Skip to content
Jerusalem Center for Public Affairs
Strategic Alliances for a Secure, Connected, and Prosperous Region

Proportionality in Modern Asymmetrical Wars

Filed under: Al-Qaeda and Global Jihad, Israel, Israel Defense Forces (IDF), Radical Islam
Publication: Special Reports

Download the pdf

Executive Summary

  • Asymmetrical conflicts are fought between a state following the laws of armed conflicts or international humanitarian law, and organizations that almost never follow these rules and have very little incentive to do so. While the Geneva Conventions and their protocols were framed in an era of “classic” military engagements, when wars were fought between nations and by armies that observed the rules of armed conflict, we should examine whether these norms are suited to modern armed conflicts.
  • In practice there exist two very different approaches to the interpretation of the principle of proportionality: the human rights model, which gives preference to the interests of civilians who might be harmed by military action, and the contractual model, which gives precedence to state interests. Yet a third approach may be more suitable: the administrative model, based on respect for the professional discretion of the commander in the field, with some necessary limitations.
  • The concept of proportionality permits military personnel to kill innocent civilians – provided that the intended targets of the operation are enemy forces and not civilians.
  • In October 1993 a force of U.S. troops was caught inside the streets of Mogadishu, Somalia. Some 18 American soldiers were killed in a battle where women and children were being used as shields. Over one thousand civilians are believed to have been killed by American fire.
  • In its air campaign against Serbia in 1999, NATO adopted a policy of zero risk to its soldiers. This meant that pilots flew at a relatively high altitude, which also enlarged the risk to civilian lives on the ground. The number of civilian deaths in the NATO campaign was around 500.
  • American forces first attempted to retake the city of Fallujah, one of the centers of the Iraqi insurgency, in April 2004. The operation was halted after a few days due to the large number of civilian casualties. In November 2004, U.S. and Iraqi forces attempted to take Fallujah again. Reports of the number of people killed range from a few hundred to several thousand. The tactics used by U.S. forces included the use of white phosphorous. Vast quantities of fire power were employed in an urban setting known to house civilians, in order to protect the lives of American soldiers.
  • As the uses of force in Somalia, Kosovo, and Iraq show, Western armies are very concerned about protecting the lives of their soldiers, and to that end are willing to risk many civilian lives. They also find acceptable the notion that civilian lives can be forfeited in order to attain important military goals.
  • Proportionality cannot be detached from the question of responsibility: which side created the situation in which civilians find themselves? From a military as well as a moral perspective, the onus clearly lies with the party that chooses to fight from within civilian concentrations.
  • Once the non-state actor internalizes the fact that his foe is committed to the protection of civilians, even at the expense of military actions, that actor will use civilians as shields that might protect him from enemy assaults. This is precisely the tactic adopted by Hamas and Hizbullah.
  • Israel’s Gaza operation clearly shows that Israeli commanders successfully followed the requirements of the administrative model of the principle of proportionality. According to newspaper and oral reports, the IDF did require commanders to take humanitarian law into account in the planning stages of the operation. Moreover, legal advisors were involved in the planning of many operations and provided advice regarding specific targets. The right questions were asked, checks were made, and the incidental damage to civilians was on the whole limited.


Envision a state involved in an armed conflict with a non-state actor – a terrorist organization, a militia, or an organized armed band. The non-state actor embeds itself within the civilian population. Its fighters dress as civilians; they hide their weapons and equipment in civilian houses and places of worship; they launch rockets from school precincts. Moreover, these combatants deliberately fight from within the civilian population. Every time they are attacked, they seek protection by surrounding themselves with civilians (who voluntarily or under duress participate in these actions).

Recent armed conflicts have increasingly accorded with the model described above. Such, for instance, was the U.S. experience in Somalia, and later in Iraq and Afghanistan. Such was the Sri Lankan experience with the Tamil Tigers, such was the Russian experience in Chechnya, and the Georgian experience in South Ossetia. Such was the Israeli experience in Lebanon (2006) and Gaza (2008-9). This study looks at the special context of these conflicts. It also seeks to draw more general conclusions about the nature of the law of war with regard to modern armed conflicts.

Armed conflicts of this type have sometimes been termed “asymmetrical”1 – an adjective used principally with reference to the fact that the protagonists are a state, with all its might and force, and an organization with few heavy arms and a limited number of fighters. But such conflicts are also asymmetrical in a more complicated sense: they are fought between a state, in possession of sound reasons for following the laws of armed conflicts (LOAC) or international humanitarian law (IHL),2 and a high incentive and organizational obligation to do so, on the one hand, and on the other hand, an organization that almost never follows these rules and has very little incentive to do so.

Asymmetrical conflicts are fought between a state following the laws of armed conflicts or international humanitarian law, and organizations that almost never follow these rules and have very little incentive to do so.

States involved in these conflicts mostly attempt to follow, or are expected by the international community to follow, IHL as detailed in customary international law, in the Geneva Conventions, and in other sources of applicable international law.3

However, it has become increasingly difficult to abide by these laws, mainly because of the novel nature of the problems that constantly arise. This brief review will only deal with two of the most prominent of such problems:

  • The first is how to apply the rule forbidding indiscriminate attacks on a civilian population when the enemy deliberately operates from within that environment. Direct attacks against civilians are of course always forbidden. However, what are the appropriate norms that a state should apply when the only possible way of fighting the enemy involves risking the lives of civilians whom the enemy is using for its own protection?
  • A second problem arises from the fact that non-state actors are not susceptible to the range of formal and informal sanction which may be used against states. Since international law is not policed effectively, non-state actors may readily assume that their violations of the laws of war, including those mentioned above, will not be punished by law. For example, they may target civilians of the state actor in the knowledge that there exists very small chance that they will be punished for doing so by any international judicial body. Consequently, while one side to the conflict behaves in accordance with IHL, the other considers itself to be free of the limitations imposed by these rules.4

Consideration of these and similar issues has motivated some scholars and politicians to call for the redefinition or reinterpretation of the rules of armed conflicts. The Geneva Conventions and their protocols, runs their argument, were framed in an era of more “classic” military engagements, when wars were fought between nations and by armies that observed the rules of armed conflict. The norms that may have been suitable in such situations are not suited to modern armed conflicts.

Changes to IHL may indeed be required, although it will prove very difficult to actually convince states to adopt them. However, before tabling dramatic changes, it might be useful to evaluate the current state of IHL, and ask whether some of the problems discussed above cannot be solved within the existing framework, without a need for reconstruction.

This study begins (Part I) by outlining the principle of proportionality in IHL, which has proven to be the major source of contention regarding the implementation of IHL in the asymmetrical conflicts described above. Understanding the problems involved in the proportionality principle constitutes the first step towards solving this issue.

Parts II and III analyze the different interpretations given to the principle of proportionality by state practice, judicial decisions, and in studies undertaken by academics and NGOs. In practice there exist two very different approaches to the interpretation of this principle: the human rights model, which gives preference to the interests of civilians who might be harmed by military action, and the contractual model, which gives precedence to state interests.

In practice there exist two very different approaches to the interpretation of the principle of proportionality: the human rights model, which gives preference to the interests of civilians who might be harmed by military action, and the contractual model, which gives precedence to state interests.

Part IV outlines an approach which may be more suitable: the administrative model, based on respect for the professional discretion of the commander in the field, with some necessary limitations. This part explains what this model means, the implications of this interpretation, and applies these principles to the Israeli operation in Gaza.

Part I: Proportionality and Its Requisites

Proportionality in IHL is a difficult concept for field commanders, legal experts and philosophers to analyze, and much more so to accept. On the one hand, it permits military personnel to kill innocent civilians – provided that the intended targets of the operation were enemy forces and not civilians. Not even knowledge in advance that civilians might be hurt outlaws an operation – unless the estimated civilian casualties are excessive relative to the military advantage that the prospective attack seems likely to confer. On the other hand, the principle of proportionality limits military action even when a legitimate military target is attacked, when the attack may cause excessive damage to civilians. Hence, an exact understanding of the norm is required.

A. What Is Proportionality in IHL?

Although the term “proportionality” does not explicitly appear in any IHL treaty,5 it boasts a long pedigree within the laws of war, according to some scholars dating back to the Middle Ages.6 However, even as the laws of war and armed conflict developed, states were very reluctant to limit their freedom of operation in armed conflict by the principle of proportionally which, as we shall see, possesses the potential to severely limit a commander’s flexibility regarding the use of force. So, while certain facets of the principle of proportionality were adopted in specific contexts,7 there existed no general pronouncement on the subject.

A major breakthrough in this area was not attained until the Additional Protocol to the Geneva Conventions of 1977 (First Additional Protocol).8 This document includes several specific clauses that are considered to embody the concept, of which the clearest is Article 51(5). The article as a whole warrants extensive citation, not least because it also embodies the principle of distinction.

Art. 51 – Protection of the Civilian Population

1. The civilian population and individual civilians shall enjoy general protection against dangers arising from military operations. To give effect to this protection, the following rules, which are additional to other applicable rules of international law, shall be observed in all circumstances.

2. The civilian population as such, as well as individual civilians, shall not be the object of attack. Acts or threats of violence the primary purpose of which is to spread terror among the civilian population are prohibited.

3. Civilians shall enjoy the protection afforded by this section, unless and for such time as they take a direct part in hostilities.

4. Indiscriminate attacks are prohibited. Indiscriminate attacks are: (a) those which are not directed at a specific military objective; (b) those which employ a method or means of combat which cannot be directed at a specific military objective; or (c) those which employ a method or means of combat the effects of which cannot be limited as required by this Protocol; and consequently, in each such case, are of a nature to strike military objectives and civilians or civilian objects without distinction.

5. Among others, the following types of attacks are to be considered as indiscriminate [and therefore prohibited – AC]: (a) An attack by bombardment by any methods or means which treats as a single military objective a number of clearly separated and distinct military objectives located in a city, town, village or other area containing a similar concentration of civilians or civilian objects; and (b) 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.

As thus contextualized, the principle of proportionality, as it appears in Article 51(5), constitutes a form of indiscriminate attack. This might seem a little odd, since disproportionate attacks are not directed against civilians. The framers of the protocol posited a similarity between a foreseen damage to civilians in excess of the military advantage and a deliberate attack on civilians. We shall return to this point later.

As can be seen, the principle of proportionality, as embodied in Article 51(5) and other similar articles in the First Protocol, is based on two complementary ideas.9 One is that measures have to be taken to limit the harm that efforts to attain military goals will cause to civilian populations. Geographically, the military target should be defined in the narrowest possible way. Similarly, the attacking power should consider whether there exists a method of attaining the military objective with less or no damage to the civilian population. Thus employed, proportionality constitutes a logical extension of the principle of distinction: everything possible should be done to target exclusively military objectives.10

The second section of Article 51(5) is more innovative and embodies a second facet of proportionality. It makes the balance between military necessity and humanitarian interests horizontal rather than vertical – i.e., instead of military necessity justifying any damage to civilians, it orders the attacking power to audit his proposed operation, comparing the foreseeable damage to the civilian population with the expected military advantage.11 It requires the army to relinquish the effort to gain a military advantage if its attainment threatens to cause disproportionate harm to the civilian population. Damage to the civilian population becomes prohibited once it is seen to be excessive in relation to the military advantage. This equation, which requires the commander to carry out extremely delicate calculations in the heat of battle, has generated much confusion and controversy.

