A lethal misconception, in Gaza and beyond: disguising indiscriminate attacks as potentially proportionate in discourses on the laws of war

Written by

In recent years and, despite the reality, in the last weeks, a trend among laws of armed conflict scholars has consolidated. This trend is in antithesis to the choral denunciation of mass civilian victimization as unlawful and criminal in other conflict theatres (such as, agreeably, in relation to the war against Ukraine). In contrast, bombing campaigns in Gaza have attracted and continue to attract the highest volume of academic commentaries emphasizing the limits of and derogations to civilian immunity from attacks. The ‘unfortunate/tragic but probably proportionate civilian loss of life’ formula has been repeated by countless analyses.

Notwithstanding the structural illegalities of this conflict, it is not new to read abundance of legal commentaries embracing the prism of the potential lawfulness of the killing of crowds of Palestinian civilians. However, what is new is that the embracement of this same prism of validations continues unabated while the killing of besieged, displaced, healthcare-denied and increasingly starving civilians in Gaza assumes shocking proportions by the minute. This trend continues despite top Israeli officials announced this offensive denying the existence itself of innocent civilians, and despite IDF spokespersons made clear this campaign had no accuracy priority.

This post (synthetizing some arguments of a longer article) submits that the described legitimizing prism reproduces some serious misconceptions, particularly on how we should construe the relationship between distinction and proportionality under customary IHL.

Many analyses seem to have so far unduly overexpanded proportionality into the realm of distinction, contributing to the current failure of most IHL and international criminal restraints. Even recent analyses debating in bello proportionality of IDF attacks, each killing and injuring several hundreds of civilians, such as the bombing of the Jabalia refugee camp in Northern Gaza (see e.g., inter alia, Manea), do not mention once the prohibition of indiscriminate attacks, and seem to imply that violations of distinction do not occur unless “direct and deliberate attacks” against civilians or civilian objects are launched.

I do not disagree with the facts indicated by Andrea. However, assuming Hamas members omitted to take passive precautions (determining this must include feasibility considerations, see Schmitt), and assuming tunnels are lawful targets (which can be neutralized or captured via less devastating methods), this does in no way render Palestinian civilians less civilian, less protected, or admissible targets of proximity. It is aerial attacks we are discussing. There is no way, I think, to bomb underground tunnels with GBU-31 warheads without knowingly flattening densely populated neighbourhoods with hundreds of civilians that will certainly be killed.

The customary prohibition of indiscriminate attacks, as a consequence, becomes crucial, and reminds lawyers and armed actors that distinction does not only prohibit direct and deliberate attacks on civilians and civilian objects, it also prohibits attacks renouncing to discern between them and lawful targets. This prohibition does not apply exclusively to the clearly indiscriminate attacks by Hamas. It is applicable to all parties to the conflict, included the IDF. What I’ll try to submit here is that attackers, in all conflicts, might well breach not only IHL, but even war crimes law without deliberately targeting solely civilians and civilian objects as such.

Customary IHL rules governing targeting include: 1) distinction, that requires parties to conflicts to always distinguish, to the maximum extent feasible, between combatants and civilians, and between military objectives and civilian objects, posing an absolute prohibition of direct attacks against the latter; 2) proportionality, inoperative unless distinction is respected (more infra), prohibits even attacks directed at lawful targets in case they “may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof [civilian harm], which would be excessive in relation to the concrete and direct military advantage anticipated”; 3) precaution, corollary of both distinction and proportionality, imposes a legal duty of constant care, in the conduct of military operations, to spare civilian and civilian objects, to verify that objects to be attacked are lawful targets, and take “all feasible precautions to avoid, and in any event to minimize, incidental” civilian harm. These three fundamental rules can perform their protective functions only if respected cumulatively, bound in a threefold synallagmatic nexus presiding to the legality of attacks. At the same time, they maintain separate spheres of operativity, of distinct operational importance, delineating obligations to be respected in sequence.

How does this sequence work? How do we decide whether attacks raise just proportionality and not, instead, previous distinction concerns, particularly when civilians are killed and maimed en masse? Is there something that needs to be preliminary scrutinized in order to gain access to proportionality evaluations?

