Ukraine Insta-Symposium: Certain (Para-)Military Activities in the Crimea: Legal Consequences for the Application of International Humanitarian Law

by Remy Jorritsma

[Remy Jorritsma (LL.M.) is a lecturer and teacher at the Department of International and European Law of Maastricht University.]

This contribution intends to demonstrate that Ukraine and Russia are involved in an international armed conflict, triggering the application of International Humanitarian Law (IHL). In particular, this post explores two relevant issues: the question of valid consent, and the legal qualification of the recent military hostilities.

In the Security Council meeting of 1 March 2014 the representative of Russia asserted that

[the Prime Minister of Crimea] went to the President of Russia with a request for assistance to restore peace in Crimea [which] appeal was also supported by Mr. Yanukovych, whose removal from office, we believe, was illegal.

Possibly such consent did not bring about an international armed conflict. Indeed, an international armed conflict (incl. occupation) does not exist when a host State allows another State to carry out armed activities on, or exercise control over, the territory of the host State (cf. SC. Resolution 1546(2004)).

However, the view that a semi-autonomous province and/or deposed Head of State can validly invite foreign troops against the express wishes of the central government makes no sense in light of the well-established principle of non-intervention. By virtue of their office the incumbent Head of Government/State and Foreign Minister are responsible in matters of a State’s foreign relations (Arrest Warrant, ICJ Reports 2005, §53). Neither the local government of the Crimea nor former president Yanukovych should be regarded as competent to issue valid consent to the presence of foreign armed forces, unless the central government of Ukraine agrees to this.

Given the present state of Ukraine, this outcome is not affected by the alleged unconstitutional nature of the ousting of former president Yanukovych. In the Tinoco case (1923) Costa Rica advanced the argument that the Tinoco government had not been a de facto government because of its unconstitutional origin. Sole arbitrator Taft rejected this, noting (at p. 381) that

[it would be a contradiction in terms] to hold that within the rules of international law a revolution contrary to the fundamental law of the existing government cannot establish a new [de facto] government.

Yanukovych’ claim to the presidency of Ukraine should be given little legal credit. His claim is opposed by an effective central authority that, by discharging regular administrative functions and controlling the governmental apparatus, including police and armed forces, is widely recognized as de facto government. In such a case, sparse non-recognition based on the alleged unconstitutional nature of the new government does not outweigh wide recognition based on effective control. Any presence of and action by foreign troops beyond the limits of consent given by the Ukrainian government in the past (eg in form of the Black Sea SOFA) must be regarded as hostile and possibly triggers the application of IHL.

According to common Article 2 of the Geneva Conventions of 1949 the application of International Humanitarian Law is applicable in three situations which amount to international armed conflicts:

(1) formally declared war;

(2) partial or total occupation, even without armed resistance; and

(3) any other (read: de facto) armed conflict.

President Putin asserted the right to invade Ukraine and even received parliamentary approval to use military force. Of course, such rhetoric is not, contrary to certain assertions by Kiev, tantamount to an explicit and formal declaration of war on Ukraine.

Instead, as from the beginning of March, IHL has become applicable as a combined result of the occupation of the Crimea and accompanying factual hostilities. Although it cannot be said with absolute certainty, it is reported that the so-called ‘local self-defence forces’ are in reality Russian armed forces who have removed their insignia,  in which case their actions are by default attributed to Russia. In addition, Russia may even bear responsibility for acts committed by organized armed groups that lack a formal relation with Russia. The Appeals Chamber of the ICTY held that, in order to attribute acts of paramilitary groups to a State it has to be shown that the State exercises overall control over the course of their operations (see Prosecutor v. Tadić, Judgment, 15 July 1999, §131). Whether a State resorts to occupation or wages inter-State conflict by using its regular forces, or indirectly by using non-State actors as proxies to act on its behalf, the legal result is the same. Any hostile action undertaken by  organized armed non-State groups in the Crimea is imputable to Russia if and to the extent that Russia exercises the required degree of operational control.

A situation of occupation as described in Article 42 of the Hague Regulations of 1907 exists when there is a hostile substitution of territorial power and authority; it is irrelevant ‘whether or not [the occupying power] had established a structured military administration’ (Armed Activities in the Congo, ICJ Reports 2005, §173). Through its military manoeuvres and presence Russia qualifies as occupying power: it has been able to establish territorial control and is demonstrably able to exercise its authority in the Crimea without the consent of the central government of Ukraine.

In addition, now that the first (warning) shots on the Crimea have been fired, I would submit that IHL applies as a result of the existence of a de facto state of armed conflict between Ukraine and Russia. The Commentaries to common Article 2 suggest that, regardless of the number of victims or the intensity of hostilities, an international armed conflict comes into being as a result of

‘[a]ny difference arising between two States and leading to the intervention of members of the armed forces’.

This low threshold of application is nowadays still maintained by the International Committee of the Red Cross, and is followed in the case law of the ICTY (see Prosecutor v. Tadić, Jurisdiction Decision, 2 October 1995, §70: ‘whenever there is resort to armed force between States’) and the ICC (see Prosecutor v. Katanga, Judgment, 7 March 2014, §1177, adopting the Tadić definition).

