Based on the ICRC’s firsthand observations in and around cities in conflict across the globe, an evolved form of one of the oldest methods of warfare – siege and encirclement – remains a persistent feature of today’s urban battles. Civilians trapped within besieged areas or those displaced from them endure some of the most horrific humanitarian conditions.

In this post, ICRC Legal Adviser Abby Zeith takes a closer look at contemporary urban siege and encirclement and the civilian harm that they cause, how international humanitarian law (IHL) regulates such methods of warfare, and why states and their policymakers and militaries need to do more to understand, prepare for, and mitigate civilian harm caused by such operations in the future.

For many, the term siege may conjure up historic images of the devastation of Cartagena, Babylon or Leningrad, where civilians and besieged forces were trapped defending themselves within a walled city, castle, or other fortification after being isolated by encircling forces. Those cut off were often left with few options but to capitulate, be subjected to assault and bombardment, or succumb to starvation or illness caused by extreme deprivation.

As seen in recent urban conflicts throughout Africa, Asia, Europe and the Middle East, siege and other forms of encirclement tactics still command a central place on today’s modern battlefield. Research indicates that, since the early 1990s, more than 60 urban sieges of various sizes and duration have occurred. However, political considerations and the ways of fighting in cities have evolved when compared with the siege warfare of a largely bygone era. So, too, has the legal framework that governs them.

The notion of “siege” and “encirclement”

IHL treaties refer to ‘besieged’ or ‘encircled’ areas but do not define them (e.g. Art. 27 of the 1907 Hague Regulations, Arts. 15(2) GC I, 18(2) GC II and 17 GC IV).

Siege – or encirclement, as often referred to in some military warfighting doctrine – can be described as “a tactic to encircle an enemy’s armed forces, in order to prevent their movement or cut them off from support and supply channels” (ICRC 2019 IHL Challenges Report, p. 23). Similar descriptions have been found in some military law of armed conflict (LOAC) manuals (e.g., New Zealand [8.10.1]; United States [5.19.1]).

The objective of an urban siege is usually to isolate and attrite a besieged force to the point at which it is either forced to surrender or conduct a withdrawal, or the besieged force can otherwise be “destroyed” (to use an oft-cited “mission task verb” commonly referred to by militaries).

Historically, this has been primarily through indiscriminate starvation of entire cities in conjunction with persistent attacks to deny access to reinforcements, weaken the besieged defenses, and induce surrender. In contemporary conflicts, while encirclement deprivation still remains a feature, besieging forces often attempt to seize control of a besieged urban area and clear it by killing, capturing, or forcing the withdrawal of besieged forces through intense hostilities.

A siege that does not attempt to capture an area through assault may be aimed at obtaining a military advantage in relative safety for the besieging forces, as occurred in Basra in 2003. Although much less commonly employed today, such tactics can avoid them the hazards of urban fighting and, under certain conditions, may limit the heavy civilian casualties and damage to civilian infrastructure often associated with fighting in populated areas. That said, such operations still give rise to civilian casualties and destruction, as well as other forms of civilian harm and deprivation.

Conversely, most of today’s sieges involve attempts to seize an urban area and defeat its defenders through assault accompanied by heavy bombardment and devastating consequences for civilians. This is particularly the case if the besieged forces ensconced within a city are left with few options other than to fight an attritional urban defense or surrender. Professor of War Studies Anthony King has characterized such battles as “Inner-Urban Sieges of the 21^st Century” or “localized micro-sieges” from below street level to the airspace above. He notes:

Today, cities envelop the armed forces. Armies are simply not big enough to surround whole cities. Battles for cities now take place inside cities themselves, as contracted forces converge on decisive points. Because forces have shrunk, the urban battle has coalesced into a series of localized micro-sieges in which combatants struggle over buildings, streets and districts. Instead of battlelines bisecting an entire city, sieges explode at particular locations. The urban battlefield is punctuated by localized fights.

There may even be localized sieges taking place within a broader besieged area as was the case with the Azovstal Steel Works during the siege of Mariupol.

