The Definition of an “Attack” under the Law of Armed Conflict
As in any branch of international law, examining the meaning of a law of armed conflict (LOAC) term, and its incorporation in the Rome Statute of the International Criminal Court, should look first to the treaties—and “attack” is no different. Indeed, as is well known, the notion of “attacks” is explicitly defined in Article 49 of the 1977 Additional Protocol I as “acts of violence against the adversary, whether in offence or in defence.” A seemingly broad definition, it is certainly clear and concise. One might thus be tempted to end an inquiry of the law there.
But in this post, we detail why such an approach would be mistaken. For it fails to consider the notion of an “attack” in Geneva Law (which protects persons and objects in the power of the enemy), which applies more broadly than does the Hague Law regulation of the conduct of hostilities. This omission is particularly serious in the context of a non-international armed conflict, where the geographical scope of hostilities is tightly circumscribed.
Thus, it is important to situate the Article 49 definition in its context. Placed in Part IV of the Protocol entitled “Civilian population,” this is the second of the two parts in the Protocol that are devoted to Hague Law. Part II of the Protocol, on the other hand, which concerns “Wounded, sick and shipwrecked,” is Geneva Law—the mainstay of international humanitarian law (IHL), where that term is correctly delineated. Contrary to what is sometimes asserted, neither of the two 1977 Additional Protocols ends the traditional distinction between Hague Law and Geneva Law. Rather, they each combine both strands in a single instrument.
Significantly, when Article 49 of Additional Protocol I was drafted the notion of “an attack” was already incorporated in the 1949 Geneva Convention I, including in Article 19. Article 19 stipulates that “Fixed establishments and mobile medical units of the Medical Service may in no circumstances be attacked, but shall at all times be respected and protected by the Parties to the conflict.” Thus we have what is ostensibly a Hague Law term—it was first employed in the Regulations annexed to the 1899 Hague Convention II and the 1907 Hague Convention IV—appearing in 1949 in what is generally cast as a pure Geneva Law treaty. How are we to interpret the use of this term in Geneva Convention I? Are we to define it retrospectively and ex post facto on the basis of the definition in 1977 Additional Protocol I?
Two factors especially caution against such an approach. First, the language of the provision in Article 19 of Geneva Convention I itself is quite broad, requiring that military medical facilities “shall at all times be respected and protected.” The travaux préparatoires make it clear that the intent was not to limit the reach of the Article geographically or temporally to the conduct of hostilities, but was meant to be comprehensive, as is generally the case with Geneva Law. Although that particular provision applies specifically in international armed conflict, the International Criminal Tribunal for the former Yugoslavia (ICTY) rightly applied Geneva Law across the whole of the territory of a party to an armed conflict. The ICTY Appeals Chamber observed that “the rules contained in [Common] Article 3 also apply outside the narrow geographical context of the actual theatre of combat operations” (para 69). So it is with the remainder of the Geneva Conventions and in customary law.
Second, while the valuable Commentaries of the International Committee of the Red Cross (ICRC) of 1952 and 2016 do not address the issue directly, their interpretation of the provision in Article 19 does provide valuable clues as to its breadth. The 2016 Commentary declares, for example, that
seizure [of a military medical unit] may impede the functioning of the facility and the continued provision of medical care for the wounded and sick. In addition, it will lead to the loss of the facility’s specific protection and moreover, upon fulfilling the relevant criteria under humanitarian law, to its also becoming a military objective, thus endangering any wounded or sick people and medical personnel within it. (para 1800)
Whether or not an “attack” is governed by Hague Law (and even without “acts of violence,” as such), seizure of the facility, or even just the theft of its medical contents, would thus fall within the scope of the provision and—we would argue—also within the scope of an attack outlawed by Geneva Law. It is not insignificant that special protection is afforded to hospitals under the law of armed conflict over and above the general protection afforded to all civilian objects.
Our broader interpretation of the notion of attack in Geneva Law is supported by reference to the protection of cultural property. The ICRC’s ground-breaking study of customary IHL/LOAC did not explicitly consider the definition of “attack,” much less its definition in customary law (if such can be deemed to exist). That said, in its Rule 38 (“Attacks Against Cultural Property”), the ICRC determined that a customary rule applicable in all armed conflict was that “Special care must be taken in military operations to avoid damage to buildings dedicated to religion, art, science, education or charitable purposes and historic monuments unless they are military objectives.” This notion of “military operations”—far broader than that of “hostilities”—provides a more accurate rendition than Article 49 of Additional Protocol I of the notion of an attack under IHL with respect to cultural objects and medical facilities.
Theft of objects of particular cultural value also falls within this broad definition of an attack. Once stolen, they may be lost to humanity for ever. Rule 40 of the ICRC study of customary IHL stipulates that each party to the conflict must protect cultural property and that, as a consequence, “Any form of theft, pillage or misappropriation of, and any acts of vandalism directed against, property of great importance to the cultural heritage of every people is prohibited.” Again, special protection under LOAC is provided to cultural property for good reason. What is lost or destroyed in such a case cannot be replaced.
In sum, the fact that objects are stolen rather than destroyed and the fact that a building is pillaged, not shelled, is not determinant. This is because the cultural or medical value demanding of special protection may not be the building itself but, manifestly, the object or equipment it contains. Beyond the scope of an attack in the conduct of hostilities—as defined by Article 49 of Additional Protocol I—an attack as defined under Geneva Law applies throughout the territory of every party to any armed conflict to prohibit all seizure, theft, desecration, and despoliation of specially protected objects and the precious items they hold. All military operations against an adversary and its civilian population and civilian objects are thereby constrained and civilian immunity protected and preserved.
What we propose does not undermine or alter in any way the definition of “attack” for the purposes of Hague Law and the rules governing the conduct of hostilities. What it does do is to reflect the broader scenario of, particularly, non-international armed conflict where the geographical area of hostilities is narrow, and ensures the general application of special protection reflected in Geneva Law. We submit that this is consonant with the object and purpose of Geneva Law and of IHL in general. And, as the International Criminal Court Appeals Chamber has justly held, reference to the “established framework of international law” permits “recourse to customary and conventional international law … to ensure an interpretation of article 8 of the Statute that is fully consistent with, in particular, international humanitarian law.”
Christof Heyns is Professor of Law, University of Pretoria, and a member of the United Nations Human Rights Committee.
Stuart Casey-Maslen is an Honorary Professor at the Centre for Human Rights of the University of Pretoria
Thomas Probert is Head of Research of the international collaboration “Freedom from Violence” at the University of Pretoria.
 While “attack” is used both with respect to crimes against humanity and in the crime of aggression, in neither case is the differing meaning of the term in these branches of international law relevant to the IHL discussion.
 Prosecutor v. Ntaganda, ICC-01/04-02/06 OA5, Judgment on Appeal against the “Second decision on the Defence’s challenge to the jurisdiction of the Court in respect of Counts 6 and 9,” ¶ 53 (June 15, 2017) (citing Prosecutor v. Lubanga, ICC-01/04-01/06 A 5, Judgment, ¶ 322 (Dec. 1, 2014)).
“Attack” Symposium Posts
- Symposium Intro: The ICC Considers the Definition of “Attack” by Ronald Alcala & Sasha Radin
- A Radical Reimagining of the Concept of “Attack” by Mike Newton
- Upcoming: Motive and Control in Defining “Attacks” by Dick Jackson
October 27, 2020
October 29, 2020
by Dick Jackson
November 11, 2020
November 17, 2020
November 24, 2020