Israel’s military campaign against Hamas has just now passed the six-month mark. Its consequences for the civilian population have been devastating. “Business as usual” is no longer an option for third States providing material support to Israel in the form of arms and other military equipment. They must undertake not only a moral and political, but also importantly legal assessment of the implications of that support.

This set of calculations is particularly important following the order on provisional measures of January 26, 2024, in which the International Court of Justice (ICJ) ordered Israel to “take all measures within its power to prevent the commission of all acts within the scope of Article II” of the Convention on the Prevention and Punishment of the Crime of Genocide (Genocide Convention), and its subsequent order of March 28, 2024, to “[e]nsure with immediate effect that its military does not commit acts which constitute a violation of any of the rights of the Palestinians in Gaza as a protected group under the Convention on the Prevention and Punishment of the Crime of Genocide, including by preventing, through any action, the delivery of urgently needed humanitarian assistance.”

The ICJ’s finding that, prima facie, claims advanced by South Africa may fall within the scope of the Genocide Convention (para. 30), and that at least some of the rights alleged by South Africa are plausible (para. 54), must have shaken even the most ardent of Israel’s supporters. Indeed, the legal risks for those providing material support to Israel are significant (see here and here). For example, Nicaragua has recently initiated proceedings against Germany before the ICJ concerning the material support Germany has been providing to Israel in alleged breach of the Genocide Convention and the Geneva Conventions. Lawsuits targeting government decisions to allow the export of arms to Israel have also been brought before domestic courts (e.g. in the Netherlands; the United Kingdom; the United States; Canada; Denmark and, most recently, Germany). The Court of Appeal in The Hague ordered the Netherlands to cease export and transit of F-35 fighter jet parts to Israel given the “clear risk” of violations of International Humanitarian Law (IHL) in Gaza (see here and for discussion see here, here and here). Governments are also making changes to their policies on arms exports to Israel (e.g. Canada recently announced that it would not approve new arms export permits in favor of Israel “until [it] can ensure full compliance with [its] export regime”; Spain did not authorize any sale of military material to Israel since 7 October 2023).

This article examines the legal criteria that States apply when deciding whether to deliver arms to a State that may be committing serious violations of international law. Similar considerations also apply to arms transfers to non-State actors, including organized armed groups.

At the outset, I discuss the recent proceedings before the English courts as an example of the application of domestic criteria for exports of arms. I then analyse the most relevant primary obligations concerning arms transfers, namely under the ATT, as well as secondary obligations in the Articles on State Responsibility (ARSIWA) that broadly reflect customary international law.

On February 19, 2024, the High Court in London in R (Al-Haq) v. Secretary of State for Business and Trade refused an application for permission to apply for judicial review in respect of the U.K. Secretary of State for Business and Trade (SST) decision to continue supplying military and dual-use equipment to Israel. The SST’s decision was based on the Strategic Export Licensing Criteria (SELC) (recently amended), which provide legal guidance for the application of the Export Control Act 2002. The decision was reached with reference to the Arms Trade Treaty (ATT) and other primary obligations (namely under the Genocide Convention and IHL). Specifically, the SST decided on December 18, 2023:

(a) not to suspend extant export licences for military and dual-use equipment being exported to Israel either directly or where Israel is the final-destination, (b) not to stop granting export licences, but rather (c) to keep her decisions about whether or not to grant, revoke, or refuse licences under continuing, careful review, in view of the current hostilities in Gaza (Summary of Grounds of the SST (SoG) [1]).

Because the SELC replaced the criteria originally adopted while the U.K. was member of the European Union (EU), and do not differ significantly from the criteria applied by other States, the relevance of these proceedings extends far beyond the UK. The High Court’s Order is also consistent with the position that the English courts took in recent challenges by the Campaign Against Arms Trade (CAAT) concerning the continued exports of arms to Saudi Arabia and their alleged use in Yemen ([2017] EWHC 1754 (Admin); [2019] EWCA Civ 1020 (CAAT 1); and [2023] EWHC 1343 (Admin) (CAAT 2)).

