The al-Zawahiri Strike and the Law of Armed Conflict
Just after daybreak on July 31, the United States killed Ayman al-Zawahiri, Osama bin Laden’s former deputy and the current leader of al-Qaeda. Two Hellfire missiles struck him on the balcony of his home in Kabul. Nearby residents reported a loud blast along with an impact tremor. The building, however, remained largely intact, other than the mangled balcony (see photos). There were no signs of the burning or charring associated with a traditional missile blast, nor have any casualties been reported.
This first reported case of an “over-the-horizon” (OTH) airstrike since the withdrawal of U.S. forces from Afghanistan in August 2021 ended a months-long effort using multiple intelligence sources to locate, target, and strike al-Zawahiri. Although officials have not specified the type of Hellfire used, various news outlets have reported that the operation was conducted using the nascent AGM-114R9X Hellfire variant (e.g., here and here). Nicknamed the “ninja bomb,” “sword bomb,” or “flying Ginsu,” it is laden with razor-sharp metal blades designed to shred targets instead of a traditional explosive warhead.
In this post, we assess the al-Zawahiri strike from the perspective of compliance with the law of armed conflict (LOAC). The strike illustrates how specific “means (i.e., weapons and weapon systems) and methods (i.e., tactics) of warfare” enable the United States to verify an intended target and limit harm to civilians and civilian objects (hereinafter “collateral damage”), thereby ensuring compliance with LOAC. It must be cautioned, however, that the means and methods used in this operation might not be appropriate in other situations. Nevertheless, the strike exemplifies how to design and execute a strike operationally to implement LOAC requirements and limitations.
Before turning to those issues, we note that the post does not examine the legal basis for conducting the attack in the first place (the jus ad bellum). Nor does it examine whether the United States and al-Qaeda are still involved in a non-international armed conflict and, if so, whether the LOAC applies in Afghanistan following the U.S. withdrawal (see here, here, and here). Instead, assuming the situation is a non-international armed conflict between a State and a non-State group like al-Qaeda, it examines three issues at the heart of the law of targeting: qualification as a lawful target, the requirement to take precautions in attack, and the rule of proportionality.
During a non-international armed conflict, it is permissible to attack members of an “organized armed group” (OAG)(see, e.g., NIAC Manual, §§ 1.1.2 and 2.1.1; ICRC Customary International Humanitarian Law Study, Rule 1; DoD Law of War Manual, § 17.5). This includes its leadership, for as noted in the DoD Law of War Manual, “leaders of non-State armed groups are also subject to attack on the same basis as other members of the group. There is no objection to making a specific enemy leader who is a combatant the object of attack” (DoD Law of War Manual, § 5.7.4).
An ongoing debate exists about whether only members of an OAG who have a so-called “continuous combat function in the group” (e.g., collecting intelligence, engaging in combat, etc.) are targetable at all times (see here at 21-24). However, a leader who is a decision-maker or otherwise directly involved in an OAG’s operational activities has such a function (i.e., command and control), thereby making that debate irrelevant in the al-Zawahiri case. It is clear that, under LOAC, al-Zawahiri was a lawful target.
Precautions in Attack
Under Article 57 of the 1977 Additional Protocol I (AP I) to the 1949 Geneva Conventions, parties to a conflict must take certain “precautions” before an attack to avoid collateral damage. Although the treaty only applies during international armed conflict and only binds its Parties, which the United States is not, it is widely recognized as generally reflecting the customary law on the subject (see, e.g., NIAC Manual, § 2.1.2).
The ICRC usefully set forth a fair articulation of the rule in its 2005 Customary International Humanitarian Law study.
In the conduct of military operations, constant care must be taken to spare the civilian population, civilians and civilian objects. All feasible precautions must be taken to avoid, and in any event to minimize, incidental loss of civilian life, injury to civilians and damage to civilian objects (Rule 15).
