Common Article 1 Does Prohibit Complicity in IHL Violations, Through Arms Transfers or Otherwise

Written by

Last week, over on Articles of War, I read with great interest a post on Common Article 1 (CA1) of the Geneva Conventions by my good friends Mike Schmitt and Sean Watts. Their post, building on their prior work, argues that CA1 should not be understood as having any external dimension. It comments in that regard on the case brought by Nicaragua against Germany before the International Court of Justice, in which the Court is now deliberating on the request for provisional measures.

Anything that Mike and Sean write is always going to be thoughtful and considered. But I must confess that I was somewhat perplexed after reading their post, on two counts.

First, in their post Mike and Sean claim that their argument is strictly confined to matters of law and not policy. But they nonetheless express regret in the post that Germany did not expressly argue in favor of their own position that CA1 has no external dimension, which they regard as being correct both historically and today. This to me very much seems to be a matter of policy. Frankly, my sense is that Germany knows better than external observers what legal position works for its own interests, in the context of this litigation or otherwise.

Second, Mike and Sean reject any external dimension of a CA1. In doing so, they seem to conflate two distinct situations – that of a state actively assisting violations of international humanitarian law (IHL) by a third party, and that of a state simply failing to exercise its influence, to the best of its ability, on a third party to induce it to comply with IHL.

But these two scenarios are very different, morally and legally. The first scenario is one of action, of active complicity; the second is one of omission, of a failure to act. The two should not be conflated. Both in ethics and in law there is a substantial distinction between a situation in which someone (say) gives a gun to another, which the latter then uses to kill an innocent, and a scenario in which a bystander passively observes a murder without attempting to stop it. Moreover, Mike and Sean do accept that some complicity-type rules do apply to arms transfers, but they don’t really explain why an IHL-specific complicity rule is not necessary and why CA1 is not the best fit for it (it is both needed and exists, as I will explain).

On the first point, Mike and Sean argue that:

Although correct in conceding the unsettled scope of Common Article 1 obligations, we regard Germany as mistaken in accepting the premise that Common Article 1 imposed any such [external] obligation. Doing so thrust the issue into the realm of facts. It would have been on firmer ground to assert there is no such Common Article 1 obligation in the first place.

Again, I find the proposition that Germany is somehow ‘mistaken’ in the approach it has espoused to be a bit confusing. At the outset it needs to be acknowledged that, while the scope of CA1 is indeed unsettled and controversial, arguing that CA1 has an external dimension is hardly some marginal view outside the mainstream. On the contrary, for Robin Geiss, for example (on whose work Mike and Sean otherwise rely), ‘today, it is clear that Common Article 1 also has an external compliance dimension’ (para. 19).

Further, while I have no idea what the views on the scope of CA1 are within the different departments of the German government, it is reasonably clear that the view that CA1 has an external dimension has become increasingly dominant in Europe, and in many other corners of the world. Thus, a document endorsed by the EU Council says, quite matter of factly, that (at 55):

Common Article 1 of the Geneva Conventions is generally interpreted as conferring a responsibility on third party states not involved in an armed conflict to not encourage a party to an armed conflict to violate international humanitarian law, nor to take action that would assist in such violations, and to take appropriate steps to cause such violations to cease. They have a particular responsibility to intervene with states or armed groups over which they might have some influence. Arms producing and exporting states can be considered particularly influential in “ensuring respect” for international humanitarian law due to their ability to provide or withhold the means by which certain serious violations are carried out. They should therefore exercise particular caution to ensure that their export is not used to commit serious violations of international humanitarian law.

The obligations set out in the Common Position 2008/944/CFSP are consistent with those set out in Article 7, paragraph 3, of the Arms Trade Treaty.

Similarly, to give one more example, in 2013 the UN Human Rights Council adopted a resolution (24/35) which in its op. para. 3 urged ‘all States to refrain from transferring arms to those involved in armed conflicts when said States assess, in accordance with their applicable national procedures and international obligations and standards, that such arms are sufficiently likely to be used to commit or facilitate serious violations or abuses of international human rights law or international humanitarian law.’  42 states – including Germany – voted in favor of this resolution. One state – the United States – was the sole negative vote, with four other states (Kuwait, Mauritania, Qatar, United Arab Emirates) abstaining.

