Interstate use of armed force in third states: what are the implications for conflict classification?

Written by

The events that are rapidly unfolding in the Middle East lately are a cause of grave concern in an increasingly violent and volatile world. On April 1, an Israeli airstrike on the Iranian consulate in Damascus killed a senior commander of the Iranian Revolutionary Guard Corps, as well as several other officers. Apart from the jus ad bellum issues that are raised, there is also a host of intriguing questions regarding the classification of similar situations under IHL.

Use of armed force against a state outside its territory: when is an IAC triggered?

First, when state A uses force against targets that could be deemed to represent state B outside the latter’s territory, the nature of the target is decisive in determining whether there is an international armed conflict (IAC) between the two states. According to the prevailing view, IHL of IACs becomes applicable from the moment a state resorts to armed force against another state, irrespective of the intensity of such force or whether there is an armed response by the second state. When this occurs within the territory of the state against which force is used, the target is usually immaterial.

However, the interstate character of the violence becomes more difficult to verify when a strike happens in a third state C. On one end of the spectrum, it seems straightforward that force against the armed forces, military bases or warships of state B triggers an IAC between the two states, regardless of these targets’ location. On the other end of the spectrum, it is hard to conceptualize that targeting nationals of state B abroad can give rise to an IAC. An intermediate zone is occupied by diplomatic missions and certain high-ranking officials of state B, the circumstances of the targeting of whom need to be carefully assessed before reaching a conclusion about a potential IAC classification.

One scholar (pp. 375-384) has suggested placing the ‘external manifestations’ of a state in concentric circles depending on how close their ties are to the state in question, with a state’s nationals at the outer end of this schema. The more removed a manifestation is from the circle’s centre, the looser its connection with the state and the more difficult (if not impossible) it becomes for a use of armed force against it outside this state to trigger an IAC. It has also been argued (p. 189) that, outside a state’s territory, only attacks against its armed forces can give rise to an IAC. Besides, the difficulty of attributing the use of armed force to state A in some cases should also be factored into the aforementioned considerations.

In view of the above, it is plausible that the targeting of officials of Iran’s armed forces, including a senior commander, by Israel in Syria’s territory, renders IAC law applicable in the relations between Iran and Israel. (The questions of whether a first strike is itself regulated by IHL, whether this was indeed a first strike or part of a pre-existing conflict, as well as the legality of this strike under IHL, IHRL and jus ad bellum are separate.)

Lack of consent by the third state and its impact on classification

Second, let’s assume that state C does not consent to the use of force on its territory by state A against state B. In 2020, a US drone strike killed Iranian General Qassem Soleimani in Baghdad, without Iraq’s consent for the use of force in its territory. The jus ad bellum aspect aside, was there an IAC classification both between the US and Iran and between the US and Iraq?

A few years back, the launch of the updated ICRC Commentary on the First Geneva Convention opened this discussion with regard to extraterritorial non-international armed conflicts (NIACs). The ensuing exchanges provide insights that are useful for our scenario too. According to the ICRC and some scholars, any unconsented to use of armed force by a state in the territory of another state automatically triggers an IAC between the two, even if the target of such force is a non-state armed group (NSAG). This force is directed against a state’s territory and affects its civilian population, including when it is equally part of a parallel NIAC with the targeted NSAG.

Arguments both in favor and against this view have been put forward. On the one hand, its supporters argue that IAC law casts a (more) protective net over the entire situation, notably when the latter does not qualify as a NIAC in view of a lack of organization by the NSAG or/and intensity of fighting between the NSAG and the intervening state. Besides, ascertaining the purpose behind a use of force, i.e. whether it is only aimed at a NSAG or (also) targets the territorial state, is a difficult enterprise and introduces a subjective element to conflict classification. On the other hand, its opponents support that adding an IAC in this context does not reflect the identity of the parties fighting on the ground and is hard to put in practice. Additionally, consent for them pertains to jus ad bellum and should not have an impact on classification.

A logical corollary of the position that unconsented to use of armed force on the territory of another state automatically triggers an IAC between the intervening and the territorial state is that, in our scenario (state A v. state B on C’s territory), and as long as state C’s consent is absent, an IAC occurs between states A and C. Even though state A targets state B, C’s territory is also subject to armed force and its civilian population may be incidentally harmed by it. In its new Opinion Paper (pp. 11-12), the ICRC confirms that the IAC classification is not altered by the fact that the state at the origin of such armed force claims that it is directed ‘only against another party it is fighting within the framework of another armed conflict’, arguably irrespective of whether this other armed conflict is an IAC or a NIAC.

When this armed force is limited to the targeted killing of B’s officer(s), the A v. C IAC classification may appear inconsequential, as the armed violence may very well cease after the strike. After the killing of Soleimani, Iraq protested the violation of its sovereignty by the US, but good policy reasons commanded to not invoke IHL at an early stage even in the relations between the US and Iran, let alone in the ones between the US and Iraq. Besides, an IAC will in any case exist between states A and C, if C responds with armed force, if it declares war against A, or in case state A occupies (part of) state C’s territory, even if this occupation does not meet with armed resistance. For those who adopt a functional definition of occupation, the mere presence of state A’s boots on the ground, as opposed to the use of a drone or an airstrike, triggers the application of some rules of occupation law.

