Appealing the High Court’s Judgment in the Public Law Challenge against UK Arms Export Licenses to Saudi Arabia

Written by

In May of this year, the Court of Appeal granted leave to appeal the 2017 High Court ruling in Campaign Against the Arms Trade v Secretary of State for International Trade.

The case and the High Court’s 2017 judgment have already received some commentary (see here). Simply put, the case concerns a public law challenge against the government’s continued approval of licenses for arms exports to Saudi Arabia, on the grounds, inter alia, that the Secretary of State’s conclusion that there is not a ‘clear risk that the items might be used in the commission of a serious violation of international humanitarian law’, in the context of the Saudi-led coalition in Yemen, was irrational (criterion 2(c) of the Consolidated Criteria set out in European Council Common Position 2008/944/CFSP and adopted by the Secretary of State as the policy to be followed in granting or refusing export licenses). The High Court found against the Campaign Against the Arms Trade, concluding that the determinations made by the Secretary of State permitting continued export of arms to Saudi Arabia were rational.

The present post focuses on what is submitted is an error of law made both by the government in its determinations as to whether to grant or refuse export licenses and by the High Court in its judgment. Specifically, both the government and the High Court appear to have mistakenly taken the view that a certain subjective mens rea threshold necessarily applies before one can say that there has been a serious violation of international humanitarian law. This issue has received some commentary in relation to the High Court hearings and judgment (see here and here). It is the purpose of the present post to set out clearly why no such subjective threshold (whether intention, recklessness, negligence or otherwise) applies at least as a general matter. 

The Government’s Position and the High Court Judgment

The government’s submissions indicate a clear preoccupation with the question of whether or not particular acts alleged to be inconsistent with humanitarian law were intentional. For example, the regular updates on international humanitarian law in Yemen that are produced by the Middle East and North Africa Directorate in the Foreign and Commonwealth Office and that were sent to the Foreign Secretary contain frequent references to evidential insufficiency regarding a purported general subjective element in the determination of compliance with humanitarian law: ‘[t]he update notes that intent is a key element in assessing International Humanitarian Law compliance, and acknowledges that there is often insufficient information to determine intent’ (para 153 of High Court judgment). Indeed, in its skeleton argument (at para 38), the government argued that ‘serious violation’ was identical to ‘war crime’ (and thus incorporated the latter’s mens rea requirements).

Importantly, this arguably appears to have had an impact on the High Court’s approach to the law, as its judgment too suggests a preoccupation with a purported subjective element in its assessment of the evidence. This is notwithstanding the Court’s clarification early in the judgment that ‘serious violations of international humanitarian law’ and ‘war crimes’ are not identical concepts; the Court nonetheless seemed to take the view that a subjective standard (even if not intention) was still one component of the former concept: ‘[i]n our view, the generic term “serious breach” would include reckless as well as deliberate or intentional acts’ (para 18). Indeed, the Court still proceeded in some instances entirely to conflate the relevant international humanitarian law standards (which, as will be shown, do not generally contain a subjective threshold) and international criminal law standards (which do contain a subjective threshold, as they set out the conditions for individual criminal responsibility as opposed to State responsibility) (see, e.g., para 24 drawing on the international criminal law standard for proportionality). The impact of such an approach on the Court’s interpretation and application of the law is clear in the following statement that wrongly attributes a subjective test to the international humanitarian law rule on distinction:

The fact that civilian casualties have occurred does not without more mean that a breach of International Humanitarian Law has taken place, still less a serious breach. Customary international law and International Humanitarian Law have long recognised that civilian casualties in military conflicts will occur. The ‘Principle of Distinction’ prohibits intentional attacks against civilians …  (para 208) (emphasis added)

This reflects a clear error of law: international humanitarian law prohibits attacks against civilians, whether intentional, reckless, negligent or simply mistaken (as shown below). The potential importance of this error for the Court’s conclusions is clear when it states in its concluding paragraphs that the ‘Secretary of State was rationally entitled to conclude as follows: (i) the Coalition were not deliberately targeting civilians …’ (para 210).