On the other hand, this “first shot” approach has recently been called into question. In its final Report the ILA’s Use of Force Committee suggested (at p. 13) that short-lived or low-intensity confrontations between states were excluded from the scope of application of IHL:

state practice [since 1945] indicated that states generally drew a distinction between on one hand, hostile actions involving the use of force that they treated as “incidents”, “border clashes” or “skirmishes” and, on the other hand, situations that they treated as armed conflicts.

Mary Ellen O’Connell, the Chair of the Committee, notes that the ICRC position may be based on policy rather than law. In my view, however, it appears to be exactly the other way around. The Committee report  was only aimed to arrive at a “general” definition of armed conflict (see p. 3, at n. 7). It based its conclusions on a coalesced overview of inter-state and internal conflicts, doing injustice to the various existing types of armed conflicts and incorrectly conflating their distinct substantive criteria. Moreover, before taking into account subsequent State practice to interpret common Article 2, such practice must have duly constituted the agreement of the parties regarding its interpretation and thus be accompanied by the requisite opinio juris regarding its interpretative value. Affected belligerent states may well have treated minor incidents as not amounting to international armed conflicts out of political motives (eg to prevent escalation or loss of face) or for practical purposes (e.g., because the limited engagements did not cause any victims), rather than out of a strict sense of legal interpretation.

Instead, the exclusion of border clashes and other low-intensity (yet intentionally hostile) inter-State confrontations from the concept of  armed conflict stands in contrast to the widespread acceptance of the Tadić definition for the very purpose of classifying non-international ánd international armed conflicts (see e.g., the summary of the debate on Article 2(b) of the Draft articles on the effects of armed conflicts on treaties, §206-213). Therefore, and to avoid these inter-State hostilities from taking place in a legal vacuum, IHL must be respected as from the moment of the actual opening of hostilities between Ukraine and Russia.

Unfortunately Russia has resorted to a mixture of legal and extralegal arguments to exonerate itself. That being said, the application of IHL rests on factual criteria relating to the identity of the parties and the character of hostilities. As a result of the current situation Ukraine and Russia must now be regarded as bound by, on one hand, customary international humanitarian law and, on the other hand, obligations undertaken by them in treaties applicable to international armed conflicts, most importantly the four Geneva Conventions of 1949 and its First Additional Protocol of 1977.

10 Responses

  1. Despite Tadic, the ICJ had recognized that effective control is not the test regarding attribution or imputation of non-state actor armed attacks to a state (which would be relevant to when an armed conflict engaged in by non-state actors is attributable to a state and internationalizes the conflict). Instead, the ICJ recognized a “substantial involvement” test relevant also to whether aggression has occured.  see  at pages 424-434 and references cited.
    Under GC art. 47, Russia cannot lawfully annex occupied territory.

  2. Remy,
    Do you buy the ICJ’s distinction between Tadić (overall control) and Nicaragua (effective control) so that former applies when classifying conflicts and the latter to attribution for state responsibility purposes? In your post you seem to adopt Tadić for state responsibility, a proposition rejected in the ICJ’s Genocide Case which stuck to Nicaragua and effective control. Personally, I think the ICJ’s supposed distinction is artificial and untenable ( <a></a&gt; ), but I’d be interested to know your thoughts.
    Related to this is the notion of occupation by proxy (through the control of territory of State A by non-state armed groups whose acts can be attributed to a State B). You seem to accept that this is permissible and could well apply to the situation in the Crimea. This would be in line with the ICTY Trial Chamber’s decision in Blaškić (paras 148-150). However, this theory of ‘proxy occupation’ was rejected wholesale by another ICTY Trial Chamber in Naletilić and Martinović (para. 214). What’s your view on this split?

  3. Manuel, thank you for your questions. In short, no I don’t buy it. To me it is indeed an artificial attempt to mask the fact that the ICJ and the ICTY/ICC are head-on colliding on this point. In my view the acts of classification of conflict and attribution of responsibility are intimately related because of the basic understanding that international armed conflicts are fought between forces under responsible command of the State (cf. Art. 4(a)(2)(a) GCIII, which is equally applicable to regular armed forces covered by Art. 4(a)(1) GCIII; Art. 43(1) API; and Art. 91 API).   
    Before speaking of interstate hostilities or internal strife you need to establish which parties you are ultimately, and legally, dealing with. The same goes for determining whether territorial control is exercised by or on behalf of a State (occupation), or not (in which case it may be an APII conflict). The logical sequence of events would be to classify armed conflicts only áfter attribution of behavior took place. Only this way is it possible to discern the true identity of the belligerents and attach legal consequences to their behavior, holding otherwise would be to put the cart before the horse.
    Accepting that classification has its own (stricter) test which is distinct from attribution of conduct leads to the absurd result that a State may get involved in an international armed conflict by reason of exercising overall control over an organized armed group, but then without being responsible for its actions. This is unacceptable in light of IHL’s emphasis on responsible command (see also Tadić, Judgment AC, §94), and at odds with the general structure of the law on State responsibility (which places attribution befóre establishment of a breach or legal consequences). It seems to me there is no test for classification as such; the test is one of attribution of conduct for the purpose of establishing whether such attributed conduct presents a breach of international law or otherwise brings about consequences under international law (such as the law of occupation).
                 Although I agree with the TC Judgment in Naletilić and Martinović that for occupation to exist, ‘a further degree of control is required’ (§214), this is missing the point and certainly cannot be taken to mean that occupation by proxy as a concept is completely discarded. The standard of control over territory is governed by Art. 42 Hague Regulations of 1907, whereas the standard of control over non-State actor behavior is governed by Art. 8 ARSIWA. These are two completely distinct matters, and I would submit that an occupations exists whenever a State exercises overall control over an organized non-State actor (for unorganized individuals effective control is still the required test), but only if the non-State actor has the required degree of (effective) control over the territory. Examples of such occupation by proxy may be found in Northern-Cyprus and Nagorno-Karabakh. 

