Targeting Non-State “Mixed Groups”
“Now I want you to remember that no bastard ever won a war by dying for his country. You won it by making the other poor dumb bastard die for his country.”
General George Patton, May 31, 1944, Addressing the 6th Armored Division
Sounds simple, but deceptively so, because first you have to determine who qualifies as the enemy’s “poor dumb bastard.” Identifying who is targetable as a matter of law is an issue that has long-challenged government policymakers, commanders, military legal advisers, international humanitarian law experts, and the soldier in the field. It remains one plagued with controversy.
In the abstract, the targetability of individuals under international humanitarian law is fairly straightforward. During an international armed conflict, combatants, members of organized armed groups, and civilians who “directly participate in hostilities” are subject to direct attack, while collateral harm to them during an attack on another military objective does not factor into the proportionality analysis or the precautions in attack requirement. Although the legal notion of combatancy does not exist in a non-international armed conflict, the rules are similar. Members of the armed forces and other government agencies engaged in the fighting, dissident armed forces, members of organized armed groups, and direct participants are lawful targets.
In the contemporary battlespace, as many readers know from personal experience, application of these rules has become highly complex. In great part, this is because of the pervasive involvement of non-State actors in modern conflict, as well as their organizational sophistication and combat wherewithal. The diversity of such groups renders simple solutions to dealing with them in a manner that is consistent with international humanitarian law elusive.
In a recent Duke Journal of Comparative and International Law article, I examine when individuals who perform intelligence functions for non-State groups may be attacked. Here, I will narrow the aperture to one aspect of that topic: When are members of non-State groups that are comprised of distinct subordinate organizations—some of which engage in hostilities, while others perform functions that are either unrelated to, or attenuated from, the fighting—targetable? I label such groups, like Hamas, Hezbollah, or ISIS, “mixed groups.”
Narrow and broad notions of organized armed groups
Of course, many non-State groups, particularly terrorist organizations, exist solely for the purpose of engaging in hostilities. When this is the case, the group in its entirety comprises an “organized armed group” (OAG). An OAG is a group that is well-organized, albeit not necessarily to the level of the armed forces; is armed in the sense of engaging in hostilities, especially by conducting operations that qualify as “attacks” under international humanitarian law (IHL); and conducts operations that have a nexus to the armed conflict.
Whether all members of an OAG are targetable is the subject of some controversy. The ICRC, in its Interpretive Guidance on the Notion of Direct Participation in Hostilities, suggests that only those members who have a “continuous combat function” in the group are targetable based on their membership; others may be attacked only “for such time” as they take a “direct part in hostilities,” as when they take up arms even though their duties in the group do not include fighting or other combat related functions.
The United States, correctly in my view, takes the broader position that OAG members are targetable while they remain members. For instance, the Army/Marine Corps Commander’s Handbook on the Law of Land Warfare (FM 6-27/MCTP 11-10C) provides
Membership in the armed forces or belonging to armed groups, that is, affiliation with these groups, makes a person liable to being made the object of attack regardless of whether he or she is taking a direct part in hostilities. Moreover, the individual, as an agent of the hostile group, may be assigned a combat role at any time, even if the individual normally performs other functions for the group. Thus, with limited exceptions, combatants may be made the object of attack at all times, regardless of the activities in which they are engaged.
Similarly, the DoD Law of War Manual observes that, “[l]ike members of an enemy State’s armed forces, individuals who are formally or functionally part of a non-State armed group that is engaged in hostilities may be made the object of attack because they likewise share in their group’s hostile intent.”
“Mixed Groups”: Groups with fighting and non-fighting components
However, some non-State groups do not exist solely to engage in hostilities. In such cases it could be inappropriate to consider every member of the group as targetable. Hamas, for example, must govern Gaza. This entails running schools, providing social and medical services, engaging in law enforcement and judicial activities, maintaining public infrastructure, and other functions normally provided by governments. But at the same time, Hamas fields fighting organizations such as the al-Qassam Brigades. Similarly, ISIS found itself in the position of performing governance functions when it occupied wide swaths of Syria and Iraq. Many members had little to do with these groups’ combat operations.
In my estimation, members of such non-State “mixed groups” are targetable when they are members of the group’s “overall organized armed group,” a concept explained below. If not, attacking them is subject to the limitations and requirements of the direct participation rule.
