Under Siege: LOAC Obligations of the Besieged Party


| Jul 5, 2024


Editors’ Note: Articles of War previously published a longer work on the subject of States’ responses to malicious or hostile actions as both a post and an inaugural Lieber Institute White Paper. The post below addresses the subject of besieged forces’ legal obligations in similar length and detail. With this work, we add a second entry to our Lieber Papers series. Look for other Lieber Papers in the future to appear initially as posts that will later be organized separately at the Lieber Institute website and available for download as formatted documents. We will offer Lieber Papers as focused studies of subjects of enduring relevance to our field. Lieber Papers will differ somewhat from our routine offerings in their length, thoroughness, and relative independence from events of the day. But they will share our dedication to careful, informed, and relevant scholarship.

The conflicts in the Middle East and Ukraine have drawn renewed attention to how the law of armed conflict (LOAC) governs siege, a tactic nearly as old as warfare itself (see, e.g., here, here, here, and here). Yet, the legal obligations the parties shoulder during sieges remain somewhat opaque – a point recently reiterated by the International Committee of the Red Cross (ICRC).

Most of the legal analysis appearing in recent discourse has focused on the legal obligations of the besieging force, particularly those associated with starvation. What has largely been missing is a comprehensive examination of legal obligations shouldered by the besieged party. That is unfortunate, for it has more direct control over the civilian population and its well-being than its opponent. In this Lieber Paper, we hope to balance the discourse by outlining the former’s LOAC obligations. As will become apparent, LOAC imposes significant burdens on and limits the operational options of the besieged party. And because sieges of urban areas have the greatest potential for affecting civilians—thereby triggering robust LOAC duties—we focus on them.

It must be cautioned that the besieged party also may owe human rights obligations to those in the besieged area, a topic we do not address. Moreover, we intentionally avoid direct reference to ongoing events in the Israel-Hamas conflict and the war between Russia and Ukraine. Views on these hostilities have become so emotive that it is becoming hard to keep attention focused objectively on the law. Thus, we address the law in the abstract, and in that regard, only LOAC. But first, it is helpful to offer a bit of context.

Siege Warfare

Sieges are military operations designed to encircle and cut off an area where the enemy is located. As Sean Watts has observed, “control of access to a besieged area is the sine qua non of siege.” The logic of most sieges is simple. Urban assault favors the defender, generally necessitating a numerical advantage. Therefore, sieges are an appealing tactic for securing an enemy force’s submission, or at least neutralization, while, inter alia, avoiding a bloody fight and freeing up forces for use elsewhere. They may also serve as a tool for containing hostilities by keeping adversarial forces confined. The besieging force can pursue its goals by various means, including starvation, blocking the resupply of military supplies necessary to defend the besieged area, and simply preventing breakout or reinforcement.

Urban sieges have been employed frequently since ancient times. Historically notable examples include: the siege of Jerusalem by Roman General Titus in 70 AD to put down the First Jewish Revolt; the 1870-71 siege of Paris by German forces that was the turning point in the Franco-Prussian War; the 872-day siege of Leningrad by German and Finnish forces that led to the deaths of between 800,000 and 1.5 million soldiers and civilians; and the brutal 2022 Russian siege of Mariupol.

These and other urban sieges are responses to an array of situations. In an excellent RUSI Journal article, Amos Fox has highlighted three. In the first, a weaker force seeks refuge in an urban environment to “neutralise or at least offset the strength of its adversary.” The intent is to leverage the fact that defenders enjoy an operational advantage in an urban fight. The second situation involves “an insurgency or partisan movement [that] operates in the city and among its people to generate and maintain support.” Finally, in the third situation, one party simply “defends from an urban area, whether intentionally or by getting caught in urban terrain before being able to move beyond that area.”

The common factor in all three is that the besieged force is collocated with the civilian population. From the former’s perspective, this has both negative and positive potential. On the one hand, the civilian population will consume food and other essentials that could otherwise prolong the besieged force’s ability to withstand the siege. Thus, there is an incentive to allow, even force, civilians to evacuate the area.