Sieges cause unspeakable human suffering

Siege and other forms of encirclement are often accompanied by bombardment including from a variety of heavy explosive weapons and intense fighting between besieging and besieged forces, creating constant danger for trapped civilians. Various impediments to the passage of commodities – such as checkpoints, armed forces’ counter-mobility activities, interdiction fire – mean that supplies essential for civilian survival are often prevented from entering a besieged area. Little or no electricity and other degraded essential services are also characteristic features. Families are forced to make impossible choices with the little food and water and health care available. Factors such as age, gender, and disabilities, may exacerbate difficulties in accessing scarce resources. The disturbing humanitarian consequences are cumulative and long-lasting. They can include hunger, malnutrition, dehydration, illness and disease, injury – often including long-term mental harm – and death.

While having similar features, no two sieges look the same. They often differ in terms of size, scale, location, nature, intensity, and parties involved. Importantly, the humanitarian consequences suffered by those trapped will depend upon inter alia the intensity and location of the hostilities; nature of movement restrictions imposed upon both civilians and wounded and sick combatants; availability of local coping mechanisms; the resilience of those interrelated essential service systems on which people depend to meet their needs and live in safety in urban environments such as electricity, health, water and wastewater treatment, solid waste disposal, as well as market systems that provide food and essential items, communication, financial systems, transportation for people and goods; the extent to which humanitarian relief and essential commodities can regularly enter and civilians can leave the area; and the prevalence of smuggling and bribery.

A point that is often forgotten is that the consequences for civilians are not simply limited to the force, terrain, or civilian population and infrastructure ensconced within a besieged or encircled area. To quote a military expert, “like the squeezing of a balloon, a siege in one location redistributes misery and suffering to other locations.”[1] Those who flee besieged areas lose their homes and livelihoods when they are uprooted from their communities. Too often, they are forced to depend upon the support of host families and communities at a time when the hosts are already impacted by the conflict themselves. As was the case in Mosul, large influxes of people from urban areas can exacerbate pre-existing problems of employment and markets, and of housing, infrastructure, access to land, waste management and other public services, in the neighborhoods affected. The scale of needs generated by large-scale urban displacement typically extend well beyond humanitarian organizations’ capacity to meet them.

In addition, military doctrine, supported by current operational practice, indicates that basing, sustainment and fire support locations for both parties as well as actors that support them, are most likely located outside the encircled area. Such objectives may even be in other urban terrain. The risks posed towards the civilian population and civilian objects upon which they rely at these locations from the effects of attacks by the besieged/encircled party must be factored into operational planning and mitigated accordingly.

How does IHL regulate contemporary urban siege and encirclement?

The way belligerents conduct contemporary urban siege or encirclement over the years has evolved for a host of reasons. So, too, has the legal framework governing them.

First and foremost, international law developments (notably those after World War II) have significantly restricted what belligerents are permitted to do during urban sieges. Undoubtedly, the type of siege warfare as traditionally executed, which failed to discriminate between civilians and the enemy’s armed forces, is effectively prohibited today (for an insightful exploration of historical law and customs governing siege and their evolution, see here).

Today, sieges must be directed exclusively against an enemy’s armed forces, such as besieging an area where there are only enemy forces or blocking their reinforcement or resupply. Unfortunately, civilians are often trapped within when towns or other populated areas are besieged or encircled, causing unspeakable suffering.

Recently, the ICRC published a brief explainer on how IHL offers vital protection to these civilians by imposing limits to what the parties can do during such sieges. This explainer summarizes ICRC’s much more detailed legal position published in its 2019 IHL Challenges Report. A few points must be highlighted in particular:

1. Belligerents must allow civilians to leave besieged or encircled areas.

Throughout history, besieging and besieged forces have prevented civilians from leaving besieged areas. For the besieging forces, the main purpose was often to hasten the surrender of the besieged forces, because civilians depend upon on the same supplies as the enemy forces. In fact, at the Nuremberg trials, the practice of using artillery to prevent civilians from leaving a besieged area was deemed an extreme, but not unlawful, measure.

Such practices no longer reflect the current state of the law. The law has evolved even beyond the essential, but limited, provisions of the Geneva Conventions on the evacuation of specific categories of vulnerable people from besieged areas (i.e., Art. 15 GC I, Art. 18 GC II and Art. 17 GC IV).

For the reasons elaborated by ICRC in much more detail in 2019 (pp. 23-24), a combination of several prohibitions (i.e. attacking civilians, indiscriminate attacks, human shields, starvation of the civilian population as a method of warfare), as well as the rules stemming from the principle of precautions, make very clear that: civilians must not be trapped in sieges today, and both parties must allow civilians to leave the besieged and encircled areas.