In my opinion, the answer is undoubtedly ‘yes’. Neglecting what entitles attackers themselves to lawfully engage in proportionality assessments can have devastating consequences (which I think we are tragically witnessing), fundamentally altering how IHL strikes its normative balance between military necessity and humanity. Logically and normatively, it remains impossible assessing proportionality when attackers have most likely launched indiscriminate attacks. It is also dangerously misleading, nurturing the belief that whatever the entity of calculated mass killing of civilians, it can always be ‘proportionate’ if high value military targets are among the people and objects struck. This is very far from a systematic reading of IHL. Let me better articulate this point.

Indiscriminate attacks are those that “a) are not directed at a specific military objective; b) employ a method or means of combat which cannot be directed at a specific military objective; or c) employ a method or means of combat the effects of which cannot be limited as required” by IHL. These attacks are unlawful because they violate the basic rule of distinction (or discrimination). Now, in at least the latter two types of attacks prohibited (under ‘b’ and ‘c’), all attackers might claim that their aim was to strike the lawful targets, and that they did not directly intend to shell the protected persons or objects, but this was an unfortunate necessity. We must convene, however, that IHL categorically rejects the admissibility of this claim, because even these types of attacks are “of a nature to strike military objectives and civilians or civilian objects without distinction”. The nature of the means or methods employed, therefore, is enough to render irrelevant that attackers only had the primary purpose of striking lawful objectives, so long as they also accepted, by using those means or methods, to strike a number of protected persons and objects alongside the lawful targets, indistinctly.

More: consider the prohibition of area bombardments. Customary IHL forbids attacks that “by any methods or means […] treat 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”. This prohibition confirms that even when several lawful targets are the directly intended objective of the attack, it remains absolutely (not relatively!) prohibited to strike them via means or methods that involve the shelling of a similar concentration of protected objects or persons interspersed among the lawful targets (even if several). This is particularly the case in densely populated areas (see par. 1973).

Indisputably, therefore, IHL also considers the calculated, indirect shelling of high concentrations of civilians and civilian objects as fully able to breach the prohibition of indiscriminate attacks, in violation of distinction/discrimination.

Now, even disproportionate attacks accept to hit civilians and civilian objects, also indirectly. Hence, the question arises: how disproportionate attacks differ from “other types” (as art. 51(5) AP I states) of indiscriminate attacks?

Disproportion (i.e. excessiveness of the foreseeable incidental civilian harm, in comparison to the concrete and direct military advantage anticipated) trespass the legal red lines separating civilians from combatants only: 1) indirectly, because civilians and civilian objects in these cases are not a direct target; in itself, however, this aspect would not sufficiently differentiate these attacks from other indiscriminate attacks. In area bombardments, the directly intended targets are still military objectives, potentially many, but the means or methods of the attack are such that they will inevitably impact these targets alongside a similar concentration or number of protected objects or persons. Therefore, disproportionate are those attacks that blur the mentioned protective red lines not only indirectly, but also 2) incidentally, because these attacks do not contemplate the shelling of similar concentrations (let alone of an incommensurably higher concentrations or numbers) of protected persons and objects alongside the lawful targets, but rather remain prohibited autonomously because they accept a foreseeable side consequence of the attack on civilians or civilian objects that might be excessive.

That extensive civilian harm is just foreseen, but not directly pursued as such, therefore, is clearly not a reason to exclude the indiscriminate nature of an attack. It is, thus, not enough to rush into debating proportionality.

Effectively, it would not be enough even to exclude the war crimes of intentionally attacking civilians or civilian objects. Some clarifications on this profile. Authors (see e.g. Ohlin, Heller, Haque) have long debate the relationship between the war crime of intentional attacks on civilians and civilian objects as such, the war crime of disproportionate attacks, and the role of recklessness. ICC jurisprudence (unlike that of the ICTY on unlawful attacks) has excluded that this means rea threshold is sufficient to perfect crimes. However, knowingly killing and maiming entire crowds of civilians, plus destroying multiple civilian objects, sits way above this threshold. Indirect/oblique intent, in fact, is different from broth recklessness and dolus eventualis. Under the latter, a perpetrator consciously disregards a substantial and unjustifiable risk that the material element will result from his conduct. Under the former, indirect/oblique intent (valid means rea under the Rome Statute), a perpetrator is considered to intend not only the pursued result of an action but even other anti-juridical results, when they are both virtually certain to occur and appreciated by the suspect. In cases such as the attack on the Jabalia refugee camp, virtual (if not absolute) certainty is surely met by the shelling of an extremely densely populated neighbourhood, killing and injuring potentially thousands of civilians.