On the other hand, the presence of the civilian population and civilian infrastructure limits the besieging force’s operations de jure and de facto. It does so de jure because the besieging force is not relieved of its LOAC obligations, particularly the requirements to consider harm to civilians and civilian objects in its proportionality and feasible precautions in attack analyses. These obligations apply even when a besieged force is using the civilian population as human shields (see below). Moreover, the prohibition on indiscriminate attacks can preclude area targeting that might otherwise be useful against an enemy embedded in urban terrain.

Probably of even greater effect is the de facto impact of the presence of civilians and civilian infrastructure. After all, the optics of civilian suffering and destruction of the urban terrain at both the domestic and international levels, even when lawful, can exert a chilling effect on the besieging force’s operations. Indeed, the besieged force will often leverage this reality through intentional lawfare tactics, such as creating situations designed to cause collateral damage and mistakes, that are hard to counter.

Finally, the besieging force may have obligations that the besieged force can exploit. For instance, if there is an obligation to allow relief supplies (we believe there generally is), the besieged force can divert part of them to its own use, thereby extending its ability to withstand the siege. This may affect the besieging force’s obligation to allow the passage of humanitarian relief. Still, as a practical matter, it may be difficult for the besieging force to confirm the besieged force is doing so (or convince others that this is the case). And even where it is clear the besieging force is diverting relief supplies, blocking relief supplies on this basis can appear cruel, thereby weakening support for the besieging party.

The Law

Although there sometimes might be an incentive to exploit the presence of civilians and civilian infrastructure, LOAC nevertheless imposes obligations and prohibitions on the besieged force. These include taking so-called “passive precautions,” facilitating humanitarian relief, the prohibition of shielding, and marking protected infrastructure.

Passive Precautions

A besieging force conducting attacks into a besieged area is obligated, inter alia, to take feasible “active precautions” to minimize injury or death of civilians and damage to civilian property, so long as the measures do not forfeit military advantage (see, e.g., Additional Protocol I, Article 57, ICRC Customary International Humanitarian Law (IHL) study, rules 15-21; Department of Defense (DoD) Law of War Manual, § 5.11). However, the besieged party is no less obligated to take measures designed to reduce the risk to the civilian population from anticipated or ongoing enemy attacks. These are known as “passive precautions.”

In treaty law, the 1977 Additional Protocol I to the 1949 Geneva Conventions, Article 58 (“Precautions against the Effects of Attacks”), sets forth the defender’s passive precautions obligation for parties to the instrument:

The Parties to the conflict shall, to the maximum extent feasible:

(a) without prejudice to Article 49 of the Fourth Convention, endeavour to remove the civilian population, individual civilians and civilian objects under their control from the vicinity of military objectives;

(b) avoid locating military objectives within or near densely populated areas;

(c) take the other necessary precautions to protect the civilian population, individual civilians and civilian objects under their control against the dangers resulting from military operations.

There are four crucial points about passive precautions. First, the examples found in Article 58 are illustrative, not exclusive. The besieged force is obliged to take all measures that would enhance the protection of the civilian population from the effects of attack, so long as those measures are feasible under the circumstances. For instance, if feasible, the besieged force should establish shelters for the civilian population.

Second, it is sometimes suggested that a siege is not an “attack,” as that term is understood in LOAC, and that, therefore, the obligations apply only to kinetic attacks conducted during the siege. In this regard, Article 49(1) of Additional Protocol I provides a well-accepted definition of attacks: “‘Attacks’ means acts of violence against the adversary, whether in offence or in defence.” As illustrated in the cyber context, an attack need not be a violent act (see Tallinn Manual 2.0, Rule 92, which States widely accept). Instead, it is an act that has violent consequences.

Though this represents a relatively recent shift in how the definition is interpreted, its broad acceptance evinces significance beyond the cyber domain. Therefore, in our view, the passive precautions obligations apply even if the besieging force is not conducting kinetic attacks into the besieged area, so long as there is an intent to starve individuals (whether military or civilian) therein or starvation or other physical suffering is foreseeable. In any event, the passive precaution obligation attaches in the face of prospective attacks; so long as there is a reasonable likelihood of kinetic attack, it would apply irrespective of whether sieges are attacks.

The third point is that this broad obligation is subject to a condition of feasibility to the maximum extent. The notion of feasibility is an apt illustration of LOAC’s balancing of military and humanitarian considerations; it is almost always the crux of the matter with respect to compliance with both active and passive precautions.