The idea of proportionality is even more unclear because it appears within an article that outlaws indiscriminate attacks – a prohibition based on the principle of distinction. IHL requires that the attacking army target only military objectives, installations, etc.12 However, the principle of proportionality states that even if an attack is directed against a military target (and hence not prohibited by the principle of distinction), it might still be prohibited if not proportional – i.e., if the attack directed at the military target would cause greater harm to civilians than the military advantage gained. Hence, distinction and proportionality are two separate concepts, with proportionality limiting the permitted scope even of attacks that are otherwise allowed.13

B. Questions about Proportionality

As noted earlier, in Article 51(5) the concept of proportionality appears in its most radical sense: the article requires armed forces to forgo some actual military advantage if the incidental civilian suffering can be expected to exceed the military gains. Of the several questions to which this interpretation gives rise,14 the most salient relate to the impossibility of balancing rights and interests. It is difficult enough to evaluate the value of human lives against military advantages.15 But even if this hurdle is overcome, there remain several practical questions concerning the application of the norm of proportionality with respect to attacks that affect civilian targets. The formula of proportionality requires the interpretation of several different terms. In all cases the interpretation is debated.

(1) The Definition of Attacks

What exactly is meant by the term “attack”? Does it imply that before firing his rifle each and every soldier should assess the possible consequences of his actions, and estimate the possible incidental damage that his shots might cause? Or is there a requirement for a more cumulative test, which takes into consideration the overall objectives of the military operation and the incidental damage involved?

The protocol provides little guidance as to this issue, and the interpretations submitted by state parties in their declarations differ considerably. Probably, the most extreme cases could easily be excluded – i.e., we do not have to adjudicate every single shot fired, nor limit examinations of proportionality to the overall conflict.16 At issue are “military operations”– a given tactical process during the course of which several armed activities take place.17 Nevertheless, the grey area remains considerable. Advocates of a strict interpretation of the term “attack” favor focusing on the specific offensive, while those supporting a more flexible interpretation tend to accept a more liberal understanding of the term – the entire tactical operation.

(2) Military Advantage

Lack of clarity is similarly characteristic with respect to the protocol’s reference to “military advantage,” which does not specify the sort of military advantage that might be taken into consideration. The text does attempt to limit the application of the term to military advantages that are “concrete and direct,” which the International Committee of the Red Cross (ICRC) interprets to mean “substantial and relatively close” with regard to causation.18 Hence, advantages that are very general and possess no clear causal connection to the occurrence should be disregarded. This limitation seems to exclude several major military operations of the recent past.19

Interpretations submitted by several states stress that the military advantage taken into consideration should be that which results from the action as a whole, and not simply from one of its isolated or particular components.20 This viewpoint seems to substantiate the view that the advantage gained should be assessed not as resulting from a single military action, but from the campaign as a whole. Of course, some limits should be put on this understanding,21 but what they are remains unclear.

(3) Which Civilians Should Be Taken into Account?

“Civilians” too is a very imprecise term. Does it include non-military personnel employed on a military base? When the enemy intentionally places its facilities within a civilian population, for the explicit goal of protecting his armed forces, should the lives of civilians here be taken into account by the attacking army?

(4) Civilian vs. Military Casualties

How does one measure the excessiveness of civilian casualties with regard to possible danger to the life of soldiers? In other words, does the protection of the lives of one’s soldiers constitute a permissible criterion in the equation? How much weight should be given to the protection of the other party’s civilian population? Are casualties to be measured on a precise one-to-one basis?

Several states have explicitly made known their position that the term “military advantage” includes the security of the attacking force.22 The committee appointed by the prosecutor of the International Criminal Tribunal for Former Yugoslavia (ICTY) concluded that states are permitted to protect their soldiers by resorting to an aerial campaign, even though it places a greater number of civilian lives at risk.23 However, whether or not that conclusion reflects a correct statement of international law is much disputed by international lawyers.24

What about the lives of one’s own civilians? How should they be measured against the lives of the enemy’s civilian population? Is a state required to risk the lives of its own citizens in order to protect those of the other side? Similar questions arise with respect to targeted killings of terrorists.25 Clearly, commanders of such operations must take into account the likelihood that their implementation endangers anyone in the immediate vicinity of the terrorist who is targeted. However, they must also weigh the threat that the terrorist, unless killed, will inevitably pose to the intended victims of his planned suicide bombing. Is there a gauge for comparing the price to be paid by the two sides?

The Future Is Unclear

The most obvious difficulty generated by the concept of proportionality relates to the feasibility of rational choice in battlefield situations. In essence, the concept of proportionality assumes that the attacker has an option. He can either take action, thereby bringing to bear his full military advantage, or he can relinquish a small part of his military advantage, in order to significantly limit the damage that he will cause to civilians. Confronted with that choice, proportionality requires the attacker to choose the latter course.

In reality, however, military operations very rarely offer such a clear-cut choice. In most cases, the dangers and potential of the battlefield are unknown. The commanding officer simply cannot make an informed assessment of the harm to the civilian population vis-a-vis the gains from a specific military attack. Hence, the problem with proportionality is that it seems to place an unrealistic burden on the attacker to make a cost-benefit analysis in the midst of a battle.

The problem with proportionality is that it seems to place an unrealistic burden on the attacker to make a cost-benefit analysis in the midst of a battle.

The result of this situation is that, as the special report to the prosecutor of ICTY regarding the NATO campaign in Yugoslavia pointed out: “[i]t is much easier to formulate the principle of proportionality in general terms than it is to apply it to a particular set of circumstances.”26 In fact, one may argue that the inability to offer more precise guidelines derives from the very nature of the principle of proportionality. It is an open-ended legal standard designed to accommodate an indefinite number of changing circumstances, not a hard and fast set of rules.

Hays Parks, who perhaps more than any other scholar has invested time on this issue, concluded in 1990:

Following more than a decade of research and meetings of international military legal experts, who are anxious to implement the language contained in protocol I, to the extent that it advances the law of war and the protection of civilian population, there remains a substantial lack of agreement as to the meaning of the provisions in protocol I relating to proportionality. This is a rather disconcerting situation, given that other lawyers claim that this principle is part of Customary International Law.27

In the two decades since those words were written, relatively little has been done to rectify that situation.

Part II: Human Rights vs. Contractual Models: Two Understandings of Proportionality

A. The Balance of Proportionality

In a recent study Eyal Benvenisti suggested that a tension between two views is embedded in the norms and principles of IHL. Armies view IHL as a “compact between rival armies” in an attempt to limit the atrocities of war.28 On the other hand, humanitarian organizations, international war crimes tribunals, and other international organizations view the aims of IHL as that of protecting civilians, regardless of their nationality.29 This tension is reflected in the attempt to understand the principle of proportionality, and is evident also in the interpretation of proportionality.

B. How NGOs View Proportionality

(1) The General Approach

From the perspectives of those who emphasize the need for complete protection of civilians, proportionality is indeed problematic. The respect for human rights, and life, rejects the possibility that injury caused by armies to civilians might be classified as collateral damage. In this view, nothing validates the subjugation of civilians to lethal attacks simply because they happen to be situated in the vicinity of a military target.30 Any such prospect fails to respect the rights of human beings as individuals, and punishes them for a crime they did not commit and could not possibly avert. However, the necessities of war dictate that some form of attack against hybrid (military and civilian) targets would be allowed; otherwise, the enemy could station all its military resources near a civilian population and thus render them immune to attack.

Proportionality, then, is an exception to the general rule of protecting human life.31 Attacks that harm civilians are allowed only because no possible alternative exists. A complete ban on all civilian casualties would be impossible to implement, and would only result in a growing disregard of IHL rules. Therefore, according to this narrative, the protocol adopts a very strict attitude when allowing a limited exception to this prohibition. That exception too, however, should be interpreted in the most restrictive way possible, and should never serve as a license to kill civilians other than under the strictest limitations.

This negative tendency towards proportionality was articulated even when the principle was written into the First Additional Protocol. At the 1977 international conference, states voiced their opposition to any kind of statement that would allow the killing of civilians, even if it was incidental to the military advantage.32

A similar tendency can be found in the commentary to Additional Protocol I published by the ICRC. When interpreting Article 51 the authors of the commentary state:

The idea has also been put forward that even if they are very high, civilian losses and damages may be justified if the military advantage at stake is of great importance. This idea is contrary to the fundamental rules of the Protocol; in particular it conflicts with Article 48 (Basic rule) and with paragraphs 1 and 2 of the present Article 51. The Protocol does not provide any justification for attacks which cause extensive civilian losses and damages. Incidental losses and damages should never be extensive [my italics–AC].33

As Dinstein notes, the authors of the commentary here confused the terms “excessive,” which appears in the protocol, and “extensive.”34 This change, however, was based not on a mistake, but on the ideologically-based conviction that IHL may not be used to cause an extensive quantity of civilian deaths, no matter what states insist on writing into the protocol.

(2) Fact-Finding and NGO Missions

A similar attitude towards the principle of proportionality appears in reports by NGOs whose members consider it their mission to protect human rights, and by certain fact-finding bodies commissioned by the UN. The reports on armed conflicts submitted by such bodies actually provide a dual service regarding the application of the laws of war. First, and in accordance with their primary function, they attempt to verify the facts on the ground. At the same time, however, most fact-finding missions and NGO reports also attempt to apply the law to the specific facts, and indicate whether or not the law had been violated.

As far as the question of proportionality is concerned, the function of these missions is even more complicated. Since (as we have seen) the law itself is not clear, members of the missions have to interpret the law before they can apply it to the specific case at hand. As I have already suggested elsewhere,35 it is the nature of the reports submitted by such bodies that they try to avoid ambiguities in the law. If indeed the principle of proportionality is ambiguous, and actually relies to a considerable extent on the discretion of the specific commander in the field, then the report cannot say very much about whether or not it was violated. Hence, many reports evidently attempt to avoid the question of proportionality altogether. Amnesty International’s and Human Rights Watch’s reports on the Second Lebanon War constitute a clear case in point36 Their underlying assumption (based almost entirely on evidence supplied by Lebanese villagers who had a clear interest in denying any Hizbullah activities in their localities) is that Israeli forces were never fired upon from villages. Hence, all Israeli attacks on villages were by definition aimed at civilian targets, and illegal per se. In almost no case did Amnesty accept the Israeli position that shots had been fired from civilian dwellings, thus rendering them legitimate military targets, and that therefore the true legal question concerns proportionality, not distinction.

