The Encyclopedia is a project of the Geneva Academy of International Humanitarian Law and Human Rights launched on 2 December 2013. The Enyclopedia aims to provide accurate, up-to-date information on weapons, the effects of their use, and their regulation under public international law, in a format that is accessible to non-specialists.+ Find out more
Submunitions are typically small, unguided explosive devices that are designed to explode prior to, on, or after impact. Depending on the model, the number of submunitions dispersed or released by a parent container can vary from several dozen to more than 600. The 2008 Convention on Cluster Munitions prohibits, among other activities, the use, production, transfer, and stockpiling of cluster munitions as defined under that treaty.
The 2008 Convention on Cluster Munitions prohibits states parties from using, developing, producing, acquiring, stockpiling, retaining, or transferring cluster munitions ‘under any circumstances’. The definition of a cluster munition for the purposes of the Convention and its scope are set out in its Articles 1 and 2 of the treaty.
Each state is required — within eight years of becoming a party to the Convention — to destroy all stockpiled cluster munitions under its jurisdiction and control. This deadline can be extended for an additional four years and further extensions of four years may also be granted in exceptional circumstances. States may retain a limited number of cluster munitions and explosive submunitions for the development of and training in detection, clearance, or destruction techniques, or for the development of cluster munition counter-measures.
Each state party must also clear its own territory or other territory it controls of cluster munition remnants Unexploded submunitions are submunitions that have been dispersed and have landed, but have failed to explode as intended. Unexploded bomblets are similar to unexploded submunitions but refer to ‘explosive bomblets’ which have been dropped from a fixed-wing aircraft dispenser but have failed to explode as intended. Failed cluster munitions are cluster munitions that have been dropped or fired but the dispenser has failed to disperse the submunitions as intended. Abandoned cluster munitions are unused cluster munitions that have been left behind or dumped, and are no longer under the control of the party that left them behind or dumped them. See 2008 Convention on Cluster Munitions, Art. 2§§4, 5, 6, 7, and 15. For more information, see the entry on the 2008 Convention on Cluster Munitions.within 10 years of becoming party to the Convention. If a state is unable to meet this deadline, it may request extensions from a Meeting of States Parties or a Review Conference for additional periods of up to five years at a time.
Unexploded or abandoned submunitions or cluster munitions also fall within the ambit of the 2003 Protocol on Explosive Remnants of War to the 1980 Convention on Certain Conventional Weapons. States parties to this Protocol are under an obligation to mark and clear unexploded submunitions or abandoned cluster munitions. According to Article 3(2) of the Protocol:
After the cessation of active hostilities and as soon as feasible, each High Contracting Party andshall mark and clear, remove or destroy explosive remnants of war in affected territories under its control.
The use of the phrase ‘as soon as feasible’ means that there is no specific deadline for these obligations to be fulfilled.The term High Contracting Party has the same meaning as 'state party'. Party to an armed conflict includees non-state armed groups that are party to an armed conflict.
The use of cluster munitions as a means or method of warfare is regulated by the rules of international humanitarian law govering the conduct of hostilities, such as the rule of distinction, including the prohibition on indiscriminate attacks, the rule of proportionality, and the rule of precautions in attacks. These rules are customary norms of international humanitarian law and apply to all states, including states not party to the 2008 Convention on Cluster Munitions, and to all means or methods of warfare, including weapons that fall outside of the scope of that Convention.
Eritrea-Ethiopia Claims Commission: Claim by Ethiopia for the use of cluster munitions against Mekele by Eritrea in 1998
The Commission held that Eritrea conducted four sorties involving cluster munitions that resulted in two strikes hitting Mekele airport and two strikes hitting the Ayder neighbourhood in Mekele. Nevertheless, the Commission was not prepared to draw the conclusion urged by Ethiopia that the attacks on civilians had been intentional, as it was not convinced that Eritrea deliberately targeted a civilian neighbourhood. However, the Commission expressed serious concerns about the manner in which the bombing operations were carried out. It declared that the failure of two out of four bomb runs to come close to their intended targets ‘clearly indicates a lack of essential care in conducting them, compounded by Eritrea’s failure to take appropriate actions afterwards to prevent future recurrence.’ Accordingly, the Commission found Eritrea liable for failing to take all feasible precautions to prevent two of its military aircraft from dropping cluster bombs in the vicinity of Ayder school and its civilian neighbourhood and for the resulting injuries and deaths.
IACtHR, Case of the Santo Domingo Massacre v. ColombiaCase No. 12.416
The Inter-American Commission on Human Rights had reported that on 13 December 1998 the Colombian Air Force was alleged to have bombed the hamlet of Santo Domingo in the department (province) of Arauca. A helicopter dropped an AN-M1A2 cluster munition containing six submunitions, which resulted in 17 civilians being killed, including four boys and two girls, and 27 civilians injured, including five girls and four boys. Colombia sought to attribute the deaths to a bomb placed by FARC guerrillas.