A well-accepted explanation of feasibility is that found in Article 1(5) of Protocol III (incendiary weapons) to the Convention on Certain Conventional Weapons (CCW): “‘Feasible precautions’ are those precautions which are practicable or practically possible taking into account all circumstances ruling at the time, including humanitarian and military considerations.” This formula is replicated in numerous other instruments (e.g., CCW, Amended Mines Protocol; CCW, Protocol V) and a number of statements made on ratification of Additional Protocol I.

Notably, there was some concern among States as to what maximum feasibility, as it appears in Article 58, would entail, particularly when it came to the placement of defensive military structures in urban areas. For instance, during deliberations at the Diplomatic Conference that drafted Additional Protocol I, the French delegate voiced a “keen sense of anxiety” about restricting States’ prerogative to place military objectives in the vicinity of civilians in densely populated countries, suggesting the rule could not be followed “unless it were accepted that French territory would not be defended” (Official Records, vol. VI, CDDH/SR.42, p. 213). The Swiss and Austrian delegates agreed it “might prove prejudicial to a country’s national defence” (p. 214).

But the Italian Delegate countered that the words “to the maximum extent feasible . . . clearly show the real aim of this rule: this is not a question of absolute obligations, but, on the contrary, of precepts that should be followed if, and to the extent that, the particular circumstances permit” (p. 232). This interpretation found wide acceptance (see, e.g., statements by the Federal Republic of Germany, United Kingdom, Netherlands, p. 214, Canada, p. 224, Republic of Korea, p. 234-235, Cameroon, p. 239) and reflects how the requirement is understood today.

Thus, for example, an ICRC Commentary to Article 58 acknowledges that “a Party to the conflict cannot be expected to arrange its armed forces and installations in such a way as to make them conspicuous to the benefit of the adversary.” But at the same time, and as noted in the DoD Law of War Manual, “A wanton disregard for civilian casualties or harm to other protected persons and objects is clearly prohibited” (§ In our view, avoiding proximity to important or highly frequented civilian facilities is an appropriate point of reference in this regard. Still, drawing the line between feasible and unfeasible measures in an area constricted by siege may pose significant challenges and will depend on the attendant circumstances.

Fourth, note that the passive precautions obligation includes the protection of civilian objects. Again, there is no prohibition on using civilian objects for military purposes during a siege unless they enjoy special protection, as with medical facilities, although such use should be necessitated by military exigencies. And in an urban fight, the use of civilian structures for military purposes will be pervasive. Yet, to the extent a besieged party can take precautions to avoid damage to them, most notably by avoiding co-location that is not justified for military reasons, it must do so.

The ICRC’s Customary IHL study characterizes these obligations as customary law binding on all parties to both international and non-international armed conflicts (rules 22-24). In the Kupreškić case, the International Criminal Tribunal for the former Yugoslavia (ICTY) came to the same conclusion (para. 524). And the fact that the United States, a non-party to Additional Protocol I, includes provisions on passive precautions in its DoD Law of War Manual is telling (§ 5.14), although it sometimes articulates said provisions in hortatory instead of obligatory terms. Less certain is the customary status of passive precautions during non-international armed conflict. Nevertheless, the ICRC has offered a colorable argument to this effect, one supported by a variety of sources (see explanation accompanying rule 22).

As guidance to the armed forces, the DoD Law of War Manual is a useful source for illustrating passive precautions, which it explicitly acknowledges apply to a besieged party (§ 5.19.5). Like Article 58, the Manual emphasizes that “[i]t may be appropriate to avoid placing military objectives, such as the armed forces, in urban or other densely populated areas, in order to reduce the risk of incidental harm to the civilian population.” It cautions, however, that “legitimate military reasons often require locating or billeting military forces in urban areas,” citing the examples of needing to “take advantage of existing facilities, such as facilities for shelter, health and sanitation, communications, or power” (§ 5.14.1). In our estimation, there is unquestionably a customary international law obligation to minimize civilian casualties in a besieged area by staying away from them to the maximum extent feasible in the circumstances.