A decade later, the DoD Law of War Manual acknowledged the rule’s binding nature for U.S. forces: “Combatants must take feasible precautions in planning and conducting attacks to reduce the risk of harm to civilians” and other protected persons and objects (DoD Law of War Manual, § 5.11; see also § 5.2.3).
The emphasis on feasibility is critical, for an attacker need not take infeasible precautions. As the DoD Law of War Manualexplains, “The standard for what precautions must be taken is one of due regard or diligence, not an absolute requirement to do everything possible. . . . Feasible precautions are those that are practicable or practically possible, taking into account all circumstances ruling at the time, including humanitarian and military considerations” (§ 18.104.22.168). It goes on to observe that factors such as risk to mission, risk to force, and the benefit gained in relation to the cost expended (e.g., time, money, resources, etc.) matter. What is feasible, therefore, depends on what is reasonable based on a totality of the circumstances.
There are many ways to minimize harm to civilians during an attack on a lawful military objective. The DoD Law of War Manual offers examples such as warning of an attack, altering attack timing, selecting weapons, assessing anticipated collateral damage, identifying areas where civilians are less likely to be present, and canceling or suspending an attack based on new circumstances that raise collateral damage concerns (§ 5.11; see also Army and Marine Corps’ Commander’s Handbook on the Law of Land Warfare (FM 6-27/MCTP 11-10C), ¶ 2-82).
With respect to the al-Zawahiri strike, however, two stand out. Both were included in Article 57 of Additional Protocol I [art. 57(2)(a)(i) and (ii)] and have their own rule in the Customary International Humanitarian Law Study (Rules 16 and 17) — verification of the target and selection of methods and means of warfare. The DoD Law of War Manual includes the latter under “weaponeering” and reflects the former in its example of taking precautions to assess risk to civilians (§ 22.214.171.124), which necessitates verifying that the target is a lawful one.
Target Verification and the al-Zawahiri Strike
Rule 16 of the ICRC’s Customary International Humanitarian Law study provides, “Each party to the conflict must do everything feasible to verify that targets are military objectives.” (Rule 16)” (see also AP I, art. 57(2); U.K. LOAC Manual, ¶ 5.32.9; and here at 153-65). There is no reason to conclude the United States believes the rule is not customary, so long as the feasibility component, which includes operational good sense, is considered.
Multiple aspects of the al-Zawahiri strike reflect compliance with the obligation. News reports describe an exhaustive intelligence campaign to identify al-Zawahiri’s suspected location and confirm his identity (e.g., here and here). Upon suspicion that he would attempt to reunite with his family in Kabul after the U.S. withdrawal, the intelligence community narrowed its search to a secured neighborhood the Taliban protected. Necessarily using all sources of intelligence (e.g., human, signal, image, etc.), intelligence officers then compiled an array of information to develop an “actionable” target package. It presumably included such details as the layout of the building, composition of construction materials, and its structural integrity. It would also have incorporated an assessment of the “pattern of life” in the area throughout the day.
Pattern of life informally refers to intelligence analysis that uses historical patterns of human activity to predict future behavior. Such an analysis is especially critical when striking a target in which, or near which, there may be civilians. For instance, analysis could focus on which persons, including a target or civilians, would likely be in an area of interest at a given time. In the context of a residential building, relevant considerations include the varying uses of rooms, work and rest schedules, travel routines, daily rituals, social activities for the building’s inhabitants, and routine visitors (see the 2014 version of the U.S. Army ATP 2-33.4, Intelligence Analysis, ¶¶ 5-120–5-122). In this case, planners went so far as to build a wooden model of the target area for senior leader briefings to visually and physically depict how civilian casualties could be avoided.
