As noted by Alex Whiting in his piece last week, the law of armed conflict, or international humanitarian law (IHL), contains broad principles and prohibitions that are applied to a set of concrete facts in the context of any armed conflict. This analysis happens both ex ante—as a target set is being identified and an attack is launched—and ex post—when a completed operation is being evaluated for its compliance with IHL, including the war crimes prohibitions.

The Saudi-led coalition’s August 11 attack on a bridge in Yemen helps to exemplify the difficulty in applying these concepts to real-world examples.  The details of the attack remain rather sparse (some of the best reporting on this topic is here).  What we do know is that the bridge in question was located on the main road from the port city of Hodeidah to the capital of Sanaa.  In addition, we know, and presumably the attackers also knew, that an enormous amount (90% to be exact) of the food coming from the U.N. World Food Program and other humanitarian actors apparently traversed that bridge en route to civilians in need.  This food aid is helping to keep the Yemeni civilian population alive: over half of the country’s population (14 million souls) are suffering from malnutrition and food insecurity.  According to UNICEF, more than 300,000 children under the age of 5 have acute malnutrition, which can lead to death and long-term developmental, cognitive, and other impairments. It’s been reported that the United States considered the bridge to be a piece of vital infrastructure and thus included it on a “no strike” list.  The State Department has accordingly denounced the attack:

We have seen those reports, and if the bridge was deliberately struck by coalition forces, we would find this completely unacceptable. The bridge was critical for the delivery, as you note, for humanitarian assistance. Destruction will further complicate efforts to provide assistance to the people of Yemen.

The spokesperson later clarified that her statement was meant to be a “condemnation”—strong language when it comes to international diplomacy. At the same time, she also rebuked the most recent (of many) Saudi attack on a Médecins sans Frontières hospital. The International Committee of the Red Cross and Amnesty International, among others, have blamed Saudi Arabia for deliberately targeting hospitals and medical facilities since that country first overtly intervened in the conflict in March 2015.

This post walks through the analysis necessary to evaluate the attack on the bridge and concludes that the attack in question should have been precluded had the attackers respected the principle of proportionality in light of the facts as we know them. The post concludes with a quick discussion of the potential war crimes that could be charged depending on the facts that emerge and other legal implications of the attack.

International Humanitarian Law Principles

The key concepts for evaluating the legality of the destruction of the bridge are found in Additional Protocol I to the four Geneva Conventions of 1949. Unlike its “parent” treaties, API has not yet achieved universal ratification (there are 174 members at present, including Saudi Arabia), but many of its terms are considered to be part of customary international law (CIL) by the International Committee of the Red Cross (ICRC), states parties, and even non-party states.   Although API by its terms applies to international armed conflicts, many the most important rules apply across the conflict spectrum by way of CIL and appear in other IHL treaties addressed to non-international armed conflicts (NIACs), like the one in Yemen. For example, the Convention on Certain Conventional Weapons and its Protocols embody verbatim many of the same targeting rules as contained in API.

The Principle of Distinction

The principle of distinction is foundational to IHL. As applied to objects, this principle dictates that:

The parties to the conflict must at all times distinguish between civilian objects and military objectives. Attacks may only be directed against military objectives. Attacks must not be directed against civilian objects.

The first question is thus: was the bridge a military objective?  If not, then the attack is per se unlawful (unless it could be shown to be truly accidental, which was clearly not the case here).  Even if the bridge can be characterized as a military objective, the attack on it may still be unlawful if any collateral damage to the civilian population was disproportionate to the military advantage to be gained.

Military Objectives v. Civilian Objects

The definition of “military objective,” as set forth in Article 52(2) of API, is generally cited as definitive. (Note, however, that there is some lingering debate about whether the definition is broad enough to cover war-sustaining objects as discussed in greater detail in Ryan Goodman’s new paper as well as my and Kenneth Watkin’s analyses on these pages). The standard definition of military objective reads:

Attacks shall be limited strictly to military objectives. In so far as objects are concerned, military objectives are limited to those objects which

  • by their nature, location, purpose or use make an effective contribution to military action and
  • whose partial or total destruction, capture or neutralization, in the circumstances ruling at the time, offers a definite military advantage.

This definition contains two prongs addressed to two sets of actors: the attacker and the defender. The desired target must make an effective contribution to the military action of the defender (by its nature, location, purpose or use); and its destruction, capture, or neutralization must offer a definite military advantage to the attacker.  In practice, these two prongs tend to collapse into each other given that it is difficult to imagine one prong being satisfied without the other.