In determining that an armed conflict was ongoing in the area, the Court rejected Colombia’s claims finding that among, other things, Colombia had violated the right to life of those killed by the bombing (Article 4 of the 1969 American Convention on Human Rights) as well as the right to physical, mental, and moral integrity of those injured by the bombing (Article 5(1) of the Convention). In part, the Court conducted its analysis on the basis of the fundamental rules of international humanitarian law, indicating that Colombia had seemingly violated the rule of distinction and concluding that it had violated the rule of precautions in attacks by using cluster munitions that were insufficiently accurate.
ECtHR, Behrami and Behrami v. France
The applicants alleged that Gadaf Behrami’s death and Bekir Behrami’s injuries in a submunition explosion were caused by the failure of the French KFOR troops to mark and/or clear the unexploded submunitions that the Kosovo Protection Force (KFOR) had known to be present on the site in question. They relied on Article 2 (right to life) of the 1950 European Convention on Human Rights (ECHR).
The Court considered that the supervision of demining fell within the mandate of the UN Mission in Kosovo (UNMIK). It concluded that since KFOR was exercising lawfully delegated Chapter VII powers of the UN Security Council and since UNMIK was a subsidiary organ of the UN created under Chapter VII, the impugned inaction was, in principle, ‘attributable’ to the UN, which had a legal personality separate from that of its member states and was not a state party to the ECHR. It therefore declared the claim inadmissible.
ICTY, Martić caseCase No. IT-95-11
This case concerned the firing of submunitions from an M-87 Orkan rocket against the city of Zagreb in May 1995. Each rocket launched on 2 and 3 May 1995 contained a cluster warhead loaded with 288 KB-1 submunitions. These submunitions are ejected at a height of 1,000 metres above the targeted area. Upon impact, each submunition explodes and releases 420 steel ‘pellets’, meaning that each rocket releases around 120,000 pellets. The lethal range of each pellet is ten metres. In its judgment, the ICTY’s Trial Chamber noted:
that the weapon was fired from the extreme of its range. Moreover, the Trial Chamber notes the characteristics of the weapon, it being a non-guided high dispersion weapon. The Trial Chamber therefore concludes that the M-87 Orkan, by virtue of its characteristics and the firing range in this specific instance, was incapable of hitting specific targets. For these reasons, the Trial Chamber also finds that the M-87 Orkan is an indiscriminate weapon, the use of which in densely populated civilian areas, such as Zagreb, will result in the infliction of severe casualties.ICTY, Prosecutor v. Milan Martić, Judgment (Trial Chamber) (Case No. IT-95-11), 12 June 2007, §463.
At first sight, the Trial Chamber appears to affirm that the Orkan is an inherently indiscriminate weapon (the use of which would always be unlawful under international humanitarian law). A closer reading, however, indicates that ‘the firing range in this specific instance’ made it an indiscriminate weapon, while its use in ‘densely populated civilian areas’ would lead to many casualties. Accordingly, a different (i.e. shorter) firing range and a dissimilar environment for its use could potentially have led to a different outcome.The Trial Chamber (§462) found that the distance from which the rockets were fired was close to the maximum range (50 kilometres) of the M-87 Orkan, at which the dispersion error is about 1,000 metres in every direction, with the area of the dispersion of the bomblets on the ground being about two hectares (20,000 square metres). This understanding appears to have been confirmed by the Appeal Chamber’s judgment in the Martić case in which it stated that ‘The Trial Chamber concluded that the M-87 Orkan was used as an ’
Last updated on: 10 December 2013
The precise origin of cluster munitions is the subject of debate. According to one commentator,Virgil Wiebe, 'Introduction: Development and Use of Cluster Munitions, and the Negotiation of the Convention on Cluster Munitions' in Nystuen and Casey-Maslen (eds.),The Convention on Cluster Munitions: A Commentary, OUP, 2010 their development, at least conceptually, can be tracked as far back as the fifteenth century, to the time of Leonardo da Vinci who produced designs of an exploding cannonball that would shatter into smaller bomblets.
In the twentieth century, the United States (US) Army Air Corps developed and produced explosive bomblets released directly from aircraft dispensers in the mid-1920s and refined their use during the 1939–45 War in the South-west Pacific.
The Germans dropped incendiary cluster munitions during the Spanish Civil War in 1937 and both Axis and Allied forces used cluster munitions in significant numbers during the 1939–45 War. During the conflict in South-east Asia in the 1960s and 1970s, the USA is believed to have dropped a total of nearly 400 million submunitions on Cambodia, Lao PDR, and Vietnam. Of these three countries, Lao PDR absorbed the heaviest bombing with cluster munitions. On average, US forces flew one bombing run every eight minutes for nine years.