(3) Case Study: The Goldstone Report

The question of proportionality is raised in the Report of the UN Fact-Finding Mission on the Gaza Conflict (the Goldstone Report),37 which dealt with the Israeli operation in Gaza in December 2008-January 2009.38 For the most part, however, this report too follows the convention outlined above and therefore attempts to avoid complex discussions of the issue. For example, the report continuously rejects Israel’s claims that mosques served as places of storage for military supplies, and that shots were fired from them.39 Notwithstanding Israel’s provision of several independent, and even Palestinian, sources to the contrary,40 the Goldstone Report simply rejects all claims that hospitals and ambulances were used for military purposes.41

Against this background, one can conclude that, as a rule, reports submitted by international NGOs and some UN missions reflect a very strong presumption against a state-based attacker who claims that a civilian facility had been put to military use and had thus become a military target. States submitting a claim of that sort have to be able to provide conclusive proof, of a credibility close enough to that which would be required in order to prove the assertion in a court of law. This seems an absurd requirement, and it will be discussed later.

Even the Goldstone mission was unable to deny that in some cases civilian premises did indeed constitute legitimate military targets, and that therefore the principle of proportionality should be applied.

The first instance in reference to which the Goldstone Report raises the issue of proportionality concerns the Israeli attack on elements of the Gaza police force on the first day of the operation, December 27, 2008.42 Members of the Goldstone mission rejected Israel’s claim that all police officers in Gaza are by definition members of Hamas.43 However, they could not deny that several members of the Hamas police force were indeed members of the Izz ad-Din al-Qassam Brigades (the military wing of Hamas) and hence personally constituted legitimate targets of Israeli attack. As the Goldstone Report concedes, whether or not the presence of these persons within the police force permit Israel to target the entire police framework is a question controlled by the rule of proportionality. Were this principle taken seriously, one would have expected to see: (1) a detailed breakdown of the names of the policemen killed in the Israeli attacks; (2) an attempt to verify which of them could – and could not – be classified as legitimate military targets; and (3) an application of this information to the proportionality equation in some way. Not unexpectedly, the mission was incapable of conducting so detailed a study and, therefore, could only limit itself to general claims and statements of uncertainty. But instead of dropping the matter due to lack of sufficient information, the mission – without any discussion whatsoever of the issues itemized above – went on to claim that Israeli actions were disproportional. Indeed, this conclusion is pronounced in just one sentence.44 Once again, it can only be assumed that the mission proceeded from the presumption that any civilian killed is killed illegally. In order for the attacker to be excused of a charge of illegal behavior, it is his duty to provide strong and concise evidence in his defense.

The Goldstone Report discusses the principle of proportionality in somewhat greater detail within the context of the shelling near the UN school at al-Fakhura on January 6, 2009.45 Confronted with hard evidence, in this case the mission reluctantly accepted that shots might have been fired from the street near the school in the direction of an IDF force advancing in the area.46 The Palestinians used 120mm mortars and, according to an Israeli report, endangered the lives of soldiers. After locating the source of fire, about 80 meters from the UN school, the Israeli force responded with mortars, after which the firing ceased. The number of people killed in this incident is not clear, but the mission assumes that they numbered around 33. The Israeli claim, on the other hand, is that far less people were killed, though naturally Israel was not in possession of all the information. The mission could not ascertain the identities of all the fatalities.

Furthermore, while Israel claimed to have exchanged mortar fire in order to protect the lives of Israeli soldiers, the mission had no knowledge about the specific danger to the soldiers that was avoided.47 The mission accepts that a state has a right to take actions to protect its own soldiers, but nowhere in the report is there an attempt to verify the risks that soldiers should take, in view of possible loss of civilian life. In other words, the mission was almost totally ignorant when it came to the variables that ought to be applied to the proportionality equation. All it knew is that among those killed in this incident were two women and two children (one of them a 13-year-old boy) from a family whose house had been bombed earlier that day by the IDF.48 The mission also assumed, once again without any factual proof, that Israel could have attained the same military result by employing weapons which would have caused fewer civilian fatalities.49 The mission concluded that the use of mortars was in this case disproportional. So, even though it knew close to nothing about the incident, the mission still decided that a violation of the principle of proportionality had occurred.

This incident also illustrates several other points which frequently appear in international mission reports that have a bearing on proportionality. One is the rejection of the reasonable assumptions made by the commander in the field. Not a single fact adduced by the Goldstone Report on the incident indicates why a reasonable commander could not have assumed that the use of mortars was in this case proportional. On the contrary, the Israeli report shows, and the facts on the ground prove, that before giving orders to fire the commander took all necessary precautions to verify that the UN school, in which refugees from other areas were located, would not be harmed.50 Significantly, not one of the fatalities was killed within the school compound. Hence, instead of analyzing the commander’s state of mind (as should have been the case), the mission focused on the results of the attack.

A second issue concerns the shelling. As noted earlier, proportionality should be adjudged in the light of the full spectrum of the “tactical process” underway, and not on the basis of a single shell or bomb. To read the Goldstone Report, however, is to gain an impression that the reasons for the military unit’s presence in the area are not at all important. All the mission analyzed were the results of a specific round of shelling. It neither possessed nor sought information regarding the military advantage that the IDF military unit engaged in this particular operation was supposed to attain.51 This too seems to be a common failing in NGO reports, which appear to work on the proposition that military advantage can be assessed only in relation to the evaluation of a specific target.

Lastly, despite paying lip-service to Israel’s right to protect its soldiers, the analysis presented by the mission appears to proceed from the position that lives of soldiers do not really constitute part of the proportionality equation. Without explicitly saying so, the report implies that the danger to soldiers should never be considered equal to dangers to civilians. Civilians always count for more.52

In sum, what the Goldstone and NGO reports offer is the human rights model of the proportionality principle. This term reflects the perspective that there exist no real differences between the requirements for an assessment of proportionality in general human rights law (which applies universally to all cases of rights abuse by governments) and those of a military conflict. The conflation of the two spheres is exemplified in the strong presumption that every civilian death in a military conflict is illegal. It is further exacerbated in the conditions allowed to a state to counter this presumption. In order to do so, a state has to demonstrate – almost beyond doubt – that the target was a legitimate military target; the specific military gains it actually received from the specific use of power; and that the military gains did not simply consist of a reduction of the risk to the lives of soldiers.53 This degree of proof accords with that expected of a government that seeks to justify the use of force in conditions of domestic unrest. It seems that when analyzing the principle of proportionality, NGOs tend simply to take the legal norms which apply in times of internal unrest and apply them to military conflicts.

However, the human rights model as described here is a deviation from treaty law, pronouncements of judges, state practice, morality, and logic.

The human rights model of the principle of proportionality is a deviation from treaty law, pronouncements of judges, state practice, morality and logic.

C. The Contractual Model

While NGOs and fact-finding missions attempt to apply the human rights model, some commentators suggest that states actually view their obligations in accordance with a contractual model. As mentioned, Eyal Benvenisti suggested that armies view IHL as a “compact between rival armies” in an attempt to limit the atrocities of war.54

Scholars who support that view claim that if non-state actors do not follow IHL, then it makes no sense for the other side to weaken itself by unilaterally observing rules that its enemy will not recognize. In its extreme version, this view rejects the relevancy of IHL in its entirety to all situations of armed conflicts – a view that seems to underlie many decisions taken by the Bush administration in the wake of the attacks perpetrated on September 11, 2001, especially with respect to the status of al-Qaeda and Taliban combatants kept in Guantanamo.55

Yet even a less radical version of the contractual model (i.e., one that does not go as far as to reject altogether the applicability of IHL) would consider the rule of proportionality to constitute a limitation on the right to belligerent reprisals. A note about the role of reprisals in IHL is in place: There is a long-standing debate within IHL over whether any kind of reprisal against civilians is permissible. It is clear that in the past, reprisals were considered an acceptable means of enforcing IHL obligations, even though the right of reprisal was restricted by several important rules, which allowed it to be used only in a very limited set of circumstances.56 It is also clear that the use of belligerent reprisals is prohibited against persons defined as “protected” by the Geneva Conventions of 1949.57 Whether reprisals against civilians not otherwise protected by the Geneva Conventions are actually prohibited in international law is hotly contested. Additional Protocol I prohibits reprisals against civilian populations altogether.58 The UN General Assembly has explicitly stated that they are forbidden,59 and many military manuals prohibit them. The ICTY in its Kupreskic case implied that it views reprisals against civilians as prohibited by customary international law.60 However, several states – some of which, like Israel, are involved in conflicts – are not party to this protocol,61 while others – even though they are parties to the protocol – have submitted reservations to this specific article.62 It is clear that in at least one armed conflict, the Iran-Iraq war in the 1980s, armed reprisals against civilians did in fact take place.63 Yoram Dinstein explicitly rejects the notion that the prohibition on reprisals against civilians has become customary,64 and even the ICRC study on customary international law takes a cautious stance, declaring that the “international community is increasingly opposed to the use of violation of international humanitarian law as a method of trying to enforce the law.”65

Sometimes states, and especially those scholars who advance the contractual model, view the principle of proportionality, at least in part,66 as a response to a prior violation of the laws of armed conflict by another party, and hence as a form of reprisal. Even if a policy of reprisals is disallowed, when one party to the conflict uses civilians as shields, and launches attacks from within civilian populations, it could be said that it coerces the other side to attack targets in ways that would necessarily harm civilians. Under such circumstances, an attack on civilians could be understood as a form of reprisal.67

Where the specific application of the principle of proportionality is concerned, proponents of the contractual and human rights models would arrive at very different conclusions. One possible case in point arises when attention focuses on the categories of civilians to be counted when auditing the proportionality balance. Hays Parks, a leading proponent of the contractual model, maintains that when counting the civilians who might be incidentally injured, the attacking army should simply disregard those deliberately placed in danger by the defending side.68 In other words, when the non-state disregards the lives of civilians, the attacking side should be able to do so too.69 Indeed, Parks concludes that no violation of proportionality exists until the quantity of civilians killed clearly outweighs the importance of the declared military objective, and is high enough to generate suspicions that the civilian target was the primary objective of attack.70 As we have seen, supporters of the human rights model completely ignore the question of who is responsible for the risk to civilian lives.

Part III: Evaluating the Evidence – Which View Is Correct?

The earlier parts of this study identified two competing perspectives on proportionality: the human rights model and the contractual model. What support for either of these models can be found in treaties, judicial decisions, state practice, and moral claims?

A. International Criminal Law and Judicial Decisions

One possible way to clarify the ambiguities in the term “proportionality” is to examine its use in pronouncements handed down by judicial bodies and in treaties where it is discussed. It must be born in mind, of course, that some such references will occur in the context of international criminal law and that discussions of individual criminal responsibility will not be exactly commensurate with the general meaning of proportionality in IHL.71 Even so, their utility as guidelines for an understanding of the term cannot be discounted.

(1) The International Criminal Court

The emergence of proportionality as part of codified IHL has also led to its inclusion within the growing body of international criminal law. We shall immediately see how this played out in the decisions of international criminal courts. However, that proportionality has become part of international criminal law is also clear from Article 8(2)(b)(iv) of the Rome Statute of the International Criminal Court.72 This text includes as a war crime:

Intentionally launching an attack in the knowledge that such attack will cause incidental loss of life or injury to civilians or damage to civilian objects or widespread, long-term and severe damage to the natural environment which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated.