Another besieged force passive precautionary measure the Manual highlights is “concentrate[ing] the wounded and sick and civilians in areas remote from military objectives or in areas neutralized by arrangement with the enemy,” a subject one of us has addressed in Articles of War. The Manual also notes that “it may be appropriate to mark protected buildings to indicate their protected status to enemy forces” (§ 5.19.5). We address that option below, concluding that doing so may be obligatory in some circumstances.

However, perhaps the most significant passive precautionary measure that can be taken to protect the civilian population during a siege is to remove civilians from the vicinity of military objectives, an obligation expressly set forth in Article 58. This may involve the besieging force relocating civilians to lessen the risk they face or moving them out of the besieged area altogether.

The DoD Law of War Manual echoes this obligation. In particular, it observes that “[t]he evacuation of civilians from areas likely to be attacked is advisable when there is immediate danger and where it would be likely to involve less hardship and danger to civilians than leaving them in place” (§ 5.14.2). This is often going to be the case in a siege, if only because of the risk of famine. The Manual advises that the conclusion of “local agreements for the removal of civilians from besieged or encircled areas” should be considered. Relatedly, note that Geneva Convention I, Article 15, specifically provides that “local arrangements” may be made for “the removal or exchange of wounded and sick from a besieged or encircled area, and for the passage of medical and religious personnel and equipment on their way to that area.” Geneva Convention IV, Article 17 further expands this obligation to include a larger group of protected persons, stating that “Parties to the conflict shall endeavour to conclude local agreements for the removal from besieged or encircled areas, of wounded, sick, infirm, and aged persons, children and maternity cases.” These provisions are, like the rest of the Conventions, now customary in character (Nuclear Weapons, para. 79).

In its Commentary on Article 58, the ICRC makes the same point:

It is clear that authorities with a sense of duty will endeavour to remove the civilian population from areas where the risk of attack is greatest. Sometimes only certain categories of the population may be removed in this way: children, mothers, the elderly, the sick etc. Evacuation requires preparatory measures, often taken even in peacetime. Sometimes the whole of the population is evacuated.

The ICTY came to an identical conclusion in Galić, where the Trial Chamber stated, “The parties to a conflict are under an obligation to remove civilians, to the maximum extent feasible, from the vicinity of military objectives” (para. 61; on evacuation, see also Biggerstaff).

In our view, it is clear that a besieged party must allow civilians to leave the besieged area. Indeed, it is obliged to try to evacuate them to the maximum extent feasible in the circumstances and must remain open to negotiations with the besieging party for doing so. Thus, there is no justification for preventing their departure, for instance, to heighten their suffering and thereby increase external pressure on the besieging force. Furthermore, even if not evacuated from the besieged area, civilians should be moved away, to the maximum extent feasible, from immovable military objectives. For example, there is no prohibition on using civilian infrastructure, such as homes or apartment buildings, for military purposes if necessary (with certain exceptions like medical facilities). But once that occurs, the force using the structure must evacuate civilians from it, so long as reasonably practical (e.g., a building used spontaneously during an ongoing firefight versus a building used to store military equipment or billet military personnel).

The question remains whether the population may be compelled to leave the besieged area, for example, to preserve the supplies there for the besieged force. In this regard, recall the reference to Article 49 of Geneva Convention IV in Article 58 of Additional Protocol I. Article 49 prohibits forcible displacement of protected persons (which includes the civilian population) from occupied territory, except as necessary for the security of the population or “imperative military reasons.” Violation is a grave breach of the Convention (art. 147) and a war crime (International Criminal Court (ICC) Statute, arts. 8(2)(a)(vii) and (b)(viii)). Additionally, the ICRC’s study of Customary IHL emphasizes, correctly so, that “all possible measures must be taken in order that the civilians concerned are received under satisfactory conditions of shelter, hygiene, health, safety and nutrition and that members of the same family are not separated” (rule 131).

The fact that the rule does not apply to unoccupied territory suggests there is no prohibition on compelling civilians to leave besieged, unoccupied territory, even for reasons unrelated to their safety or military expediency. We believe it is self-evident, however, that the besieged force may not order involuntary evacuation if the health and safety of the evacuated population would suffer. As acknowledged in the DoD Law of War Manual, “The evacuation of civilians from areas likely to be attacked is advisable when there is immediate danger and where it would be likely to involve less hardship and danger to civilians than leaving them in place” (§ 5.14.2, emphasis added).