1. Domestic U.K. SELC criteria for arms transfers

The U.K. SST’s most recent decision not to suspend arms transfers to Israel was essentially based on the application of the following SELC criteria:

Criterion 1(b)

Respect for the UK’s international obligations and relevant commitments…The Government will not grant a licence if to do so would be inconsistent with, inter alia: …

(b) the UK’s obligations under the United Nations Arms Trade Treaty.

Criterion 2(c)

Respect for human rights and fundamental freedoms in the country of final destination as well as respect by that country for international humanitarian law. Having assessed the recipient country’s attitude towards relevant principles established by international human rights instruments, the Government will:

(c) Not grant a licence if it determines there is a clear risk that the items might be used to commit or facilitate a serious violation of international humanitarian law, the Government will also take account of the risk that the items might be used to commit or facilitate gender-based violence or serious acts of violence against women or children.

For our purposes, the most important disagreements between the parties in R (Al-Haq) v. Secretary of State for Business and Trade concerned the nature of the assessment to be conducted by the SST under each of these criteria and the required threshold of knowledge.

Al Haq argued that Criterion 1(b) does not require a predictive assessment and sets a “lower” standard than Criterion 2(c). Criterion 1(b) requires the SST to examine whether the grant of a licence would be inconsistent with the U.K.’s obligations under Art. 6(2) of the ATT, which in turn implicates the obligation to prevent genocide under Art. I of the Genocide Convention (Statement of Facts and Grounds for Judicial Review (SoF) [124]). Relying on the ICJ’s judgment in Bosnian Genocide, Al Haq took the position that the duty to prevent is engaged where the State is merely aware of a “serious danger” that genocide might occur (ICJ, Bosnian Genocide [431-432]), and that “there is no requirement to prove actual knowledge of genocide or that there has been the commission of genocide by an offending State” (SoF [44]). According to Al Haq, “it does not need to be proven that the UK could have prevented the genocide or that its failure to act was causatively linked to the commission of genocide. It is sufficient that it had the means to contribute to this goal and that it manifestly refrained from using them” (SoF [45]).

The U.K. government, for its part, argued that both criteria require a predictive assessment as to the future use of exports (Summary of Grounds of the SST (SoG) [57]). According to the government, Criterion 1(b) sets a “very high” standard, requiring “knowledge that the arms or items would be used in the commission of genocide,” whereas “there [was] no factual basis for any such conclusion” in the present case (SoG [56]). Moreover, “there is no violation of the Genocide Convention (as is required by Art. 6(2) of the ATT) unless genocide (or other acts listed in Art. III of the Genocide Convention) has actually occurred” (SoG [19]). For the government, Criterion 1(b) implicates Art. 6(3) of the ATT and “deals specifically with the risk that exported arms or items might be used in the commission of genocide” (SoG [56]). On this view, the SST’s decision was rightly focused on Criterion 2(c) “in circumstances where the real concern relates to Israel’s use of licensed equipment in armed conflict” (SoG [57]). The SST relied heavily on the judgments mentioned above in CAAT 1 and, particularly, CAAT 2, finding that a “considerable degree of respect … must be accorded to the Secretary of State in this particular context” (SoG [9]).  The following excerpts of the U.K. government’s submission to the court demonstrate how the SST has been applying these precedents in its assessments:

a. … [T]he main factors which govern a proper assessment of the Criterion 2c risk are the willingness and ability … to comply with IHL. … (CAAT 2, Judgment [108]);

b. The test requires a clear risk of a serious violation of IHL. In our view clear does not simply mean something which is not theoretical… “Clear” connotes a concrete risk for which there is evidential support (albeit that it is a nuanced value judgment). It means that the risk must be clear from evidence which forms a meaningful basis for making an assessment of whether a violation might occur in the future. Insofar as that assessment is based on past breaches or possible breaches, that means that there must be sufficient information available to make a meaningful judgment on that question. The mere fact of civilian casualties or that strikes hit a humanitarian or civilian installation does not equate, in this context, to a “possible” breach of IHL if there is insufficient information to form a view on the “why” question. … (CAAT 2, Judgment [124(1)]);

c. … [T]he mere fact of prior breaches does not equate to a risk of future breaches, still less a clear risk of a serious breach (CAAT 2, Judgment [124(4)]) (SoG [7]).