Several points in this pronouncement are worth noting: To begin with, there are the changes in the formulae. While Additional Protocol I required the incidental damage to be “excessive” in relation to the military advantage, the Rome Statute now specifies damage that would be “clearly excessive.” This addition might help to clarify two of the questions already referred to. First, it specifies that criminal liability will only be granted when the damage to civilians is deemed much greater than the military advantage. In addition, and secondly, the term “clearly” goes to the question of the requirement from the attacker. For criminal responsibility to be attributed to the attacking party, it has to be proven that he possessed clear knowledge of the consequences.

Another important point which might be clarified in this provision is the use of the word “overall” in the last sentence. The equation has to be made not with regard to a specific attack, but in relation to the overall military advantage anticipated. The use of this term seems to agree with our prior assumption that a military operation cannot be viewed in isolation.

One additional interesting point is that proportionality as a criminal offence is mentioned only in the context of an “international” armed conflict. While this in no way guarantees that the requirement of proportionality does not apply to non-international conflicts,73 it certainly hints that there could be certain differences in the application of this principle between these kinds of armed encounters.

(2) Judicial Decisions

Criminal prosecutions for violations of the principle of proportionality are almost non-existent, and only in the Blaskic case was a conviction in a criminal tribunal reached on that basis. Even then, the conviction was overturned on appeal, although on different grounds.

The Blaskic Case of the ICTY74

The Blaskic case dealt with an incident in the war between Croatia and Bosnia in 1993. The trial chamber of the ICTY declared that the Croatian attack on the Muslim area in the town of Vitez on April 1993 constituted a war crime because a disproportional amount of force had been employed.

Superficially, this seems to have been a very problematic decision. Partly, this was because it seems to have implied that the fact that many civilians were killed is in itself a proof that the attacks were disproportionate.75 If that was so, then the court seems to have created a new norm, a need for the proportionality of consequences. But when the decision is studied in its full context, a different picture emerges. It then becomes clear that the court was using proportionality in a completely different manner – as an institutional tool to refute a clearly false claim by the attacking party.

Courts are supposed to decide on matters that are by their nature impossible to verify. For example, in cases involving accusations that an attack was illegal because it ignored the principle of distinction, the court is required to decide whether the attacking party intended to cause harm to civilians. The intention of the attacking army is of course something that is very tricky to prove. Hence, the court is required to use proportionality as a probative mechanism and, in the Blaskic case, to show that the attack on Vitez was undeniably directed against civilians. To that end, the court cited the staggering number of civilian casualties, and especially the almost complete absence of military casualties. The real claim of the court is the violation of the principle of distinction; proportionality was used here only as an additional proof for this assertion.

To sum up this point: International criminal law as it stands does not support the human rights model of interpreting the principle of proportionality. On the other hand, neither does it provide support for the contractual model. While there is no mention of reprisals in the Rome Statute, it also does not form an excuse for committing any of the actions forbidden in the Rome Statute.

 B. How States Behave

One of the classic foundations of international law is state practice. State practice is both part of the definition of customary international law and an interpretive tool to understanding commitments in international law. Hence, state practice should be evaluated while trying to understand the ways in which international law actually operates. In the context of warfare, however, assessments of the behavior of states are notoriously difficult to attain, especially since few states explicitly explain their decisions and the procedures by which they were arrived at.

One method employed to solve these difficulties, and that is employed for instance by the ICRC study,76 is to look at military manuals as reflections of state practice. Although this method can, and has, been criticized (on the grounds that it conflates two questions: how states behave and what they think is the law [opinion juris]),77 it nevertheless offers some promise.

As already detailed (above in Part I), the military manuals produced by several states emphasize a number of points with regard to the use of proportionality. Military manuals stress that the commander makes decisions based on the information that he may reasonably be considered to possess at the relevant time.78 They also emphasize how important it is to assess military gains from the perspective of the operation in its entirety.79

Only by studying specific examples is it possible to ascertain whether even the armies of liberal democratic states have in recent conflicts operated in accordance with the guidelines provided in their manuals in matters affecting proportionality. (The behavior of states such as Algeria, Pakistan, Saudi Arabia or Russia – all members of the Human Rights Council – hardly merits investigation in this respect.)

(1) The Battle of Mogadishu80

When in October 1993 a force of U.S. troops was caught inside the streets of Mogadishu, Somalia, American forces deployed relatively large numbers of troops and firepower in a rescue effort. Eighteen American soldiers were killed in the course of the operation, while on the other side, according to the testimony of Ambassador Robert Oakley, the U.S. special representative to Somalia:

My own personal estimate is that there must have been 1,500 to 2,000 Somalis killed and wounded that day, because that battle was a true battle. And the Americans, and those who came to their rescue, were being shot at from all sides…a deliberate war battle, if you will, on the part of the Somalis. And women and children were being used as shields and some cases women and children were actually firing weapons, and were coming from all sides. Sort of a rabbit warren of huts, houses, alleys, and twisting and turning streets, so those who were trying to defend themselves were shooting back in all directions. Helicopter gun ships were being used as well as all sorts of automatic weapons on the ground by the U.S. and the United Nations. The Somalis, by and large, were using automatic rifles and grenade launchers and it was a very nasty fight, as intense as almost any battle you would find.81

Of course, there was never an evaluation of how many of the Somali dead were civilians, and how many were Somali militiamen operating under the instructions of Aidid. However, Captain Haad, of the Somalian militias, in an interview on American public television, said 133 of the SNA militia were killed.82 If accurate, such estimates imply that over one thousand civilians were killed by American fire. Absent an in-depth study, it is of course impossible to assess whether the principle of proportionality was indeed preserved in all stages of the battle. However, the proportion between Somali civilian fatalities and those of U.S. soldiers provides some indication of the relative importance that each side attached to its own soldiers.

In October 1993 in Somalia, 18 American soldiers were killed in a battle where women and children were being used as shields. Over one thousand civilians were believed killed by American fire.

(2) The NATO Serbia Air Campaign

In contrast to the Mogadishu incident, the NATO air campaign against Serbia in 1999 has indeed been the subject of detailed study.83 The NATO attack was launched as a measure of humanitarian intervention after the Serbian government began to expel the Muslim residents of Kosovo.84 Hence, the campaign was not a war of self-defense, and NATO member states feared that even a small number of casualties would undermine domestic public support for the campaign. In order to limit casualties, NATO adopted a policy of zero risk to its soldiers.85 In the context of the air campaign this meant that pilots flew at a relatively high altitude. This tactic certainly reduced the danger of their falling victim to anti-aircraft fire. But since it also increased the chance of bombing inaccuracies, it also enlarged the risk to civilian lives on the ground.

A committee appointed by the prosecutor of the International Criminal Tribunal for Former Yugoslavia was mandated to assess whether any NATO actions during the course of the campaign aroused suspicion of war crimes.86 The committee decided that the zero-risk guideline did not constitute a violation of the norm of proportionality, even if it did result in higher risks for civilians.87

In its air campaign against Serbia in 1999, NATO adopted a policy of zero risto its soldiers, which also enlarged the risk to civilian lives on the ground. A committee appointed by the International Criminal Tribunal decided that this policy did not constitute a violation of proportionality.

The committee also assessed the legality of the air attack on the Serbian TV building on April 23, 1999, which caused between ten and seventeen civilian deaths. The report is at pains to clarify several points. One is that the attack has to be seen as part of larger attempt to destroy all command, control and communications facilities, and that the bombing of the TV station should not be studied in isolation. Secondly, much of the blame for the events should be attributed to Serbian leaders who failed to evacuate the building, even after they received advance warning of the attack.88

The number of civilian deaths in the NATO campaign was around 500,89 a relatively small number. However, most of the campaign was conducted against an enemy which generally operated within the confines of the laws of war.90

More than anything else, the Kosovo campaign has come to symbolize the fact that states think about the legality of their military operations. Each target in the Kosovo campaign was approved prior to being attacked and lawyers played a significant role in the approval process.91 A prior review of targets has become an integral part of war.

(3) American Operations in Fallujah: Operation Vigilant Resolve (April 2004) and Operation Phantom Fury/Al-Fajr (November-December 2004)

Fallujah is a city in Iraq of approximately 300,000 inhabitants, most of them Sunni Muslims. In 2003 and the beginning of 2004 it became one of the centers of the Iraqi insurgency, and it was clear to the U.S. that the city had to be retaken in order to restore order to the region.

The first attempt to take Fallujah is known as “the first battle of Fallujah,” or Operation Vigilant Resolve, waged in April 2004.92 This operation was not successful. The Marine units that participated in this operation were halted after just a few days in their attempt to retake the city of Fallujah. Moreover, problems in planning the operation caused a large number of civilian casualties, which resulted in internal Iraqi pressure to halt the operation.

In November 2004, U.S. and Iraqi forces attempted to take Fallujah again.93 Most reports agree that the majority of the city’s inhabitants – perhaps as many as 270,000 out of 300,000 – escaped. Many of those who remained were combatants. The number of people killed was never verified, with reports ranging from a few hundred to several thousand.94 The tactics used by U.S. forces in this operation included raids and the use of white phosphorous and air power. Most importantly, vast quantities of fire power were employed in an urban setting known to house civilians, in order to protect the lives of American soldiers.

In 2004, American forces operating in Fallujah, Iraq, used white phosphorous. Vast quantities of fire power were employed in an urban setting known to house civilians, in order to protect the lives of American soldiers.

Several conclusions may be drawn from the case of Fallujah. First, a Western army cannot, and does not, ignore civilian casualties, even if the reason for the risk to civilians was that the enemy, the non-state organization, took refuge within the civilian population. The first battle of Fallujah was simply halted when the number of civilian casualties became too high.95

On the other hand, when the military advantage of the operation is extremely important – such as quashing the insurgency in Iraq – states do not shy away from using massive power, even when the fighting takes place within a civilian population and there is a significant risk to the lives of civilians. Even when a state initially abandons an operation because of risk to civilians, it has no choice but to come back to the same place later, and take the same risk, simply because of military necessity.

To sum up: although it is clear that even democratic states do not accept the human rights model as a practical guide, neither do they adopt the contractual model. As the uses of force in Somalia, Kosovo, and Iraq show, Western armies are very concerned about protecting the lives of their soldiers, and to that end are willing to risk many civilian lives. They also find acceptable the notion that civilian lives can be forfeited in order to attain important military goals. At the same time, however, Western armies are certainly aware of the need to protect the lives of civilians on the other side, even when responsibility for their endangerment lies with the non-state organization that the Western army is fighting.

C. The Moral Analysis

At first sight, the human rights approach to proportionality appears to possess a moral advantage over any other model. After all, what could be more morally reprehensible than the killing of innocent civilians? Most human rights analyses of the issue indeed stop at that point, and go no further.