My point is simply this – one would think Germany is perfectly entitled to choose a legal position which is at the very least reasonable, even if Mike and Sean think it to be incorrect, which aligns with the views of most of it allies, and which Germany may decide to espouse in various other situations. Germany probably knows best what’s good for Germany, if I can put it that way. And Germany almost certainly knows what’s best for it in the context of this particular ICJ litigation. It was the ICJ itself that twenty years ago affirmed the existence of an external dimension of CA1 (Wall Advisory Opinion, para. 158). Why would Germany now seek to overrule that holding – which is almost inevitably going to be reaffirmed and perhaps even strengthened in the currently pending Palestine advisory opinion – when it thinks that it has a stronger case in ‘the realm of facts’? Bearing in mind the caliber of the counsel representing Germany in proceedings before the Court, under the supervision of its undoubtedly well-advised Foreign Ministry, I really don’t see how they would have been on ‘firmer ground’ to adopt the strategy that Mike and Sean argue for. Arguing a case on the facts sometimes really is optimal – we shall see soon whether that was so here, when the Court delivers its order on provisional measures.

Which brings me to my second point. While I understand entirely why Mike and Sean argue that CA1 has no external dimension, it still seems to me that this argument doesn’t really engage with the distinction between situations of complicity and those of culpable omission. And this distinction is crucial. The degree of wrongdoing is much higher in the former scenario than in the latter. Further, all the former scenario requires for compliance is abstention, not assisting a party committing violations of IHL; it is much less onerous an obligation than a positive duty.

 As the 2016 ICRC Commentary puts it:

158  Pursuant to common Article 1, the High Contracting Parties have certain negative obligations, which means they must abstain from certain conduct. In particular, they may neither encourage, nor aid or assist in violations of the Conventions.[77] It would be contradictory if common Article 1 obliged the High Contracting Parties to ‘respect and to ensure respect’ by their own armed forces while allowing them to contribute to violations by other Parties to a conflict. Accordingly, the International Court of Justice recognized in 1986 the 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’.[78] This obligation ‘not to encourage’ has also been expressly acknowledged by the High Contracting Parties themselves.[79]

159  In addition, under general international law States are responsible for knowingly aiding or assisting another State in the commission of an internationally wrongful act.[80]According to the ILC, this requires that ‘the relevant State organ intended, by the aid or assistance given, to facilitate the occurrence of the wrongful conduct’ (emphasis added).[81] The subjective element of ‘intent’ is unnecessary, however, for the purposes of common Article 1. In line with the rationale laid out in the preceding paragraph, common Article 1 does not tolerate that a State would knowingly contribute to violations of the Conventions by a Party to a conflict, whatever its intentions may be.

Only later, at paras. 164 et seq., does the ICRC discuss the positive duty to exert influence.

Again, there is a world of difference between a scenario in which a state sells weapons to another state while knowing that these weapons might be used to commit violations of IHL, and a scenario in which a state simply sits idly by while IHL is being violated. In the first scenario the assisting state is an accomplice, in the second merely a passive bystander.

In their post, Mike and Sean do accept the ICJ’s Nicaragua holding that there is a prohibition on encouraging other actors to commit violations of IHL, and they also endorse the aiding and assisting rule in Article 16 of the ILC Articles on State Responsibility:

Thus, although not basing its conclusion on Common Article 1, the International Court of Justice concluded in its Paramilitary Activities (Nicaragua) judgment that the United States was “under an 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. 46). Indeed, as noted above, parties to a treaty are required to act in “good faith” (VCLT, art. 26). This duty encompasses the doctrine of “abuse of rights,” by which parties “shall abstain from acts calculated to frustrate the object and purpose and thus impede the proper execution of the treaty” (Villiger, p. 367, emphasis added). And it is equally clear that a State that aids or assists another State in violating international law is responsible for doing so if it knew of the attendant circumstances and the aid or assistance was “given with a view to facilitating the commission of that act [and it did so]” (Articles on State Responsibility, art. 16; ILC Commentary at 66, emphasis added).

A couple of points needs to be made here. The ICJ in Nicaragua did not expressly base its encouragement holding on CA1. But it also most certainly didn’t base it on a duty to act in good faith, nor on any doctrine of ‘abuse of rights’ – much flimsier foundations doctrinally, it needs to be said, than a negative duty implied in CA1 and its customary equivalent. Moreover, international law does not actually prohibit states from encouraging other states to commit violations of international law. Instigation is not generally seen as a form of prohibited complicity in the wrongs of other states (see the extensive discussion of this in Jackson, 2015). Whatever the source of the encouragement point in Nicaragua was, it’s likely in a rule that is specific to IHL, rather than the application of some kind of general principle.

Further, it makes no sense to prohibit encouragement, but not prohibit more material forms of aid and assistance. If, in other words, encouragement of a serious violation of IHL is unlawful, then so it the provision of weapons that facilitate such violations (cf. the EU position quoted above).