Widespread calls for the safe and unhindered evacuation of civilians during hostilities from cities in Ukraine and in United Nations Resolutions relating to Syria, demonstrate broad support amongst states for such an imperative. Several legal experts (academic, military, humanitarian) also support the existence of an obligation to allow civilians to leave besieged and encircled areas.

Despite this, diverging views still appear in military LOAC manuals. Several manuals seem to expressly confirm the existence of the obligation (e.g. Denmark (§2.12), France (§5.5.4); Israel; United States (§5.19.4.1)). Unfortunately, other manuals only seem to refer to the limited Geneva Convention provisions addressing evacuation of certain vulnerable groups (e.g., New Zealand), while other still leave it to the commander’s discretion entirely as to whether to allow civilians to leave (e.g. Australia, Canada, United Kingdom) which, according to the ICRC, is an outdated view which longer reflects the current state of the law. Of concern also is the fact that some other LOAC manuals don’t even mention siege at all.

States – particularly those undertaking LOAC manual revision processes – are urged to give further attention to this issue, including by asserting in no uncertain terms that civilians have the right to leave besieged and encircled areas.

2. Civilians fleeing or evacuating from a besieged or encircled area are always protected.

Civilians fleeing, or being temporarily evacuated from, a besieged or encircled area are protected. They must not be attacked. In case of displacement, all possible measures must be taken to ensure these civilians receive satisfactory conditions of shelter, hygiene, health, safety and nutrition and that members of the same family are not separated. Should the besieging party decide to screen displaced persons for security reasons, such measures must be conducted with full respect for IHL and human rights law.

Evacuations are strictly limited by IHL. Forcible evacuations can only take place where the security of the civilians involved or imperative military reasons so demand, and may only be temporary. Displaced persons, whether they have fled or have been evacuated, have a right to voluntary return in safety to their homes or places of habitual residence as soon as the reasons for their displacement cease to exist. As hostilities during sieges entail a high risk of incidental civilian casualties, temporary evacuations may be necessary, and even legally required, but sieges must not be used to compel civilians to permanently leave a particular area.

3. Civilians and wounded and sick remaining in besieged or encircled areas for whatever reason must be protected

Those civilians and wounded and sick who remain in a besieged area for whatever reason continue to be protected. Both the besieged and besieging forces remain bound by important rules protecting civilians against the effects of hostilities, in particular the principles of distinction, proportionality and precautions, the prohibition of human shields, and the rules affording specific protection to the wounded and sick.

In addition, the IHL rules on starvation and on relief operations and humanitarian access are designed to ensure – in combination – that civilians are not deprived of supplies essential to their survival or made to starve.

The so-called “isolation imperative” in siege

It has been suggested that historical accounts (e.g. Leningrad, Grozny and Sarajevo) confirm that “absolute isolation” – which includes sealing off encircled forces from lines of operation, communication, and logistical support as well as control of access to the besieged area – is the “sine qua non” of an effective siege. Proponents of this conclusion seem to suggest anything short of this would give an adversary the lifeline necessary to sustain its defensive posture potentially extending the siege indefinitely and precluding a decisive victory.

According to this view, the military imperative of achieving and maintaining complete isolation (e.g. primarily physical but could also include psychological and electronic facets), is the most important consideration in a siege and, arguably, the basis for the “limited reach and reduced humanitarian promise” of the IHL that governs their conduct. This so-called “isolation imperative”, it is argued, might be invoked by a besieging force as a basis for rejecting offers of humanitarian relief or a refusal to permit the temporary evacuation of civilians from besieged areas. In the view of some belligerents, assumptions as to the anticipated military advantage from achieving complete isolation may even impact upon besieging force decisions concerning target selection, proportionality, and precautions in attack.

However, as will be demonstrated below, “isolation imperative” arguments seeking to privilege military considerations at the expense of humanitarian imperatives in contemporary urban siege and encirclement must not be accepted without further assessment of their legality, military validity, and practicality.