As the question before the High Court was taken in the context of judicial review, the test was whether the SST’s decision was irrational. The Court found that it was not [5]. The SST was entitled to regard her conclusion as to Criterion 2(c) as governing the approach to other criteria [6]. In particular, the Court held that the Government did not err in law when it considered “Israel’s understanding of its obligations as an element in determining whether there is an intention on the part of Israel to comply with its obligations” [9]. This is so notwithstanding the fact that the SST had aired concerns about Israel’s commitment to IHL. To give just one example:

The assessment further noted that, although Israel accepted that it was under an obligation to facilitate (but not to provide) humanitarian assistance in Gaza, the Israeli Response gave no detail of the reasons for restricting the quantity of supplies of food, water and medical supplies. The decision of the Israeli Cabinet of 18 October 2023 had linked the supply of humanitarian assistance to the release of hostages. The assessment noted that the absence of further explanation raised concerns regarding the commitment to comply with the obligation not to arbitrarily deny access to humanitarian assistance and was relevant to an assessment of Israel’s overall commitment to IHL. However, Israel’s position was that it is acting in accordance with what it believes to be the relevant legal obligations in relation to humanitarian assistance and it is therefore possible that this is a case of disagreement about what the law requires, rather than an intentional disregard of IHL (SoG [43]).

The Court also found that there was no error of law as to the ATT and the Genocide Convention, holding that “the difference between the parties’ positions turns not on the interpretation of the legal provisions but on the view as to whether there is a risk of genocide” [10]. Ultimately, the SST’s decision could not be reproached as irrational where it was being kept “under review” [10]. The claimant has since sought to reverse the decision of the High Court. What is more, hundreds of U.K. judges and lawyers, including four former justices of the U.K. Supreme Court, have recently called upon the U.K. government to “suspend the provision of weapons and weapons systems to the Government of Israel.”

2. Primary rules of international law – the ATT

States should carefully consider different primary rules of international law in deciding whether to deliver arms or other military equipment when there are reasons to believe that such supplies may be used in the commission of international crimes or other serious violations of IHL or International Human Rights Law (IHRL). These primary rules may overlap and complement each other. For instance, an arms transfer could amount to a breach of Art. I of the Genocide Convention (duty to prevent genocide), Art. III(e) of the Genocide Convention (complicity in genocide) or Common Art. 1 of the Geneva Conventions; at the same time, it could constitute an independent breach of Art. 6(3) of the ATT. However, the required thresholds and conditions to establish the existence of a violation of each of these obligations may differ.

This article focuses on the ATT because it provides the most detailed and up-to-date legal framework governing arms transfers in international law, while importing other relevant primary rules, including the prohibition of genocide and other international crimes. It also complements and gives specific expression to the overarching obligation of each State to respect and ensure respect for IHL and IHRL (Principles of the ATT).

Art. 6(2) ATT provides that:

A State Party shall not authorize any transfer of conventional arms covered under Article 2 (1) or of items covered under Article 3 or Article 4, if the transfer would violate its relevant international obligations under international agreements to which it is a Party, in particular those relating to the transfer of, or illicit trafficking in, conventional arms.

Art. 6(3) ATT provides that:

A State Party shall not authorize any transfer of conventional arms covered under Article 2 (1) or of items covered under Article 3 or Article 4, if it has knowledge at the time of authorization that the arms or items would be used in the commission of genocide, crimes against humanity, grave breaches of the Geneva Conventions of 1949, attacks directed against civilian objects or civilians protected as such, or other war crimes as defined by international agreements to which it is a Party.

Art. 7(1) ATT provides that:

If the export is not prohibited under Article 6, each exporting State Party, prior to authorization of the export of conventional arms covered under Article 2 (1) or of items covered under Article 3 or Article 4, under its jurisdiction and pursuant to its national control system, shall, in an objective and non-discriminatory manner, taking into account relevant factors, including information provided by the importing State in accordance with Article 8 (1), assess the potential that the conventional arms or items: (…)

(b) could be used to:

(i) commit or facilitate a serious violation of international humanitarian law;

(ii) commit or facilitate a serious violation of international human rights law;

At the outset, what is striking about these provisions is that, at least on their face, they are silent on the required degree of risk of violations of the obligations they reference, the definition of what amounts to a serious violation of IHL, and on the requisite nature or degree of knowledge. The wording reflects, as it often does, a compromise between competing views expressed during their negotiation.