Finally, relying on Article 16 of the ILC Articles is all well and good, but it can’t be the end of the story. There has been much scholarship in recent years on complicity in international law. If there’s one thing that all of these authors likely agree on, it’s that the quality of the ILC’s product when it comes to Article 16 is not what it could have been. This is particularly the case with the lack of terminological precision in the commentary (e.g. the ‘with a view to facilitating’ language) and the contradiction between the text of the provision and the commentary when it comes to the mental (fault) element of the rule.

In my work on complicity and intelligence sharing, I have extensively examined the fault element of the Article 16 complicity rule, and the fault element of other, domain-specific complicity rules (see here and here). To my mind, there is little doubt that an IHL-specific complicity rule exists – whether under the umbrella of CA1 or otherwise. Such a rule is necessary for two reasons, which both distinguish it from the one in Article 16 of the ILC Articles.

First, it applies to state assistance given to non-state armed groups (by its own terms, Article 16 does not). Second, the fault element of the rules is different. At least as far as complicity in serious violation of IHL is concerned, complicity under CA1 does not require an intention to facilitate a wrong (however understood). Rather it can be based on conscious risk-taking akin to recklessness – the state providing assistance to its partner while consciously disregarding a risk that the partner would commit a serious violation of IHL would be liable for complicity if the partner does, in fact, commit such an act and the aid provided (e.g. weapons) does facilitate it. If the risk ultimately materializes – the facilitated harm occurs – the assisting state will become complicit in it.

Bottom line: if (say) Iran provides money, missiles, other weapons or technology to Hezbollah, and in doing so knows that there is a substantial risk that Hezbollah would use the aid provided to direct attacks Israeli civilians, which Hezbollah then proceeds to do, then Iran would be complicit in Hezbollah’s violations of IHL and incur state responsibility under the negative duty in CA1. Or, if Iran or North Korea provided artillery shells or drones to Russia, knowing that there was a substantial risk that Russia would use this aid to direct attacks against civilians or civilian objects in Ukraine, which Russia then did, then Iran and North Korea would be complicit in Russia’s violations under CA1.

Similarly, if the United States, or Germany, or the United Kingdom, sold bombs or other weapons to Israel while knowing that there was a substantial risk that Israel would use those weapons to direct attacks against Palestinian civilians in Gaza, and this then happened, the assisting states would be liable for CA1 complicity. In all of these cases the responsibility of the complicit state would depend on whether the assisted state actually committed a wrongful act. The responsibility is not inchoate, but hinges on the wrong occurring and on the aid provided having causally contributed to it. In none of these cases, however, would the assisting state need to act with the purpose of facilitating the wrongful act, or even be virtually certain that the act would be facilitated – subjective risk-taking suffices.

Whether any of this has actually happened on the facts in the examples mentioned above, and what level of risk was subjectively appreciated by the assisting states in these examples, is not something I am giving any views on here. The point of principle, I hope, is the one that comes across. Also, to be clear, I remain entirely agnostic on the positive obligation to exert influence under CA1. The arguments made by both sides of this controversy, including those of the ICRC and most certainly those of Mike and Sean, are eminently reasonable. I simply don’t have a firm view on this controversy. What I do have a firm view on, however, is that the disagreements about the positive duty should not obscure the relevance of the negative obligation. In fact, from what I can observe, the practical importance of situations in which states are providing assistance to third parties even in the face of a risk that they might be committing violations of IHL greatly outweighs those at which a positive duty to exert influence might be at stake.

Print Friendly, PDF & Email

Leave a Comment

Your comment will be revised by the site if needed.


Adil Haque says

April 15, 2024

Hi Marko,

Great post. Just adding that Germany has voted in favor of multiple General Assembly resolutions affirming the positive obligation, as well as the Declaration of 17 December 2014 adopted by the Conference of High Contracting Parties to the Fourth Geneva Convention which invoked the positive obligation.

Mike and Sean don't place much weight on UNGA resolutions, which I think explains both their position and why their argument would lose badly at the ICJ.



Lawrence Hill-Cawthorne says

April 15, 2024

Thanks for the great post, Marko. I suppose the reason for conflating the purported negative and positive obligations under common article 1 is because Mike and Sean view it not as speaking to a State's conduct vis-a-vis other States/non-State actors (the external dimension) but rather their own population (the internal dimension).