The decision to employ an unmanned aircraft system (UAS, see generally here at GL-7 and Figure III-14, also sometimes labeled, inter alia, a remotely piloted aircraft or drone) as the attack platform further facilitated positive target verification. A UAS offers several benefits over other strike capabilities, such as a manned aircraft, a ship-launched cruise missile, or a raid. For example, UAS strikes are often more acceptable (the military term of art for a course of action that balances mission accomplishment against the potential loss of human life; see DoD Joint Publication 5-0, p. III-41) because there is no physical risk to the personnel remotely piloting the aircraft. Additionally, UASs have reduced visual, audio, and thermal signatures compared to jets and missiles, allowing them to evade detection more easily and remain in the operational area longer.
Such factors allow those who plan and execute attacks to engage in target verification with less risk than would otherwise be the case; a commander who is not worried about the loss of a pilot or a high-value resource (e.g., a piloted aircraft) may be more willing to take chances than would otherwise be the case. For instance, diminished risk means a UAS can loiter in the target area longer than a manned aircraft to better identify the target and to assess the likelihood of collateral damage. And an attacker can maneuver closer to the target to get more accurate verification or a more direct shot. In this case, for instance, the UAS crew executed the strike while al-Zawahiri was alone on the balcony.
Indeed, the mechanical reliability and mission endurance of UASs make them an attractive option when planners anticipate a brief window of opportunity for a strike, as was the case here. For example, the MQ-9A Reaper, MQ-1C Gray Eagle, and MQ-1 Predator, three UAS platforms that employ Hellfire missiles, have flight times of approximately 27, 42, and 24 hours, respectively. Moreover, UAS crews can relieve one another mid-mission, so as a practical matter, endurance is limited by fuel capacity and environmental conditions—not, as with manned aircraft, aircrew stamina. Accordingly, compared to alternatives, UASs buy more time during which the target and possible collateral damage may be reliably verified and assessed.
And the combination of greater loiter time and less risk permits more accurate verification of the target using on-board UAS sensors. The Reaper and Gray Eagle, for instance, are equipped with advanced electro-optical/infrared imaging sensors that use multiple spectrums of light to provide daytime, low-light, and nighttime detection and recognition capability (see, e.g., here). As in the al-Zawahiri operation, this allows for greater certainty when confirming the target, even at high altitudes. And a key benefit of a UAS using on-board sensors is that this can be done in real-time, which may not always be the case with other weapon systems.
Choice of Method and Means of Warfare
Rule 17 of the Customary International Humanitarian Law study provides, “Each party to the conflict must take all feasible precautions in the choice of means and methods of warfare with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects.” The DoD Law of War Manual offers the example of “weaponeering (e.g., selecting appropriate weapons, aim points)” (§ 5.11). Selecting a missile, for example, that will minimize collateral damage illustrates choice of a “means” of warfare. Selection of the aim point, which can enhance the effect on the target or avoid collateral damage, exemplifies the choice of a “method” as a precaution in attack.
In deciding to attack al-Zawahiri while he was on the balcony, rather than, for instance, destroying the building in which he was present, the United States chose a method that minimized risk to civilians, in this case, his family and other non-fighters who might have been present. The ability to verify the target and assess risk to civilians, as described above, made this possible.
However, it was the means of warfare used that best illustrates this precaution in attack. As the DoD Law of War Manual notes, attackers can take feasible precautions by “selecting munitions of appropriate size and type, as well as appropriate aim points, while offering the same or superior military advantage in neutralizing or destroying a military objective” (§ 5.11.6). Accordingly, this attack employed an AGM-114 Hellfire missile launched by a UAS.
There are two principal Hellfire variants: the Longbow, typically employed as a “fire and forget” anti-armor missile, and the Hellfire II semi-active laser (SAL) guided missile. Hellfire IIs can be launched from select UASs, including Reapers, Predators, and Gray Eagles, and guide on laser energy reflected off a target designated by on-board or other systems (e.g., another aircraft or ground personnel). The DoD classifies the Hellfire II as a “precision strike” weapon; as discussed below, its precision capabilities were self-evident in this case.