In terms of the first prong, the proposed target must make an “effective” contribution to military action, but not necessarily a “direct” one. An impact could thus be indirect but still effective. The term “military action” is not defined, but is clearly broader than the concept of a “military operation.” A particular target can offer a military advantage that is felt over a lengthy period of time; it can also affect military action in areas other than in its immediate vicinity.  At the same time, it is not entirely clear how far removed from military activities the object’s contribution can be to still qualify as a military objective susceptible to direct attack.

A feature like a bridge can be a military objective by means of its location, use, or purpose.  “Purpose” is often conceived of as “intended future use” in contradistinction to “use,” which implies current utility.  The difficulty of knowing, at the time target sets are being drawn up, the intended future use of some objects without extremely granular intelligence should not be underestimated.  Targeting a civilian object because it might potentially make an effective contribution to military action at some future date is obviously problematic. Theoretically, every civilian object could—at some point in time and under some circumstance or another—be used for military action.  Rendering such objects universally targetable on the basis of a commander’s ability to articulate some future military use would vitiate the principle of distinction altogether.

In terms of the second prong of the definition, the advantage gained to the attacker in destroying the proposed target must be substantial and relatively proximate rather than hypothetical, speculative, or distal. It is often stated that the military advantage can be evaluated by the attack as a whole and not necessarily isolated components of it, although how far this reaches remains somewhat controversial.

IHL treaties and articulations of CIL contain a presumption of civilian status that applies both to persons and things.  In cases of doubt about how to classify and object or potential target, it shall be presumed that it is not being used to make an effective contribution to military action.  The U.S. Department of Defense has taken the position that this presumption is not part of CIL, however, because such a rule would shift the burden of determining the precise use of an object from the defender to the attacker (i.e., to a party that does not exercise control over the object).  Specifically, the Department of Defense’s Law of War Manual at § states:

Under customary international law, no legal presumption of civilian status exists for persons or objects, nor is there any rule inhibiting commanders or other military personnel from acting based on the information available to him or her in doubtful cases.

Although there is no definitive list of legitimate military objectives, categorical examples include weapons caches, barracks, and military transports. Conversely, places of worship, schools, museums, monuments, and hospitals—so long as they have not lost their civilian character by being employed in military action—are considered civilian objects.  The ICRC lists “civilian means of transportation” as an example of “prima facie civilian object[s].”

Back to the bridge: From the facts available in open-source accounts, it does not appear that the Houthi rebels were using the bridge for military action at the time it was attacked.  That said, hindering an opponent’s ability to transport material, armaments, or personnel; fragmenting an adversary by isolating units from each other and command and control apparatus; and cutting off transportation routes would clearly offer the attacker an advantage.  Indeed, the DoD’s Law of War Manual at § uses the destruction of a bridge to exemplify the concept of military objective:

the military advantage in the attack of an individual bridge may not be seen immediately (particularly if, at the time of the attack, there is no military traffic in the area), but can be established by the overall effort to isolate enemy military forces on the battlefield through the destruction of bridges.

It was on this rationale that NATO forces targeted the bridge over Grdelica Gorge in Kosovo in 1999 during Operation Allied Force. That bridge formed part of a re-supply route being used by Serb forces. Unfortunately, the attack also hit a passenger train that came into view after the first bomb had locked.  The Yugoslavia Tribunal’s Office of the Prosecutor issued a Report on the Operation that determined that further investigation was not warranted because the bridge was a legitimate military objective and the passenger train was unfortunate collateral damage.

It seems plausible that the Houthi rebels would use the bridge in Yemen for military purposes, given that it was on the main road between the port and the capital. The rebels have controlled the strategically-vital Red Sea port (the second most important after Aden) and parts of Sanaa on and off during the conflict, and a commander could easily explain the military necessity of closing off transportation routes between rebel-held areas. (A 2015 map showing evolving areas of control is here; I can’t seem to locate anything more current).  All these facts would potentially transform the bridge into a legitimate military objective.

Incidentally, if the bridge in Yemen were historic or culturally symbolic, it might also benefit from the protection accorded to cultural property, which—like civilian objects generally—generally enjoys immunity from attack unless it is used for military purposes.

Dual Use Objects

Dual use objects—objects with both military and civilian use—exist somewhere in between these military/civilian archetypes (although it is insisted that is no intermediate legal category).  Elements of an adversary’s transportation, power-generation, or communications infrastructure are regularly considered to be dual-use and may be attacked only if they are also military objectives per the definition above.

That said, dual use objects are subject to a proportionality calculus: they may only be attacked so long as the harm to the civilian population is not excessive when compared with the expected military advantage. The U.S. Department of Defense’s Law of War Manual states at §

If an object is a military objective, it is not a civilian object and may be made the object of attack. However, it will be appropriate to consider in a proportionality analysis the harm to the civilian population resulting from the destruction of such a military objective.

