Motive and Control in Defining Attacks

by | Nov 11, 2020

Attacks. Destruction in Homs, Syria, 2012. Bo yaser

In the appellate case of The Prosecutor v. Bosco Ntaganda, the Appeals Chamber of the International Criminal Court asked for briefs from “qualified publicists with the requisite expertise, who are interested in submitting observations on” the definition of “attack” as used in Article 8(2)(e)(iv) of the Rome Statute.

I was part of a group of five retired judge advocates and current scholars—led by Professor Geoff Corn, and including Eric Jensen, Chris Jenks, and Jim Schoettler—that accepted the request and submitted a brief on the issue. Two of us appeared before the Appeals Chamber on October 14 to submit to questions from the judges.[1] Each author of our briefing team is deeply experienced in advising commanders how to apply the LOAC in operational settings. We thought it important to weigh in on this issue as in the future, practitioners will obviously be influenced and military commanders will be judged by the opinions of international criminal courts.

In the interest of not repeating the analysis discussed by other authors in this series, I will limit this post to a discussion of two key elements of our brief. We argue—contrary to the assertions of the prosecutor in the appellate chamber—that acts against protected objects and the destruction of objects within the control of a party to the conflict do not qualify as attacks. They are distinct and separate violations of the law of armed conflict. We also suggest that motive plays a role as an element that distinguishes “attacks” from other harmful acts.

Case Background

Ntaganda was the leader of a group of non-State actors in the on-going conflict in the Democratic Republic of the Congo. After ejecting opposing forces from an area of hostilities, his group damaged a church and looted a hospital, taking up residence in one and seizing medical equipment in the other. The Trial Chamber acquitted Ntaganda of committing an “attack” on these protected objects, and the Prosecutor appealed the judgment. The Appeals Chamber asked amicus curiae to answer two series of questions:

How is “attack” defined under international humanitarian law, particularly in the context of cultural property and hospitals? What are the differences between the concepts of ‘attack’, ‘conduct of hostilities’ and ‘combat action’? What is the difference between ‘attack’ and ‘act of hostility’?

What does the term “attack” mean in article 8(2)(e)(iv) of the Statute? Does it cover acts such as pillaging and destruction? Would it cover acts committed in the course of a ratissage operation, conducted shortly after the takeover of a town?

 “Attacks” versus “Acts of Hostility”

In our view, while all attacks are acts of hostility, not all acts of hostility are attacks.

In the context of targeting in both international and non-international armed conflicts, an act of hostility that qualifies as an “attack” triggers LOAC applicable to targeting, including the rules relating to distinction (such as the obligation to refrain from targeting objects that do not meet the definition of a military objective), the obligation to take feasible precautions in the attack, and the principle of proportionality.

Other acts of hostility involving the destruction of property on land in an armed conflict trigger the application of different LOAC rules but are not “attacks.” For example, Article 23(g) of the Hague Regulations and its corollary customary rule, applicable in both international and non-international armed conflicts, prohibit the destruction of enemy property unless the act is imperatively demanded by the necessities of war. And seizure and appropriation of enemy property is governed, inter alia, by Article 23(g), the prohibition against pillage, and the rule on appropriating booty of war. These are all, according to State practice (as evidenced by para 5.17 of the U.S. Law of War Manual), “outside the context of attacks.”


As a threshold matter, acts of violence directed at persons or objects under the dominion and control of a belligerent—to include persons who are hors de combat—do not qualify as attacks. This notion of control as it relates to the assessment of what qualifies as an attack is distinct from other uses of the term in LOAC and international criminal law. Acts of harm against persons or objects in the power of the belligerent—meaning the person or object is under the physical control, dominion, or restraint at the hands of the belligerent (sometimes characterized as “possession” or “power”)—are governed by other LOAC rules. This is clear from an examination of the official records of negotiation of Additional Protocol I (AP I), as outlined in our brief.

For example, acts of harm in violation of the LOAC obligation to protect—such as murder, torture, or assault on an individual “in the hands of a party to the conflict”—does not constitute an attack in the sense of Article 49(1) of AP I. Yet, those individuals are in the physical control of an adversary.