The problem with that approach is that there is in fact much more to be said. Specifically, there exist at least three separate reasons why – from a moral viewpoint – the premise of the human rights approach is open to question, especially in cases of asymmetrical armed conflicts.

Proportionality cannot be detached from the question of responsibility: which side created the situation in which civilians find themselves? In an abstract sense, there may be no correct answer to this question. From a military as well as a moral perspective, however,96 the onus clearly lies with the party that chooses to fight from within civilian concentrations. This circumstance does not, of course, absolve the attacking force from its obligations to protect civilian lives. But it does shift moral responsibility for the situation thus created to the other side.

Proportionality cannot be detached from the question of responsibility: which side created the situation in which civilians find themselves? From a military as well as a moral perspective, the onus clearly lies with the party that chooses to fight from within civilian concentrations.

Second, and perhaps more important, are the consequences of the human rights model perspective. Once the non-state actor internalizes the fact that his foe is committed to the protection of civilians, even at the expense of military actions, he will be bound to use civilians as shields that might protect him from enemy assaults. That is precisely the tactic adopted by Hamas and Hizbullah, and one of the reasons for the fact that they situate their headquarters, fighters, and armaments within civilian concentrations.97 In effect, then, an effort to protect civilians in one case places them in danger further down the line.98

On the other hand, the contractual viewpoint also raises moral problems. Some writers, as we have already seen, insist that civilians endangered by the deliberate actions of the other side should not at all be taken into account in the proportionality equation.99 I do not consider this position to be correct as a matter of state practice, nor do I believe that it can be defended morally. If the enemy intentionally places its headquarters within a hospital, it is unthinkable that a liberal state would simply ignore all potential civilian casualties in the facility, claiming that responsibility lies entirely with the enemy. Some middle ground should be found between these views.

Finally, and most complex of all, is the moral issue raised by the relative value of the lives of soldiers and civilians. On this point, Walzer and Margalit best articulate one extremity of the spectrum of views. Their premise is that the most basic rule of armed conflict is the protection of civilians, a protection which soldiers by definition forfeit. Since the lives of soldiers are therefore worth “less” than those of civilians, of both sides, soldiers must almost always be placed at risk rather than civilians.100 Kasher and Yadlin adopt a diametrically contradictory position and maintain that the lives of soldiers should almost never be put at risk, even at the price of risking the lives of civilians on the other side.101 At the heart of this argument lies the contention that states owe debts to their soldiers, especially those who have been subject to mandatory enlistment. Hence, their lives must be judged to be “worth more” than those of enemy civilians.

Here, too, both extreme positions seem to me to be incorrect. Lives of soldiers have not been made inconsequential simply because of the draft. On the other hand, states are duty-bound to provide enemy civilians, too, with some protection. The most appropriate solution to the dilemma thus posed may be to allow the entire issue to be governed by the principle of proportionality. The following section will attempt to show how that suggestion might be implemented.

Part IV: The Administrative Model

The ambiguity inherent in the notion of proportionality is well accepted by states and commentators alike, and was acknowledged even when the terms of Additional Protocol I were drafted.102 Moreover, judicial decisions and formal declarations interpreting this article are scarce. Those that exist have contributed very little in the way of clarification, which has also not been advanced by analyses of moral claims and state practice.

Here we seek to help remedy this situation by offering a new model for the interpretation of proportionality. We begin by examining the Israeli Supreme Court decision in the targeted killings case, and then generalize the analysis.

A. The Targeted Killings Case

The decision of the Israeli Supreme Court sitting as the High Court of Justice (HCJ) in the targeted killings case103 constitutes one of the very few exceptions to the rule of judicial silence regarding proportionality. In this case the court declared the targeted killing of terrorists to be legal under certain specific conditions. Principal among the limitations placed by the court is the need to minimize the “collateral damage” sustained during the course of targeted killing operations by civilians not taking direct part in hostilities (referred to by the HCJ as “innocent civilians”).104 The court appreciates that the application of the test is riddled with uncertainty:

[O]ne must proceed case by case, while narrowing the area of disagreement. Take the usual case of a combatant, or of a terrorist sniper shooting at soldiers or civilians from his porch. Shooting at him is proportionate even if, as a result, an innocent civilian neighbor or passerby is harmed. That is not the case if the building is bombed from the air and scores of its residents and passersby are harmed. The hard cases are those which fall into the space between the extreme examples.105

With relation to “hard cases,” the court offered only limited guidance, referring again to ambiguous or subjective considerations: 1) the desired military advantage has to be both “direct and anticipated”;106 2) a balance has to be maintained between the “state’s duty to protect the lives of its soldiers and civilians” and its “duty to protect the lives of innocent civilians harmed during attacks on terrorists.”107

While this part of the decision is of little help in clarifying the parameters of proportionality, far more useful are the institutional elements that the court introduced in those of its judgments that appear to relate to the application of the proportionality principle. Justice Aaron Barak (then President of Israel’s Supreme Court) posited in his judgment that targeted killing operations ought to be made subject to ex ante and ex post examination or investigation. With relation to ex ante review, Barak required that prior to the attack a “meticulous examination” be conducted of every case that potentially could give rise to collateral damage.108 This requirement seems to correspond to the precautionary obligations introduced by Article 57 of the First Additional Protocol.109 Even more significant is Barak’s introduction of the concept of ex post review in the targeted killings cases – a review process that is ultimately subject to judicial supervision.110

Hence it seems that for the Israeli Supreme Court, the solution to the ambiguity of the term of proportionality lies in investigations, both before and after the attack. This, however, seems to be problematic – what use are investigations if the parameters of proportionality are not clear? What should be investigated when it is not clear how the decision should have been taken?

B. Proportionality as Reasonableness

Investigations and Reasonable Commanders

Proportionality, like many other open-ended terms in law, is concerned with reasonableness. Most states that have expressed opinions on this matter seem to assume that there exists some standard of proportionality which the reasonable commander must apply in accordance with his knowledge of the field.111 This, of course, is a very general standard and one which is very hard to implement.112 Does there exist a gauge that would facilitate an estimate of what a reasonable commander would decide?

Clearly the answer is negative. However, this question is neither novel nor unique. Similar issues commonly arise whenever courts review the actions of administrative bodies. Most governmental agencies are expert in their field of operation, and the courts are reluctant to dispute the decisions of the agencies in their areas of expertise. Instead, when courts review decisions of governmental agencies, the question that they ask is whether the decisions taken deserve to be considered reasonable.113 The test for reasonableness is in the main procedural:114 Did the agency follow the correct procedure? Was it in possession of all the relevant data? Did it give a proper hearing to all views?115 This is the only way courts can decide whether the actions of the agency were reasonable. It is the usual practice of administrative courts to give at least some deference to the substance of the agency’s decision.116

A substantively similar process takes place with regards to proportionality in IHL. What we really want to know is whether the commanders in the field, when making their decisions, took into account the likelihood of civilians being hurt. We cannot possibly judge whether the decision ultimately taken was correct; we do not possess all the required information, and even if we did, we wouldn’t know which parameters to apply. The best we can do is judge the decision-making process.

Naturally, then, ex ante review is required. A military operation should be initiated only if we can be sure that an appropriate investigation as to the amount of collateral damage to civilians will be carried out. Of course, this requirement carries different meanings in different contexts. In a pre-planned attack of a large scale, it mandates gathering all available information and subjecting the planned operation to in-depth analysis. By contrast, should an immediate decision be required during the course of an operation already under way, an entirely different level of both information gathering and decision-making would be applied.

A military operation should be initiated only if we can be sure that an appropriate investigation as to the amount of collateral damage to civilians will be carried out. Most armies in the West are using legal advisors to verify that such a review is undertaken.

Ex ante review is a most important facet of any military operation, and especially so when civilian casualties are involved. This is one of the basic requirements of the First Additional Protocol, and it seems that most armies, in the West at least, are indeed using legal advisors exactly in order to verify that such a review is undertaken.117 Whatever the context, the important point to verify is that the question was asked. If we apply the same assumptions to military commanders as to administrators, and hence approach them as reasonable persons, that is the most we can ask.

C. Ex Ante Review Is Not Enough

Commanders, however, are not equivalent to administrators. Most obviously this is because of the differences in their functions. Officials who work in administrative agencies service their own communities and deal with citizens of their own country. Hence, an assumption that they will behave reasonably is, so to speak, reasonable. Field commanders are different; their function is to fight the enemy. Hence, we should be much more careful in assuming that they take the interests of the lives of enemy civilians into account.

Second, even those commanders who are “reasonable” will only reach a reasonable answer if they ask the correct questions. How can we be sure that such is indeed the case? In matters pertaining to administrative law, that is precisely the task of the courts; by subjecting many administrative issues to judicial review both before and after their occurrence, they verify that the administrators did indeed ask the correct questions prior to embarking on a course of action. However, courts are reluctant to intervene in military operations prior to their initiation.118 After all, judicial review takes time and could involve a postponement in the timing of the action. And no court is happy to shoulder the responsibility for whatever damage a delay might cause. Moreover, judges are noticeably hesitant to intervene in military matters even after the fact, since they consider their knowledge of the situation to be inferior to that of military commanders.

Even the Israeli Supreme Court, which has shown an exceptional and unprecedented degree of willingness to intervene in military matters, has usually avoided intervening in specific military operations while they are in progress, and has limited its review to general comments about appropriate courses of action.

In sum, therefore, it must be concluded that ex ante review cannot provide assurance that the action taken did indeed comply with the requirements of proportionality. There exists no guarantee that commanders are acting reasonably, and that the actions are subject to adequate judicial review prior to their initiation. What is required, therefore, is a further form of review: ex post investigations.

D. Ex Post Investigations

Ex post review ensures that the actions of the commander will eventually be examined. As such, it constitutes an additional influence on his decision-making process prior to the operation. A commander who is aware that his actions will be monitored after the fact is likely to take care that he gives due consideration to all possibilities when reaching a decision to act. Ex post review can mean several things: usually, armies use internal investigations as a means of evaluating the effectiveness of their mission. In many cases, armies employ the same method in order to investigate allegations of war crimes. However, if the investigators of the commander form constituent elements in the same chain of command, there is little chance that the investigation would yield trustworthy results.

A commander who is aware that his actions will be monitored after the fact is likely to take care that he gives due consideration to all possibilities when reaching a decision to act.