Back to the bridge: even if the bridge in Yemen was a proper military objective (through its current or potential use transporting Houthi personnel or materiel), the attack may still be unlawful if disproportionate force was used in light of the military advantage to be gained and the concomitant harm to the civilian population.

The Principle of Proportionality

The principle of proportionality dictates that:

Launching an attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated, is prohibited.

The precise formulation of this test indicates that at least in international criminal law (putting state responsibility to the side for the moment), the legality of an attack is evaluated based upon what the commander knew about the expected harm and advantage anticipated ex ante rather than in hindsight.  As a result, the actual harm caused to civilians and civilian objects is technically not the touchstone of whether an attack was disproportionate.  The results of an attack may, however, be highly relevant to inferring the intent of individuals involved in planning and implementing the attack.  In other words, if the harm to civilians is excessive, it may be inferred that civilians were actually the intended target notwithstanding the proximity of a plausible military objective.  The rules of state responsibility may also be implicated where there is significant harm to a civilian population by state action in the aggregate, regardless of any particular commander’s advanced knowledge.

Finally, the formulation of proportionality dictates that the anticipated harm must be “excessive” when compared to the advantage to be gained, which arguably sets a rather high bar.  (Indeed, some formulations—including that within the Statute of the International Criminal Court—add the modifier “clearly” before “excessive”).

Duty to Take Precautions and Protect the Civilian Population from the Effects of Attacks

In order to be in compliance with these obligations, warring parties are required to take “all” feasible precautions to minimize civilian harm. According to the ICRC’s CIL Study:

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.

In this regard, states must make affirmative efforts to protect the civilian population from the effects of attacks.

A range of precautions are available to belligerents, including target verification exercises, considerations about the timing of an attack, avoiding populated areas, warning civilians co-located with military objectives, incapacitating versus destroying particular objectives, and selecting precise weapons and weapon systems. When a choice is possible between several military objectives that would result in the same military advantage, “the objective to be selected must be that [which] may be expected to cause the least danger to civilian lives and to civilian objects.”

Calculating Proportionality

Calculating proportionality is difficult and depends on balancing the relative risk of collateral damage with the anticipated military advantage.  Such a calculus depends, of course, on assigning values to these two sets of variables. The risk of incommensurability is ever present in this exercise: there may be no universal agreement on the relative “weight” to be accorded to factors that would be placed on each side of the scale. How such weight is assigned will inevitably depend upon the perspective, background, and priorities of the observer. In general, the standard in international criminal law is one of a “reasonable commander,” but applying this standard can still result in indeterminacy and disagreement.

Although it is obvious that proportionality takes into account actual physical harm to civilians and civilian objects, it is unclear to what degree commanders are expected to calculate future hardships or other grave consequences that would be experienced by the civilian population following the destruction of a dual use object.  For example, destroying a power grid may offer an immediate military advantage by incapacitating an opponent, but it may also severely impact the civilian infrastructure causing longer-term and more diffuse harm, including to schools, water treatment facilities, refrigeration units, and hospitals.  The U.S. Joint Chiefs of Staff’s 2013 publication on targeting thus emphasizes:

If the attack is directed against dual-use objects that are legitimate military targets but also serve a legitimate civilian need (e.g., electrical power or telecommunications), then this factor must be carefully balanced against the military benefits when making a proportionality determination.

The same may be true of elements of a transportation system.  For example, the graceful Mostar Bridge (Stari Most), which dated from 1566 and the Ottoman occupation, once spanned the River Neretva in Bosnia-Herzegovina. It was destroyed in 1993 during the dissolution of the former Yugoslavia. In Prosecutor v. Prlic, the International Criminal Tribunal for the former Yugoslavia (ICTY) ruled that although Bosniak forces used the Mostar Bridge as a supply line—such that it constituted a legitimate military target—its destruction by Bosnian Croat forces caused disproportionate harm to the Muslim civilian population of Mostar. These civilians were effectively cut off from supplies and their sources of livelihood, which exacerbated the war-time humanitarian situation. In particular, the tribunal noted that the destruction of the bridge made

it impossible for [civilians] to get food and medical supplies resulting in a serious deterioration of the humanitarian situation for the population living there.

Although it did not specifically invoke the prohibition against attacking cultural property, the tribunal also took into account the cultural and symbolic value of the bridge, which had united communities for centuries, and the psychological impact of its collapse. In light of the bridge’s crucial civilian utility, the militants deemed responsible for its destruction were convicted of the war crime of carrying out the “wanton destruction of cities, towns or villages or devastation not justified by military necessity,” a crime set forth in Article 3 of the ICTY’s Statute.