3. The requirement of “knowledge” that the arms “would be used” in the commission of genocide, crimes against humanity or war crimes

Art. 6(3) of the ATT is one such compromise. It resulted from unsuccessful attempts by some States to limit its scope to: (a) situations where the purpose of the transfer was to facilitate commission of genocide, crimes against humanity or certain war crimes; and (b) grave breaches or serious violations of Common Art. 3 of the 1949 Geneva Conventions which would have excluded most violations that occur during the conduct of hostilities (OUP Commentary [6.12]). Other States and international organizations, including the ICRC, regarded such an approach as depriving the prohibition of any practical effect. For example, the ICRC argued that the purpose limitation would render the provision inoperable because it would effectively require an admission by a State that they were transferring arms for the purpose of facilitating war crimes.

As a result of this debate, the knowledge requirement was adopted. However, the ATT does not define “knowledge” nor the applicable standard thereof (OUP Commentary [6.14]). Thus, recourse is often made by analogy to standards elaborated in international criminal law, even though these cannot and should not be simply transposed to regimes that are primarily concerned with the responsibility of States (cf. Jørgensen). The more relevant source would be secondary norms of State responsibility.

As will be shown below, knowledge of the circumstances of the internationally wrongful act is required to arrive at a finding of State responsibility for aid or assistance. Art. 6(3) ATT likewise requires knowledge on the part of the transferring State. However, contrary to the general international law position on the responsibility of assisting State, a State may be found responsible under Art. 6(3) ATT without an underlying war crime or other offence listed therein being committed (OUP Commentary [6.85]). The existence of a risk that such offences may take place and the knowledge thereof at the time when the transferring State authorizes the transfer would suffice for the finding of a breach.

I now turn to the concept of risk and risk assessments in the context of the ATT.

4. The assessment of risk when authorizing arms transfers and exports

The terms “would be used” in Art. 6(3) implies an assessment of risk by the transferring State. In fact, the phrase “substantial risk” figured explicitly in earlier drafts of the provision, but was abandoned in favor of the present more general wording (see Chair’s Draft Paper of 14 July 2011) (OUP Commentary, [6.11]).  Here, it is worth noting that the “clear risk” language found in the U.K.’s statutory SELC criterion 2(c) and discussed in R (Al-Haq) v. Secretary of State for Business and Trade is not derived from the ATT but from the EU Common Position on Arms Exports and the User’s Guide. These instruments have had a “harmonizing effect” on the criteria adopted by other States such as Bosnia and Herzegovina, Canada, Iceland, North Macedonia, Montenegro, and Norway, long before the adoption of the ATT. The User’s Guide provides a reference point on how States should assess whether there is a “clear risk,” which involves:

an inquiry into the recipient’s past and present record of respect for international humanitarian law, the recipient’s intentions as expressed through formal commitments and the recipient’s capacity to ensure that the equipment or technology transferred is used in a manner consistent with international humanitarian law and is not diverted or transferred to other destinations where it might be used for serious violations of this law (User’s Guide [2.13]).

The requirement that a risk assessment be undertaken can also be derived from the surrounding context, namely Art. 7, which governs all other exports of arms that do not trigger the prohibitions of Art. 6. According to Art. 7(1) of the ATT, if a proposed export of arms is not prohibited on the grounds set out in Art. 6, a State, before transferring such arms, shall carry out a risk assessment exercise on terms explicitly set out in Arts. 7(1) and 7(3) that, among others, inquire into the end use of weapons.

