The McGowan Davis/Schabas Inquiry: The UN Legal Pogrom

Israel and the Rules of International Humanitarian Law:
Military Necessity, Humanity, Distinction, Proportionality, and Precaution

The Principle of Proportionality

The principle of proportionality requires that the anticipated loss of life and damage to property incidental to attacks must not be excessive in relation to the concrete and direct military advantage expected to be gained.

Applicable Law


A. Treaty Law


Article 51, Additional Protocol I of the Geneva Conventions:


"Indiscriminate attacks are prohibited... Among others, the following types of attacks are to be considered as indiscriminate...
(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."

Article 57(2)(a)(iii) , Additional Protocol I of the Geneva Conventions:


"With respect to attacks, the following precautions shall be taken:
(a) those who plan or decide upon an attack shall:...
(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."

Article 57(2)(b) , Additional Protocol I of the Geneva Conventions:


"[A]n attack shall be cancelled or suspended if it becomes apparent that the objective is not a military one or is subject to special protection or that the attack 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."

B. Customary Law


ICRC Customary Rule 14:


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

ICRC Customary Rule 18:


"Each party to the conflict must do everything feasible to assess whether the attack 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."

ICRC Customary Rule 19:


"Each party to the conflict must do everything feasible to cancel or suspend an attack if it becomes apparent that the target is not a military objective or that the attack 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."

C. Learned jurists and commentary


Yoram Dinstein,The Conduct of Hostilities Under the Law of International Armed Conflict, Cambridge University Press, February 26, 2004:


"The fact that a lawful target has been properly identified as such is not conclusive of a targeting process. A crucial consideration - in addition to identification of a lawful target for what it is - is proportionality in terms of the collateral damage expected from an attack. Collateral damage represents: (a) incidental losses or injury to civilians; (b) destruction of or damage to civilian objects; or (c) a combination of both. . . . Current customary [IHL] definitely confirms the precept that an attack against military objectives expected to cause disproportionate collateral damage to civilians or civilian objects, in relation to the military advantage anticipated, is unlawful. This is the core of the principle of proportionality . . . . In the words of Judge R. Higgins, in her Dissenting Opinion in the Nuclear Weapons Advisory Opinion: The principle of proportionality, even if finding no specific mention, is reflected in many provisions of Additional Protocol I to the Geneva Conventions of 1949. Thus even a legitimate target may not be attacked if the collateral civilian casualties would be disproportionate to the specific military gain from the attack.

It must be grasped that:

  1. A military objective does not cease being a military objective only because its attack would be expected to cause disproportionate collateral damage to civilians or civilian objects. The point is that, notwithstanding the unambiguous identification of an object as a military objective, its attack will still be illegal if the incidental injury to civilians or damage to civilian objects is expected to be disproportionate. The principle of proportionality provides 'a further restriction' by disallowing attacks against impeccably lawful targets owing to the envisaged disproportionate injury/damage to civilians or civilian objects.

  2. Proportionality has nothing to do with injury to combatants or damage to military objectives. [IHL] does not require any proportionality between combatants' losses on the two warring sides: the losses inflicted on enemy combatants and damage to military objectives may be immeasurably greater than the counterpart casualties and destruction suffered at the enemy's hand. . . . Proportionality in collateral damage is strictly limited to injury/damage to civilians or civilian objects.

  3. Even collateral damage to civilians and civilian objects is by no means determined by purely crunching numbers of casualties and destruction on both sides. The yardstick of proportionality is more complex, requiring a balance between the anticipated gain (in military terms) and pain (to civilians who fall prey to collateral damage) or destruction (of civilian property). . .

The practical significance of the principle of proportionality can only be fully appreciated when it is borne in mind that almost every object (which is not a military objective by nature) may be transformed into a military objective through use, purpose or location...The requirement of identification of an object as a military objective is, consequently, outstripped (and, in some sense, eclipsed) by the need to comply with the principle of proportionality. Proportionality is the true guarantee of robust civilian protection from the effects of attacks in wartime."

Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia, June 13, 2000:


"48. The main problem with the principle of proportionality is not whether or not it exists but what it means and how it is to be applied. It is relatively simple to state that there must be an acceptable relation between the legitimate destructive effect and undesirable collateral effects. For example, bombing a refugee camp is obviously prohibited if its only military significance is that people in the camp are knitting socks for soldiers. Conversely, an air strike on an ammunition dump should not be prohibited merely because a farmer is plowing a field in the area. Unfortunately, most applications of the principle of proportionality are not quite so clear cut. It is much easier to formulate the principle of proportionality in general terms than it is to apply it to a particular set of circumstances because the comparison is often between unlike quantities and values. One cannot easily assess the value of innocent human lives as opposed to capturing a particular military objective. . . .
50. The answers to [questions about proportionality] are not simple. It may be necessary to resolve them on a case by case basis, and the answers may differ depending on the background and values of the decision maker. It is unlikely that a human rights lawyer and an experienced combat commander would assign the same relative values to military advantage and to injury to noncombatants. Further, it is unlikely that military commanders with different doctrinal backgrounds and differing degrees of combat experience or national military histories would always agree in close cases. It is suggested that the determination of relative values must be that of the "reasonable military commander". Although there will be room for argument in close cases, there will be many cases where reasonable military commanders will agree that the injury to noncombatants or the damage to civilian objects was clearly disproportionate to the military advantage gained."

Yoram Dinstein, "Distinction and Loss of Civilian Protection in International Armed Conflict," U.S. Naval War College International Law Studies, Volume 84, International Law and Military Operations, 2008:


"It follows that the key to robust civilian protection lies, perhaps, less in the fundamental requirement of concentrating attacks on identifiable military objectives and more in the complementary legal condition of observing proportionality in the effects of the attack. This means, as prescribed in Article 51 (5)(b) of Protocol I, that-when an attack against a military objective is planned-incidental losses to civilians or civilian objects (usually called 'collateral damage') must not be expected to be excessive in relation to the concrete and direct military advantage anticipated. Intentionally launching an attack in the knowledge that it will cause incidental loss of life or injury to civilians or damage to civilian objects which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated is a war crime under Artide 8(2)(b)(iv) of the Rome Statute of the International Criminal Court... Undeniably, what is deemed excessive is often a matter of subjective appraisal, which takes place in the mind of the beholder (always remembering that the appraisal must be done in a reasonable fashion ). The difficulty is that military advantage and civilian casualties are like the metaphorical apples and oranges: a comparison between them is an art, not a science. Civilian losses can be counted, civilian damage can be surveyed and estimated, but how can you quantify a military advantage on a measurable scale? Additionally, since the entire process is a matter of pre-attack evaluation and expectation, it must be acknowledged that it is embedded in probabilities. What is to be done if 'the probability of gaining the military advantage and of affecting the civilian population is not 100 percent but lower and different'?

All the circumstances must be factored in. Thus, the bombardment of a hospital or a church used by the enemy may be given a green light if the actual number of patients or worshippers on site is negligible, whereas, should the numbers be disproportionate, the attack may have to be aborted. However, there is a difference between the cases of, say, one mosque where the minaret is used by a single enemy sniper and another serving as a command and control center of an armored division.

Taking out the sniper must not entail a substantial civilian price tag, but the elimination of a key command and control center is a different matter. It has to be borne in mind that 'excessive' is not interchangeable with 'extensive.' Some scholars take that position, but it is based on a misreading of the text. If the strategic and military value of a military objective is exceedingly high, significant collateral civilian losses resulting from an attack may well be countenanced.

