13 Oct Robson on Common Article 1’s Duty to ‘Ensure Respect’ for the Geneva Conventions: A Critique
In her post on Ensuring Respect for the Geneva Conventions: A More Common Approach to Article 1, Verity Robson discusses the ICRC’s recent commentaries to Common Article 1 (CA1) of the Third Geneva Convention (GC3) and argues, despite largely verbatim recitations of the previous two commentaries on the GCs by the ICRC, that this august body had finally removed some perceived doubt regarding ‘whether states have a legal duty – or simply a power – to prevent and end breaches [of the GCs] by others’ under CA1’s duty to ‘ensure respect’ for the GCs in ‘all circumstances’. Robson concludes that this new ‘clarification’ by the ICRC implies that states have no such duty, which she supports by a brief examination of CA1 and the apparent application of the law of treaties.
The ICRC’s Use of Elegant Variation
As mentioned above, Robson observed that the ICRC noted for the first time in its commentaries to CA1 of GC3 that ‘there is disagreement’ among states as to whether they have a legal duty to prevent and end breaches by other members of the GCs in support of her proposition that there was no conventional obligation of prevention in CA1 of GC1-3. While the commentaries to Article 1 of GC3 do say, ‘There is disagreement as to the legal nature of the positive component of the duty to ensure respect by others because the content of the obligation is not clearly defined and its concretization to a large extent left to the High Contracting Parties,’ the commentaries to Article 1 in GC1–2 both say, ‘Some have expressed doubts as to the legal nature of the positive component of the duty to ensure respect by others because the content of the obligation is not clearly defined and its concretization to a large extent left to the High Contracting Parties.’
While positive obligations will be discussed in greater detail below, the change from ‘Some have expressed doubts’ in the commentaries to Article 1 of GC1-2 and ‘There is disagreement’ in the commentary to Article 1 of GC3 represents not a paradigm shift in the ICRC’s interpretation of the plain and ordinary meaning of the duty to ‘ensure respect’ for the GCs, as Robson contends, but merely an unfortunate example of ‘elegant variation’ in legal drafting on the part of the ICRC. ‘Elegant variation’ refers to the use of synonymous words or phrases in writing in order to avoid repetition and thus tedium for the reader. While generally disfavored in legal writing because it can, in some circumstances, create ambiguity of important terms in legal texts, such is not the case here. The plain and ordinary meaning of the two brief phrases – ‘Some have expressed doubts’ and ‘There is disagreement’ – are the same and thus equivalent in meaning, in that they both acknowledge that a few scholars and one dissenting judge in the Wall opinion disagree that the duty to ‘ensure respect’ for the GCs imposes positive obligations on states. (para. 169, n. 93(I); para. 191, n. 93(II); para. 202, n. 95(III)). While it should be acknowledged that Robson’s name is added to the short list of minority voices in footnote 95 in the commentary to CA1 of GC3, nothing else about the provision is new: every other word within this sentence is the same across all three commentaries. Moreover, all three commentaries to CA1 of GC1-3 on the positive obligations associated with the duty to ‘ensure respect’ of the GCs (which will be discussed in greater detail below) are the same: they all endorse the due diligence standard the scope of which was promulgated by the ICJ in the 2007 Bosnian Genocide case.
States’ Negative and Positive Obligations under CA1 of the GCs
Regardless of whether Robson devoted sufficient attention to the aforementioned three words in the commentaries, she clearly did not address the most crucial aspects of the ICRC’s commentaries: namely, the negative and positive obligations incumbent upon states to ‘ensure respect’ of the GCs with respect to CA1.