The idea of a subjective threshold clearly played an important role in the government’s and High Court’s reasoning. It is clear that there is no explicit reference to a subjective threshold in criterion 2(c) itself (quoted above). Any such threshold would therefore need to come from one of three sources, each of which is implicated in criterion 2(c): general international law (specifically the law of State responsibility), international humanitarian law, or the notion of ‘serious violations’. It will be shown that under none does such a threshold exist as a general matter.

General International Law

In many areas, international law is not concerned with questions of intention, recklessness, negligence or other ‘subjective’ standards of responsibility. There are exceptions (see, e.g. Article II of the 1948 Convention Against Genocide, on which see here), though it is instructive that no general rule was set out in this regard in the International Law Commission’s 2001 Articles on State Responsibility (ASR), with the choice instead being left to specific (primary) rules of international law. Indeed, as stated in the commentary to Article 2 ASR, ‘[i]n the absence of any specific requirement of a mental element in terms of the primary obligation [e.g. the rules of international humanitarian law], it is only the act of a State that matters, independently of any intention’ (para 10, Commentary to Article 2 ASR). The general position under international law has been set out clearly by James Crawford:

Once more, it should be stressed, State responsibility is predicated on a principle of “objective” liability, in the sense that once the breach of an obligation owed under a primary rule of international law is established, this is prima facie sufficient to engage the secondary consequences of responsibility. Unless otherwise provided, no delinquency, culpability or mens rea need be proved … And this conclusion is desirable as a matter of policy, since the “intention” underlying State conduct is a notoriously difficult idea, quite apart from questions of proof (Crawford, State Responsibility: The General Part (CUP, 2013), p 61–2).

This illustrates the non-exceptional nature of ‘objective’ responsibility in international law, that is, the ubiquity of international legal obligations the breach of which does not depend on any subjective element such as intention or recklessness. Indeed, against this background, one might presume the absence of any such element without a clear indication in the particular primary rule at issue. In the case of criterion 2(c), any such subjective element would have to be manifest in either the particular rules of international humanitarian law or in the notion of ‘serious violations’ thereof.

International Humanitarian Law

The quotes above indicate that the government and High Court took the view that the relevant rules of international humanitarian law themselves contain a subjective element. This, however, does not appear to be the case at least for certain of the most important relevant rules. There are a number of rules of humanitarian law that are in issue in this particular context, including the principles of distinction (between lawful and unlawful targets) (Articles 48, 51(2) and 52(2) of Additional Protocol I to the Geneva Conventions, Article 13(2) of Additional Protocol II) and proportionality (between expected civilian casualties and anticipated military advantage) (Articles 51(5)(b) and 57 of Additional Protocol I). Notwithstanding the limited application of the so-called Hague rules to non-international armed conflicts as a matter of treaty-based humanitarian law, it is uncontroversial that they apply as a matter of custom (see 2005 ICRC Customary IHL Study).

Rather than examine all of the relevant rules, let us consider briefly the principle of distinction, given that this was one of those rules to which the High Court attributed a certain mens rea requirement (see above). With respect to the principle’s application to individuals, Article 13(2) of Additional Protocol II states ‘[t]he civilian population as such, as well as individual civilians, shall not be the object of attack.’ The ICRC Customary Law Study expresses the principle in the following terms: ‘The parties to the conflict must at all times distinguish between civilians and combatants. Attacks may only be directed against combatants. Attacks must not be directed against civilians.’ The rule is expressed in objective terms, with no reference to the intention or state of mind of the particular actor. Nor is a subjective element present in any of the definitions of the terms used in these sources (e.g. the definitions of ‘civilian’ in Article 50(1) of Additional Protocol I and Rule 5 of the ICRC Study are objective—and the controversy over the definition of ‘civilian’ in non-international armed conflicts does not concern this point).

There is also no indication in the travaux of the treaty rules on distinction that a subjective element was intended. On the contrary, the context of these provisions confirms their objective content. For example, Article 85(3) of Additional Protocol I includes as a grave breach (i.e. those violations that entail individual, in addition to State, responsibility) ‘making the civilian population or individual civilians the object of attack’, only ‘when committed wilfully’. The specific inclusion in the criminal law standard of a mens rea requirement as a supplement to the basic principle of distinction (the humanitarian law standard) confirms the absence of such a requirement in that basic principle itself.