By contrast to Art. 6(3), which requires that the arms “would be used” in the commission of the relevant crimes, Art. 7(1)(b) entails a lower threshold, i.e. one only needs to show that the arms could be used to “commit or facilitate” the relevant violations. Accordingly, “[t]his means that the weapons may be one or more steps removed from the actual violation” (OUP Commentary [7.35]). While Art. 7(1)(b) is silent on the specific character of the risk, Art. 7(3) nonetheless suggests that the risk of negative consequences in Art. 7(1) must be “overriding” (“If, after conducting this assessment and considering available mitigating measures, the exporting State Party determines that there is an overriding risk of any of the negative consequences in paragraph 1, the exporting State Party shall not authorize the export”). This implies showing a “sufficient degree of likelihood” that arms might be used to commit or facilitate the commission of serious violations of IHL.

5. Serious violations of IHL

To recall, Art. 7(1) requires an assessment, prior to authorization of a transfer, whether the items being transferred could be used to commit or facilitate a serious violation of IHL. The ICRC has a clear position as to what constitutes serious violations of IHL. Following the adoption of the ATT, the ICRC published Protecting Civilians and Humanitarian Action Through the Arms Treaty, where it states:

“Serious violation of IHL” is another term for “war crime” and encompasses grave breaches of the 1949 Geneva Conventions and Protocol I of 8 June 1977 additional to the Geneva Conventions, as well as the other war crimes listed in the Rome Statute of the International Criminal Court and those defined under customary IHL. Serious violations of IHL include wilful killing, torture or inhuman treatment, taking of hostages, pillage, rape, directing attacks against civilian objects or civilians not taking direct part in hostilities, and directing attacks against hospitals, ambulances, or medical staff using the distinctive emblems of the Geneva Conventions.

This statement aligns with the prevailing view that led to the adoption of the wording “a serious violation of international humanitarian law” in Art. 7(1) of the ATT. During the negotiations of the ATT, the ICRC explained that serious violations of IHL cover grave breaches of the 1949 Geneva Conventions and 1977 Additional Protocol I, war crimes under Art. 8 of the 1998 Rome Statute of the International Criminal Court, and other war crimes under customary IHL (OUP Commentary [7.39]). This raises the question of whether the prohibition in Art. 6(3) of the ATT is narrower than the scope of “a serious violation of IHL” because it only refers to “grave breaches of the Geneva Conventions of 1949, attacks directed against civilian objects or civilians protected as such, or other war crimes as defined by international agreements to which it is a Party.”

Be that as it may, we should not lose sight of the fact that arms transfers may independently breach the obligations of States Parties to the Geneva Conventions, which enjoy universal ratification. Under Common Art. 1 to the Geneva Conventions, States Parties to the Geneva Conventions shall not only “respect” but also “ensure respect” for the Conventions in “all circumstances.” This entails at least a duty not to aid or assist in the commission of IHL violations, including through arms transfers, as well as a duty to ensure that warring parties comply with their IHL obligations at all times (see ICJ, Wall Advisory Opinion [158]-[159]; Commentary to Common Art. 1 to the Geneva Conventions; Hill-Cawthorne; Milanovic; cf. Goodman; but see Schmitt & Watts).

6. The commission of genocide or other acts under Art. III of the Genocide Convention

Finally, as to the question whether the transfer of arms would be used in the commission of genocide or other acts under Art. III of the Genocide Convention such as complicity, the U.K. government was correct in R (Al-Haq) v. Secretary of State for Business and Trade when it asserted that this inquiry falls within the purview of Art. 6(3) of the ATT and not Art. 6(2). However, the U.K. government’s position that “[a]ctual knowledge that genocide is taking place or would take place is required” (SoG [20]) and that genocide “must occur for there to be a violation of the obligation to prevent” (SoG [18]) could have been more nuanced, to say the least.

