Non-Recognition and Non-Assistance

The Advisory Opinion of 19 July 2024 on Israel’s occupation of Palestine must have had international law experts advising foreign ministries around the world working extra hours. The International Court of Justice (ICJ) not only made it crystal clear that Israeli occupation is illegal in every respect – by itself a challenge for Western foreign offices as they face reproaches for double standards. The Court also added a number of paragraphs detailing the legal consequences of the Advisory Opinion for UN Member States (paras. 273-279). A common element in Advisory Opinions, this section appears at first sight to contain a rather detailed list of “dos” and “don’ts”. However, on a second reading, the passages harbor a host of unresolved legal questions. They range from deep theory issues implicating the legal basis of third States’ obligations, to more practical ones concerning the limits of non-assistance and non-recognition, particularly with respect to military cooperation, and the possibility of sanctions.

Legal Basis: Erga omnes or ius cogens?

Having found Israel in violation of the right to self-determination, the rules relating to the use of force, human rights law, and international humanitarian law, the ICJ recalls in the initial lines of the section on States’ obligations that these rules have an erga omnes character.

This statement is perplexing, as others have noted before (here and here). To qualify an obligation as having erga omnes character is generally understood to refer to the question of standing: States other than the injured State may file a suit against the State allegedly violating the obligation in question, since the obligation is the concern of “the international community as a whole”, to quote the ICJ’s famous Barcelona Traction judgment of 1970. The genocide cases against Myanmar and Israel, initiated by the Gambia and South Africa respectively, provide recent, much-discussed examples.

By contrast, it is a widespread conviction that obligations for third States derive only from violations of ius cogens. In this sense, Judge Tladi, truly an expert on issues of peremptory international law, argues in his Declaration that the Court – which is on the record for its long-standing reticence in recognizing ius cogens – should have referred to the peremptory character of the rules violated by Israel (para. 28-30). In his view, erga omnes addresses standing, while the peremptory character defines the scope of an obligation. In support of this position, Judge Tladi invokes Art. 41 of the ILC’s Articles on State Responsibility, which obliges States to refrain from recognizing a situation arising from a violation of peremptory international law. Moreover, his declaration echoes Judge Higgins’s Separate Opinion in the 2004 Wall Advisory Opinion (para. 38-9). In Judge Tladi’s view, it is all the more surprising that the Court relied on erga omnes as it recognized the peremptory character of the right to self-determination in the very same Advisory Opinion (para. 233).

However, in support of the judgment, Judge Cleveland in her Separate Opinion points to the Court’s previous case law, which uniformly based third States’ obligations arising from violations of international law on the erga omnes character of the violated rules. Of importance in this regard is the Chagos case (para. 180), where the Court recognized the customary character of the right to self-determination, but refrained from attributing peremptory character to it, basing its findings on third States’ obligations instead on the erga omnes nature of the right to self-determination. In the 2004 Palestine Wall Opinion, the Court even provided a detailed analysis of the erga omnes character of self-determination and international humanitarian law (para. 154). The case law, it seems, is strikingly consistent, although at odds with the view of Judge Tladi and many voices in the literature.

Who is right? Perhaps the question requires no further resolution. In fact, it seems entirely consistent with the rationale behind erga omnes duties to base third States’ obligations on them. Their very purpose is not just to serve as a cause of action, but also to define the scope of an obligation. In Barcelona Traction, the Court brought up the concept of erga omnes obligations for the first time to distinguish obligations incumbent upon the international community as a whole from bilateral obligations. The Court recognized erga omnes obligations as a new category of international obligations because of their substance, their fundamental, quasi-constitutional significance for the international legal order. The implications for standing are only derivative of that character.

The essence of this is that erga omnes and ius cogens are two sides of the same coin, actually two overlapping concepts with only marginal, terminological differences between them. In this sense, it is telling that the ICL 2022 draft conclusion on peremptory norms recognizes that all peremptory norms have erga omnes character (for evidence, see the 2004 Wall Opinion, para. 157), while avoiding the often-heard opposite conclusion that not all erga omnes rules had peremptory character. I doubt one will find an erga omnes rule without peremptory character. In fact, this would hardly make sense. For it is the multilateral, quasi-constitutional character of peremptory rules which makes it impossible for States to derogate them without the consent of the international community as a whole; and it is that very same character which grants standing to third States.