In non-international armed conflict, Additional Protocol II, Article 17, which technically applies only to those conflicts meeting the strict criteria outlined in the instrument, disallows the involuntary displacement of the civilian population except for the same reasons as Article 49. Lest the prohibition be characterized as applicable only to parties to the instrument, and then only when it is in operation, Article 8(2)(e)(viii) of the ICC Statute treats “[o]rdering the displacement of the civilian population for reasons related to the conflict, unless the security of the civilians involved or imperative military reasons so demand” as a war crime in non-international armed conflict. Thus, in our opinion, it necessarily follows that such an action is likewise a LOAC violation in all non-international armed conflicts.

Finally, Article 57(1) of Additional Protocol I sets forth a further obligation in the context of precautions generally: “In the conduct of military operations, constant care shall be taken to spare the civilian population, civilians and civilian objects.” The article is the subject of much uncertainty. Does it apply to all military operations or only those qualifying as attacks? Does it bind all parties or only the attacking force? What does it require in practice?

These complex questions are beyond the scope of this piece. However, one of us has considered the obligation in greater depth, concluding that “Commanders and others with control over military operations cannot lawfully ignore any possibility of an impact on the civilian population. They may be able to dismiss the possibility very quickly, but the consideration thereof must always be part of the operational analysis.” We agree that the obligation is customary, that it applies to all parties and all military operations, and that it at least prohibits a besieged force from disregarding the effect of those operations on the civilian population.

Humanitarian Relief 

Starvation of the civilian population as a method of warfare is unquestionably forbidden, although the parameters of the obligation are the subject of significant debate. As already pointed out, the prohibition on starving civilians has been extensively discussed in the context of the besieging actors’ obligations. However, we are of the view that a besieged force is no less bound to refrain from intentionally using the starvation of civilians entrapped with it as a method of warfare. For instance, a central question is whether the besieging force must allow relief supplies into the besieged area, and, if so, under what circumstances (see, e.g., Watts, Corn/Gillard, Dannenbaum; for in-depth analysis, see Akande/Gillard, Watts, Dannenbaum). The flip side of that question is whether the besieged party must do so. Because the besieged party is co-located with the civilian population and may be the sole authority in the area, its role in facilitating humanitarian relief is crucial.

Article 23 of Geneva Convention IV requires parties to an international armed conflict to “allow the free passage of all consignments of medical and hospital stores” for civilians and “the free passage of all consignments of essential foodstuffs, clothing and tonics intended for children under fifteen, expectant mothers and maternity cases.” It reasons that the besieged party has a corresponding duty to refrain from activities that would frustrate the delivery of qualifying relief to the persons singled out in the article.

The besieged party’s failure to comply with that duty has immediate consequences concerning the delivery of humanitarian relief and may result in individual criminal responsibility of those involved. The key immediate consequence is that the besieging party need not allow such relief to enter the besieged area if it has “serious reason for fearing” that: 1) the besieged forces will use the supplies in question; 2) they will offer the besieged party’s military effort or economy a “definite advantage . . . through the substitution of the above-mentioned consignments for goods which would otherwise be provided or produced by the enemy or through the release of such material, services or facilities as would otherwise be required for the production of such goods”; or 3) control over its distribution will be ineffective. The DoD Law of War Manual is in accord (§ 5.19.3). Additionally, the willful impediment of relief supplies can amount to a war crime in international and non-international armed conflicts (arts. 8(2)(b)(xxv) and 8(2)(e)(xix) of the ICC Statute), though to date only fourteen States have accepted the latter amendment to that effect.

Geneva Convention IV also imposes duties on Occupying Powers when occupied territory is besieged. Article 55 obliges them to ensure “the food and medical supplies of the population.” Article 59 provides that “[i]f the whole or part of the population of an occupied territory is inadequately supplied, the Occupying Power shall agree to relief schemes on behalf of the said population, and shall facilitate them by all the means at its disposal.” These obligations are broader than Article 23 in that they extend beyond medical supplies for civilian use and do not contain a limitation as to the individuals for whom other relief may be provided.