The duty to ensure respect for the GCs is comprised of both negative and positive obligations. According to the ICRC, the High Contracting Parties (HCP) have negative obligations to ‘neither encourage, nor aid or assist in violations of the Conventions.’ (para. 158(I); para. 180(II); para. 191(III)). This includes aid and assistance through the provision of arms, training, logistics, intelligence, material, and military support. (paras. 162-163(I); paras. 184-185(II); paras. 195-196(III)). As the commentaries to CA1 of GC1-3 rightly point out, ‘It would be contradictory if [CA1] obliged the [HCP] to “respect and to ensure respect” by their own armed forces while allowing them to contribute to violations by other Parties to a conflict.’ (para. 158(I); para. 180(II); para. 191(III)). Thus, contra Robson, CA1 is not only replete with obligations, but also relatively unconstrained in terms of territorially limitations (although arguably not as unconstrained as the duty to prevent genocide). As the ICRC cited by way of example in the Nicaragua crisis, the ICJ recognized the US’s negative obligation ‘not to encourage persons or groups engaged in the conflict in Nicaragua to act in violation of the provisions of Article 3 common to the four 1949 Geneva Conventions’. (para. 158(I); para. 180(II); para. 191(III)). (For a more detailed discussion on this topic, go here). However, despite the ICJ’s holding in the Nicaragua case, Robson does not mention states’ negative obligations to ‘ensure respect’ for the GCs in her post. Instead, she conducts a ‘treaty interpretation’ of CA1 in light of the aforementioned three words in the commentaries to GC3, capped off with an analogy about erga omnes obligations. (Those obligations will be discussed in greater detail below.)
The ICRC describes the positive obligation to prevent violations of the GCs under CA1 in a manner that will be familiar to anyone conversant with the 2007 Bosnian Genocide case, and the ICJ’s discussion of the primary obligation to prevent genocide and the secondary obligations of due diligence, the latter of which in turn incorporates the three elements of the ‘capacity to effectively influence’ degree of control, the ‘knew or should have known’ standard of knowledge, and the triggering obligation of ‘conduct’, ‘means’, or ‘best efforts’. The ICRC’s commentaries discuss in depth the same due diligence standard from paras. 430-431 and 438 of the Bosnian Genocide case, as highlighted by inter alia the following paragraphs:
The [HCP] also have positive obligations under common Article 1, which means they must take proactive steps to bring violations of the Conventions to an end and to bring an erring Party to a conflict back to an attitude of respect for the Conventions, in particular by using their influence on that Party. This obligation is not limited to stopping ongoing violations but includes an obligation to prevent violations when there is a foreseeable risk that they will be committed and to prevent further violations in case they have already occurred. [para. 164(I); para. 186(II); para. 197(III)]
* * *
States remain in principle free to choose between different possible measures, as long as those adopted are considered adequate to ensure respect. The duty to ensure respect is to be carried out with due diligence. As noted above, its content depends on the specific circumstances, including the gravity of the breach, the means reasonably available to the State, and the degree of influence it exercises over those responsible for the breach. Unlike the negative obligation described above, it is an obligation of means, i.e. the [HCP] are not responsible for a possible failure of their efforts as long as they have done everything reasonably in their power to bring the violations to an end. [para. 165(I); para. 187(II); para. 198(III)]
In other words, the ICRC’s ‘due diligence standard’ – a legal term of art not mentioned in Robson’s post – demonstrates that states have not only a duty to prevent serious violations of IHL (most likely grave breaches of the GCs), but also a duty to bring those violations to an end (within their respective means and capacities to influence the relevant violators, provided they have the requisite knowledge of the aforesaid onerous behavior). Moreover, states must do ‘everything reasonably in their power’ (see here and here) – or what James Crawford calls their ‘best efforts’ (Crawford, at 140) – to ensure the ultimate respect for the GCs by terminating the offensive behavior. As mentioned repeatedly above, this is contrary to Robson’s interpretation of CA1 to the GCs in light of the ICRC’s complete commentaries to GC3, a problem that is further compounded by her comments on erga omnes obligations and state responsibility.
Robson’s Seeming Conflation of Erga Omnes Obligations and Jus Cogens Norms
While I will not repeat fully Robson’s interpretation of erga omnes obligations, I will note some basic tenets about obligatio erga omnes that are relevant to this analysis. Erga omnes are obligations owed to the international community as a whole. (Barcelona Traction, p. 32, at para. 33). As Crawford notes, erga omnes obligations provide inter alia legal standing for a third state to assert the rights of an injured party against a wrongdoing state before an international court of law. (Crawford, at 242-245). If Robson is simply implying with her analysis of erga omnes that such obligations fail to provide states with a right to unilateral humanitarian intervention, I agree. However, if that is Robson’s intention with this section, it is a straw-man argument. While the duty to ‘ensure respect’ of the GCs is unquestionably erga omnes in character (Wall opinion, p. 136, at paras. 146, 158), it cannot provide for a right – or more importantly, a duty – of humanitarian intervention unless the duty is also a jus cogens norm, capable of overcoming the prohibition of the use of force, an issue Robson does not address in her post.