There has been some practice suggesting an intention requirement within the principle of distinction. Most explicitly, Israel has suggested, with reference to the Additional Protocol I standard, that ‘a commander‘s intent is critical in reviewing the principle of distinction during armed conflict’ (Ministry of Foreign Affairs, The Operation in Gaza, 27 December 2008 – 18 January 2009: Factual and Legal Aspects, para 110). As demonstrated above, however, this is not accurate: whilst intent is fundamental in the case of individual criminal responsibility, the humanitarian law standard, a violation of which engages the State’s responsibility, does not contain any such subjective threshold. It is also worth emphasising that such limited practice could not approach the threshold required in order to effect a modification in the law through subsequent practice.

Other relevant rules of international humanitarian law do potentially incorporate a certain subjective element. The principle of proportionality, for example, prohibits attacks that ‘may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated’ (Article 51(5)(b) of Additional Protocol I). References to ‘may be expected’ and ‘military advantage anticipated’ indicate the relevance of the information available at the time of the attack. However, even this does not necessarily set out a truly subjective test. The Trial Chamber of the International Criminal Tribunal for the former Yugosavia, for example, in Prosecutor v Galić (at para 58) viewed this as setting a reasonable person test.

‘Serious Violation’

In the absence of any generally-applicable subjective standard either in the law of State responsibility or international humanitarian law (where, as noted, at least certain rules, such as the principle of distinction, are purely objective), such a general standard could only otherwise come from the reference in criterion 2(c) to ‘serious violations’ of humanitarian law. This position was also argued by the UK government in its skeleton argument: ‘the term “serious violation” has a particular meaning as a matter of IHL and is synonymous with “war crimes” and “grave breaches”’ (para 38). Furthermore, as noted, the High Court took the view that the notion of ‘serious violation’ incorporated a certain subjective test (para 18 of the High Court judgment).

This particular point has been explored well in a previous blog post, and I will not repeat the points made there. It suffices to repeat here that the notion of ‘serious violations’, though associated with, is not identical to the notion of ‘war crimes’. As the International Criminal Tribunal for the former Yugoslavia noted in its Tadić jurisdiction decision, ‘serious violation’ is defined as ‘a breach of a rule protecting important values, and the breach must involve grave consequences for the victim’ (para 94). This was one amongst other (e.g. mens rea) conditions for individual criminal responsibility to follow from a violation of international humanitarian law; the categories of ‘serious violation’ and ‘war crime’ are thus not coterminous. Similarly, the notion of ‘serious violations’ of humanitarian law in Article 7 of the Arms Trade Treaty, on which criterion 2(c) is based, does not appear to have been intended to refer exclusively to war crimes (and thus incorporating a subjective element) (see Clapham, Casey-Maslen, Giacca and Parker, The Arms Trade Treaty: A Commentary (OUP 2016), paras 7.39 to 7.49).

Concluding Remarks

Both the government and the High Court appear to have wrongly taken the view that the reference in criterion 2(c) to ‘serious violations of international humanitarian law’ includes a general subjective standard, whether one of intention, recklessness or otherwise. At times, this appeared to lead to a conflation of international humanitarian law and international criminal law. This is not an accurate reflection of the law. Given the absence of a general subjective standard in the law of State responsibility, and the entirely ordinary omission of a fault-based standard for responsibility in international law, any subjective standard would have to be expressed clearly in either the relevant primary rules of international humanitarian law or in the concept of ‘serious violations’ thereof. It is clear that there is no general subjective element in the relevant rules of humanitarian law, and certainly in the case of the principle of distinction the test is entirely objective. Moreover, it seems clear that the concept of ‘serious violations’ was not intended as incorporating any mens rea requirement.