First, while isolation may indeed be a military priority, it cannot be considered in a vacuum as others also assert. It must be considered in the light of the contemporary legal framework. These obligations apply to all means and methods of warfare; there is no exception for sieges. Irrespective of whether a siege or encirclement is perceived to be the “decisive” battle in the context of an overall campaign or large-scale military operation, belligerents must always faithfully comply with the rules governing the conduct of hostilities which aim at protecting civilians and civilian objects. This is relevant not only when planning and deciding upon attacks against military objectives within or from a besieged urban area (notably whether certain precautions may be feasible and on questions of proportionality in attack) but it also concerns decisions regarding the restrictions on humanitarian relief and the conduct of evacuations.

Furthermore, belligerents must account for the fact that it is actually civilians who bear much of the brunt of sieges’ tactics. Importantly, leveraging the suffering of civilians trapped in besieged area is impossible to reconcile with contemporary IHL. In the ICRC’s view, the indiscriminate use of starvation as a method of warfare, i.e. – where the deprivation of food and water or other things necessary for survival cannot be or is not directed exclusively at enemy armed forces – is prohibited. Nor can a besieging force justify deliberate mass starvation of civilians by claiming its specific purpose was only to starve enemy fighters who were also within the area.

What’s more, the commander of a besieged force who is not able to provide the supplies essential to the survival of the civilian population under its control and the commander of a besieging force must allow and facilitate rapid and unimpeded passage of humanitarian relief for civilians remaining in the besieged area (e.g. France (§5.5.4); Canada §614(7): Israel). This is subject to the parties’ right of control and ability to impose temporary and geographically limited restrictions required by military necessity at the time and place of ongoing hostilities.

Second, some accounts (e.g. Dannenbaum pp. 389-390 and Lattimer) challenge so-called empirical claims that unlawful “starvation methods” – including a complete block on the entry of essential supplies – are necessary to overcoming enemy forces within well-defended urban areas. Such views query a binary framing that the only options available to armed forces are “all out military assault or comprehensive encirclement starvation”. Such accounts imply there may be alternative approaches to “contain enemy forces, undermine their capacity to launch an offensive from that location, and preclude their deployment elsewhere”.

Furthermore, besieging forces who seek to justify restrictions on humanitarian relief or refuse to allow civilians to leave besieged areas based upon an assumption that sieges have historically relied upon entrapped civilians to hasten the surrender of besieged forces (because they rely upon the same supplies as the enemy forces), appear to be disconnected from the reality of how contemporary urban warfare is conducted in which encirclements are frequently accompanied by combined arms assaults into an urban area.

Third, if “absolute isolation” was supposedly the “sine qua non” of a successful siege from a military perspective, then, it would be doubtful as to whether urban sieges could still be regarded as a tactic likely to succeed given the practical challenges with isolating a modern city today. Even military experts are likely to tell you that it is a common misconception that modern sieges require full encirclement or complete isolation. In fact, they rarely do. It is said that whether a siege is porous or aim at being impermeable may depend upon inter alia the political and military goals, available forces and capabilities, and size of the encirclement. For example, lack of full encirclement might be due to insufficient personnel or materiel resources to completely encircle a city or a deliberate decision on how a particular actor wants a siege to evolve.

Policymakers, strategists, and military practitioners should do more to avoid civilian harm in contemporary urban sieges

As noted at the beginning of this post, research indicates that, since the early 1990s, belligerents have either employed or been drawn into some 60 urban sieges across the globe and at all levels of operations.

Sieges or other forms of encirclement may also be part of a larger operational plan: for instance, they can be used to isolate dislocated enemy forces ensconced within a particular area. Often, they are the foreseeable consequence of a decision to take the fight to – or defend from within – urban terrain. Today’s urban siege may be terrain focused (i.e., the primary objective is to control strategic areas) or enemy focused (i.e., the key objective is to defeat – and not merely contain –  enemy forces). Sometimes they are a combination of both. Whether a siege is a tactical, operational, or strategic-level operation should impact various aspects thereof including their planning and in turn the precautions that can feasibly be taken.

