcome alone, but how clearly the Court laid out the test it was applying.

Before indicating provisional measures, the Court applies a well-defined legal test. First, it must be satisfied that it has prima facie jurisdiction, meaning there must be a reasonable basis to conclude that the dispute falls within its jurisdiction at this preliminary stage, without deciding the issue conclusively. Second, the rights claimed by the applicant must be plausible under the relevant treaty or rule of international law. In this case, the Court considered whether the rights asserted under the Genocide Convention were capable of protection through provisional measures. Third, there must be a real and imminent risk of irreparable prejudice to those rights, making the indication of provisional measures urgent. This requirement ensures that provisional measures are reserved for situations where delay could cause serious and irreversible harm. These three elements had already taken shape in earlier cases, including Bosnia and Herzegovina v. Serbia and Montenegro, but Gambia v. Myanmar is where the Court applied them together as a single, structured test.

The decision shows that provisional measures are granted only after the Court applies a careful and rigorous legal standard. That rules out one explanation for what comes next: the compliance problems examined in the next section are not a product of unclear law. The same structured test was applied by the ICJ in South Africa v. Israel in 2024, which is where the question of compliance becomes most pressing.

IV. Binding Yet Unenforced

The ICJ has now made clear, twice over, that provisional measures bind states and that it will not indicate them lightly. Yet compliance with these orders still varies once they are issued. Part of the reason is structural: international law has no central authority that can force a state to comply with a judicial order, the way a domestic court can call on police or bailiffs. Compliance instead depends on whether states are willing to meet their obligations, and on the diplomatic and political pressure applied around them.

The proceedings in South Africa v. Israel illustrate this challenge. On 26 January 2024, the ICJ indicated provisional measures after finding that the legal conditions for such relief had been satisfied. As the situation developed, the Court issued a second order on 28 March 2024, followed by a third order on 24 May 2024, each addressing requests for additional or modified provisional measures. Without drawing conclusions about the conduct or intentions of either party, the sequence of these three orders is significant. It shows that provisional measures are not always self-executing, and that the Court may be required to revisit the same dispute as circumstances evolve. Three orders in one case, rather than one, point to a gap between what the Court directs and what happens on the ground.

This gap traces back to how the international legal system is built. The ICJ has no independent enforcement arm comparable to a domestic court’s; it cannot fine a state or send anyone to compel compliance. Article 94(2) of the UN Charter lets a party bring non-compliance to the Security Council, but that route runs straight into politics—a permanent member’s veto can block action, including action against itself or an ally. The legal authority behind provisional measures is therefore only part of the picture; whether states actually follow through depends just as much on political will. The real challenge, then, sits less in the law itself than in what exists to enforce it.

V. Conclusion

The ICJ’s jurisprudence on provisional measures shows real legal maturity. Article 41, as LaGrand interpreted it, makes clear that these orders bind states, and later decisions—especially The Gambia v. Myanmar—set out a clear, rigorous test for granting them. What this maturity hasn’t produced is a matching level of practical effectiveness. The problem is not that the law is unclear; it’s that no reliable mechanism exists to make states comply with what the Court has ordered.

Closing this gap calls for institutional fixes, not doctrinal ones. A structured state-reporting requirement, triggered once provisional measures are indicated, would let the Court and the wider international community track compliance rather than assume it. Where the Security Council route is blocked by politics, wider use of General Assembly resolutions could add a layer of scrutiny and pressure that doesn’t depend on a single veto.

Provisional measures will only matter in practice if courts’ authority to declare the law is eventually matched by the international community’s willingness to enforce it. Until that changes, orders that are binding on paper will keep falling short of being binding in fact.

References

1. Statute of the International Court of Justice art. 41, June 26, 1945, 59 Stat. 1055, 33 U.N.T.S. 993.

2. LaGrand (Ger. v. U.S.), Judgment, 2001 I.C.J. Rep. 466 (June 27). https://www.icj-cij.org/case/104

3. Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosn. & Herz. v. Serb. & Montenegro), Judgment, 2007 I.C.J. Rep. 43 (Feb. 26). https://www.icj-cij.org/case/91

4. Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Gam. v. Myan.), Provisional Measures, Order of Jan. 23, 2020. https://www.icj-cij.org/case/178

5. Application of the Convention on the Prevention and Punishment of the Crime of Genocide in the Gaza Strip (S. Afr. v. Isr.), Provisional Measures, Orders of Jan. 26, 2024, Mar. 28, 2024 & May 24, 2024. https://www.icj-cij.org/case/192/provisional-measures

6. U.N. Charter art. 94.