This prompts the question why the ICJ is much more reluctant to recognize the peremptory character of rules than their erga omnes effect. One can only speculate. One possibility is that the Court simply started with erga omnes in 1970 and saw it unnecessary to add to this case law another, overlapping yet controversial category that would contribute little substance. Another possibility is that the Court’s reasoning corroborating the erga omnes character has mostly been deductive, deriving their character from their substance rather than from the consent of member States. Peremptory international law is at least in theory as much based on consent as on the fundamental significance of the rules in question – as the 2022 ILC draft demonstrates in conclusions 1 and 2. Consent, however, is much more difficult to establish in a judgment. The ICJ is notorious for sloppy reasoning concerning practice and opinio iuris corroborating “simple” customary rules. This gets all the more problematic for peremptory international law as the line dividing it from “mere” custom is hard to pin down in practice. Therefore, the Court may have made a wise choice to stick to erga omnes as far as possible and to avoid the trouble of tracking State consent.

The Limits of Non-Recognition and Non-Assistance

As ambiguous as the legal basis of duties of non-recognition and non-assistance is their precise scope. At first sight, the Court seems to carve out third States’ duties in this respect with some level of detail. At closer inspection, however, it remains utterly unclear where to draw the line. On the one hand, it is evident that any act containing a recognition of Israel’s occupation as legal is ruled out. This comprises recognition of occupied Jerusalem as Israel’s capital. Moreover, direct forms of assistance to occupation, such as deliveries of military equipment specifically destined to control the Occupied Palestinian Territory (OPT), or economic cooperation with Israeli actors, private or public ones, on the OPT, would be prohibited. On the other hand, the Court certainly did not have a BDS-style full boycott of Israel on its mind as it emphasizes the need to distinguish between Israel and the OPT. Also, non-recognition and non-assistance must not harm Palestinians (cf. Namibia Advisory Opinion, para. 125).

In between these poles, there is much uncertainty. Many forms of assistance to Israel, such as technological cooperation, may at least remotely benefit its capacity to control the OPT. As a general rule, I believe that good faith, proportionality as a general principle of law, and the duty to cooperate encased in Art. 2(5) of the UN Charter require States to weight risks as they review their cooperation with Israel. The more serious the risk that some form of cooperation will contribute to illegal occupation, and the more direct the relation between the cooperation in question and illegal occupation, the more States are bound to discontinue existing projects and refrain from starting new ones.

The Future of Military Cooperation

An example for a high-risk field would be military cooperation. In this respect, joint ventures in the production of weapon systems should be a matter of utmost concern as it seems impossible to exclude with certainty that such weapons, or the knowledge acquired through cooperation, will be used to uphold illegal occupation. The German weapons industry has a lot of joint ventures with Israeli producers – and the German government would do well to use all legal means available, including arms control and foreign trade control legislation, to disentangle, minimize, or freeze such relationships as long as occupation is ongoing.

Moreover, any arms deliveries require full guarantees that the equipment will not be used to uphold occupation, as well as effective mechanisms to control respect for these guarantees. Particularly problematic in this respect are weapon deliveries destined for Israel’s war effort in Gaza. The Court’s Opinion does not cover the period after 7 October 2023. However, even if one assumes that Israel has been exercising legitimate self-defense after 7 October, such self-defense might slowly morph into a new state of occupation that becomes indistinguishable from occupation prior to 7 October – and might therefore be subject to the same legal challenges upheld by the Court. From this point of view, only punctual forms of assistance might remain legal where the impact on occupation can be minimized. Examples might include forms of cooperation targeted specifically at Hamas personnel and equipment associated with the massacre of 7 October 2023.

By contrast, member States are held to cooperate with the UN General Assembly (UNGA) and the Security Council in carving out a peaceful solution. Part of this solution includes protecting the civilian population. The UNGA has established UNRWA for this purpose. Defunding this organization any further would violate the letter of the Court’s opinion – and the spirit of international law.

Sanctions

The obligations breached by Israel’s illegal occupation are owed to the international community as a whole. For that reason, third States may only claim cessation of the violation and reparations for Palestine as per Art. 48(2) of the Articles on State Responsibility (ASR). This recalls debates on the sanctions imposed by third States on Russia after its invasion into Ukraine, particularly the freezing and subsequent sequestration of Russian central banks assets. As it has been argued in this context, third States enforcing peremptory (or erga omnes) obligations may only resort to legal means, also called retorsions. No reprisals may take place, unless the Security Council specifically authorizes member States to do so.

A different situation emerges where specific treaty provisions may be invoked to justify reprisals including the suspension of treaty privileges. Al Tamimi has made the case for the EU to activate the human rights clauses of the EU-Israel Association Agreement. Moreover, one might rely on Article XXI(c) of the GATT to justify trade measures. While taking such measures to force Israel’s compliance with the ICJ Opinion is not a binding legal obligation under the UN Charter, one might give the security exception of the GATT a wider reading to comprise measures aiming at the maintenance of international peace and security in alignment with UN policies and efforts. With the Security Council having fallen back into hibernation, rendering ICJ decisions effective may be one of the last straws to rely on for those interested in maintaining a universal order of peace.