In cases involving accusations of human rights violations, several courts have expressed opinions as to how an investigation should be conducted. The most expansive description is that of the European Court of Human Rights in the Isayeva case,119 which concerned the death of several hundred Russian-Chechnyan civilians during the armed conflict in that region. The European Court of Human Rights decided that the death of civilians provided prima facie grounds for claiming violation of the right to life, and deemed the internal Russian investigation that exonerated all participants to have been insufficient. The court specified that in order for an investigation in these matters to be considered adequate, four criteria had to be met:

1. The formal and practical independence of the investigators from the persons whose actions they were examining

2. Effectiveness: the ability of the investigation to lead to effective remedies including, where appropriate, criminal investigations

3. Promptness of the investigation

4. Availability of public scrutiny120

There exists no clear international rule which declares these to be the only possible criteria. Neither is there a precise formula as to their respective weights and how they are to be applied. However, they do provide the general basis for the type of investigation that should be initiated into operations involving civilian casualties121 and that will contribute to the general adherence to the rules of IHL. With specific reference to proportionality: an ex post investigation, conducted in accordance with the guidelines set down in the Isayeva case, is likely to cause soldiers and commanders to take into account the requirements of proportionality, and especially the need to consider the “collateral damage” when they plan or carry out an attack. Moreover, it seems that in order for such an investigation to be effective, the committee of enquiry should include military personnel capable of assessing the reasonableness of the actions undertaken by the attacking force.122

E. The Implications of the Administrative Model of Proportionality

The administrative model for proportionality, as detailed here, focuses on the professional soldier as the ultimate decision-maker on questions of proportionality. It is based on several foundations:

The Importance of Asking the Questions: The assumption lying behind the administrative model is the fact that a commander who asks himself the correct questions would provide better results in terms of protecting innocent lives. This assumption is based on the fact that the ethics of the military profession require that armed conflict will be mainly directed against soldiers and combatants. Hence, if we leave the decision to soldiers, with a review of the decision-making process, they would arrive at results which tend to protect human lives.

The Importance of Understanding the Effects: The suggested model is compatible with the general aims of the military. It is based on the fact that the commander ordering an attack should have an understanding of the results of his actions.

Command Responsibility: A very important application of the institutional ingredient of proportionality may be felt not in cases alleging direct responsibility of soldiers and commanders, but rather in command responsibility cases.123 Article 28 of the ICC Statute imposes responsibility on commanders who did not prevent international crimes from occurring, despite the fact that “owing to the circumstances at the time” they “should have known” about their occurrence. That being the case, a robust ex ante review could significantly extend the exposure of commanders to such negligence-based responsibility (in addition to the knowledge-based responsibility discussed above). In fact, where circumstances so warrant, it can be argued that commanders should insist upon effective ex ante review and might incur criminal liability for failing to do so. Furthermore, since Article 28 also criminalizes failures on the part of commanders to punish soldiers for violations that had already occurred, improved ex post investigations could introduce significant pressures on commanders to order criminal prosecutions of subordinates involved in attacks entailing “clearly excessive” consequences. Here again, failure to order an investigation might serve in itself as the basis of responsibility in appropriate circumstances.

Taking Responsibility into Account: As mentioned earlier, responsibility is an integral part of the proportionality equation. In applying proportionality to a specific case, the commander may ask himself who is responsible for the possible incidental loss of lives. If indeed it is the enemy that is responsible, then this might allow the commander more flexibility in the application of the principle of proportionality. Even then, the commander should still take into account the possible incidental loss of civilian lives. The balance that the commander has to strike is between avoiding the danger that the enemy would turn civilians into human shields, on the one hand, and that too many innocent civilians will be killed, on the other.

Protecting the Lives of Soldiers: As we have seen, all armies protect the lives of their soldiers, even at the risk of lives of civilians on the other side. The commander may take this into account. However, it seems clear that policies of “zero casualties,” when they might result in the death of too many civilians, should not be used. A reasonable commander should also take the lives of civilians of the other side into account.

Proportionality and Criminal Responsibility: The institutional elements of proportionality developed here might be useful in determining a violation of proportionality requirements in the criminal law context too. The quality of any ex ante review might be relevant in ascertaining the mental state of soldiers and commanders carrying out military operations, in the sense that an improved decision-making process might seriously curtail the ability of the commanding ranks involved in the review process, or exposed to its findings, to claim ignorance of the anticipated disproportional consequences of their actions. In other words, the criminal trial would inquire whether the attackers asked themselves questions relating to proportionality. At the same time, low-ranking soldiers in the field engaged in military operations could, perhaps, rely on their knowledge that an effective review process exists in maintaining the reasonableness of their belief that their actions were indeed proportional.124

F. Application: Israel and the Gaza Operation

Here it might be useful to review the case of Operation Cast Lead, Israel’s operation in Gaza between December 27, 2008, and January 18, 2009, not in order to determine whether Israel’s operation was fully in compliance with international law, but rather to examine whether the assumptions made in this study are plausible.

On the whole, the evidence regarding the Gaza operation clearly shows that Israeli commanders successfully followed the requirements of the administrative model of the principle of proportionality. According to newspaper and oral reports, the IDF did require commanders to take humanitarian law into account in the planning stages of the operation.125 Moreover, legal advisors were involved in the planning of many operations and provided advice regarding specific targets.126 The right questions were asked, checks were made, and the incidental damage to civilians was on the whole limited. As detailed above, the Goldstone report makes several claims regarding deliberate attacks against civilians. These claims still await a proper Israeli response. Notwithstanding these claims, in those instances where it is agreed that an Israeli commander was indeed within the sphere of the principle of proportionality, the behavior of the IDF seems to have been completely reasonable and totally within the parameters of legitimate use of proportionality as detailed in this study.

In the Gaza operation, the IDF required commanders to take humanitarian law into account in the planning stages of the operation. Legal advisors were involved in the planning of many operations and provided advice regarding specific targets. The right questions were asked.

One further question of importance is the total number of civilian casualties. The actual number of casualties is debated. Figures released by Hamas report 1,417 Palestinians killed; of whom 236 were combatants, 255 members of Hamas security forces, and 926 civilians. A further 5,300 Palestinians were wounded.127 A report issued by the IDF speaks of 1,166 Palestinian fatalities, of whom at least 709 were combatants and 295 uninvolved civilians. The remainder are listed as undetermined.128 On the Israeli side there were 13 fatalities: ten soldiers and three civilians. On the one hand, looking at the total number of civilian casualties seems to be totally irrelevant, since the principle of proportionality requires evaluation of each tactical operation by itself. On the other hand, public opinion seems to be mesmerized by the total number of civilian casualties.129

For argument’s sake, let us suppose that the total number of casualties is important.While the Israeli figures are saddening, since they speak of three to four hundred civilian casualties, they seem to be completely within the parameters of reasonable application of force according to historical examples. Moreover, even if the numbers suggested by the Palestinians are correct, this does not immediately mean that the quantity of civilian casualties is unreasonable or disproportional. Both historical evidence and moral claims point to the fact that when the enemy chooses to fight from within the civilian population, the large number of civilian casualties is the enemy’s responsibility. However, a final opinion in this matter must await the final tally of civilian casualties in this conflict.

Naturally, and as stressed in this study, the application of the norm of proportionality should also be tested in ex post investigations. These investigations are indeed taking place.130 If indeed a specific unreasonable military action has taken place, the commanders should be punished and the lessons must be learned.


States which face non-state actors in military armed conflict tend to claim that changes to IHL/LOAC are required. This study attempts to show that a proper interpretation of the existing principles of IHL might sometimes provide satisfactory answers to the problems posed by non-state actors.

Its main argument involves the principle of proportionality. This principle is best understood as an administrative or institutional principle intended to be based on the reasonable discretion of the commander in the field. This discretion is not unlimited – it is guided by respect for human life and civilian immunity. It should be reviewed in advance to make sure that the proper questions are asked. It should be reviewed after the actions take place in order that mistakes might save future commanders from similar mistakes.

Proportionality in international humanitarian law is a complicated principle to apply. However, this does not make this norm unimportant, or inapplicable. This work shows that a proportionality principle, properly defined, could serve as a reasonable and appropriate guiding principle even in asymmetrical wars against terrorist organizations, non-state actors, and all those who do not respect the laws of war. This has been done in the past, and should likewise be done in the future.