Article 69(1) of Additional Protocol I further extends, for States parties, the obligation to include “clothing, bedding, means of shelter, other supplies essential to the survival of the civilian population of the occupied territory and objects necessary for religious worship.” Both instruments provide that these obligations are to be complied with “to the fullest extent of the means available,” although they are limited to territory under siege that the besieged party occupies. The DoD Law of War Manual adopts the text of Geneva Convention IV, Article 59 verbatim (§ 11.17.1).

Additional Protocol I widens the Article 59 obligations to unoccupied territory in its Article 70:

1. If the civilian population of any territory under the control of a Party to the conflict, other than occupied territory, is not adequately provided with the supplies mentioned in Article 69 [supplies essential to the survival of the civilian population], relief actions which are humanitarian and impartial in character and conducted without any adverse distinction shall be undertaken, subject to the agreement of the Parties concerned in such relief actions.

. . .

2. The Parties to the conflict and each High Contracting Party shall allow and facilitate rapid and unimpeded passage of all relief consignments, equipment and personnel provided in accordance with this Section . . . .

Although debate is ongoing as to whether concluding an Article 70 agreement is obligatory, and if so, when (see, e.g., Oxford Guidance and Watts), we concur with the ICRC’s Commentary to the article, which explains that refusal may not be “arbitrary or capricious” (para. 2805). In most cases, therefore, the besieged party must either provide for the care and feeding of the civilian population or allow humanitarian relief efforts.

The besieged party bears two enumerated obligations concerning the delivery of humanitarian relief. First, Article 70(3)(c) prohibits it from diverting “relief consignments from the purpose for which they are intended” or delaying their delivery. Indeed, if it does so, the besieging party, as noted, may be relieved of its obligation to allow the humanitarian relief through. Second, Article 70(4) obligates the parties to “protect relief consignments and facilitate their rapid distribution.” This would include, for instance, taking action to prevent looting of the supplies, to the extent that protecting them is feasible in the circumstances.

As noted above, Geneva Convention IV, and therefore Article 23, generally reflects customary law. In Rule 55 of its Customary IHL study, the ICRC asserts that Article 70 does so too, setting forth its basis for the conclusion in the study’s practice section. That rule states, “The parties to the conflict must allow and facilitate rapid and unimpeded passage of humanitarian relief for civilians in need, which is impartial in character and conducted without any adverse distinction, subject to their right of control.”

By contrast, the DoD Law of War Manual limits its discussion of humanitarian relief beyond occupied territory to the terms of Geneva Convention IV (§§ 11.14-11.17). The UN Security Council (UNSC) seemed to acknowledge the distinction between party and non-party States in its first Resolution 2417 (2018) on food insecurity in times of armed conflict. There, it demanded that “all parties to armed conflicts fully comply with their obligations under international law, including international human rights law, as applicable, and international humanitarian law, in particular their obligations under the Geneva Conventions of 1949 and the obligations applicable to them under the Additional Protocols thereto of 1977” (emphasis added). Though subtle, this may hint towards the UNSC’s reluctance, or at least that of certain members, to recognize Article 70 as customary in character. Thus, a degree of uncertainty surrounds the issue of customary status.

Somewhat greater uncertainty exists concerning a purported analogous customary rule in non-international armed conflict. In treaty law, Additional Protocol II, Article 18, provides: “If the civilian population is suffering undue hardship owing to a lack of the supplies essential for its survival, such as foodstuffs and medical supplies, relief actions for the civilian population which are of an exclusively humanitarian and impartial nature and which are conducted without any adverse distinction shall be undertaken subject to the consent of the High Contracting Party concerned.” Two factors restrict the scope of this provision and its applicability to the besieged force.

First, Additional Protocol II is limited in its material scope of application to non-international armed conflicts in which the organized armed group in question “exercise[s] such control over a part of [the High Contracting Party’s] territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol” (art. 1(1)). Relief actions in non-international armed conflicts that do not fall within the scope of the instrument’s applicability remain uncodified.

Second, as one of us has previously discussed, the issue of who qualifies as a High Contracting Party concerned is unclear (see also Bothe et al., New Rules, p. 801). An amendment to the second paragraph of the draft article led to the deletion of the phrase “in any territory under the control of a Party to the conflict” and replaced the words “the Party or Parties concerned” with the more restrictive “High Contracting Party concerned” (Official Records, vol. IV, Amendment CDDH/440, p. 106). The change was adopted by consensus, which alludes to the value the delegates placed on States’ right of control in their territory (Official Records, vol. VII, CDDH/SR.53, p. 146). The uncertainties are, therefore, amplified if the besieged forces do not belong to the State on whose territory the hostilities take place (as will often be the case). In this regard, we caution that whether and, if so, how the relevant rules bind armed non-State actors is a fraught question exceeding the scope of this piece.