  4. Correction: “has its own (stricter) test” should read: “has its own (less demanding) test”.  
    How is it ever possible to classify a conflict without resorting to the law of State responsibility? By the nature of things, attribution of conduct for purpose of classification of conflict must be governed by the same standard as attribution of conduct for the purpose of State responsibility.

  5. Manuel & Remy: too many writers have made the mistake that the ICJ’s effective control test regarding state responsibility for specific war crimes or human rights violations is the same test for attribution of non-state actor armed attacks to a state for the purpose of self-defense.  The test in the latter circumstance includes the “substantial involvement” test.
    Also Remy, note that the test for pow status under GPW art. 4(A)(1) and (3) is merely membership.  Art. 4(A)(2) only applies to certain militia and volunteer groups and not all states are parties to Geneva Protocol I, nor is art. 43(1) thereof proven CIL.

  6. I wasn’t really talking about attribution for self defence purposes, but since you raise it Jordan, I always had a hard time figuring out why the ICJ thinks its useful to have a myriad of different tests applying to a bunch of different circumstances. One test for attribution for state responsibility, another for classifying conflicts, another for self defence, and yet another for the acts of specific individuals (e.g. unilateral Latino Control Assets in Nicaragua) etc. It starts to look like a jumbled attribution soup – i.e. a little ridiculous

  7. Remy,
    There’s an interesting paper on occupation by proxy in the forthcoming edition of the JICJ that’s worth checking out: 
    Unfortunately it doesn’t really go into the Blaškić/Naletilić and Martinović split (which by the way has not yet been resolved by the ICTY Appeals Chamber – might come up in the Prlić appeal though). I take your point that Naletilić and Martinović could be read as not wholly rejecting occupation by proxy per se, but at the same time I find it a little difficult to swallow the proposition that a state can exercise effective control over a territory when it merely exercises overall control of an armed group. Could a state impose its will on foreign territory if it is not capable of issuing specific commands/instructions to the relevant armed group? I’m not sure.. Just because the armed group exercises effective control over territory, I don’t think it necessarily follows that this is automatically translated to a foreign state when it only has overall control over that group.

  8. Jordan, possibly different fields of law have different standards of attribution (cf Art. 55 ARSIWA), but this is a discussion for a different topic altogether. My contribution was merely to show that IHL applies.
    I was making a more general point that wars are fought on the understanding that the belligerent parties bear responsibility, not just for their regular armed forces, but also for other units and groups (Rule 149 CIHL Study). While perhaps not all of Art. 43(1) API may represent CIHL, I believe that this is certainly the case for such an essential element as responsible command which can be traced back to both the Art. 2 Hague Convention of 1907 and Art. 1 of its Regulations. With respect, arguing that Art. 43(1) API is not CIHL at all is I believe partly incorrect (at least as far as responsibility is concerned), and for the remainder not relevant to the point that I was making.
    Manuel, my main concern is not so much the level of control, but the view that classification and attribution are treated as coming from two different planets. I understand the ICTY’s preference for overall control (and the State-centered ICJ’s reluctance towards it), but to me the standard of attribution of conduct is one thing, the required level of territorial control is another. Both come into play when facing direct or possibly indirect occupation, but from two different angles. In addition, what constitutes “effective” for the purpose of attribution is not the same as for occupation; for attribution it requires specific instructions, for occupation it requires substitution of control in the post-invasion stage (for which specific instructions may actually not even be necessary).

  9. I understand your point Remy, but do you really think that states can be said to exercise effective control over territory if they are not able to enforce their will over the non-state armed groups who control said territory? I’m not sure that overall control is enough to show this, but of course I suppose it depends on the facts of the specific case.

Trackbacks and Pingbacks

  1. […] // Remy Jorritsma prüft die Anwendung des humanitären Völkerrechts auf der Krim bei opinio juris. Sina Etezazian fragt sich auf opinio juris, ob die Ukraine ein Recht auf Selbstverteidigung […]