A jus cogens norm is peremptory norm of general international law accepted and recognized by the international community of states as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character. (VCLT, at art. 53). As I’ve addressed at length elsewhere (here, here, here, here, and here), in order for a norm to be identified as jus cogens, it must satisfy the following elements. First, a very large majority of states must accept and recognize the norm in question. This acceptance and recognition is evidenced by the existence of a norm of customary international law or the conclusion of a multilateral treaty of a fundamentally norm-creative character. Second, the candidate customary norms or treaty provisions must be comprised of a special subject matter, one that protects the overriding interests and fundamental values of the international community as a whole. (See here.) Third, the non-derogability of the norm must equate to the non-bilateralizability of the norm; meaning, no two states may contract their way out of the norm unless they do so with a norm having the same peremptory character. Finally, and most importantly, ‘non-deroragability’ refers to the fact that the norm in question cannot be subordinated to another norm if it is to achieve peremptory status. Here, in light of the aforementioned test, the duty to ‘ensure respect’ for the GCs does not rise to the level of jus cogens.
It is true that 196 states, or a ‘very large majority of states,’ have ratified the Four GCs. These Conventions, like the Genocide Convention, consist unquestionably of a special subject matter that protects the overriding interests and fundamental values of the international community as a whole. As a result, the first two steps of the jus cogens test are undoubtedly met. However, the third step is not. According to the ICRC Commentaries, ‘the [HCP] undertake to act, jointly or individually, in co-operation with the [UN] and in conformity with the [UN] Charter’ when they are discharging their duty to prevent serious violations of the GCs. Thus, unlike the duty to prevent genocide, the duty to prevent violations of the GCs is subordinated to another norm: the UN Charter. As such, the duty to prevent serious violations of IHL, such as grave breaches of the GCs, fails the third part of the jus cogens test and is therefore not a jus cogens norm. Nevertheless, it must be emphasized that the standard negative and positive obligations under CA1 noted above still apply, meaning that even though jus cogens obligations are inapplicable, there still exist regular conventional and customary obligations that are binding upon states, which are subject only to the exception of ‘decisions’ imposed by the UN Security Council pursuant to Chapter VII of the Charter.
Robson concludes her post with a few observations on state responsibility that are accurate in themselves. Because she failed to address the full ICRC commentaries on CA1 to the GCs, though, she missed a very important discussion on and distinction between the primary obligations under the GCs and the secondary obligations under ARSIWA, a distinction that is crucial when it comes to the issue of intent. While she does not mention express provisions of ARSIWA in her post, given her reference to ‘aiding and assisting’ it is safe to assume that she is referring to Article 16, which provides as follows:
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.
In his commentaries to Article 16 of ARSIWA, Crawford observes that a ‘[s]tate is not responsible for aid or assistance under article 16 unless the relevant [s]tate 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 [s]tate’. (Crawford, at 149). The ICRC notes that the ‘subjective element of “intent” is unnecessary, however, for the purposes of [CA1]’. (para. 159(I); para. 181(II); para. 192(III)) (emphasis added). Instead, in light of its primary character, ‘[CA1] does not tolerate that a [s]tate would knowingly contribute to violations of the Conventions by a Party to a conflict, whatever its intentions may be’. (para. 159(I); para. 181(II); para. 192(III)) (emphasis added). Thus, the prohibition on aid and assistance under CA1 to the GCs is determined without consideration of the difficult notion of intent under Article 16 of ARSIWA. This is in accordance with the distinction between primary and secondary norms that undergirds the international order, as emphasized by the ICRC (see para. 160(I); para. 182(II); para. 193(III)). Robson does not discuss this critical distinction between primary and secondary norms vis-à-vis intent and her analysis and ultimate conclusion suffer as a result. Instead, she seems content to say, ‘as a matter of policy, of course, a state may choose to exert its political influence in order to prevent or bring to an end breaches by others.’ Such a statement by a legal representative of a veto-wielding member of the UN Security Council is disheartening, to say the least.