Specifically, the U.K. government’s formulation is lacking for several reasons. First, a State may have knowledge imputed to it. It is enough if the State, for example, is aware, “or should normally have been aware, of the serious danger that acts of genocide would be committed” (ICJ, Bosnian Genocide [432]). Second, the knowledge requirement may be met by knowledge of “the circumstances surrounding the eventual prospective recipients of the arms” using primarily “intelligence and information about the likely future behaviour of the recipients” (OUP Commentary [6.93]). In this context, States are to exercise due diligence and take all measures reasonably available to them to prevent genocide. The required degree of due diligence that States are expected to exercise when reviewing their existing arms export licenses or granting new ones to Israel is demanding, not least in light of the recent orders of the ICJ. Third, it is not necessary to show that the Art. 6(3) violations would occur only if the transfer has been made. It is sufficient to show that the arms in question would be used to carry out those violations (OUP Commentary [6.93]). Fourth and fundamentally, the U.K. government’s position appears to ignore the rationale of ATT in that the obligations in Arts. 6 and 7 may be breached regardless of whether a genocide or other offences occur. In this regard, the ATT has a broader reach than the framework under general international law in respect of a State facilitating the commission of an internationally wrongful act by another State, where indeed the responsibility is only engaged if the genocide is ultimately committed (ICJ, Bosnian Genocide [431]).

7. Secondary rules of international law – Articles on State Responsibility

In addition to specific primary obligations under the ATT considered above, a State may engage its responsibility in connection with the conduct of another State, if the former provides assistance that facilitates the commission of an internationally wrongful act by the recipient. Art. 16 of the Articles on State Responsibility (ARSIWA) is commonly regarded, including by the ICJ, as reflecting customary international law:

A State which aids or assists another State in the commission of an internationally wrongful act by the latter is internationally responsible for doing so if:

(a) that State does so with knowledge of the circumstances of the internationally wrongful act; and

(b) the act would be internationally wrongful if committed by that State.

Three aspects are worth highlighting for the purposes of this discussion.

First, contrary to the overarching logic of the ATT provisions, State responsibility for aid or assistance under general international law does not arise in connection with the risk of an underlying violation by the other State. Responsibility for aid or assistance is only engaged once the principal wrongful act is committed with the means that have been provided. This is the result of the derivative nature of responsibility for aid or assistance. Further, while the ILC Commentary to Art. 16 suggests that assistance must be “clearly linked” to the wrongful act, and it is “sufficient” if the assistance contributed “significantly” to the act, at the same time the Commentary indicates that “the assistance may have been only an incidental factor in the commission of the primary act, and may have contributed only to a minor degree, if at all, to the injury suffered.” As I argued elsewhere, it would be inconsistent with the gist of the ILC’s codification work on Art. 16 of ARSIWA and the relevant State practice to qualify the relationship that must exist between the assistance and the principal act by reference to the relevant tests of causation as these apply in determining the content of State responsibility, including reparation (Lanovoy). Yet, that is the argument that Germany appears to have advanced in the recent proceedings brought by Nicaragua and, which was in my view one of its weakest points on law in an otherwise persuasive pleading (CR 2024/16, at 35, [9]-[10] (Peters)).

Second, to mitigate the potentially expansive scope of responsibility for aid or assistance, the ILC included a requirement of dual illegality in Art. 16(b) of ARSIWA, i.e. a finding of responsibility can only be made if the principal act would have been wrongful if it had been committed by the aiding or assisting State itself. That said, it is not necessary that the obligation breached be derived from the same source for the two States (Crawford). In practice, this requirement may be problematic because the scope of primary rules may differ as between conventional regimes or between conventional regimes and the current state of customary international law. For example, if the aiding or assisting State A is a party to the ATT and all the other major arms control treaties, but the recipient State B is not, or is a party to only some of those treaties, it may not necessarily be possible to arrive at a finding of responsibility on the part of both States.

Third, knowledge of the circumstances of the internationally wrongful act is the test for the responsibility for aid or assistance to be engaged. This test is controversial because the ILC’s commentary to Art. 16 of ARSIWA suggests that knowledge of the putative accomplice should be construed as intention, i.e. that “the relevant State organ intended, by the aid or assistance given, to facilitate the occurrence of the wrongful conduct and the internationally wrongful conduct is actually committed by the aided or assisted State” (ILC Commentary to Art 16, [5]). State practice and opinio juris, including the comments made during the drafting of Art. 16 of ARSIWA and subsequent debates in the context of a similar provision in the draft articles of responsibility of international organizations, are not conclusive (Lanovoy). While some States have taken the view, including the United Kingdom and the United States and, most recently, Germany, that both knowledge and intention are required (CR 2024/16, at 36 [13] (Peters)), others argued in favour of a lenient approach to the knowledge test, namely that the constructive knowledge could meet the required test for the responsibility to be engaged under Art. 16 (e.g., The Netherlands). Scholars have taken diametrically different approaches at times. Some argue that knowledge implies intent, relying too much on the Bosnian Genocide judgment, which drew rather incautiously in my view a direct analogy between Art. III(e) of the Genocide Convention and Art. 16 of ARSIWA [419] (Aust; Crawford).

