20 Dec On Forcing Civilians to Remain in Hostile Zones: The Prohibition of Shielding and the Corresponding Obligations of an Attacking Party Under International Humanitarian Law
[Chris McQuade is a Senior Teaching Fellow in Law at the University of Portsmouth. He holds a PhD in International Law from the University of Sussex and researches in the fields of public international law, international humanitarian law and international and domestic human rights law.]
In response to the October 7 attack by Hamas, the Israeli army has engaged in an intense military campaign in the Gaza strip over the past three months. As the campaign has escalated in its ferocity, so too has criticism of the Israeli response (among other posts, see here and here and here). Israel’s efforts to justify the impact of their campaign on Gaza’s civilian population are pervaded by a familiar theme. ‘Hamas uses innocent civilians as a human shield’ claimed Israeli Prime Minister Benjamin Netanyahu in 2014, in response to criticism of the then rising death toll during Operation ‘Protective Edge’. On 7 November 2023, in an interview with ABC News, Netanyahu once again claimed that Hamas uses ‘their civilians as human shields’ when asked whether he was concerned by reports that close to 10,000 civilians had been killed in Israel’s current campaign in Gaza. Similar claims were advanced with respect to Operation ‘Cast Lead’ in 2009, and in defence of Israel’s response to the 2021 March of the Return protest at the Gaza/Israel border.
These claims position Israel as the moral actor fighting to rid Gaza of a terrorist organisation that doesn’t respect their own civilians, with the direct implication that this mitigates responsibility for the destruction the Israeli response has caused thus far. The allegation of human shielding, however, is a precise violation of IHL, and such claims should be appropriately interrogated. Equally, the legal responsibility of the Israeli army is governed by the obligations incumbent upon them as a party to the conflict, not by the conduct of an opposing armed force.
Amidst the current bout of hostilities, one of the most persistent claims concerns the alleged practice of Hamas threatening to shoot civilians attempting to leave hostile areas of Gaza. Netanyahu stated in November that ‘Hamas is doing everything to keep [civilians] in harm’s way’ and indicated that they were preventing civilians from evacuating at gunpoint. This echoes the views of the Israeli army communicated in October via X (formerly Twitter), suggesting that Hamas were preventing civilians from evacuating to Southern Gaza. This particular allegation, characterised by Schmitt as a clear example of shielding, is noticeably different in form when compared to other alleged examples of Hamas’s shielding conduct. Suggestions that they have fired rockets in close proximity to civilians and have embedded military objectives within the civilian population have long been documented, but claims that Hamas have prevented evacuation have not been prominently addressed as a mode of shielding.
The remainder of this post will first set out the international law that prohibits the use of human shields, before assessing whether the reported practice of forcefully preventing civilians from evacuating hostile zones represents a violation. It will then consider the impact, if any, that a determination of such conduct as shielding has on the obligations of Israel as an attacking party.
The Prohibition of Human Shields under IHL
Both the Third and Fourth Geneva Conventions of 1949 prohibit the use of prisoners of war and ‘protected persons’ as human shields respectively, but Additional Protocol One to the Geneva Conventions (API) provides the most explicit prohibition of shielding, extending it to all civilians. Article 57 (1) reads:
The presence or movements of the civilian population or individual civilians shall not be used to render certain points or areas immune from military operations, in particular in attempts to shield military objectives from attacks or to shield, favour or impede military operations. The Parties to the conflict shall not direct the movement of the civilian population or individual civilians in order to attempt to shield military objectives from attacks or to shield military operations.
I shall skirt around the complex questions of how and why this prohibition applies to the conflict at hand by relying on the International Committee of the Red Cross’s widely accepted position that the use of human shields is prohibited customarily in both international (IAC) and non-international armed conflicts (NIAC). They indicate that shielding ‘requires an intentional co-location of military objectives and civilians or persons hors de combat with the specific intent of trying to prevent the targeting of those military objectives’(ICRC, Customary IHL study, Rule 97). The International Criminal Court also lists human shielding as a war crime in IACs, with the essence of their definition tied to the intention to use or otherwise take advantage of the presence of civilians to shield military targets (Elements of Crimes, Article 8(2)(b)(xxiii)). Although it is not explicitly listed as a war crime in NIACs, it would at the very least represent inhuman treatment, which would amount to a war crime in NIACS. Such was the case at the ICTY in Blaskić, where shielding was held to cause the requisite degree of mental suffering to constitute inhuman treatment (para. 716).
Does Forcing Civilians to Remain in Hostile Zones Violate the Prohibition of Shielding?
Some might question whether forcefully preventing civilians from leaving a general area is representative of shielding, as the conduct differs from more typical examples. Indeed, the paradigmatic example involves the movement of protected individuals to a military objective with the goal of preventing an attack on that objective. Such was the case during the 1990 Gulf War where Iraqi forces placed foreign nationals in strategic sites throughout Baghdad. The conflict in the Balkans witnessed similar incidents, most notably when the the Army of the Republika Srpska chained UNPROFOR personnel to military objectives to deter NATO attacks (Prosecutor v Karadžić, para 5860-5877).
The difference in form between preventing civilians from evacuating areas of hostilities and placing civilians in or around military targets, was an obstacle for the Report of the Secretary General’s Panel of Experts (POE) on Accountability in Sri Lanka. They were presented with facts indicating that the Liberation Tigers of Tamil Eelam (LTTE) forced civilians to remain in the Vanni region of Sri Lanka as it came under increasingly intense attack by the Sri Lankan Army. The POE described this not as human shielding, but as a ‘human buffer’ (p. 50). Much like the accusations levied against Hamas, accounts addressed by the POE suggested that the LTTE had adopted a policy of shooting anyone who tried to leave (p. 65).