Turning once again to the bridge in Yemen, it seems likely that an authoritative body would agree that the bridge constituted a military objective if it were in fact being used by Houthi rebels to transmit supplies or personnel.  At the same time, its essential role in the provision of humanitarian assistance to civilians should also have been considered ex ante in a rigorous proportionality analysis.  This crucial functionality may explain why the United States reportedly put the bridge on a “no strike” list.  The principles of proportionality and the obligation to consider precautions would dictate that the Saudi-led coalition should have explored other ways to incapacitate the rebels’ ability to resupply themselves, for example, by targeting Houthi vehicles or supply depots.

Potential War Crimes

Not every violation of IHL gives rise to individual criminal responsibility. Some violations—such as the failure to take precautions—implicate only state responsibility under the law as it currently stands.  Nonetheless, states remain under a continuing duty to prevent breaches of these rules as well.

CIL and a number of IHL treaties do, however, consider deliberately, indiscriminately, or disproportionately attacking civilian objects to be a war crime.  The ICRC in its CIL Study lists the following war crimes involving civilian objects that are potentially relevant to the attack on the bridge depending on whether the bridge was in fact in use by the rebels:

  • Making civilian objects, that is, objects that are not military objectives, the object of attack;
  • Extensive or wanton destruction of property, not justified by military necessity and carried out unlawfully and wantonly;
  • Launching an attack in the knowledge that such attack will cause incidental loss of civilian life, injury to civilians or damage to civilian objects which would be clearly excessive in relation to the concrete and direct military advantage anticipated; and
  • Making buildings dedicated to religion, education, art, science or charitable purposes or historic monuments the object of attack, provided they are not military objectives.

If evidence emerges that the attack was undertaken with the intent of terrorizing or harming civilians (e.g., through blocking humanitarian aid to Houthi supporters), an additional set of war crimes is potentially actionable:

  • Unlawfully inflicting terror on the civilian population;
  • Using starvation of civilians as a method of warfare by depriving them of objects indispensable to their survival, including by impeding relief supplies;
  • Making persons or objects involved in a humanitarian assistance or peacekeeping mission in accordance with the Charter of the United Nations the object of attack, as long as they are entitled to the protection given to civilians or civilian objects under international humanitarian law;
  • Launching an indiscriminate attack resulting in loss of life or injury to civilians or damage to civilian objects;
  • Launching an indiscriminate attack resulting in death or injury to civilians, or an attack in the knowledge that it will cause excessive incidental civilian loss, injury or damage;
  • Wilfully causing great suffering or serious injury to body or health.

These crimes find varying degrees of expression in the IHL treaties, the statutes of the international tribunals, and domestic law across the conflict spectrum. The mens rea also differs depending upon the crime; some crimes require a showing of intent; others, mere recklessness (the conscious disregard of a substantial and unjustified risk in the language of the Model Penal Code).  For example, the crime of wanton destruction of property not justified by military necessity requires a showing that the perpetrator acted with the intent to destroy the property or “in reckless disregard of the likelihood of its destruction.”

Deliberate attacks directed against the civilian population can also rise to the level of persecution as a crime against humanity if a racial, religious, ethnic or other group is targeted.

Other Legal Implications

The possibility of war crimes prosecutions depends, of course, on the existence of a court with jurisdiction and the power to charge a range of NIAC war crimes.  There will also be state responsibility implications, both for Saudi Arabia and its allies and other states that may be providing assistance. The Principles of State Responsibility state at Article 16:

A State which aids or assists another State in the commission of an internationally wrongful act by the latter is internationally responsible for doing so if:

(a) That State does so with knowledge of the circumstances of the internationally wrongful act; and

(b) The act would be internationally wrongful if committed by that State.

Finally, the commission of war crimes will also impact international arms sales to the warring parties.  The 2014 Arms Trade Treaty (ATT), for example, states at Article 6(3) that:

A State Party shall not authorize any transfer of conventional arms … if it has knowledge at the time of authorization that the arms or items would be used in the commission of genocide, crimes against humanity, grave breaches of the Geneva Conventions of 1949, attacks directed against civilian objects or civilians protected as such, or other war crimes as defined by international agreements to which it is a Party.

The ATT already has over 80 ratifications. The United States has signed but not ratified the treaty, but many of the treaty’s provisions are already consistent with U.S. arms control/export laws.  Human rights groups and other non-governmental organizations have condemned arms sales to Saudi Arabia and its coalition partners, particularly by ATT states parties such as France, Germany, and the United Kingdom.