Does this also mean that knowingly killing one or a small number of civilians or damaging civilian objects while targeting high value targets always amount to the offences under art. 8(2)(b)(i) and (ii) RS? It does not. The issue here is not with civilian harm that can be reasonably considered incidental. The issue is conduct of indiscriminate warfare accompanied by claims that killing (no matter how many) crowds of civilians of another national group in aerial attacks is always incidental and never excessive. Thus, should attacks calculating killing crowds of civilians (and razing multiple civilian objects) be scrutinized under the war crime of disproportionate attacks (art.8(2)(b)(iv) RS), or rather under the war crimes of intentionally attacking civilians and/or objects? The answer depends on the possibility to qualify the virtually certain civilian harm as extensive and preponderant in comparison to collateral military advantages. We need, in good substance, a legal parameter allowing us to qualify the civilian harm calculated (even if not directly pursued) in order to understand whether its entity was so extensive as to affirm that attacks were blind to discrimination.

To reasonably ponder this expected harm under the sole realm of proportionality, IHL requires it to be incidental. Incidental civilian harm remains to be avoided or minimized even when not excessive under mandatory precautions. Some military lawyers tend to interpret incidentality as accidentality (civilian harm not pursued). This is an exceptionally narrow interpretation of the term, re-translating the concept as ‘harm incidental to attacks.’ Crucially, incidentality in the wording of AP I provision is never relative to attacks, but always and unequivocally referred to the civilian harm.

Literature dedicating specific attention to the incidental harm side of proportionality assessments has synthetically affirmed that the requirement that the civilian harm must be incidental means that “it must occur in the course of an attack directed against a military objective”, while abstaining from other clarifications about the meaning of incidentality in its systematic normative context. This temporal understanding of incidentality appears reductionist. That civilian harm occurs in the course of an attack directed against a military objective, in fact, is only sufficient to exclude that the attack was directly intended to strike civilians or civilian objects as such, and nothing else.

The fact that the relevant prognostic evaluation is an ex ante assessment, in addition, does not mean that incidentality can be reduced to a purely subjective determination of attackers (eventually attackers devaluing as less worthy of protection the life of neighbouring civilians). A reasonably well-informed (par. 58) agent not only can but actually must foresee, as a minimum, the consequences of an action in their objective measure, before being able to assign any value to them.

Admitting non-incidental, extremely extensive, preponderant civilian harm expected as such to be balanced by high military advantages anticipated dissolves a fundamental boundary of proportionality evaluations and projects them beyond their legal limits. IHL does not accepts calculated mass maiming and killing of large crowds of civilians, certain extensive destruction of civilian infrastructure, to be balanced by high values assigned by attackers to specific targets. In 1987 it was already clear that this view is against IHL: “[t]he 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; the Protocol does not provide any justification for attacks which cause extensive civilian losses and damages. Incidental losses and damages should never be extensive” (see par. 1980).

The interpretative misconception discussed, in sum, risks to rewrite IHL as framework enabling the vindication as ‘collateral damage’ of unlimitedly indiscriminate killing and destruction.

Thus, in my opinion (with a coordinated, systematic reading of IHL and war crimes law), to be passible of proportionality assessment, civilian harm must be anticipated as incidental from the point of view of attackers in a threefold sense. It will need to be anticipated as incidental:

  1. in relation, and in contrast to the main result of the attack pursued, from the volitional point of view.
  2. in relation, and in contrast to the main result of the attack foreseeable, from the cognitive/prognostic point of view of the representation of the consequences of the action.
  3. in relation, and in contrast to the main result (of military character) of the attack objectively, from the material point of view, which needs as a minimum to be logically distinguishable and preponderant in comparison to the, secondary and indirect, impact on civilians and civilian objects.