There are multiple Hellfire II subvariants with different characteristics and functions. The K and N subvariants use explosive warheads or penetrating, metal-augmented charges to attack armored or soft-skin vehicles, buildings, or other structures. The R subvariant, on the other hand, is a multipurpose weapon system that allows the user to select specific warhead effects based on the applicable target. Although the United States has not acknowledged the particular munition used in the Zawahiri strike, the characteristics of the strike suggest that a modified Hellfire R—the R9X—was used. While the munition’s exact specifications are not publicly available, analysts have assessed its capabilities by piecing together details of prior strikes.
Most notably, the missile lacks an explosive warhead. Instead, the payload section is equipped with metal blades that swing open moments before impact to smash, slice, or shred the target (see a detailed image of the blade hinges from a detonated missile here). The kinetic energy from the approximately 5-foot, 100-pound, subsonic projectile can even penetrate thin protective covers such as the skin of vehicles or buildings. Photographs of the aftermath of vehicle strikes in 2017 and 2019 indicate the blades extend the weapon’s diameter to approximately 5 feet. Given the lack of blast or fragmentation damage characteristic of conventional high-explosive missiles, the munition is generally only capable of targeting individuals or tightly packed groups.
In the LOAC context, this limitation is a positive feature. It makes the R9X an ideal weapon for limiting collateral damage. Due to the weapon’s concentrated engagement radius, the United States was able to engage al-Zawahiri on his balcony while minimizing harm to any civilians who might have been in the building or otherwise nearby. External photographs of the building following the attack show no burn or fragmentation damage and no apparent damage to the building’s structural integrity. Only the balcony appears to have been affected. These effects are similar to previous strikes in Syria, where the R9X has penetrated vehicle roofs to kill those inside without placing civilians in the area at significant risk (see, e.g., here and here).
Thus, the United States reduced the risk of harm to civilians during the al-Zawahiri strike by both “selecting a munition of appropriate size and type” and an “appropriate aim point” (i.e., the balcony) as is required by LOAC when feasible in the circumstances (DoD Law of War Manual, § 5.11.6).
Once all feasible measures under the circumstances have been taken to minimize collateral damage without sacrificing military advantage, attackers must consider the rule of proportionality. Codified in Articles 51 and 57 of Additional Protocol I for Parties to the instrument during international armed conflict, the rule is customary in character for all armed conflicts. As described in the DoD Law of War Manual, it requires an attacker to “refrain from attacks in which the expected loss of civilian life, injury to civilians, and damage to civilian objects incidental to the attack would be excessive in relation to the concrete and direct military advantage expected to be gained” (§ 5.10; see also Customary International Humanitarian Law, Rule 14; NIAC Manual, § 126.96.36.199).
Compliance with the rule of proportionality is based on what the attacker reasonably believed when the attack was planned, approved, and executed, not the ultimate results. Given the significant anticipated military advantage of permanently taking the leader of an enemy-organized armed group off the battlefield and the reasonable expectation that there would be little collateral damage, the strike on al-Zawahiri clearly complied with the rule of proportionality.
The al-Zawahiri strike is a textbook example of how to execute a surgical strike against a lawful target after having taken feasible precautions under the circumstances to minimize collateral damage, and in compliance with the rule of proportionality. It must be cautioned, however, that every operation must be assessed individually and contextually; feasibility, after all, includes considerations of military sensibility. The fact that these measures were possible in this operation should not be interpreted as suggesting that they will always be feasible.
Michael N. Schmitt is the G. Norman Lieber Distinguished Scholar at the United States Military Academy at West Point. He is also Professor of Public International Law at the University of Reading; Professor Emeritus and Charles H. Stockton Distinguished Scholar-in-Residence at the United States Naval War College; and Strauss Center Distinguished Scholar and Visiting Professor of Law at the University of Texas.
Major William C. Biggerstaff is a military professor at the Stockton Center for International Law at the U.S. Naval War College, where he co-teaches a course on the Law of Armed Conflict.
Photo credit: TSGT Scott Reed, U.S. Air Force