 * * *


* This article was prepared for the Jerusalem Center’s Global Law Forum series in 2010.
1. David Kretzmer, “Rethinking the Application of IHL in Non-International Armed Conflicts,” 42 Isr. L. Rev. 8 (2009).
2. Some, especially in the U.S., use these two terms (LOAC and IHL) as synonyms. Traditionally, the laws of war were called the Laws of Armed Conflict. This signifies their general goal – to regulate armed conflicts according to pre-agreed forms. During the second half of the twentieth century the terms were changed and the name of this area of law became International Humanitarian Law. Clearly, this change also shifted the focus of the area of law from agreement between armies to protection of civilians. For a general description, see A.M. Slaughter and L. Helfer, “Why States Create International Tribunals: A Response to Professors Posner and Yoo,” 95 California Law Review 899 (2005).
3. For detailed descriptions of IHL, see Yoram Dinstein, The Conduct of Hostilities Under the Law of International Armed Conflict (Cambridge UP, 2004); Jean-Marie Henkaerts and Louise Doswald-Beck, Customary International Humanitarian Law (ICRC, Cambridge UP, 2005).
4. There are even those who claim that such non-state actors are not subject to IHL because they are not a state. For the purpose of this study I will assume that this is an incorrect interpretation of the laws of armed conflict.
5. Note that this is not a mere oversight since there was a serious attempt, which failed, to include the term “proportionality” in the First Additional Protocol to the Geneva Conventions (1977). W.J. Fenrick, “The Rule of Proportionality and Protocol I in Conventional Warfare,” 98 Mil. L. Rev. (1982) 91, 102-107. One may note, however, that the term “proportionate” is used in the ICC Statute as a limitation on the grounds for excluding criminal responsibility. Rome Statute of the International Criminal Court, 17 July 1998, art. 31(1)(c), 2187 U.N.T.S. 90 (hereinafter ICC Statute).
6. Some scholars maintain that the concept was already part of the Christian medieval laws of war, which posited that war could be deemed to be just, and hence legitimate, only if its gains exceeded the horrors that it wrought. J.G. Gardam, “Proportionality and Force in International Law,” 87 A.J.I.L. (1993) 391, 394-395. A more concrete articulation of this standard can perhaps be seen in the Hague regulations. These adopted as a basic rule the principle of distinction, which outlawed direct attacks against non-military targets (Article 25). But this did not mean that all other attacks, for instance attacks against targets that were both civilian and military, could be undertaken without restraint. Article 23(e) prohibits employing “arms, projectiles, or material calculated to cause unnecessary suffering,” while Article 27 directs the attacking army to spare, as far as possible, buildings which possess religious or cultural importance.
7. See e.g., the St. Petersburg declaration of 1868 that condemns the “employment of arms which uselessly aggravate the sufferings of disabled men, or render their death inevitable.” “Declaration Renouncing the Use, in Time of War, of Explosive Projectiles Under 400 Grammes Weight,” 29 November/11 December 1868, reprinted in 1 AJIL Sup. 95 (1907). Similarly Article 35(2) to AP-I forbids explicitly the use of these arms. See Gardam, supra note 6, at p. 406 (“This provision codifies the preexisting customary principle and is also based on proportionality”). See also M. Bothe, H.J. Partsch and W.A. Soft, New Rules for Victims of Armed Conflict: Commentary on the Two 1977 Protocols Additional to the Geneva Conventions of 1949 (1982) 195-197; Y. Sandoz, C. Zwinarski and B. Zimmermann (eds.), Commentary on the Additional Protocols of 8 June 1977 to the Geneva Conventions of 12 August 1949 (1987), 401-402.
8. Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts (Protocol I), 8 June 1977, 1125 U.N.T.S. 3 (hereinafter AP-I or First Additional Protocol).
9. IHL traditionally applies to conflicts between two states: international armed conflicts. In recent years, more and more of IHL is applied to conflicts between states and non-states: non-international armed conflicts. According to the ICRC study of customary IHL, the principle of proportionality applies to non-international armed conflicts too. J. Henckaerts and L. Doswald-Beck, Customary International Humanitarian Law (ICRC, Cambridge UP, 2006).
10. See generally, Larry May, War Crimes and Just War (Cambridge UP, 2007), 211-227.
11. M. Cohen-Eliya, “The Formal and Substantive Meanings of Proportionality in the Supreme Court’s Decision Regarding the Security Fence,” 38 Isr. L. Rev. (2005) 262, 288-289.
12. J. Gardam, Necessity and Proportionality in International Law (Cambridge UP, 2004), 93.
13. R. Higgins, Problems and Process: International Law and How We Use It (Oxford UP, 1994).
14. For a more detailed analysis of debates concerning the application of the rule of proportionality, see J. Gardam, supra note 12, Chap. 4.
15. Of course, this question arises every time human rights are balanced against other state interests, even in domestic settings. In attempting to resolve this dilemma, constitutional jurisprudence courts generally adopt one of two possible courses of action. The first is to lay down a clear rule which declares in advance when the interest will overcome the right. E.g., security interests take precedence over freedom of speech in cases of clear and present danger. The other course is to leave the application of the general standard to the courts. Article 51 has adopted neither route – hence the problem. As we shall see, the ICC has employed both.
16. Examining the whole war would, of course, mean that there would be no distinction between proportionality in jus in bello and in jus ad bellum.
17. L. Doswald-Beck, “The Value of the 1977 Geneva Protocols for the Protection of Civilians” in: Armed Conflict and the New Law: Aspects of the 1977 Geneva Protocols and the 1981 Weapons Convention, Michael A. Meyer, ed., at 156-7.
18. Y. Sandoz et al., eds., Commentary to the Additional Protocols, supra note 7, at p. 684.
19. Gardam, supra note 12 at 101. Judith Gardam suggests that the application of this construction would probably have made illegal the American defoliation campaign in the Vietnam War. In this campaign the U.S. army destroyed several of Vietnam’s forests so as to prevent the Vietcong from using them as cover. In that case the military advantage was to be achieved only in the long term, and even then only in a piecemeal manner.
20. See J. Henckaerts and L. Doswald-Beck, Customary International Humanitarian Law, supra note 9, vol. I, at 49, and the practice of Australia, Belgium, Canada, France, Germany, Italy, Netherlands, New Zealand, Spain, United Kingdom, and United States cited there.
21. Taken to the extreme, this would mean that the entire military operation should be the basis for assessment. That, however, could lead to a blurring of the line between jus ad bellum and jus in bello. What states mean is that the military advantage from a military operation should be assessed as a result of a policy. This perspective will be examined in greater detail below.
22. Henckaerts and Doswald-Beck, supra note 20, at p. 50.
23. “Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia,” 39 I.L.M. 1257 (2000).
24. See e.g., Amnesty International, “NATO/Federal Republic of Yugoslavia, ‘Collateral Damage’ or Unlawful Killings? Violations of the Law of War by NATO during Operation Allied Force,” 6 June 2000, available at
25. For a discussion of this issue, see H.C.J. 769/02, The Public Committee Against Torture in Israel v. Gov’t of Israel, judgment of 14 Dec. 2006, at para. 21, (Allowing the use of the practice of targeted killing provided the conditions of proportionality are met). But see A. Cassese, Expert Opinion on Whether Israel’s Targeted Killings of Palestinian Terrorists is Consonant with International Humanitarian Law, (claiming that the policy of targeted killing is forbidden per se).
26. “Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia,” 39 I.L.M. 1257 (2000).
27. Hays Parks, “Air War and the Law of War,” 32 Air Force L. Rev. 1, 175 (1990).
28. See E. Benvenisti, “Human Dignity in Combat: The Duty to Spare Enemy Civilians,” 39 Isr. L. Rev. (2006) 81, 82.
29. Id.  at 83.
30. For a detailed description of the attitude of human rights law to the loss of life, see David Kretzmer, supra note 1, at 14-17.
31. In the negotiations over Additional Protocol I, several states suggested deleting the last sentence of Article 51(5)(b) (from the words “which is expected”), thus completely prohibiting any attack which the attacking party knows would cause suffering to civilians, even if it is directed at military targets. See ICRC commentary on AP-I, supra note 7 at p. 683.
32. The Romanian representative was apparently one of the major opponents of the insertion of proportionality into the article. His delegation had abstained in the vote on Article 50 [now 57] and had voted against paragraph 2(a)(iii) and 2(b), which embodied the “rule of proportionality” that his delegation had always opposed. Article 50 [now 57] introduced into humanitarian law a concept which was contrary not only to humanitarian principles but to the general principles of international law. It amounted to legal acceptance of the fact that one part of the civilian population was to be deliberately sacrificed to real or assumed military advantages and it gave military commanders the power to weigh their military advantage against the probable losses among the civilian population during an attack against the enemy. Military leaders would tend to consider advantage to be more important than the incidental losses. The principle of proportionality was therefore a subjective principle which could give rise to serious violations. Accidental losses among civilians must be reduced to a minimum through scrupulous application of the Geneva Conventions. All precautionary measures must be taken to protect the civilian population before embarking on an attack. In no circumstances should legal provisions give parties the right to dispose of human lives among the civilian population of the adversary. Modern international law prohibited aggression and only wars of defense against aggression were permitted. The rule of proportionality was therefore against the principles of international law. Quoted in Fenrick, supra note 5, at 105-6.
33. Commentaries to the Additional Protocol of the Geneva Conventions, p. 626.
34. Yoram Dinstein, supra note 3, p. 120.
35. A. Cohen, “The Lebanon War and the Application of the Doctrine of Proportionality,” in: The Doctrine of Proportionality (ICFAI University Press, 2009).
36. Amnesty International, “Israel/Lebanon: Out of All Proportions, Civilians Bear the Brunt of War” (hereinafter: Amnesty report), available at; and Human Rights Watch, “Fatal Strikes: Israel’s Indiscriminate Attacks Against Civilians in Lebanon” (hereinafter: HRW report), available at
37. Available at
38. While the Goldstone Report is not an NGO report, but was rather commissioned by the UN Human Rights Council, I think it is reasonable to see it as part of the “family” of NGO reports. First, members of the mission, including Justice Goldstone himself, were affiliated with NGOs in the past. Second, an enormous amount of the report, close to half if one calculates according to citations, is based on NGO accounts of the Gaza conflict. Third, the substantive position taken by the report seems to me to be in precise accord with the views of NGOs regarding proportionality. To the best of my knowledge, no international human rights NGO took issue with the Goldstone Report in these matters.
39. E.g., paragraphs 464-5 to the final report. This was despite the fact that Israel published a picture of weapons within a mosque, and provided a detailed account of an investigation of Hamas operatives which provided precise names of mosques in which weapons were stored. See The Operation in Gaza – Factual and Legal Aspects, Israel Ministry of Foreign Affairs, July 2009, para. 164 (The Israeli Report), available at
40. The Israeli Report, para. 171-180.
41. Goldstone Report, para. 466-474.
42. Id., para. 393-438.
43. Id., para. 427.
44. Id., para. 437.
45. Id., para. 653-708.
46. In this case, as in others, the Palestinian witnesses denied that there was any shooting at the IDF from near the school. However, in this case, two different media reports from the same day (one by AP and the other by Channel 4 [UK]) also described shooting. The mission could not simply discard this evidence. Id., para. 690.
47. Of course, the reason for this is Israel’s controversial decision not to cooperate with the Goldstone mission. Yet, the decision not to cooperate does not mean, I think, that a state concedes that all factual claims of the other party are correct.
48. The family members were not hurt in the earlier bombing because they received a warning by phone and had time to evacuate their house. Goldstone Report, para. 687. Nowhere does the report claim that the house was bombed illegally. However, in an interesting twist of logic, the commission uses this incident to prove that the family members were not Hamas operatives. Otherwise, so claims the report, the IDF would have surely not warned them in the earlier bombing, and would have killed them then. Id., para. 684. The possibility that the IDF tried to avoid exactly the sad result of the death of innocent family members apparently did not cross the mission members’ minds. The possibility that Hamas shootings were intentionally undertaken with innocent civilians in the vicinity so as to protect Hamas fighters was also not thought even worthy of mentioning by the mission.
49. Goldstone Report, para. 699.
50. The Israeli Report, para. 340.
51. It should be noted that the reference here is not to the Gaza operation as a whole, but just to the specific operation in which this particular unit was engaged.