Absent these conditions, debating proportionality appears a rather armies-friendly fictio iuris, since if:

  1. the civilian harm is expected to be, from the objective point of view, the incommensurably prevalent result of the attack in comparison, quantitatively, to a secondary (collateral) impact on legitimate military targets, plus
  2. the foreseeability of the main result of impact on civilians is so high as to be virtual or (as in Jabalia) absolute certainty, then
  3. it can be said that this prevalent impact of the attack on civilians and/or civilian objects is indirectly intended, for the simple reason that it is no more a ‘collateral’ one, and – in addition – it is expected with certainty and appreciated by the attacking forces.

When specific attacks are characterised by calculations of mass killing and extensive destruction of protected persons and objects, included specially protected persons and objects, included children, proportionality appears simply inoperative under IHL.

The Jabalia attack, in sum, was an indiscriminate attack, knowingly employing a method or means of combat which could not, in the circumstances, be directed at specific military objectives singularly identified (included underground), without certainly injuring, killing, and destroying much higher concentrations or numbers of protected persons and objects.

In conclusion, authors continuing to propose the prism of validations based on proportionality (even amidst the unfolding massacres and continuing incitements to genocide) might easily verify the moral feasibility of their interpretations in light of jus in bello equal application. Would they be willing to have the same interpretations they propose followed by strategic enemies in hypothetical future wars on their national soil? Would they be willing to consider as potentially proportionate, rather than blatantly indiscriminate, aerial attacks on their cities, thousands of tons of explosives raining down on their neighbourhoods, including the schools of their children, the hospitals where their injured family members are sheltered, and the infrastructures indispensable for the survival of their national group? I suspect and hope they would not.

Print Friendly, PDF & Email

Leave a Comment

Your comment will be revised by the site if needed.


Dr Mary Jane Fox says

November 7, 2023

The only thing the actions of the IDF are proportional to is Netayahu's uncontrollable and unprofessional outrage.

Sze Hong Lam says

November 7, 2023

Aerial bombing does not by nature make it an indiscriminate attack. The attack that took place at the Jabalia refugee camp was not a WW-II-styled "area bombardment" with unguided free-fall bombs but rather an attack using the JDAMs equipped with a GPS receiver that allows it to be guided specifically toward a military target (CEP of 5m according to the specs available online). From the information available, it also appeared that a delayed fuse was being used (from the tweet of Marc Garlasco helpfully cited in this post, and also the NYT coverage cited by the earlier post of Manea), which indicates that the JDAMs used by the IDF are more likely to be the "bunker buster" variations that are specifically designed to target military targets underground (the same point was helpfully raised by Dr. Brian L. Cox in Manea's earlier post). The nature of the ammunition certainly has a huge impact on whether the attack would be by nature indiscriminate and satisfy the mental element of indirect/oblique intent. While I do agree with the author's thesis that the distinction between an indiscriminate attack and a disproportionate attack was a fine one, I am not convinced by the mere facts that (1) the bombing of the Jabalia refugee camp was an aerial attack; and (2) the attack took place in a densely populated area would automatically put the Jabalia bombing in the first category.

Matthias Zechariah says

November 7, 2023

I do not purport to critique Dr. Luigi Daniele's ideas. It is one of the best analyses I have ever seen on this platform. I am in substantial agreement with his legal analysis, though I doubt whether his conclusions in relation to the Jabalia operations are absolutely correct. I think it is too early to assess issues of proportionality of incidental damage in relation to compliance with the principle of discrimination to civilians and civilian objects from the current 'developing' operations. Suffice it to say that I agree, in principle, with H. S. Lam that 'Aerial bombing does not by nature make it an indiscriminate attack', especially where precision weapons are deployed for such military attacks. Let me conclude by asking Dr. Daniele what should be the operational approach to attacking, if necessary, an area objectively assessed to be of high military necessity and yet is interspersed with huge civilian populations deliberately planted by the adverse party to an armed conflict as civilian shields? Secondly, and as an aside, do you regard Gaza as a State under international law, or as part of a State? If the answer is in the affirmative in respect of the latter, part of what State is it? This will help us in characterizing the nature of the armed conflict between Hamas and Israel/the IDF. Thank you.