Only when a non-State group consists of clearly distinct subordinate organizations or other subcomponents would the issue of mixed groups arise. Their sub-organizations must be unequivocally identifiable as distinct. Indicia can include organization charts displaying the components as separate; different budgets; personnel assignments “into” and “out of” the individual organizations; and, of greatest relevance, a leadership structure that dominates decision-making for each. Of course, since these components are ultimately accountable to the non-State group’s leadership, at some point chains of command or authority necessarily merge. However, this should occur above the tactical or operational level of command or authority if the organization is to be considered distinct.
Absent qualifying subordinate organizations, the entire non-State group is to be treated as an OAG, even though some individual members or insufficiently distinct entities within the group engage in activities wholly unrelated to the conduct of hostilities. This raises the subsequent disputed issue of whether all of its members are targetable—which is the U.S. view—or only those having a continuous combat function—as the ICRC suggests.
The challenge with the mixed non-State groups approach is how to identify those individuals who are members of a non-State group’s OAG and therefore targetable on that basis alone. The situation is complicated by the fact that some non-State groups—as is the case with the Hezbollah and its conventional forces and Islamic Jihad terrorist wing—include multiple subordinate organizations that conduct hostilities.
Problems with separate assessments of each subordinate organization
The tendency has been to look for subordinate organizations within a non-State group that qualify individually as an OAG on their own—so called “fighting wings.” By this approach, members of these fighting wings are treated as targetable OAG members, while everyone else is assessed for targetability based upon the direct participation in hostilities criteria. Importantly, individuals in the latter category may only be attacked, and need not be considered in proportionality analyses and precaution in attack assessments, while they are engaging in the activity that qualifies as direct participation. This is especially problematic for States adhering to the—in my estimation flawed—revolving door approach to the temporal aspect of direct participation.
Moreover, applying the OAG criteria separately to subordinate organizations within the non-State group can leave entities that contribute meaningfully and directly to combat operations beyond OAG status. Consider a non-State group’s intelligence agency that serves the group’s political leadership, law enforcement authorities, and the fighting wings. If assessed on its own activities alone, it would fail to qualify as an OAG because it does not satisfy the armed criterion. This is the case even if the organization is the primary source of tactical intelligence for the non-State group’s fighting wings. Additionally, during attacks on the intelligence agency’s facilities, which are military objectives, any harm to those for whom direct participation status could not be confirmed would count as civilian collateral damage in the proportionality analysis and the requirement to take precautions in attack.
This approach also oversimplifies the rationale for making the OAG assessment, which is to identify the de jure and de facto armed forces of a party to the conflict in order to distinguish them from the civilian population in a manner that fairly balances IHL’s foundational principles of military necessity and humanitarian considerations. In the above example, the fact that the intelligence is being provided by an entity outside the fighting wing should makes no difference to that balance. If an organization provides the essential military intelligence, it should be treated as an organization for targeting purposes, not as a collection of individuals who are assessed individually, for conduct of hostilities purposes.
Finally, the approach creates an imbalance between non-State groups and the States with which they are engaged in an armed conflict. With a few exceptions, all units and services of the armed forces are considered together as the “armed forces.” Services or units thereof are not judged individually for targeting purposes. And, no component ever benefits from the more restrictive direct participation targeting regime. For example, units that play no role in an armed conflict, and will not conceivably do so, nevertheless are fully targetable as a matter of law.
The “overall organized armed group” approach
An approach that better approximates IHL’s inherent military necessity-humanitarian balancing is to treat all organizations within a non-State group that operate collaboratively and cooperatively to conduct hostilities as a single “organized armed group”—the non-State group’s “overall organized armed group,” much as all the military units of a State comprise its armed forces. Doing so would dispense with the need to individually satisfy the three conditions precedent to qualification as an OAG and avoid having its members treated differently than would be the case for members of its enemy’s armed forces.
This begs the question of how to identify those organizations within the non-State group that are to be considered part of its overall OAG, and those that are not. Obviously, groups that perform functions essential to conducting hostilities, such as providing the intelligence necessary to mount ambushes or other attacks, would qualify. Beyond such clear-cut cases, a useful point of reference is the “constitutive elements” of direct participation in hostilities set forth by the ICRC in its Interpretive Guidance.