Any planned attack-and any commensurate estimate of the number of civilians present in or near military objectives-must be based on up-to-date intelligence. The 'fog of war' is such that mistakes are unavoidable in every sizable military operation. When a legal analysis is made after the event, there is a built-in temptation to scrutinize the situation with the benefit of hindsight. But this temptation must be strongly resisted. The proper question is not whether collateral damage to civilians proved to be excessive in actuality: it is whether collateral damage could or should have been reasonably expected to be excessive at the time of planning, ordering or carrying out the attack. A reasonable expectation has to be linked to the data collated and interpreted at the time of action. Evidently, a valid evaluation of the state of affairs must be based on information that is current and not obsolete. If crucial information (say, about the absence of civilians from the vicinity of a military objective) is derived from a reconnaissance mission, the attack should follow soon thereafter since a long interval may mean that the facts on the ground have undergone a profound change."

Christopher Greenwood, "The Law of Weaponry," International Law Studies Volume 71: The Law of Armed Conflict: Into the Next Millennium, Eds. Michael N. Schmitt and Leslie C. Green, 1998:


"[E]ven if the target of an attack is a legitimate military objective, the principle of proportionality provides that it is prohibited to proceed with the attack if it: '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.'...

These principles are some of the most important cornerstones of the law of armed conflict. They were so widely disregarded during the Second World War that it was open to question whether they could still be regarded as part of the customary law. Since 1945, however, they have been reaffirmed on a number of occasions, most recently in Additional Protocol I, and were applied by, for example, the Coalition States in the operations against Iraq in 1991. Their status as part of the contemporary customary law cannot now be doubted.

While difficulties in their application remain, Protocol I has resolved a great many problems. In particular, it has clarified the principle that attacks must be directed only against military objectives by offering a workable definition of a military objective and has made clear that in applying the test of proportionality, only a 'concrete and direct military advantage,' rather than a nebulous concept such as the effect on enemy morale, is to be weighed against the effect of an attack upon the civilian population... The principle of proportionality, on the other hand, involves a balancing of the military advantages to be gained from an attack upon a military target against expected civilian losses and damage. As with the principle of unnecessary suffering, if those same military advantages can be achieved in different ways, one of which involves likely civilian casualties whereas the other does not, then the choice of the first route will entail a violation of the principle. However, the same qualifications apply here. In determining whether a commander who possesses a choice of weapons or methods of warfare should select one rather than the other, the extent to which both are truly available to him (in the light of such considerations as the likely future calls on precision munitions, the protection of his own forces and the logistic questions considered in the previous section) must be examined. The difference is that, although the security of his own forces remains an important part of this calculation, the need to reduce the risk to the civilian population means that a commander may be required to accept a higher degree of risk to his own forces.

Where the proportionality principle differs from the unnecessary suffering principle is that it is clearly established that it does not stop at the prohibition of unnecessary collateral injury and damage, but also requires a belligerent to abstain from an attack altogether, even if that means losing a military advantage which cannot be obtained by other means, if the military advantage would not be worth the expected civilian casualties and damage. The principle of proportionality is thus a more substantial constraint than the unnecessary suffering principle. Nevertheless, it remains a requirement to balance military gains against civilian losses; it does not possess an absolute character. In this respect, the Commentary on Additional Protocol I published by the International Committee of the Red Cross is misleading when it says that:

'The idea has been put forward that even if they are very high, civilian losses and damage may be justified if the military advantage at stake is of great importance. This idea is contrary to the fundamental rules of the Protocol. ... The Protocol does not provide any justification for attacks which cause extensive civilian losses and damage. Incidental losses and damage should never be extensive.'

What the principle of proportionality (as stated in both customary law and the Protocol) prohibits is the causing of excessive civilian losses and damage. By substituting the word extensive, the Commentary replaces a term which necessarily implies a balance between two competing considerations with a term which suggests an absolute ceiling on civilian losses. There is no basis in the law for such an approach."

Laurie Blank, "Asymmetries and Proportionalities," The Hill, July 29, 2014:


"Proportionality ... is a legal term with a specific legal meaning. It is one of a set of fundamental legal obligations that helps to minimize suffering during wartime. The principle of proportionality forbids attacks in which the expected civilian casualties from the attack will be excessive in relation to the anticipated military advantage gained.