References in the High Court judgment and the government’s submissions either to a general subjective threshold before one can determine that there have been ‘serious violations of international humanitarian law’ or to a specific requirement of intent or recklessness in finding a breach of the principle of distinction are thus errors of law. Indeed, this is one of the grounds on which the Campaign Against the Arms Trade is appealing (see especially Claimant’s Reply to Respondents’ Updated Statement (28 March 2018) at paras 12–15), and it is one of the grounds on which the Court of Appeal granted leave (see Court of Appeal ruling of 4 May 2018 at para 13). Of course, the test in criterion 2(c) is forward-looking, requiring the government to assess the risk that arms would be used in the commission of serious violations of humanitarian law; determining whether there is sufficient evidence to conclude that past violations have occurred is thus not exhaustive of the test. It is, however, an important consideration in applying that test, and it is therefore essential that the content of the relevant rules be set out clearly.   

Print Friendly, PDF & Email

Leave a Comment

Comments for this post are closed

Comments

Adil Haque says

November 30, 2018

Hi Lawrence,

On your view, if an attack is directed at C, and C turns out to be civilian, then the attack violates IHL, correct? And this is true even if all feasible precautions are taken prior to attack, and all reasonably available information indicates that C is not civilian but military, yes?

I'm inclined to agree but, if that's right, then is it ever lawful to export weapons to another state engaged in armed conflict? Isn't there always a "clear risk" that a weapons system "might be used" in such an attack, over the course of the conflict, even if the receiving state meticulously follows its precautionary obligations?

Many thanks,

Adil

Dapo Akande says

December 1, 2018

Hi Lawrence,

Many thanks for this post! I have a question about what it means to say that the principle of distinction is only violated when the state makes a civilian "the object of the attack" or only when the attack is "directed" at the civilian. Do those words/concepts embody a mental (or subjective) element? Or can one address those issues simply by looking at objective elements.

Imagine two scenarios.

In Scenario 1, a state fires a weapon at a group of 100 people. It is known by all relevant state agents involved that 99 people in that group are civilians but 1 is a military commander. All relevant state agents aim to kill the military commander. It is also known by all relevant actors that the weapon used will kill all 100 people.

In Scenario 2, a state fires a weapon at a group of 100 people. It is known by all relevant state agents involved that all 100 people in that group are civilians. The state does not have the objective of affecting anything or anyone other than those 100 people. It is also known by all relevant actors that the weapon used will kill all 100 people.

In Scenario 2, I think it would be fairly clear that the civilians are the object of the attack and thus there is a violation of the principle of distinction.

My question is this, has there been a violation of the principle of distinction in scenario 1? I confine my question to the principle of distinction (the state may or may not have violated the principle of proportionality, it might be depend who the 1 military commander is and what they are planning).

Has the state in Scenario 1 directed its attack at the 99 civilians? Or to put it differently, are those 99 civilians in Scenario 1 the object of its attack? If not, why not? And if the 99 civilians are not the object of the attack in scenario 1 (or if the attack is not directed at them) what differentiates Scenario 1 from Scenario 2? Is it something about the state of mind of the relevant actors?

Lawrence Hill-Cawthorne says

December 3, 2018

Hi Adil,

Thanks for your question – it’s a good one as it helps to draw out the controversy over the precise standard in the criterion. First, on my view, an attack that is directed at a civilian is a breach of the principle of distinction, even if the State agent was mistakenly of the view that the target was a combatant and even if all feasible precautions were taken in identifying the individual as a combatant. The fact that a civilian was targeted on the basis of being mistakenly identified as a lawful target will likely be relevant in terms of compliance with certain rules of IHL (such as the rules on taking all feasible precautions) but it will not be relevant for the principle of distinction itself.

Second, though it is true that there will always be a ‘risk’ of a serious violation of IHL, I think criterion 2(c) introduces two additional thresholds before which it is engaged. First, the reference to ‘clear’ risk, as opposed simply to ‘risk’, must indicate that a higher threshold was intended; any other interpretation would render the word ‘clear’ ineffective. Of course, beyond saying that the word ‘clear’ raises the threshold, little more can be said in the abstract, including how high the threshold is raised. It seems reasonable to say that past violations by the transferee State would be relevant in considering whether the ‘clear’ risk threshold is met. It might also be that a long-term armed conflict in which the transferee State is presently involved might be relevant. Second, the focus of criterion 2(c) is on the particular weapons being transferred, which might further raise the threshold of its application in particular instances.