52. The claim was articulated in an article by Michael Walzer and Avishai Margalit, “Israel: Civilians and Combatants,” New York Review of Books, vol. 56, no 8 (May 14, 2009), I shall address this claim below.
53. For a detailed support of this position, see Adil Ahmed Haque, “Killing Civilians,” available at SSRN:
54. E. Benvenisti, supra note 28.
55. For a detailed support of this position, see e.g., John Yoo, War by Other Means: An Insider’s Account of the War on Terrorism (2006).
56. Henckaerts and Doswald-Beck, supra note 20 at pp. 515-8, name five principles in customary IHL which limited the use of belligerent reprisals. (1) Purpose: only for the purpose of inducing the enemy to comply with IHL. (2) Reprisals are used as measures of last resort. (3) Reprisal action must be proportionate to the violation. (4) The decision should be taken at the highest level of government. (5) Reprisals must be terminated as soon as the other party complies with IHL.
57. Geneva Convention I, Art. 46, prohibits reprisals against wounded on land; Geneva Convention II, Art. 47, prohibits reprisals against wounded and survivors of shipwrecks at sea; Geneva Convention III, Art. 13, prohibits reprisals against prisoners of war; Geneva Convention IV, Art. 33, prohibits reprisals against civilians in occupied territories.
58. Additional Protocol I, Art. 51(8) states that: “Any violation of these prohibitions 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.”
59. UN General Assembly Res. 2675 (XXV) 1970.
60. Prosecutor v. Kuperskic (ICTY, Trial Chamber, Judgment of 14 January 2000), para. 515-536; judgment available at
61. Other states include: United States, India, Afghanistan, Eritrea, Indonesia, Iran, Iraq, Malaysia, Morocco, Pakistan, Sri Lanka and Turkey. Altogether 167 states are parties to the protocol.
62. Positions of United States and UK cited in the Kuperskic case, supra note 60, fn. 785 and 787 respectively.
63. Id., at para. 533.
64. Y. Dinstein, The Conduct of Hostilities under the Law of International Armed Conflict, 2nd ed. (Grotius, Cambridge, 2004).
65. Henckaerts and Doswald-Beck, supra note 20, at 514.
66. Of course, not all uses of proportionality are related to reprisals. Yet the basis of the situation respecting the conflicts discussed in this study, such as the conflicts in Lebanon and Gaza, posit a situation which calls for an analysis from this point of view, even if it does not encompass the entire notion of proportionality.
67. I am of course aware that proportionality does not apply exclusively in situations in which I suggest some kind of reprisal is called for. Proportionality is also relevant where a bridge which serves both military and civilian populations is the target, without involving any violation on the part of the army using the bridge. Yet, I suggest that the application of the principle of proportionality is most problematic in situations where the other party deliberately places itself within civilian population, because it is exactly where the opposing party is taking advantage of proportionality that the principle raises the most objections.
68. Parks, supra note 27, at 162-3.
69. Richard Rosen likewise suggests that a presumption of innocence should be accorded to commanders of armies confronting situations of this sort. R. D. Rosen, “Targeting Enemy Forces in the War on Terror: Preserving Civilian Immunity,” 42 Vanderbilt J. Int’l L. 683 (2009).
70. Parks, supra note 27, at 174.
71. International criminal law and individual criminal responsibility are intended to cover only the most serious war crimes (and other serious violations of international law). Hence, in the criminal context we can expect proportionality to be defined in a narrower way than in IHL in general. On the other hand, basic rules of international law require clarity (Nulle Crimen Sine Lege) and therefore the quest for clarity should begin in criminal law.
72. Supra note 5.
73. As mentioned in supra note 9, the ICRC study lists proportionality as applicable in both international and non-international armed conflicts.
74. Prosecutor v. Blaskic, Judgment of Trial Chamber of 3 March 2000.
75. Id., at para. 507.
76. Supra note 9.
77. Customary international law includes two aspects, which together create a custom – state practice, and opinion juris – what states believe is the law.
78. Supra note 20.
79. Supra note 20.
80. “Somalia: Anatomy of Disaster,” Time Magazine, October 18, 1993, available at,9171,979399-1,00.html.
81. “Ambush in Mogadishu,” Frontline interview with Ambassador Robert Oakley, available at
82. “Ambush in Mogadishu,” Frontline interview with Captain Haddad, available at http://www.pbs. org/wgbh/pages/frontline/shows/ambush/interviews/haad.html.
83. Ward Thomas, The Ethics of Destruction: Norms and Force in International Relations (Cornell UP, 2000), 161-168. Daniel L. Byman and Matthew C. Waxman, “Kosovo and the Great Air Power Debate,” 24(4) International Security (Spring 2000) 33-34.
84. Benjamin S. Lambeth, NATO’s Air War for Kosovo (RAND, 2001).
85. Thomas, supra note 83.
86. Supra note 24.
87. Id., para. 56.
88. Id., para. 71-79.
89. Thomas, supra note 83, puts the number between 477–512.
90. Thomas, supra note 83.
91. Id.
92. Ahmed S. Hashim, Insurgency and Counter-Insurgency in Iraq (Cornell UP, 2006), 31-46; Carter Malkasian, “The First Battle for Fallujah,” in War in Iraq: Planning and Execution, T.G. Mahnken and T.A. Keaney, eds. (Routledge, 2007), 163.
93. The facts here are based on “Operation Al-Fajr” in; Rosen, supra note 69; Hashim, supra note 92.
94. The website Iraq bodycount, which keeps track of all databases regarding civilian casualties, puts the number at a few hundred. Available at
95. Similar evidence to support the same conclusion may be adduced from the constant changes in Rules of Engagement and NATO directives regarding bombing in Operation Enduring Freedom in Afghanistan. As documented in several reports, the number of civilians killed in bombings in this operation was alarmingly high in 2005-2007. See, e.g., Human Rights Watch, “Troops in Contact: Airstrikes and Civilian Death in Afghanistan” (September 8, 2008), available at The result of the high number of civilian deaths is a constant attempt by U.S. and NATO commands to tighten the rules of engagement, and limit the use of especially harmful bombs. See, e.g., R. Norton-Taylor, “NATO Tightens Rules of Engagements to Limit Further Civilian Casualties in Afghanistan,” Guardian, September 9, 2008, available at There is some support to the assumption that these changes actually limit the number of collateral civilian casualties.
96. Michael Walzer, “Responsibility and Proportionality in State and Non-State Wars,” Parameters (Spring 2009), 40-52.
97. For detailed accounts, see Rosen, supra note 69.
98. Strict deontologists might reject this argument on the grounds that it analyzes the consequences of actions rather than their inherent moral value. However, and as Gabriella Blum has suggested, deontologist moral views are at any rate problematic when wars are involved because of the fact that war is always about using “bad” methods to achieve good results. Gabriella Blum, “The Laws of War and the Lesser Evil,” 35 Yale J. Int’l L. (2010). The principle of proportionality itself, even in its most limited interpretation, is by definition in contradiction to deontological principles because it evaluates the results of actions rather than their inherent values. Moreover, it seems to me a caricature of deontological views to claim that we may ignore the effects of our actions in terms of civilian deaths.
99. Parks, supra note 68.
100. Walzer and Margalit, supra note 52.
101. Asa Kasher and Amos Yadlin, “Military Ethics of Fighting Terror: An Israeli Perspective,” 4(1) J. Military Ethics 3-32 (2005).
102. William Fenrick, supra note 5.
103. H.C.J. 769/02, The Public Committee Against Torture in Israel v. Gov’t of Israel, judgment of 14 Dec. 2006,, hereinafter: Targeted Killings case.
104. Targeted Killings case at para. 45.
105. Ibid., at para. 46.
106. Ibid. This legal standard appears to derive from the language of Art. 51(5)(b) of AP-I.
107. Ibid., at para. 46.
108. Ibid., at para. 46.
109. Article 57(2)(a) of AP-I requires parties to:
(i) Do everything feasible to verify that the objectives to be attacked are neither civilians nor civilian objects and are not subject to special protection but are military objectives within the meaning of paragraph 2 of Article 52 and that it is not prohibited by the provisions of this Protocol to attack them;
(ii) Take all feasible precautions in the choice of means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss of civilian life, injury to civilians and damage to civilian objects;
(iii) Refrain from deciding to launch any 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.
110. Targeted Killings case at para. 54.
111. Fenrick, supra note 5. ICRC study, supra note 9.
112. See Amichai Cohen, “Rules and Standards in the Application of IHL,” 39 Isr. L. Rev. 41 (2008).
113. E.g., Stephen Breyer, “Vermont Yankee and the Courts’ Role in the Nuclear Energy Controversy,” 91 Harv. L. Rev. 833 (1978).
114. I do not intend to go here into the full discussion in U.S. administrative law regarding the correct parameters of review of administrative law. Suffice it to say that even after the Supreme Court’s decisions of Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 415 (1971), Motor Vehicle Mfrs. Ass’n of the United States, Inc. v. State Farm, 463 U.S. 29, 43 (1983), etc., procedural review still seems to be the court’s main tool of review of administrative action. The most vocal proponent of procedural review was Judge Bazelon of the D.C. Circuitv. E.g., Ethyl Corp. v. Environmental Protection Agency, 541 F. 2d 1 (DC Cir. 1975) cert. denied 426 US 941 (1976). Though it seems that his extreme views of exclusive procedural review were rejected, courts still mostly employ procedural review as the preferred method of review.
115. On judicial review of governmental agencies’ procedures, see generally, Richard Pierce et al., Administrative Law and Process, 3rd ed. (Foundation Press, 1999), chapter 6, p. 221. For a critical position, see Jerry L. Mashaw, Greed, Chaos and Governance: Using Public Choice to Improve Public Law (Yale UP, 1997), pp. 158-180. As said, I cannot exhaustively discuss this issue here, nor do I propose that procedural review is the only review available for courts over agency action.
116. In the United States the discussion revolved around the “hard look” view. This view is usually rejected by courts, judges and academics alike. For an extended discussion, see Pierce, supra note 115, at pp. 386-393.
117. On legal advisors in modern military conflicts, see Kenneth Anderson, “The Role of the United States Military Lawyer in Projecting a Vision of the Laws of War,” 2 Chi. J. Int. L. 445 (2003).
118. Jonathan Masur, “A Hard Look or a Blind Eye: Administrative Law and Military Deference,” 56 Hast. L.J. 441 (2005) (describing the deference administrative courts give to the military, and criticizing it).
119. Application no. 57950/00 Isayeva v. Russia, European Court of Human Rights, Judgment of February 24, 2005.
120. Ibid., para. 209-214.
121. Contrary to some claims made by NGOs, there exists no specific requirement for an international investigation, although of course such an investigation would seem to be more independent. I know of only one incident of loss of life where a government appointed a committee of investigation which included international members. That is the Saville inquiry initiated by the British government in 1998 in order to investigate the events of Bloody Sunday in Ulster in 1972.
122. Indeed, this is the practice of the commission of inquiry appointed by the UN. E.g., the commission of the UN appointed to investigate the Qana incident in Lebanon in 1996 was headed by a retired general. See Report of the Secretary’s General Military Advisor Concerning the Shelling of the United Nations Compound at Qana on 18 April 1996 (1 May 1996), U.N. Doc. s/1996/337. annex.
123. On command responsibility in general, see Yuval Shany and Keren R. Michaeli, “The Case against Ariel Sharon: Revisiting the Doctrine of Command Responsibility,” 34 N.Y.U. J. Int’l L. & Pol. (2002) 797, 816-867; M. Lippman, “Humanitarian Law: The Uncertain Contours of Command Responsibility,” 9 Tulsa J. Comp. & Int’l L. (2001-2002) 1; A.D. Mitchell, “Failure to Halt, Prevent or Punish: The Doctrine of Command Responsibility for War Crimes,” 22 Sydney L. Rev. (2000) 381.
124. ICC Statute, Art. 33(1). Note that according to the Elements of Crime, soldiers in the field are expected to make a value judgment on the proportional effects of their acts on the basis of the information available to them. International Criminal Court, Elements of Crimes, note 37, U.N. Doc. PCNICC/2000/1/Add.2 (2000).
125. E.g., Yotam Feldman, “Operation Cast Lead: The Military Advocates Let the IDF Win,” Ha’aretz [Hebrew], January 23, 2009 [my translation of the title-AC], available at .
126. Ibid.
127. Palestinian Center for Human Rights, Report, March 12, 2009, available at
128. “Majority of Palestinians Killed in Operation Cast Lead: Terror Operatives,” IDF Spokesperson, March 26, 2009.
129. The importance of the total number of civilian casualties is supported by the controversial Kuperskic decision of the International Criminal Tribunal for Former Yugoslavia. The court declared that “in case of repeated attacks, all or most of them falling within the grey area between indisputable legality and unlawfulness, it might be warranted to conclude that the cumulative effect of such acts entails that they may not be in keeping with international law.” Prosecutor v. Kuperskic (ICTY, Trial Chamber, Judgment of 14 January 2000), para. 526.
130. A separate question is whether such investigations should be undertaken by an independent commission. I do not wish to go into this discussion in this study.