Yet, as with international armed conflicts, the ICRC asserts in Rule 55 of its study of Customary IHL that the obligation to allow humanitarian relief (subject to the restrictions and limitations discussed above) is binding in all non-international armed conflicts. We find this to be a reasonable conclusion based on the support it offers, but not an unassailable one.


It is not per se unlawful to fight from urban areas under siege, even when the option of operating from other areas presents itself. However, the law is clear that intentionally doing so to shield oneself from enemy attack is a LOAC violation and a war crime. In other words, the motivation for operating from a populated area or in proximity to individual civilians or groups of civilians, such as intentionally operating from an apartment complex to deter attack, distinguishes lawful urban warfare from unlawful shielding. As noted earlier, besieged forces sometimes intentionally operate from urban areas to leverage the presence of civilians and civilian objects to their benefit.

 In treaty law, this prohibition is found in Article 51(7) of Additional Protocol I:

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.

Notably, the prohibition applies irrespective of whether the tactic is designed to affect the besieging force’s proportionality assessments, possibly precluding attack as a matter of law, or simply to complicate its operations. The latter might be effectuated, for instance, by compelling the urban population’s movement to thwart the besieging force’s maneuver or simply to cause it to hesitate to act because of the death or injury that may befall civilians. Shielding can be accomplished by putting civilians in, on, or near military objectives and targetable persons or intentionally being in their proximity with the intent to affect enemy operations.

Article 28 of the 1949 Geneva Convention IV similarly prohibits the “presence of a protected person” from being used “to render certain points or areas immune from military operations.” It is, however, limited, for it only applies to the use of protected persons, that is, those who “find themselves . . . in the hands of a Party to the conflict or Occupying Power of which they are not nationals” (Geneva Convention IV, art. 4). Thus, it would extend to fighting from a besieged urban area in occupied territory but not from one’s own or other unoccupied territory. Beyond civilians, it is also prohibited to use medical units and, although less relevant to urban operations, prisoners of war as shields (Additional Protocol I, art. 12(4); Geneva Convention III, art. 23), prohibitions that are undoubtedly of customary status.

Rule 97 of the ICRC’s Customary IHL study suggests, and rightly so in our opinion, that the prohibition on human shielding applies as customary law in both international and non-international armed conflict. Most military manuals, including the DoD Law of War Manual (§ 5.16), acknowledge this for international armed conflict. As to non-international armed conflict, the ICRC suggests, inter alia, that while Additional Protocol II (which governs certain non-international armed conflicts) does not address shielding head-on, “such practice would be prohibited by the requirement that ‘the civilian population and individual civilians shall enjoy general protection against the dangers arising from military operations.’”

Although one must be cautious about jumping from Additional Protocol II to conclusions as to the state of customary law, we find the ICRC’s analysis reasonable in light of LOAC’s object and purpose and this particular tactic. It is supported, for instance, by the widespread condemnation of the use of human shields during non-international armed conflicts (see, e.g., UN Sec. Gen. remarks; U.S. condemnation of ISIS use of shields in Mosul and Raqqa; NATO Strategic Communications COE report; Human Rights Watch interview, and sources cited here, n. 14).

The DoD Law of War Manual discusses the prohibition, again correctly so in our estimation, as extending to protected objects. It provides, “Parties to a conflict may not use the presence or movement of protected persons or objects: (1) to attempt to make certain points or areas immune from seizure or attack; (2) to shield military objectives from attack; or (3) otherwise to shield or favor one’s own military operations or to impede the adversary’s military operations” (§ 5.16, emphasis added). The Manual explains that, “[i]n particular, the civilian population, protected persons under the [Geneva Conventions], [prisoners of war], fixed medical establishments and medical units, parlementaires and other persons protected by a flag of truce, and cultural property are protected persons and objects.”