Any analysis of proportionality during conflict thus rests on several key concepts. First, proportionality applies to attacks on lawful military targets that pose a risk of incidental civilian harm. The principle of proportionality seeks to enhance protections for civilians - beyond the prohibition on deliberate attacks on civilians - by minimizing the incidental harm that is an unfortunate but inevitable result of military action.

Second, the principle of proportionality accepts that not all civilian deaths during war are unlawful. The law recognizes the complexity and fluidity of the operational environment - including the location and movement of both enemy personnel and civilians, weather, terrain, capability of weapons and personnel, enemy tactics and conduct, for example - and seeks to minimize civilian harm while accepting that such harm cannot be eliminated altogether.

Third, the essential analytical factor in the proportionality framework is 'excessive.' Attacks that are likely to cause excessive civilian casualties in light of the military gain from the attack are prohibited. Not attacks that are likely to cause any civilian casualties, nor attacks that are likely to cause some civilian casualties, nor attacks that are likely to cause civilian casualties slightly greater than the military gain.

Fourth, proportionality is a prospective analysis, as the wording of key treaty provisions highlights: 'expected' civilian casualties; 'anticipated' military advantage gained; and 'in the circumstances ruling at the time.' Commanders must assess whether the risk of civilian harm is excessive given the anticipated military advantage based on information about the target, about civilians in the area and their patterns of movement, about the weapons being deployed and their known or anticipated blast radii or other consequences, and a host of other considerations. The lawfulness of attacks then depends on whether those assessments were objectively reasonable based on the information available at the time of the attack. Hindsight has no role here.

There is no doubt that, after an attack, the urge to simply count up the casualties and declare a war crime is powerful. However, an effects-based analysis - that is, using the numbers of casualties and extent of destruction to make legal claims - is simply incorrect. The law does not require that commanders be right all the time. The law also does not require perfect accuracy in targeting. But it does require extensive steps to protect civilians and reasonable judgments about the potential harm to civilians and the actions needed to minimize that harm...

For both legal and practical reason . . . conflating asymmetry and proportionality is both inaccurate and harmful to the law of war's core purposes. Disproportionate does not mean asymmetric, nor does it mean 'more than.' And asymmetrical does not mean disproportionate, nor does it mean criminal. They are simply wholly different concepts: one a factual game of numbers and one a comprehensive legal analysis."

Kenneth Anderson and Laurie Blank, "Laurie Blank follow-up on Gaza, proportionality, and the law of war," The Washington Post, August 6, 2014:


"[P]roportionality is more than just a principle; it is a methodology for assessing lawfulness in advance through careful consideration of both the value of the military advantage and the likelihood of civilian casualties. The principle tells us what we are trying to achieve - a balance between military needs and humanitarian concerns that minimizes civilian harm as much as possible. The methodology provides guidance on how to achieve that goal - by gathering and analyzing information about both the military value of the target and the consequences to the civilian population and making choices among various operational alternatives to achieve the mission while minimizing harm to civilians...

[A]ssessing the legality of an attack that results in civilian casualties must be done prospectively, based on the information the commander knew or should have known at the time of the attack. The standard is "reasonableness" - whether a reasonable commander in the same position would determine, based on the information available at the time, that the expected civilian casualties would be excessive in light of the anticipated military advantage.

Key to this assessment is not whether the court, the media, or anyone else thinks the decision was right or would have actually made the very same decision. Nor is it whether any resulting casualties seemed or even were excessive afterwards. The controlling factor in assessing proportionality after an attack is whether the commander's determination-that the likely civilian casualties in that operation would not be excessive-was reasonable. This reasonableness assessment can only be made with a full understanding of the situation and all relevant information at the time of the attack - and, just as important - an awareness of what is considered to be reasonable in light of general practice.