Brian L. Cox says

November 7, 2023

There is enough material here to unpack and address that the comment section of the venerable EJIL:Talk! site isn't the appropriate venue to cover all relevant bases. I've saved this post in the reference folder for my ongoing dissertation project (examining the connections between LOAC and HRL in theory and practice) and will welcome the opportunity to address many of the points raised above in a long form venue. For now, I address three points: one doctrinal, one on selection of sources, and the last on the conceptual purpose of LOAC.

1) As a matter of doctrine, I encourage the author to incorporate the definition of military objectives (nature, location, purpose, or use...effective contribution...capture, destruction, or neutralization) when analyzing the bridge between distinction and discrimination. The first component of the discrimination rule (in order of listing) essentially represents the bridge between distinction and discrimination. The second discrimination component (cannot be directed) is not particularly relevant to the current analysis. The third discrimination component (the effects of which cannot be limited) ends with this clause: "as required by this Protocol". If we're using the text of AP I as the template for the discrimination rule (more on use of sources next), the "as required" component of the rule includes, among other requirements, the proportionality rule. With this framework in mind, I suggest that compliance with each of these rules described above - distinction, discrimination (the first component of which essentially restates the distinction rule, but the second and third components stand alone), and proportionality should be evaluated in sequence. If the attack is directed against a military objective (nature, location, purpose, or use), the distinction rule (and the first component of discrimination) is satisfied. If incidental damage is anticipated and a concrete and direct military advantage is expected, then the "excessive in relation to" proportionality standard must be considered. The outcome of the proportionality analysis informs (but doesn't completely subsume) the "effects of which cannot be limited" component of discrimination because proportionality is "required by this Protocol" (that is, AP I).

2) On selection of sources, the ICRC CIHL study is incredibly useful as a research tool, but methodological limitations and inherent ideological preferences suggest that caution should be utilized (even though caution rarely ever is) when referring to the study as a primary source of LOAC. AP I is widely ratified, but not universally enough that every component represents extensive and virtually uniform state practice with an accompanying opinio juris. Commentary from highly qualified publicists can be persuasive, but it remains a subsidiary source of customary law. We in the academy often seem to take on the role of the tail wagging the dog of state practice by pronouncing as "illegal" conduct that states routinely engage in. Perhaps it is the case that states are committing systemic violations of LOAC with seemingly every bomb they drop - but an alternative perspective is that states routinely comply with international law and it is our interpretations as non-state actors (that is, as subsidiary sources of law) that miss the mark.

3) On the conceptual purpose of LOAC - I didn't realize it would be the case when I first started my dissertation, but this topic has emerged as the primary conceptual frame for my study of the connections between LOAC and I/HRL. If the accepted purpose of LOAC (or, IHL depending on one's persuasion) truly is to be a set of rules that seeks, for humanitarian reasons, to limit the effects of armed conflict (as the venerable ICRC pronounces), then it makes sense to develop a progressively restrictive interpretation of the law that is decreasingly "armies-friendly" and instead increasingly focused on "denunciation of mass civilian victimization as unlawful and criminal." This progressive vision of LOAC, as commonly accepted as it may seem in public discourse, is a relatively recent development. Its roots can be traced primarily back to Pictet's effort in 1966 to merge the Hague and Geneva branches of LOAC with human rights law under the collective moniker of IHL (The Principles of International Humanitarian Law, 6 IRRC no. 66, pp. 455-69 (1966)). A competing - and, dare I say, more doctrinal - vision of the purpose of LOAC is described in the DoD LoW Manual: "prohibitions on conduct in the law of war may be understood to reflect States’ determinations that such conduct is militarily unnecessary per se." The universe of conduct that is *truly* unlawful based on the DoD formulation is infinitesimal as compared to the progressive vision of the ICRC. The 2012 attack by SSG Robert Bales against 22 civilians (which killed 16) in Kandahar, Afghanistan, for example, was a war crime – it violated the distinction rule and it was militarily unnecessary per se. The 2015 attack against the MSF trauma center in Kunduz City, Afghanistan, which killed 42 people and injured dozens more, was not a war crime – a doctrinal analysis reveals that all relevant LOAC rules were complied with (notwithstanding the derelict conclusions of the 15-6 investigator), and it was “militarily necessary” to attack what was believed to be a Taliban stronghold even though the target was *in fact* a medical facility. One may prefer to denounce an attack such as the latter as unlawful based on mass civilian victimization, but a doctrinal application of LOAC rules suggests that the mass victimization that *resulted* from the attack isn’t controlling – but rather, the intent of the attacker(s) *at the time of the strike* is. The conclusions of the latter are accurate regardless of whether the attack occurred in Kunduz, Afghanistan or in Ottawa, Ontario (where I currently reside). Those engaged in attacks during the conduct of hostilities are expected to do so on the basis of military necessity rather than individual personal emotions. Those engaged in analyzing compliance with LOAC from the outside after the fact should be held to the same standard.