This is a supportable conclusion, for although the treaty law and the ICRC’s customary rule restatement both limit prohibited shielding to civilians, the effect of special protections under LOAC is effectively to bar the use of other protected persons and objects from use for shielding; thus, such action may not technically qualify as human shielding but would nevertheless be forbidden if the purpose was to use a protected object as a shield. In our opinion, the notion of “protected object” includes civilian structures that are not being used for military purposes. By this view, basing oneself in an urban area in the hope that the potential for damage to civilian property that does not qualify as a military objective will deter the besieging force’s operations is prohibited.

Finally, shielding is a war crime during an international armed conflict, as illustrated in Article 8(b)(xxiii) of the ICC Statute; it entails “utilizing the presence of a civilian or other protected person to render certain points, areas or military forces immune from military operations.” Interestingly, the statute contains no analogous prohibition for non-international armed conflict.


Article 27 of the Regulations annexed to the 1899 Hague Convention II and Hague Convention IV of 1907 sets forth one, and possibly two, of the earliest obligations borne by a besieged party. The text of the two instruments is nearly identical; the latter provides:

In sieges and bombardments all necessary steps must be taken to spare, as far as possible, buildings dedicated to religion, art, science, or charitable purposes, historic monuments, hospitals, and places where the sick and wounded are collected, provided they are not being used at the time for military purposes.

It is the duty of the besieged to indicate the presence of such buildings or places by distinctive and visible signs, which shall be notified to the enemy beforehand.

The first paragraph is directed primarily at the besieging party, although it could be interpreted as encompassing, for instance, unnecessarily placing military objectives in or near the listed buildings and places. Indeed, so interpreting the provision would be consistent with the later LOAC obligations and prohibitions cited earlier. In any event, as explained above, treaty and customary law impose just such an obligation today.

The novel aspect of Article 27 is found in the second paragraph. LOAC provides for numerous protective emblems designed to enhance the protection of specially protected places and locations, such as hospitals (see NWP 1-14-M, Annotated, ch. 8). There is generally no legal obligation to employ them, although they do serve to enhance protection de facto.

Yet, as is evident from the second paragraph of Article 27, there is such an obligation for the enumerated buildings and places for the besieged party during sieges, as well as an obligation to notify the besieging force of having done so. And lest there be any doubt as to the obligation’s continuing force, it is well accepted that, in significant part, the Regulations continue to reflect customary law in modern international armed conflicts (see discussion here), thereby binding even those States that are not party to the Hague Conventions.


What is clear from the discussion above is that, as with all of LOAC, the obligations imposed on the besieged party are the product of a delicate balance between military and humanitarian considerations. Indeed, these obligations, which operate in favor of the besieged civilian population, were repeatedly conditioned by feasibility, a notion that considers military practicality. Feasibility depends not only on the terrain in which besieged forces maneuver but also on the circumstances of the siege. Those circumstances are, in turn, influenced by the besieging party’s conduct. However, subject to feasibility, we believe the besieged party shoulders the following obligations.

– To move military objectives and military personnel away from civilians and civilian objects.

– To refrain from conducting military operations near civilians and civilian objects.

– To create safety or other zones to which the wounded or sick and civilians can be moved to enhance their protection, particularly by securing an agreement with the besieging party.

– To evacuate wounded and sick civilians, so long as their health is enhanced.

– To evacuate the civilian population if its safety or health is at risk, so long as the location to which the evacuation occurs enhances their safety or health.

– To allow humanitarian relief as necessary to ensure civilians have access to supplies needed for their survival or health.

– To mark certain buildings and locations to enhance their protection in the face of enemy operations into the besieged area.

Finally, we believe that the humanitarian-military considerations balance of LOAC is evolving toward greater emphasis on protecting civilians in a besieged area. Some may lament this trend, but it does seem to be gaining traction, especially in light of the current conflicts.


Michael N. Schmitt is the G. Norman Lieber Distinguished Scholar at the United States Military Academy at West Point. He is also Professor of Public International Law at the University of Reading and Professor Emeritus and Charles H. Stockton Distinguished Scholar-in-Residence at the United States Naval War College.

Rosa-Lena Lauterbach is a PhD Candidate at the University of Cologne and a former Visiting Researcher at the Lieber Institute and Columbia Law School.







Photo credit: Staff Sergeant Aaron Allmon