1. The act must be likely to adversely affect the military operations or military capacity of a party to an armed conflict or, alternatively, to inflict death, injury, or destruction on persons or objects protected against direct attack (threshold of harm), and
2. There must be a direct causal link between the act and the harm likely to result either from that act, or from a coordinated military operation of which that act constitutes an integral part (direct causation), and
3. The act must be specifically designed to directly cause the required threshold of harm in support of a party to the conflict and to the detriment of another (belligerent nexus).
Although designed for the very different purpose of identifying individuals who are not members of an OAG but still targetable, the three elements nevertheless can serve as a useful analog for assessing when a distinct subordinate organization within a non-State group is appropriately considered part of the group’s overall OAG. After all, if an individual may be attacked because he or she is directly participating in hostilities, it would appear incongruent to apply a higher standard to an organization that, as a whole, is similarly involved in the hostilities.
Consider an intelligence organization that is structurally not part of the non-State group’s fighting wing, but generates operational level intelligence, such as enemy force disposition, that the fighting wing’s leadership relies on for planning purposes. The enemy is adversely affected because the intelligence enhances the likelihood of the wing’s success, thereby altering the combat balance between the two sides. The causal nexus between the organization’s generation of the operational level intelligence and the harm caused the enemy is direct. And because it is military intelligence related to the conflict, the belligerent nexus is likewise self-evident. All three conditions are satisfied.
By contrast, take the case of a distinct subordinate organization of the non-State group that produces only political intelligence used by the group’s leadership for such purposes as facilitating ongoing negotiations with the enemy or identifying external States that might offer the group aid and assistance. The intelligence is useful at the strategic level of war and has indirect bearing on the operational and tactical levels of war but does not directly affect combat operations or the ability of either side to conduct them. The requisite causal relationship is missing. Or consider an organization that generates intelligence products for law enforcement purposes. The maintenance of law and order in territory under control of the non-State group will not directly affect the enemy’s military capabilities and lacks a belligerent nexus. In both situations, it would be inappropriate to treat the intelligence organizations as part of the non-State group’s overall OAG. Accordingly, targeting individual members of the organizations would only be permissible based on their acts of direct participation and only for such time as they engaged in them.
In some cases, a distinct subordinate organization of a non-State group shoulders responsibilities that are both military and nonmilitary in character, as with a law enforcement intelligence organization that produces intelligence for purposes that range from simple crime fighting and border control to military operations or even terrorist attacks. In that it is performing some functions that qualify it as part of the overall OAG, it should be considered as such.
However, in a few cases a non-State group’s distinct subordinate organization may itself be comprised of clearly distinct components identifiable as such by the other party to the conflict. If so, only those that are involved with the conflict should be considered part of the non-State group’s overall OAG for targeting purposes. To illustrate, assume that in the previous law enforcement organization example, the entity has different departments, only one of which is involved in the production of military intelligence. It is clearly distinguishable from the other departments, for instance by operating out of a different facility and wearing different uniforms; only this department would qualify as part of the non-State group’s overall OAG.
It must be emphasized in the strongest terms that to benefit from such targeting granularity, a component of a distinct subordinate organization must be plainly distinguishable and never perform functions that qualify it as part of the overall OAG. Should it fail to meet either of these requirements, it will be considered part of the overall OAG and its members may be attacked on that basis alone.
If international humanitarian law is to be respected by States facing well-armed and highly effective non-State group OAGs, it must be interpreted in a manner that fairly balances the principles of military necessity and humanitarian considerations and that ensures a level playing field for parties to the conflict. Mixed non-State groups present particular challenges in this regard. On the one hand, overly inclusive treatment of the OAG targeting rules will fail to accommodate valid humanitarian concerns with respect to the protection of individuals who have little to do with the armed conflict. But an overly restrictive interpretation of the notion will inevitably be shunned as inequitable by governments battling such groups. The concept of “overall OAG” is a means of achieving a fair balance lying between these unacceptable extremes. I commend it to States struggling to shape operations against mixed non-State groups.
Michael N. Schmitt is the G. Norman Lieber Distinguished Scholar at the United States Military Academy. He is also Professor of Public International Law at the University of Reading, Strauss Center Distinguished Scholar and Visiting Professor of Law at the University of Texas, and Charles H. Stockton Distinguished Scholar-in-Residence at the United States Naval War College.