With respect to objects, once a party has seized an object—even if its possession is temporary or tenuous—any further damage done by that party to the object would not be evaluated under LOAC targeting rules. Rather, the relevant rules would be the “imperative necessity” standard of Article 23(g) of the Hague Regulations and its corollary customary rule, as well as the constant care obligation of AP I. For example, if a party seizes a building, it is not free to continue damaging the building or its contents, except as may be permitted by the “imperative necessity” standard. The power/possession/control over a person or an object discussed here must not be confused with control over territory, which is a separate issue. Attacks may certainly take place in a territory under the attacker’s control.


The key substantive element of an attack outlined in our brief that I would like to discuss in this forum is that the motivation for executing the act must be to cause harm to the adversary or other persons in the conduct of hostilities. It is clearly implied in the ordinary meaning of the word “attack,” as far as the definition in Article 49(1) of AP I is concerned. Moreover, the motive element is also reflected, inter alia, in the phrase “against the adversary” and in the manner the term “attack” is used in the provisions of AP I that follow Article 49. The motive element is likewise reflected in the International Committee of the Red Cross drafts that preceded the AP I diplomatic conferences, which served as the basis of defining “attack” in Article 49(1). Furthermore, the existence of the motive element is demonstrated by common military practices and how they are legally analyzed, as will be detailed below.

The motive element is critical when distinguishing between attacks and other harmful acts, including those governed by Article 23(g) of the Hague Regulations and its corollary customary rule or the constant care obligation of Article 57 of AP I. Accordingly, violent acts directed at harming the adversary (including the civilian population and civilian objects) through physical injury or destruction are “attacks” within the meaning of LOAC.

Conversely, without this motive the act is not an “attack.” For example, providing air-delivered supplies such as food or medical equipment may inadvertently cause injury or damage if the air-delivered materiel lands on a person or structure, but such delivery would not be considered an attack. Other examples include maneuver damage to roads and fields; destruction of military equipment to prevent capture by the enemy; altering buildings or land to enhance tactical advantage; or dropping excess ordnance from aircraft in an open and unpopulated area. These are not “attacks.” Many of these circumstances are discussed in LOAC literature—to include Michael Bothe, et. al, in New Rules for Victims of Armed Conflict (1982) and Noam Neuman’s 2018 law review article. These military practices, albeit destructive and sometimes occurring in the preparation for attacks, are not treated as “attacks.”


War crimes under the Rome Statute are to be interpreted under Articles 8(2)(e) and 21(1)(b) in full consistency with LOAC. Thus, the term “attack” in Article 8(2)(e)(iv) must be interpreted to be synonymous with its well-established meaning under LOAC. Accordingly, the term “attack” in Article 8(2)(e)(iv) does not hold a special meaning that is different than its meaning under LOAC. To interpret it differently would be over-inclusive and would require application of LOAC targeting rules to numerous activities that are regulated by other provisions of LOAC and regularly practiced by military units.

The legal rules pertaining to unlawful destruction of property are different than those applicable to “attacks” in various respects, including in respect to the motive for the act. And destruction of property is punishable in circumstances that involve the belligerent in control of the property, unlike attacks. Accordingly, destruction of property is addressed specifically and separately in the Statute in Articles 8(2)(a)(iv), 8(2)(b)(xiii), and 8(2)(e)(xii).

Professors Geoff Corn, Chris Jenks, Eric Jensen, and Jim Schoettler, as well as this old retired soldier,[2] submitted the amicus brief on September 18, 2020. It is our hope that it will help avoid misapplication of standards to those military commanders who are charged with the important business of distinguishing between civilian objects and military objectives and protecting civilian objects under their dominion or control.


Colonel (Retired) Richard B. “Dick” Jackson recently retired as the Special Assistant to the U.S. Army Judge Advocate General for Law of War Matters.


[1] See Chris Jenks on Opinio Juris, Motive Matters: The Meaning of Attack Under IHL & the Rome Statute. He discusses his presentation, along with Geoff Corn, before the Appeals Chamber and also covers issues similar to those addressed here.

[2] Four of the five experts are Lieber Senior Fellows.



“Attack” Symposium Posts