Published by The Lawfare Institute
in Cooperation With
In the coming days, Iraqi forces will be mounting a new advance into eastern Mosul. U.S. troops have provided air support to Iraqi and Kurdish forces throughout the Mosul offensive and will likely continue in this new effort. As recently as October, the Office of the United Nations High Commissioner for Human Rights as well as multiple news outlets reported that ISIL has forced women and children from their homes to serve as human shields, putting these civilians in between coalition forces and themselves to compel coalition troops to stand down. The use of shields presumes that the targeter is more hesitant to execute an attack due to the civilian casualties.
The illegality of using human shields is well established: Additional Protocol I (twice) and the ICC Rome Statute bar the practice, as does customary international law. So, too, do U.S. military and domestic laws. (Beth Van Schaak has compiled an exhaustive list of legal regimes in play on Just Security.) It is also undisputed that neither the United States nor any other country is necessarily barred from attacking targeted shields. In the words of the 2015 Defense Department Law of War Manual, such a prohibition “would perversely encourage the use of human shields and allow violations by the defending force to increase the legal obligations on the attacking force.” In practice, however, the United States seems to nevertheless act as if such a rule is in place: the United States military does not make human shields the objects of attack. Indeed, it takes all feasible precautions to avoid civilian injury or death. Because, as Adil Ahmad Haque at Just Security states, “Daesh’s inhumanity seems to know no bounds,” ISIL may thus be wickedly emboldened to employ the tactic again in the hopes of staving off further advances. This post examines that practice and its implications for the efficacy of shields.
Legal and Policy Frameworks
i. Legal Obligations
The U.S. government can attack a shielded target within certain confines. As Law of War expert Michael Schmitt points out, “[t]he sole express provision [of international treaty law] bearing on the attacker’s obligations in shielding situations is Article 51(8) of Additional Protocol I: ‘Any violation of these prohibitions [including by deploying human shields] shall not release the Parties to the conflict from their legal obligations with respect to the civilian population and civilians, including the obligation to take the precautionary measures provided for in Article 57.’” Parallel provisions appear elsewhere in Additional Protocol I. Though not a party to Additional Protocol I, the United States, in the 2015 Defense Department Law of War Manual, cites the Protocol’s proportionality rule in adopting the rule itself. Thus, although there is no absolute bar on attacking human-shielded targets, an attack against shielded targets remains subject to the rule of proportionality, which mandates that the harm caused to non-military personnel or objects cannot be excessive with respect to the military advantage gained by an attack on a military objective.
ii. United States Policy
How the U.S. applies proportionality analysis to human shields depends on two considerations: (1) whether the shields are voluntary or involuntary; and (2) if the shields are voluntary, whether they are combatants, direct participants in hostilities, or civilians.
First, are the shields involuntary or voluntary? This distinction is significant because it determines whether such persons enjoy equivalent protections from attack as civilians. As explained in Joint Publication 3-60, the United States treats involuntary shields no differently than civilians for proportionality analysis: “otherwise lawful targets involuntarily shielded with protected civilians may be attacked, and the protected civilians may be considered as collateral damage, provided that the collateral damage is not excessive[.]” (Note that others, including Great Britain, “discount” involuntary shields relative to civilians in proportionality analysis, attributing a lesser value to them.)
As to how such conclusions are drawn, the United States has not publicly released any official guidelines for determining whether a civilian shield is acting voluntarily or is being involuntarily forced to cover a target. Thus, it appears that U.S. forces must make this determination based on all reasonably available intelligence, the specific circumstances leading up to a strike, and subject to a good faith requirement. The on-the-ground assessment of a shield’s classification is complicated by the fog of war, under which, as the Manual says, “information is often limited and unreliable.” Consequently, targeters need only “make [these] decisions in good faith and based on the information available to them.”
b. Direct Participation in Hostilities
If the shields are voluntary, however, a second inquiry is required: are the volunteers direct participants in hostilities? Answering this first requires defining what constitutes direct participation in the context of shields. According to the 2015 Defense Department Law of War Manual, to be considered “taking part in direct hostilities,” individuals must, “[a]t a minimum, tak[e] actions that are, by their nature and purpose, intended to cause actual harm to the enemy” (emphasis added). The ICRC agrees with the U.S. position.
Other commentators and sovereigns—namely Israel—disagree: Schmitt, for example, claims that because “[a] voluntary shield takes affirmative steps to frustrate harm to [military] objects (or persons) . . . he contributes to military action in a direct causal way; it is difficult to style his behavior as anything but direct participation.” In other words, by compelling one’s adversaries not to engage an otherwise military target, a voluntary shield by definition takes a role in the course of the hostilities. Meanwhile, Additional Protocol I offers a more restrictive targeting guideline in that it contains a presumption against direct participation that the United States has not adopted.
Once an individual has been determined to be a voluntary shield, however, U.S. military doctrine makes clear that such a conclusion makes that person targetable. According to Joint Publication 3-60, Joint Targeting, “[i]n cases where civilians voluntarily act as human shields . . . those civilians may be taking a direct part in hostilities and lose protection from attack. Such civilians need not be taken into account when assessing collateral damage under the law[.]” The inverse applies, of course. When voluntary shields are not deemed to be taking part in direct hostilities, they remain subject to the rule of proportionality and are thus treated akin to involuntary shields and civilians.
In short, whether and how the U.S. military can respond to human shield use in the Mosul campaign depends on a good faith determination of the human shields’ voluntariness and threat posed in light of typical proportionality concerns. And yet, though the U.S. government’s legal obligations and its policies leave the door open to attacking shielded targets, there is a difference between that which is permissible and that which is prudent. Recent evidence suggests that the United States is not likely to target human shields even when it has the legal authority to do so.
Recent U.S. Responses to the Use of Human Shields
In 1998, after Iraq expelled American arms inspectors, Saddam Hussein was said to have been preparing for airstrikes by shielding targets with civilians. It was reported that Iraqi citizens were “volunteering” to serve as human shields to repel such strikes, however whether such participation was voluntary or coerced is up for debate. Though some civilians were killed in the bombings, no reports suggested that they were deployed as shields, let alone targeted and attacked while serving as shields. And yet, Hussein credited those “who headed for the people’s palaces, factories and other installations to be a strong shield against the unjust aggression threatening our struggling country by the forces of evil and tyranny.”
Five years later, Hussein again used human shields before the United States’ 2003 invasion in Iraq. This time, however, Westerners actually volunteered to take the place of locals and Prisoners of War being used as human shields. While shields were deployed, U.S. forces neither targeted nor attacked any shielded targets. This military restraint was extraordinarily controversial—as some argued it encouraged further use of shields and thus bolstered Saddam’s position and escalated the war—and volunteers that took part were subject to civil proceedings but were not charged criminally.
Throughout the War on Terrorism, the Taliban also employed this horrific tactic. The Taliban shielded targets as early as November 2001, though the tactic was also reportedly executed in 2006, 2007, 2008, 2009, and 2010. (The Taliban has continued doing so to fend off other adversaries, including Afghan forces, as recently as June 2016.) When the United States faced these tactics—under both the Bush and Obama Administrations—there were no reports of the United States targeting the shielded object; shield deaths appear to have occurred only where Taliban members opened fire from shielded buildings and coalition forces returned fire in self-defense, accidentally killing involuntary shields.
Though we cannot know whether the U.S. would have targeted the shielded objects but for the shields’ presence, there is nevertheless a pattern of the U.S. military not targeting such objects in recent history.
Why Hasn’t the U.S. Attacked?
There are several possible explanations for the U.S.’s previous decisions to forego attacking shielded targets. Certainly, there is a strong moral component that commanders grapple with: they simply don’t want to kill civilians—including shields—except when absolutely necessary and especially not if there are other options to achieve the necessary effects. Alternative explanations may be that the destruction of the shielded objects may not confer sufficient military advantage, non-shielded alternative targets may have conferred an equal advantage, or tactical patience may have afforded opportunities for equal or similar effects without the negative consequences associated with killing human shields.
This last answer—tactical patience—speaks to a broader and perhaps more significant explanation: strategy. Killing civilians, even if legally justified, erodes local, U.S. domestic, and international support for the operation. The 2016 Executive Order, United States Policy on Pre- and Post-Strike Measures to Address Civilian Casualties in U.S. Operations Involving the Use of Force, summarizes this point well: “protection of civilians is fundamentally consistent with the effective, efficient, and decisive use of force in pursuit of U.S. national interests. Minimizing civilian casualties can further mission objectives[.]” And as stated in Gen. David Petreus’s August 1, 2010 Tactical Directive for all Iraqi Security Assistance Forces and U.S. Forces-Afghanistan, “Every Afghan civilian death diminishes our cause.” (This was the only underlined sentence of the entire directive.)
The idea of aiding a war effort by winning the “hearts and minds” of the enemy’s citizenry was popularized during the Vietnam War (though the verbiage stems from John Adams). Targeting shielded objects, and thus killing innocent civilians, is clearly antithetical to that goal. As stated by the 2006 Counterinsurgency Field Manual (COIN FM), “[a] successful COIN operation meets the contested population’s needs to the extent needed to win popular support while protecting the population from the insurgents . . . . Indeed, excessive use of military force can frequently undermine policy objectives at the expense of achieving the overarching political goals that define success.” What’s more, not only can losing local support make the operation that much longer or more dangerous due to a lack of cooperation, but it can also be used by the enemy as a recruitment tool, as the COIN FM notes.
Attacking shielded objects would also undermine domestic support for the operation. Again, the COIN FM is instructive: “At the strategic level, gaining and maintaining U.S. public support . . . is critical. [Senior military commanders] ensure that the conduct of operations neither makes it harder for elected leaders to maintain public support nor undermines public confidence.” The commencement of foreign hostilities typically enjoys the support of the American people (the “rally around the flag effect”) and decays over time due to reminders of the war’s costs. If attacking shielded targets undermines the operation’s efficacy, as described above, the public may become critical when confronted with rising insurgent numbers and pictures of deceased soldiers being brought back to Dover Air Force Base (known as the “the Dover Test”).
Lastly, the international community may not look favorably upon such tactics. The dissemination of “images of civilian casualties” across televisions worldwide (what Schmitt refers to as the “CNN effect”) could deter the United States’ allies from backing the operation diplomatically and or militarily, as well as induce additional criticisms domestically.
Other policy sources that speak to the United States’ treatment of civilians substantiate its reluctance to attack shields and is consistent with its strategy, suggesting that the practice of avoiding shield casualties will continue. The 2013 Presidential Policy Guidance Procedures for Approving Direct Action Against Terrorist Targets Located Outside the United States and Areas of Active Hostilities (PPG), which applies when “the United States takes direct action … against terrorist targets outside the United States and areas of active hostilities,” mandates that “absent extraordinary circumstances, direct action will be taken only if there is near certainty that the action can be taken without injuring or killing non-combatants.” This appears to foreclose the ability to attack shielded targets outside of hot battle zones. And as stated by the December 2016 Report on the Legal Policy Frameworks Guiding the United States’ Use of Military Force and Related National Security Operations, the PPG “prioritizes capture operations over lethal action” and permits the latter “only when capture of an individual is not feasible and no other reasonable alternatives exist to address the threat effectively.
On a more granular level in theater, the unclassified Tactical Directive from Afghanistan in 2011 enjoins soldiers to “spare no effort in preventing civilian casualties,” even at the cost of accepting greater risk to U.S. and NATO forces. The U.S. military even includes special considerations for protecting civilians in self-defense, when a service member’s life is at risk. The most recent unclassified Standing Rules of Engagement (SROE)—albeit from 2005—bars attacking civilians “except in self defense to protect yourself, your unit, friendly forces, and designated persons or property under your control.” The next line, however, reads: “Do not fire into civilian populated areas or buildings unless the enemy is using them for military purposes or if necessary for your self-defense. Minimize collateral damage.” Throughout much of the ISIL campaign, the Administration maintained a zero civilian casualty policy in airstrikes, such that 75 percent of strikes were called off due to intolerable risk to civilians through 2015. According to news reports, that policy has since been relaxed slightly, though it is still, as stated by the Executive Order, “more protective than the requirements of the law of armed conflict that relate to the protection of civilians.”
This begs an important question: If strategy and morality militate against attacking shielded targets, why maintain the right to so in the first place? It’s simple. Proclaiming that the United States will not attack a shielded target would encourage the widespread practice. Frankly, an enemy would be foolish not to use shields under such a policy.
But an even scarier conclusion looms large: if the U.S. continues its practice of not targeting shielded objects despite being legally entitled to do so, eventually our enemies will take note of this. There may actually be finite future circumstances when attacking a shielded target could be necessary despite all of the aforementioned reasons not to do so. It is foreseeable that ISIL and future enemies will encourage the use of shields and publicly announce the presence of civilians the presence of civilians in strategic buildings, such as Command, Control, Communications, and Intelligence (C3I) centers, to take advantage of the U.S.’s unwillingness to attack. This has recently happened in other conflicts with non-state actors. For example, in 2014, the Israeli Defense Force found a Hamas combat manual that encouraged the use of human shields. If confronted with the tactic, it will be interesting to see how the incoming administration reacts: Gen. James Mattis, President-Elect Trump’s pick for Secretary of Defense, co-authored the COIN Manual alongside Gen. Petraeus and oversaw many of the tactical directives in theater, which suggests staying the course, but the President-elect has advocated new foreign policy strategy and more aggressive in-theater tactics. Only time will tell.