However, there are other cases in which this classification acquires more importance. There may be armed forces of state B permanently stationed on C’s territory after having secured the latter’s consent for their presence there (as peace forces, as part of a larger multinational force, or for any other reason). If state A engages in fighting with B’s forces on the territory of and without the consent of state C in a more prolonged manner, it is more important to determine whether an IAC exists between the intervening and the territorial state in parallel to that between the intervening and the attacked state. Who and what can be characterized as a legitimate target in the conduct of hostilities depends on who the conflict parties are and is, thus, distinct for each IAC. Similarly, who qualifies as a ‘protected person’ when in the hands of a party to a conflict will differ in each IAC, to mention but two examples of the practical significance of this question.

Additional factors may further complicate our scenario and its classification under IHL. For instance, what if state A uses force against state B in the territory of state C which is occupied by state D? Or, what if state A employs armed force against state B on the territory of state C, where state B is an occupying power? Does a lack of consent by the territorial sovereign automatically trigger an IAC between the territorial and the intervening state under these circumstances? Following the 1st of April strike, Iran directed an aerial attack against Israel on April 13. Had Iran targeted Israel outside its territory, for instance in the Shebaa farms, where Israel is an occupier, would a lack of consent from Lebanon lead to an Iran- Lebanon IAC classification (independently of the Iran v. Israel one)? To better illustrate this question, it is worth returning to the case of extraterritorial NIACs. What if a state is involved in an extraterritorial NIAC in an occupied territory and the territorial state does not consent to the use of force on its (occupied) territory? If the rationale of the position described above is that the state whose territory and population are subject to force becomes party to an IAC with the intervening state, then this IAC would involve the occupied state, since occupation does not lead to transfer of sovereignty. However, the presence and control of the occupying power on the ground further complicates things.  

Although this blog poses more questions than it gives answers, it takes the recent events as a starting point to invite for more reflection on some of the challenges of conflict classification. However, it should not deflect attention from the fact that, at these times, priority should be given to efforts for de-escalation and effective compliance with other legal regimes as well, such as jus ad bellum.

Print Friendly, PDF & Email

Leave a Comment

Your comment will be revised by the site if needed.


Dor Hai says

May 1, 2024

Many thanks for this interesting and thought provoking piece!

Some preliminary comments which popped up after reading it.

1) I would be cautious to attach too much weight to States' invocation or non invocation of IHL as an indicator for the existence of an IAC - for instance, the fact that Iraq (arguably) had some good policy reasons not to invoke IHL, is not a determinative factor as to the existence of an IAC.
2) As for your argument that when armed force is limited to a targeted killing of B's officers, the classification between A and C may appear inconsequential, I tend to disagree, chiefly since it is unclear in which instances the armed violence would cease after the strike, and anyway this analysis, (whether the violence would cease or not) by definition would be conducted ex post facto, thus leading to uncertainty as to the temporal scope of application of IHL. Hence, I am not convinced that the classification becomes more important in the case of prolonged hostilities. In my view, it is as equally important and relevant for the first strike as for a strike that is carried out in the midst of hostilities.
3) In relation to my last point, you argue that it is important that the classification between A and C becomes more pertinent where prolonged hostilities are taking place, because that would affect who and what may be attacked and who may be regarded as protected person. However, that would anyway be the case even if no armed conflict existed between States A and C. For instance, let us assume that not armed conflict existed between the States, and at some moment in time State A decides to launch an attack on the territory of State C - the legality of that attack would anyway we subject to the relevant rules of IHL - distinction, proportionality, precautions etc., no? (unless you reject the first-shot theory and argues that the first strike should be subject to IHRL, provided it is applicable).

I would love to hear you thoughts and views, and many thanks again for this very interesting piece!

Zoi Lafazani says

May 2, 2024


Τhanks for your comments.
My brief response would be that:

1) Only facts on the ground are determinative for classification. States often have interests in denying that IHL applies in a certain situation (or they may assert that it applies, while this is not the case)- their views are not relevant for classification. That is why I refer to policy, as opposed to law, as a reason for not invoking IHL (which does not exclude its application). Besides, policy does not only pertain to states; after the killing of Soleimani, several actors in the international legal arena preferred not to engage in the discussion regarding the strike in IHL terms.

2) I do not argue that classification changes depending on whether there is only one strike or more prolonged hostilities- I merely advance that the consequences of an IAC classification with the territorial state C can be better understood in the event of more sustained hostilities between states A and B on C’s territory. In the case of a single strike of A against B, the practical implications of an A v. C classification are way more limited.

3) I do not contest that IHL would apply absent an A v. C classification, but there are differences depending on whether one or two IACs are ongoing, including concerning the classification of persons and objects in CoH and who qualifies as a protected person under GCIV. For instance, in a (single) state A v. state B classification, members of C’s armed forces qualify as civilians in CoH and shall be treated accordingly.

I hope this helps and thanks for taking the time!