Best wishes
Lawrence

Lawrence Hill-Cawthorne says

December 3, 2018

Hi Dapo,

Thanks for the questions – again, I think these draw out the ‘hard’ cases that might otherwise be understood by incorporating a subjective element in the IHL rules. I do, however, still think that no such element exists at least in the principle of distinction itself. These hard cases show the situations in which a subjective element would add clarity/a principled way of categorising different actions as lawful or unlawful. But they cannot alone generate a subjective element as a matter of law, nor do they explain away the problems with including a subjective element.

First of all, though the references in the various permutations of the principle of distinction to making civilians the ‘object’ of attack or ‘directing’ an attack against civilians might be read as incorporating a certain subjective standard, I think the context of these provisions suggests otherwise. For example, as noted, Article 85(3)(a) of Additional Protocol I includes amongst the grave breaches provisions making civilians ‘the object of attack’ ‘when committed wilfully’. This seems to indicate that making a civilian the ‘object’ of attack does not equate to ‘intentionally targeting a civilian’. Similarly, the Rome Statute includes as a war crime ‘[i]ntentionally directing attacks against the civilian population’ (Arts 8(2)(b)(i) and 8(2)(e)(i)), again suggesting that ‘directing’ an attack against a civilian does not equate to ‘intentionally targeting a civilian’.

On the scenarios, these highlight exactly the circumstances in which a subjective element in the principle of distinction would help us to differentiate between the two different situations. But I don’t think a subjective element is the only way of doing so. In my view, we can assess scenario 1 without recourse to a subjective test, though the conclusion one draws over its legality depends on whether we disentangle the group into military and civilian elements. We could say that the presence of a military objective within the group targeted is the relevant factual consideration, such that the attack is directed against a military target. The presence of the military target which is hit could then be considered sufficient for saying that the attack is not directed against the civilians. This could lead to some peculiar legal results in certain cases, however. My own view is that, to the extent that the attack is simply directed at the group of 100 persons as a whole, scenario 1 would qualify as a violation of the prohibition of distinction and likely falls within the category of indiscriminate attacks if the weapon used could not be targeted at the specific military objective, that is, if the weapon was such as to treat all 100 individuals as a single target. It’s difficult to see how on principle one could differentiate scenario 1 from other situations recognised as violating the prohibition on indiscriminate attacks even when a military objective is present.
Best wishes

Lawrence

David T says

December 4, 2018

I don’t think characterizing this as a debate between objective and subjective standards is helpful. The standard is absolute but it is an obligation of means rather than ends – referring to the process of an operator distinguishing between different categories of persons/objects and directing the munition accordingly. If the process of distinction was somehow flawed or the missile subsequently malfunctioned such that it landed unexpectedly on civilians that would not necessarily render it a breach of the relevant IHL obligation. As discussed, you would look instead to whether the appropriate feasible precautions were carried out.

David T says

December 4, 2018

Does it really follow from Article 85(3)(a) of Additional Protocol that merely making a civilian the ‘object’ of attack does not equate to ‘intentionally targeting a civilian'? Could making a civilian the object of attack be made out - and distinguished from grave breaches - if feasible precautions were not taken to justify the flawed distinction/direction of the operator? Or flawed assessments that individuals were civilians? Just because grave breaches require willfulness does not mean less grave breaches a judged on strict liability principles.

Lawrence Hill-Cawthorne says

December 11, 2018

Dear David,

Many thanks for your comments and apologies for my delay in replying.

I think we perhaps disagree on the nature of the principle of distinction. I do not think that it is an obligation of conduct as opposed to result. I agree that certain of the rules on taking precautions are about conduct, but I view the principle of distinction as entirely separate therefrom as a matter of law (of course, in practice, one is more likely to comply with the principle of distinction if the correct precautions have been taken). I therefore do not take the view that compliance with the rules on feasible precautions is a defence to a breach of the principle of distinction, which I understand to be your view (though do correct me if I have misunderstood).

I take your second point that '[j]ust because grave breaches require willfulness does not mean less grave breaches are judged on strict liability principles'. I agree entirely. But we would then need to show some source of law indicating any such lower subjective standard, and my argument is that one cannot find any such source.

Best wishes,
Lawrence