According to the Cluster Munition Monitor:
Cluster munitions have been used during armed conflict in 36 countries and four disputed territories since the end of World War II. Almost every part of the world has experienced cluster munition use at some point over the past 70 years, including Southeast Asia, Southeast Europe, the Caucasus, the Middle East and North Africa, Sub-Saharan Africa, and Latin America.
Among other use by dozens of states over the past four decades, in 1991, in Iraq and Kuwait, the USA, France, and the United Kingdom dropped 61,000 cluster munitions containing some 20 million submunitions. The number of cluster munitions delivered by surface-launched artillery and rocket systems is not known, but an estimated 30 million or more dual purpose improved conventional munitions (DPICM) submunitions were used in the conflict.
Cluster munitions were also used extensively in south Lebanon in 2006. The UN Mine Action Coordination Centre of South Lebanon (MACC-SL) estimated that Israel used between 2.6 million and 4 million submunitions in its armed conflict with Hezbollah in July and August of 2006. Human Rights Watch received additional information from Israeli soldiers that pushed the estimates to between 3.2 and 4.6 million submunitions. The submunitions used were delivered by air-dropped bombs, artillery shells, and ground rockets. Human Rights Watch also reported that Hezbollah fired ground rockets containing at least 4,600 submunitions into Israel.
Since the 2008 Convention on Cluster Munitions entered into force, there have been credible allegations of new use of cluster munitions by the USA in Yemen in December 2009; by Libya and Thailand in the first half of 2011; and by Sudan and Syria during 2012. None of the alleged users was a state party to the 2008 Convention on Cluster Munitions.
Last updated on: 10 December 2013
According to Human Rights Watch/Cluster Munition Monitor, a total of 34 states have developed or produced more than 200 types of cluster munitions.
A 2009 publication by the Geneva International Centre for Humanitarian Demining (GICHD) reviewed types of submunitions depending on five issues:
There are four principal ways of delivering submunitions onto a target: tube-launched (e.g. artillery shell, mortar, or naval gun), air-dropped container, aircraft dispenser, and rocket/missile. Although most submunitions used to be air-dropped, ground-based delivery by artillery or rocket systems has become increasingly prevalent, most notably in the 1991 Gulf War, the conflict between the coalition led by the United States of America (USA) and Iraq in 2003, and the conflict in southern Lebanon in 2006.
While some submunitions are designed for anti-personnel use (typically with a fragmentation effect upon detonation), others are anti-armour, typically a High Explosive Anti-Tank (HEAT) shaped charge designed to penetrate the armour of tanks and other protected vehicles. A HEAT shaped charge incorporates a conical metal liner (usually made from copper). On detonation, the liner is forced into a high velocity molten jet, which is projected forwards into the target. The high density and hyper-velocity of this jet give it the ability to penetrate armour and other hard surfaces to a far greater depth than high explosive could otherwise achieve.
‘Dual-purpose improved conventional munitions’ (DPICM) combine anti-armour and fragmentation effects, while ‘combined effects munitions’ (CEM), add an additional incendiary element. A widely used CEM is the CBU-87 cluster munition: the BLU-97 submunitions it disperses incorporate a HEAT warhead capable of penetrating more than 200 millimetres of armour. The body of the submunition, made from internally notched steel, shatters into approximately 300 fragments, which can kill personnel, disable vehicles, and damage materiel over dozens of square metres. Also incorporated into the body of the submunition is a zirconium ring, which has an incendiary effect intended to ignite fuel and other combustible materials in the target area.
A wide range of methods are used to open the cluster munition containers and to dispense the submunitions. Once the cluster munition has been fired, launched, expelled, or dropped, the opening of the container is normally determined by a time delay or proximity fuze. The submunitions are normally dispensed by base ejection, nose ejection, or case rupture. Base ejection is most common in projectiles, but is also used in other carriers. In both nose ejection and base ejection, the fuze usually initiates a small propellant charge, which ejects the base plug or nose, and then pushes the submunitions out.
The majority of submunitions use some form of stabilisation (normally fins, a streamer, or a chute) to bring them into a nose-down attitude. In general, submunitions use spin and air resistance to actuate their arming mechanisms, preparing them to explode on impact. This system is referred to as a fuzing mechanism (or a safety and arming unit). Most submunitions are designed to detonate on impact with a hard surface. For instance, when an anti-armour submunition strikes a hard object nose first, the detonator at the rear of the shaped charge is initiated to produce the anti-armour effect. This can be achieved using a firing pin striking a stab-sensitive cap, or a ‘piezoelectric’ element, which generates an electric charge when mechanically deformed.