Much more to say in a long form venue, though I imagine that I am pushing the boundaries of a worn-out welcome with the existing word count of these reflections presented in a comments section. The post presents a productive humanitarian perspective on the practical application of LOAC, and as always, the EJIL:Talk! site is to be commended for continuing to provide such a valuable platform for the interactive exchange of ideas.

Nicolas Boeglin says

November 7, 2023

Dear Professor Daniele

Many thanks for this extremely accurate analysis you present us. It seems that some colleagues feel obliged to support Israeli official position.

May I simply recall what we read in 2009 UN Report on the Gaza military operation:

"1921. The Mission found numerous instances of deliberate attacks on civilians and civilian
objects (individuals, whole families, houses, mosques) in violation of the fundamental
international humanitarian law principle of distinction, resulting in deaths and serious
injuries. In these cases the Mission found that the protected status of civilians was not
respected and the attacks were intentional, in clear violation of customary law reflected in
article 51 (2) and 75 of Additional Protocol I, article 27 of the Fourth Geneva Convention
and articles 6 and 7 of the International Covenant on Civil and Political Rights. In some
cases the Mission additionally concluded that the attack was also launched with the
intention of spreading terror among the civilian population. Moreover, in several of the
incidents investigated, the Israeli armed forces not only did not use their best efforts to
permit humanitarian organizations access to the wounded and medical relief, as required
by customary international law reflected in article 10 (2) of Additional Protocol I, but they
arbitrarily withheld such access

Source: (available in the UN official languages)- Document A/HRC/12/48


In 2008-2009 as well as in 2014 we hear the very same official objective of Israel: to "erradicate the terrorist of Hamas".

I'm wondering how many times Israel will committ the very same mistake it committs, thinking that desporportionnate and indiscriminate campaign of bombs on Gaza civil population will make Israel safer.

If I may ask, any detailled report of the exact number of bombs launched on Gaza since October 7 at night will be extremely usefull in order to try to understand the israelian logic.

Sincerely yours

Nicolas Boeglin

Luca Pasquet says

November 8, 2023

I must thank the author for this post, which is morally necessary in a time where international law is (mis)used to legitimize the most outrageous acts against human life and dignity. Grazie.

Omar Hammady says

November 8, 2023

As far as I understand, this article does not claim that aerial bombing is in itself an indiscriminate bombing. It cannot be disqualified simply on that basis. It states that such bombing could not "in the circumstances, be directed at specific military objectives singularly identified (included underground), without certainly injuring, killing, and destroying much higher concentrations or numbers of protected persons and objects." It's indeed a fact.

Luigi Daniele says

November 9, 2023

Many thanks all for the comments.

Sze, the post does not suggest was an area bombardment as in WW II but violation of the customary prohibition correspondent to 51(4)(b) AP I.

Matthias, sure, aerial bombardments are not per se indiscriminate attacks. They are when they are calculated to strike at once lawful targets and huge civilian populations, as you write. Alternative approaches: use of ground forces (given top capabilities)?

Brian, thanks. So much to discuss (hope we’ll have the chance). Just two notes for a future chat. First, I have referenced rule 8 and 17 ICRC study next to each other precisely thinking about your initial point (alternative achievement of equivalent military advantage + duty to select less harmful means and methods for civilians).
Second, you mention several times ‘as required by the Protocol’ in 51(4)(c). Good point, I agree that this might include proportionality. Note that this is why the rule itself affirms that the relevant type of attacks *cannot* respect it either (coherently, I think, with the systematisation the post proposes).

Marko Milanovic says

November 9, 2023

A reminder by the editors that only those comments with the commenters full name will be posted, not those made anonymously or using a pseudonym.