Despite being a regular feature of contemporary urban warfare, warfighting doctrine makes scant or no reference to siege per se. As mentioned at the beginning, one is more likely to find military doctrine referring to the notion of “encirclement operations”. For example, US Army Techniques Publication 3-06, Urban Operations, [7-46]) makes a link between the urban defense and isolation tactics by an attacking force when alluding to activities typical of sieges:

Encirclement operations are operations where one force loses its freedom of maneuver because an opposing force is able to isolate it by controlling all ground lines of communication and reinforcement (ADP 3-90). In urban areas this encirclement control includes securing all relevant underground facility portals within or around the encirclement area objective. Units consider restricting services (such as water, electric, or telecommunications) to an encircled area briefly to gain positions of advantage; however, units are mindful to restore them in order to cause no undue harm to civilians.

What is clear is that when war moves to the city the fight to control or defend a city (or parts of it) very often becomes a slow grinding positional battle of attrition that resembles a siege or encirclement – regardless of the label one gives it. An urban siege may not always be what a force trains or prepares for; it may be in response to an emerging situation, and sometimes even the result of poor planning or bad tactics. Indeed, much of the widespread civilian suffering caused by such operations is often attributable to the cumulative effect of a series of tactical level decisions some of which may not always support the strategic military aims. Lack of preparedness for the inevitable exacerbates harm for civilians.

Urban warfare experts who have recognized the reemergence of the siege and its evolution blame “attempts to sidestep the siege problem in language and practice” for the inadequacy of current doctrine and, in turn, failing to help military practitioners understand and prepare for such operations. For example, Fox and Watkins express concern at the fact that militaries “lack a framework for operating in a siege from a tactical and legal standpoint” notwithstanding their own participation in several sieges in recent years, and they encourage states and their militaries to “develop siege-compatible doctrine informed by international law to better prepare themselves for the rigors of war in a siege scenario”.

Similar to the scores of other urban warfare experts, non-governmental organizations and academics,  the ICRC is of the view that a more complete reflection on the ways in which belligerents conduct contemporary urban siege and encirclement – particularly the legal and humanitarian concerns – is well overdue. To that end, the ICRC has called for states to develop specific urban warfare doctrine. Similarly, more needs to be done to train and prepare for organized civilian evacuations from besieged and encircled areas; they are complex operations that are often fraught with dilemmas for militaries and humanitarians alike. The ICRC has published a Commander’s Handbook on Reducing Civilian Harm in Urban Warfare with specific annexes on evacuations and screening, and a handbook for non-state armed groups. More recently, the ICRC published an expert meeting report and set of recommendations on measures to prevent and mitigate the indirect effects on essential services from the use of heavy explosive weapons which, as noted above, are of grave concern when used during urban siege and encirclement. We recommend these documents to political authorities, armed forces and non-state armed groups for their consideration.

Of positive note was the workshop co-hosted by Harvard Law School Program on International Law and Armed Conflict, West Point Lieber Institute and the International Committee of the Red Cross in 2022 which brought together a series of legal, humanitarian, academic and military experts to discuss some of these issues (see a summary prepared by two of the participants). The hope is that these discussions continue in years to come.

There is much that is uncertain when it comes to the future of warfare. We are still discovering the ways in which technological developments may be shaping the battlefield or changing the way in which wars in cities are fought. That said, what we do know based on our firsthand observations in and around cities in conflicts across the globe, is that an evolved form of one of the oldest methods of warfare, namely siege and encirclement, remain a persistent feature of urban warfare.

Civilians trapped within besieged areas or indeed those displaced from them will continue to endure some of the most horrific humanitarian conditions. It is imperative that states and their militaries reflect more broadly about how to prevent and mitigate the devastating humanitarian consequences caused by such methods including by ensuring that the way in which they are employed comply fully with contemporary IHL.

Author’s note: The author would like to thank Laurent Gisel and Amos Fox as well as various ICRC colleagues for their precious feedback on earlier drafts.

[1] Amos Fox, Presentation on sieges in modern war. Presented at Harvard Law School on March 31, 2022. Published online by author.

See also:

• Laura Boillot, Laurent Gisel, Paul Holtom, Frederik Siem, Dina Abou Samra, Juliana Helou van der Berg, Protecting civilians in conflict: the urgency of implementing the Political Declaration on Explosive Weapons in Populated Areas, April 22, 2024
• Ruben Steward & Celia Edeline, The NSAG handbook: helping non-State armed groups reduce civilian harm during urban warfare, March 30, 2023
• Eirini Giorgou, Explosive weapons with wide area effects: a deadly choice in populated areas, January 25, 2022