I agree with you that there are very different considerations in relation to both. In fact, an article of mine on common Article 1 was published in the ICLQ late last year. There, I make two key arguments in opposition to Mike and Sean: first, the drafting history of common article 1 does not establish a restrictive meaning, and the drafting history of API (which is entirely neglected in other studies on common article 1) provides some significant support for the wider, external dimension; second, there is very widespread State practice over the last several decades in support of the external dimension. Importantly, in contrast to the claims of Mike and Sean in their ILS piece, that subsequent practice is 'subsequent practice' in the art 31(3)(b) VCLT sense (and thus an authentic means of interpreting common article 1).

Once we accept an external dimension to common article 1, which, as the practice shows, in particular requires States to respond to IHL violations by others, many of the negative obligations follow a fortiori (in other words, if one has an obligation to respond to and bring to an end violations by others, they must themselves have obligations not to aid/assist such violations by others).

Best wishes,

Asher Rottenberg says

April 16, 2024

Thank you for the interesting debate. Could you please explain why do you emphasize that even under CA1 obligation, a violation of IHL must be actually committed by the third state, in order for the violation of CA1 to be materialized? Why isn't the mere positive assistance with the likelihood of future IHL violation sufficient to deem this assistance inconsistent with the obligation to respect and ensure respect. This might also differ CA1 violation from article 16 violation. Thank you

Marko Milanovic says

April 16, 2024

Many thanks for the comments.

Lawrence, it's true that a duty not to be complicit would a fortiori follow from a duty to prevent an act. But my point in the post is to say that even if one doesn't accept a duty to prevent (i.e. try to do so), one can still accept a duty not to be complicit.

Asher, the reason why I mention the harm occurring is that this is structurally how complicity rules work. They are parasitic on the wrong being committed by the principal. It is possible to have inchoate rules with a similar content - indeed that's the whole point of the regime in the Arms Trade Treaty. A violation of the ATT occurs as soon as an arms transfer happens despite the risk of violation, even if there is no violation of IHL ultimately. But again the complicity framing is different. This doesn't mean, however, that a decision-maker can't say ex ante that an arms transfer mustn't proceed. Cf. also the ICJ's finding in the Bosnian Genocide case that a duty of prevention is only violated if genocide actually happens. (For a more extended discussion, please see my intelligence sharing piece cited in the post).

Nicolas Boeglin says

April 18, 2024

Dear Professor Milanovic

Many thanks for this extremely interesting post.

It seems that, in recent weeks, some colleagues have felt obliged to present Nicaragua´s request against Germany baseless, for reasons I really ignore. I have observed that some German´s colleagues have also felt obliged to write that Germany will not be subject to any of the different provisional measures requested by Nicaragua.

Usually after hearings at the Peace Palace at ICJ, I always recommend to my students that prudence and patience are the best friends a jurist can have.

Here a note (from an Israel´s newspapers usually weel informed) on possible warrant from ICC to be issued:

Its quite astonishing to see ICC Prosecutor Office been so discrete since ICJ Order of January 26, and I would really be interested in reading your views on such inaction showed by ICC Prosecutor Office.

Yours sincerely

Nicolas Boeglin

Ben Clarke says

April 27, 2024

Very interesting debate. One could argue that the issuing of financial sanctions and travel bans by the US State Department following settler violence in the West Bank, and public discourse over whether to impose sanctions against IDF military units involved in alleged violations in the West Bank, is State practice pointing towards US recognition that Common Article 1 has an external dimension..

Nicolas Boeglin says

May 2, 2024

Dear Professor Milanovic

May I add to my previous message what is read in ICJ order between Nicaragua and Germany of April 30rd:

"23/.../ With regard to the Genocide
Convention, the Court has had the opportunity to observe that the obligation to prevent the commission of the crime of genocide, pursuant to Article I, requires States parties that are aware, or that should normally have been aware, of the serious risk that acts of genocide would have been committed, to employ all means reasonably available to them to prevent genocide so far as possible /.../
Further, States parties are bound by the Genocide Convention not to commit any other acts enumerated in Article III (ibid., p. 114, para. 168).

24. Moreover, the Court considers it particularly important to remind all States of their international obligations relating to the transfer of arms to parties to an armed conflict, in order to avoid the risk that such arms might be used to violate the above-mentioned Conventions. All these obligations are incumbent upon Germany as a State party to the said Conventions in its supply of arms to Israel.


In my view, paragraphs 23 and 24 answer both negatively to Germany when it requested ICJ ro remove the case. An aspect I have not really found in the numerous articles published in Germany since this order has been delivered.

Yours sincerely

Nicolas Boeglin

Note: I take this opportunity to refer you (and our dear EJIL Talk colleagues) to a short note on this ICJ Order (in Spanish, sorry), waiting for a more detailled and complete commentary to be posted by some colleague of EJIL Talk in coming days: