Anyone serious about combating ISIS and minimizing civilian casualties needs to consider the importance of the Rules of Engagement (ROE) in Iraq and Syria.  We offer our thoughts on the ROE in the hope of deepening, if not starting, conversation. First, we examine the application of ROE in light of the ISIS tactic of packing civilians into a building and then baiting an attack.  Second, we address the broader question of how the ROE, including changes to approval levels for calling in strikes, might be applied to reduce civilian casualties in urban warfare.

Feasible Precautions and ISIS Tactics

The use of human shields as a method of warfare has an unfortunately long history that has been addressed in the context of Syria at some length here and elsewhere.  ISIS has added its own bloody twist to this particularly abhorrent (and unlawful) practice.  As a coalition spokesman recently noted, “what you see now is not the use of civilians as human shields… [instead] ISIS is smuggling civilians into buildings so we won’t see them and trying to bait the coalition to attack.”  This tactic is qualitatively different than the classic use of human shields.  The very point of a human shield (voluntary or involuntary) is to signal the presence of civilians on a military objective so as to increase the burden on the attacking force—either through a changed proportionality calculation or through the imposition of more onerous precautions in attack.  Surreptitiously filling a building with civilians and then baiting an attack does not seek to make the attack more legally complicated, but rather intends (unlawfully) to cause civilian casualties for the purpose of an information operations campaign.  Given this tactic, what are the implications for the attacking force?

Before ISIS used this tactic, coalition planners and commanders would plan an attack based on their knowledge of a given target in the circumstances ruling at the time, including the construction of a building, the function of the building, nature and type of surrounding buildings, etc.   With this information, planners could estimate through a set of institutionalized procedures—the “collateral damage estimation methodology”—potential civilian casualties for a given attack.   Understanding the threat to civilians and civilian objects is critical to an informed proportionality analysis as well as effective precautions in attack.  Thus, an attack on an arms cache in a warehouse on the outskirts of town, late at night might be judged to present a relatively low risk of civilian casualties.  When, however, ISIS has clandestinely fills the building with civilians, this would change the analysis considerably.

It is a widely accepted that customary international law imposes an obligation on attacking forces to take “feasible precautions” to minimize harm to civilians and civilian objects.  This requirement is captured in Article 57 of Additional Protocol I, and US military doctrine in the Department of Defense Law of War Manual (¶ 5.11).  Article 57(1) requires that “constant care shall be taken to spare the civilian population, civilian and civilian objects.”  Article 57(2) provides several specific precautions in attack including notably, an obligation to “do everything feasible to verify that the objectives to be attacked are neither civilians nor civilian objects” and the obligation to choose a “means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss to civilian life, injury to civilians and damage to civilian objects.”  The DoD Law of War Manual utilizes similar, but distinct language:  “parties to a conflict must take feasible precautions to reduce the risk of harm to the civilian population and other protected persons and objects.” (¶ 5.2.3). (The differences in language are not relevant to our analysis here.)

Precautions in attack require verification that the object of attack is a lawful target, and verification that the target can be attacked lawfully, that is to say, proportionally.  As the DoD Manual makes clear, “the requirement to take feasible precautions in planning and conducting attacks and the prohibition on attacks expected to cause excessive incidental harm are fundamentally connected and mutually reinforcing obligations.” (¶ 5.10.5).  Verification is informed through current available intelligence. As the Bothe, Partsh, Solf commentary makes clear, “the obligation to do everything feasible … involves and obligation to assign a high priority to the collection, collation, evaluation and dissemination of timely target intelligence.”  This is incumbent on commanders at all echelons of command.

A key element of the precautions requirement is feasibility.  As recalled in the official commentary to Article 57, the drafting process involved significant debate over the phrase “everything feasible.”  Notably, as recalled in Bothe, Partsh, Solf, the drafting committee rejected an absolute standard (e.g., States shall “ensure”…) in favor of a qualified standard (States shall “do everything feasible”…).   A number of countries made reservations regarding this article to emphasize that feasibility was to be judged in the circumstances at the time.   The DoD Law of War Manual (¶ expresses a similar qualification, noting that “feasible precautions are those that are practicable or practically possible, taking into account all circumstances ruling at the time.”

We suggest that absent intelligence to the contrary, it is not feasible (or reasonable) to assume every building is filled with civilians.  It may be the tactic was employed only once by a single ISIS commander and will never be used again.  That said, additional precautions must be a taken if intelligence indicates that this tactic is being used regularly, or will be used in the future.  It is incumbent upon commanders to assess the intelligence available at the time—to include knowledge of ISIS’ tactics—in determining the nature and extent of precautions that can be taken.  The implementation of such precautions leads us to our second point of discussion:  how the ROE can be designed to reduce casualties in urban warfare.