In my view, the actual knowledge of the circumstances of the wrongful act is a sufficient test by default for the engagement of responsibility. However, evidence of intention on the part of the aiding or assisting State when authorizing the delivery of aid or assistance could indeed affect the determination of the scope and content of responsibility of that State at a subsequent stage of analysis (Lanovoy). Other scholars have suggested that the knowledge required by the test is akin to wilful blindness, i.e. “somewhere in between actual or near-certain knowledge and constructive knowledge” (Moynihan; Jackson).

Be that as it may, where scholarly writings appear to roughly concur is that the exact nature and degree of knowledge in the circumstances may vary depending on the particulars of the primary norm breached. Most breaches occur without any need to prove intent on the part of the principal or the accomplice, some may require the ascertainment of standards such as that of due diligence, and a smaller set may require the showing of intent as a precondition for establishing a breach. It is only in those cases where the finding of a breach of an obligation by the principal wrongdoing State involves a specific intent, as in the case of genocide, that the aiding or assisting State must be at least aware of that intent (ICJ, Bosnian Genocide [421]).

Finally, in circumstances where there is a serious breach of peremptory norms, all States are under an obligation not to render assistance to the maintenance of the situation flowing from that breach. Art. 41(2) of ARSIWA provides: “No State shall recognize as lawful a situation created by a serious breach within the meaning of article 40, nor render aid or assistance in maintaining that situation.” According to Art. 40(2) of ARSIWA, “[a] breach of such an obligation [i.e. under a peremptory norm of general international law] is serious if it involves a gross or systematic failure by the responsible State to fulfil the obligation.” By contrast to Art. 16, the requirement of knowledge is immaterial in this context since all States are presumed to have the knowledge of the serious breach of peremptory norms by the principal wrongdoing State (ILC Commentary to Art. 41, [11]). Moreover, the obligation not to render aid or assistance under Art. 41(2) of ARSIWA applies regardless of whether the existence of a serious breach of peremptory norm has been ascertained by an international court or tribunal, the UN Security Council or General Assembly. Where the evidence of a serious breach of peremptory norms is compelling and reliable, originating from multiple sources, third States that continue delivering arms and other equipment to the wrongdoing State do so at their own risk of engaging responsibility under international law.

8. Conclusion

This article has provided an overview of overlapping legal regimes that governments need to consider when assessing whether to authorize transfers or exports of arms and other military items to recipients suspected of committing serious violations of IHL or other international crimes. The article first zoomed in on an example of the application of the U.K. domestic criteria justifying continued arms trade between the U.K. and Israel. The article then turned to the assessment of relevant primary and secondary norms of international law.

Over time, the far-reaching character of some of the primary norms in this area, namely under the ATT, may lead States to more carefully consider decisions to authorize arms transfers and exports to third States suspected of breaching international law. On the flipside, important differences in legal tests and standards between different rules and in their application domestically (e.g. the “clear risk” requirement) may give too much leeway to States to avoid responsibility and to continue business as usual. That said, key aspects of the legal framework are clear. As the analysis above showed, elements of the U.K. and German governments’ positions in the proceedings discussed above fall outside of the settled understanding of the law. It remains to be seen whether international and domestic litigation in this area will result in stronger or weaker controls on arms transfers.

IMAGE: Israeli army vehicles move in an area along the border with the Gaza Strip and southern Israel on April 4, 2024, amid the ongoing conflict in the Palestinian territory between Israel and the militant group Hamas. (Photo by JACK GUEZ/AFP via Getty Images)