International tribunals have rarely undertaken this analysis. This may well be simply because they have no lack of much easier and obvious cases. Their dockets can easily be filled to overflowing with the staggering number of deliberate crimes against civilians that demand accountability-such as genocide, crimes against humanity, torture and a host of other categorical violations that require no difficult judgments about the reasonableness of a commander's judgment. The difficulty in translating the proportionality rule from the operational dynamic of the battlefield and the fog of war to the evidence-bound confines of the courtroom is certainly another factor. The few instances of adjudication, however, consistently reinforce both the prospective approach and reasonableness as the touchstones of the analysis. Responsible militaries, for their part, investigate and review every incident involving civilian casualties to determine whether further investigation or prosecution is warranted-and simply to improve training and implementation to mitigate civilian harm in future missions. Both internal and international inquiries have often explored, or attempted to explore, proportionality with respect to specific incidents...

The international law governing when states may use force in self-defense (the jus ad bellum) also has a requirement of proportionality, but it is quite distinct (and serves a different purpose) from the law of war rule of proportionality discussed above. This jus ad bellum rule of proportionality mandates that a state acting in self-defense in response to an armed attack can only use force that is proportionate to the needs and goals of repelling or deterring the attack. This is not a "tit-for-tat" requirement, however, limiting the state acting in self-defense to only what its attackers did. There is no obligation of symmetry between the original attack and the force used in self-defense; indeed, the force needed to repel an attack may well be disproportionate relative to the the original attack, in order to stop it and deter continuing attacks. What it must be, instead, is proportionate to the ends of stopping and deterring the original attack and further attacks...

Importantly, this rule of proportionality does not address civilian casualties. That is the task of the law of war principle of proportionality analyzed above. Unfortunately, these two concepts of proportionality are regularly conflated, leading to misunderstandings and ineffective legal analyses. First, if the bare fact of civilian casualties were to become the measure of whether the overall use of force in self-defense is lawful, the international legal framework governing the use of force in self-defense would be undermined. Any military operation causing civilian casualties would then be considered unlawful, even if a valid exercise of self-defense, emasculating state options for protecting their own civilians against attack.

Second, focusing on civilian casualties, without any legal analysis of proportionality, the targeting process or the nature of the objective attacked, simply incentivizes insurgent groups to co-mingle military personnel and assets within the civilian population and use civilians as a shield, thus causing greater and greater numbers of civilian casualties and louder claims of unjust war and war crimes. Facilitating the defending party's exploitation of the law for its own defensive and propaganda purposes in this way gravely endangers the very persons the law of war seeks to protect-the civilians caught up in the combat zone-and thus undermines the essential fabric of the law of war."

Israel Ministry of Foreign Affairs, "Behind the Headlines: Fighting Hamas Terrorism Within the Law," August 7, 2014:


"International humanitarian law also requires that any military attack be "proportionate" in the sense that expected collateral damage to civilians or civilian objects must not be excessive in relation to the military advantage anticipated to be achieved as a result of such attack. Note that the test of proportionality does not consider the actual results of the attack after the fact; rather, it is a test that looks at expected (not actual) results, given the anticipated objectives, prior to the execution of the attack. This is a complex and difficult calculation and international law relies on the commander in the field in the heat of the conflict to weigh all relevant considerations, including the security of his own forces."

The Jewish Institute for National Security Affairs, "2014 Gaza War Assessment: The New Face of Conflict," JINSA-commissioned Gaza Conflict Task Force, March 10, 2015:


"This does not require a precise symmetry between the original attack and the use of force in response – possibly to be measured by comparing casualty figures between the two sides – but rather that the measure of counterforce used is proportionate to the needs and goals of repelling the original attack. Hence, the force used in response may be significantly greater than that used in the original attack if it is necessary to stop or defeat the original attack."

Alan Baker, "The Latest Hamas-Israel Confrontation - Some Pertinent Legal Points," Jerusalem Center for Public Affairs, July 24, 2014:


"A monograph entitled 'Applying the Principle of Proportionality in Combat Operations,' published by of the Oxford Institute for Ethics, Law and Armed Conflict, states: '...harming civilians is not in itself illegal. An injury to civilians or damage done to civilian objects as a side-effect of a military operation may be permissible provided that it is proportionate to the military gain anticipated from the operation.'"