With respect to the POE, their description of LTTE conduct as a ‘human buffer’, not as a use of human shields, was based on a misreading of the prohibition of shielding under IHL (p. 421). It was their opinion that as the civilians had not been moved to military targets (applying the ICRC definition outlined above) that the LTTE’s conduct could not amount to human shielding (p. 65). Yet, human shielding is clearly inclusive of a broader range of situations. API refers to using the ‘presence’ of civilians, the ICRC grounds their understanding in the notion of ‘co-location’ and the ICC refers to taking ‘advantage of the location of one or more civilians’. None of these tie human shielding to the movement of civilians to a military objective – it can also occur in reverse. Furthermore, API Article 51(7) states that ‘parties to the conflict shall not direct the movement of the civilian population or individual civilians in order to attempt to shield military objectives from attacks…’, indicating that any controlled direction of civilians for the purposes of shielding violates the prohibition. Even when applying the ICRC definition alone, if forcing civilians to remain in a hostile zone ensures their close proximity to military objectives, then this would clearly satisfy the requirement of co-location.
Whilst the form of this conduct does not preclude it from violating the shielding prohibition, the lack of direct evidence of an ‘intent to shield’ will. However, it stands to reason that evidence of this intent cannot be restricted to only direct information revealing a party’s motivations, as this will rarely be forthcoming in complex conflict scenarios. The ICTY Trial Chamber’s judgment in Mladić offers clarity in this respect. They characterised forcing detainees to drive through a minefield at the head of a convoy as a use of human shields. They regarded the laying of mines as a military operation or armed attack and held that ‘the only reasonable inference for moving the detainees in this manner was to render the area, and/or the activities of those in the cars, or future…soldiers going through that area, immune from this military operation or armed attack’ (para. 3382). This suggests that the relevant intent can be inferred where it is the only reasonable inference to be drawn from a party’s conduct. In the present context, this prompts the question of whether there are any other plausible reasons for Hamas’s alleged actions.
Schmitt concludes that ‘it would be naive to believe that Hamas does not intend the presence of civilians to frustrate IDF operations’ and that the ‘efforts to physically prevent civilians from leaving unambiguously qualify as the use of human shields’. However, he also acknowledges that there ‘may be other reasons for the encouragement [to remain]’. Following the ICTY’s approach that intent can be inferred where it is the only reasonable inference to be made, the plausibility of alternative motivations precludes the characterisation of such conduct as shielding in the absence of specific intent – these ‘other reasons’ clearly open the door to drawing multiple reasonable inferences as to why civilians were prevented from leaving. This conduct, however, would be a clear example of a failure to take precautions against the effects of attacks, as it is incumbent upon a party in Hamas’s position to endeavour to remove the civilian population from the vicinity of military objectives, to the maximum extent feasible (API art 58; ICRC, Customary IHL study, rule 22).
What are the Obligations of an Attacking Party in a Shielding Scenario?
If Hamas can be shown to have acted with an intent to shield, they will have violated the prohibition of human shielding, but, significantly, this does not release Israel of their own obligations under IHL. The context in which the aforementioned claims of shielding have been made characterise civilian harm as the ultimate result of a Hamas’s exploitation of civilians, not as the result of an Israel’s military operations. Yet, responsibility for civilian casualties is not a zero sum game – such responsibility can be split between both parties. Whilst Hamas should bear legal responsibility for any violations of IHL they commit, so too should Israeli actions be appropriately scrutinised through the lens of relevant IHL obligations.
Attacking parties are bound to comply with the essential principles of distinction and military necessity, but shielding situations most obviously call into question the principle of proportionality. Applicable to both forms of conflict, this requires that a party launching an attack does not cause excessive civilian harm when weighed against the concrete and direct military advantage anticipated (API article 51(5)(b); ICRC, Customary IHL study, Rule 14). This is supported by the principle of precautions in attack which obliges an attacking party to take measures to avoid harm to the civilian population, to the maximum extent feasible (API article 57, ICRC, Customary IHL study, Rule 15). This can include selecting specific weaponry to avoid excessive harm, suspending attacks in light of new intelligence or providing an effective advance warning of an operation. The application of the principle of precautions in the context of northern Gaza has been discussed on this blog previously.
An attacking party must treat human shields as protected civilians as long as they take no direct part in hostilities, which involuntary shields, such as civilians forced to remain in a hostile zone, do not (ICRC, Interpretive Guidance on the Notion of Direct Participation in Hostilities; Public Committee Against Torture in Israel v Government of Israel, para. 36, but also note Kelemen’s contrasting view). As such, they must be factored into a proportionality calculation as would any other civilian. This renders shielded targets immune from attack as a matter of law, leading some to call for a revised application of proportionality when shields are used (see Rubinstein and Roznai; Artz; Dunlap). This is not a view that I conform to. In situations of shielding, it is arguably more important than ever to advocate for a steadfast adherence to proportionality, in order to safeguard the lives of those civilians who are placed in harm’s way against their will.
In summary, for conduct to violate the prohibition of shielding a specific intent to shield is required. Absent this intent, or facts that make it the only reasonable inference to be drawn from preventing civilians from evacuating hostile zones, it cannot be legally classified as shielding. If further verifiable facts were to come to light suggesting such intent is the only reasonable inference to be made, it remains the case that an attacking party’s obligations are unchanged, even when an opposing force engages in the use of human shields. In this respect, it is imperative that attackers factor human shields into proportionality calculations, as they would any other civilian. It is important to precisely identify and condemn genuine acts of human shielding on the part of Hamas, which undermine the essential protections afforded to civilians under IHL. However, it is of equal significance to uphold the enduring applicability of cardinal IHL principles to Israel’s military operations, identifying and condemning violations in kind.