ROE and Civilian Casualties

One approach could be to impose very restrictive ROE.  This is the most obvious lever to reduce civilian casualties, but it is important to consider the intended and unintended consequences of doing so.  In the instant example, an operational or strategic level commander could retain approval authority over the targeting of structures.  Alternatively, this same commander could impose additional conditions on the targeting of structures such as the requirement to use infrared sensors so as to more accurately detect humans in a structure; or, a requirement to observe the target continuously for, say, 48 hours before the attack.  An advantage to this top-down approach lies in the doctrinal distinction between different echelons of command, which U.S. Army doctrine explains as follows:

At the strategic level, leaders develop an idea or set of ideas for employing the instruments of national power (diplomatic, informational, military, and economic) in a synchronized and integrated fashion to achieve theater, national, and multinational objectives. The operational level links the tactical employment of forces to national and military strategic objectives, with the focus being on the design, planning, and execution of operations using operational art … Finally, the tactical level of warfare involves the employment and ordered arrangement of forces in relation to each other.

In short, strategic and operational commanders are focused on planning long-term operations with a view to strategic objectives, while tactical commanders are focused on day-to-day operations.  Unsurprisingly, the resources available to commanders increase up the chain of command.  By way of example, an infantry company commander will have virtually no intelligence-generating assets at their level.  At best they may have a RQ-11 Raven, a hand-held unmanned aerial vehicle (UAV) with a very short operating range and dwell time.  An infantry brigade commander will have a military intelligence company that generates intelligence as well as organic UAVs such as the RQ-7 Shadow, which has approximately four times the range and dwell time as a Raven.  An infantry division will have multiple subordinate military intelligence companies as well as an aviation brigade with several dozen helicopters and substantial UAV capabilities.  And so on up the operational chain.  In a deliberate (as opposed to dynamic) targeting situation, the assets available to strategic and operational commanders (typically Division-level commanders and above) assist with reducing uncertainty related to positive identification and the presence of civilians.

Restrictive rules of engagement may also have the unintended consequence of encouraging tactical commanders to develop other courses of action that they can approve at their level.  This potentiality raises a disadvantage of restrictive ROE.  When the ROE is made more restrictive, the mission does not change—just the execution.  Thus, rather than wait for higher-level approval in time sensitive situations, commanders may simply pursue other courses of action to achieve their mission.  These alternative courses of action may result in increased risk to friendly forces or civilians.  Consider a circumstance in which coalition forces are taking fire from a warehouse, but the ROE requires coalition commander’s approval for indirect fires on all structures.   The tactical commander may decide he doesn’t have time to request permission, and instead orders an infantry assault.  This decision is both lawful and tactically sound, yet may result in increased casualties for both combatants and civilians.

Another approach to fighting in complex urban environments would be to empower the lower level commander with more authority and less restrictive ROE.  Tactical commanders may not have the robust intelligence apparatus of a higher unit, but they have greater situational awareness with regard to ongoing operations and enemy tactics in their area of operations compared to those up the chain of operational command.  Less restrictive ROE is also in keeping with the US Army doctrine of Mission Command.  In particular, Mission Command emphasizes the use of mission orders, which, in turn, “emphasize to subordinates the results to be attained, not how they are to achieve them” in order to achieve the subordinate’s “disciplined initiative within the commander’s intent.”

In the example above, the tactical commander may be in a far better position to judge the threat to his or her forces than the approving commander, who may be located on a different continent.  Thus, counter-intuitively, a less restrictive ROE may have both military and humanitarian benefits in certain circumstances.

There are, however, disadvantages to less restrictive ROE, which have been a recent topic of discussion.  Tactical commanders, unlike strategic and operational commanders, have limited assets to resolve uncertainties related to the presence of civilians and positive identification.  Such limitations may increase the risk to civilians.  For example, enemy tactics designed to increase civilian casualties like the tactic mentioned above may go unnoticed by the tactical commander if such tactics were not used in the commander’s area of responsibility.  Yet the strategic commander may be well aware of such tactics and capable of responding in a measured way to reduce civilian casualties.  This is not to say that tactical commanders are incapable of limiting civilian casualties by exercising the requisite restraint and tactical patience.  Their decisions, however, will depend on the information received from the available assets and the time they have to make the decision.

The point of comparing these two approaches is to highlight that restricting or relaxing the ROE does not directly correlate to decreases or increases in civilian casualties.  Restrictive ROE do not always result in fewer civilian casualties and less restrictive ROE do not always result in more civilian casualties.  Since there are advantages and disadvantages to these approaches, leaders should consider both of them and objectively weigh the risks to the civilian populace in a given area of operation.

* * *

Regardless of the approach to ROE, restrictive or otherwise, the legal standard for precautions will be based on what is “feasible.”  Thus, in dynamic circumstances, information available to a strategic commander may not be available to a tactical commander, and vice versa.  Consequently, what is feasible for the strategic or operational commander may not be feasible for the tactical commander.  In any event, coalition forces in Syria should carefully consider ISIS tactics and current intelligence when determining the feasibility of precautions in attack.  Understanding the benefits and the risks involved will be critical to defeating ISIS in a complex urban environment while also limiting civilian casualties.


The views expressed here are the authors’ own and do not necessarily reflect those of the Department of Defense, the United States Army, the United States Military Academy at West Point, the Naval War College, or any other department or agency of the United States Government. The analysis presented here stems from his academic research of publicly available sources, not from protected operational information.

Image: U.S. soldiers get briefed while their unit conducts a key leader engagement during exercise at U.S. Army Joint Multinational Readiness Center in Hohenfels, Germany, April 11, 2015- Dept of Defense