Secondary fuzing mechanisms may be incorporated to initiate the submunition if the primary fuze fails for any reason, such as impact at the wrong angle. Some use ‘all-ways acting’ mechanisms (see Figure 1 below) that incorporate a ball-bearing housed in a chamber with sloping sides, meaning that they should function no matter what direction the submunition hits the ground. Sideways movement of the ball-bearing acts on the sloping surface to push a pin into a stab-sensitive composition. Like their fin-stabilised variants, most chute-stabilised submunitions produce an anti-personnel/anti-materiel effect as the body is shattered, and many of the submunitions’ exteriors are scored to produce consistent fragmentation.
Since submunitions disperse after ejection, the density of the impact ‘footprint’ (see Figure 2) is strongly depends on the speed and altitude at which the dispenser opens. In addition to leaving behind large areas contaminated with unexploded submunitions, a major humanitarian concern concerns the accuracy of targeting of submunitions during an attack.
Most submunitions free fall in a ballistic trajectory determined by a combination of factors, and can stray far from their intended target. Several anti-armour cluster munition systems now use independently targeted submunitions, which identify and fire at an individual vehicle, although with limited ability to differentiate between military and civilian vehicles. The US BLU-108 Sensor-Fuzed Weapon is an example of such a system, developed to detect and engage individual armoured vehicles without creating a wide-area antipersonnel effect. It carries 40 submunitions, instead of several hundred.
Self-destructing submunitions are designed to automatically detonate after a set period of time if they do not detonate on impact as intended. Mechanisms to ensure this automatic detonation are most commonly either electronic or mechanical.
Last updated on: 10 December 2013
The military utility of cluster munitions is largely due to their capacity to disperse explosive submunitions over a large area. Concerns about the humanitairan impact of cluster munitons arise both, due to their area effect at the time of use, and post-use, due to the high failure rate of submunitions.
Submunitions are intended for use against different targets and their effects also differ. Some are fragmentation devices intended to kill or injure personnel while others are anti-armour.
There has been an increasing trend towards combining effects in order to make submunitions more versatile; this allows the same submunition to be employed against multiple target types. ‘Dual-purpose improved conventional munitions’ (DPICM) combine anti-armour and fragmentation effects, while ‘combined effects munitions’ (CEM), add an additional incendiary element. On the military utility of cluster munitions, see, in particular, O. Dullum, Cluster weapons – military utility and alternatives, Norwegian Defence Research Establishment, Report 2077/02345, February 2008.
Submunitions have killed many thousands of civilians and have injured tens of thousands more. As the GICHD has written:
The design of cluster munitions means that they are a particular threat to the civilian population during and after use. First, their wide-area effect means an increased likelihood of civilian victims or collateral damage to civilian objects from the explosion of the submunitions dispersed by each cluster munition during an attack. This problem is worsened by the typically high number of submunitions that are delivered in a single attack. Second, the failure rate of many submunitions means that a single attack may also leave hundreds or thousands of small unexploded, but lethal devices (sometimes called ‘blinds’ or ‘duds’). Third, the sensitive fuzing system of many submunitions means that even minimal disturbance may be enough to cause them to explode. In addition to causing death and injury, the presence of unexploded submunitions endangers the safe return of the displaced and impedes livelihood activities, such as agriculture or grazing.
Cluster Munition Monitor identified a total of 17,194 cluster munition casualties in 30 states and other areas from 1999 and through the end of 2011, although the Monitor argues that a better estimate for the number of cluster munition casualties globally is between 20,000 and 54,000 casualties. It further cautions that the number of casualties that occurred during cluster munition strikes is known to be grossly underrepresented in both recorded casualties and estimates. According to the Monitor, Afghanistan, Iraq, Lao PDR, Lebanon, Cambodia, and Vietnam are the ‘worst affected’ states, each with significant numbers of cluster munition victims.
One of the major humanitarian concerns regarding the use of cluster munitions is the numbers that fail to explode as intended. According to the GICHD, it is very difficult to determine the precise failure rate of submunitions, but there is considerable evidence that the predicted failure rate suggested by a manufacturer is often much lower than the failure rate when submunitions are used. The main reason for this is the different scenarios in test and real combat conditions. Submunition failure rates are dependent on a number of factors, including:
At least as far as children are concerned, submunitions may be a greater threat than landmines. Submunitions are small and often attractive for children to pick up and play with. In 2001 in Kosovo, for example, the ICRC found that as compared to those killed or injured by anti-personnel mines, those killed or injured by submunitions were 4.9 times as likely to be under age 14. Incidents involving submunitions were also much more likely than landmines to result in death or injury to several people.
Indeed, Cluster Munition Monitor has observed that the vast majority (15,053) of reported casualties occurred after cluster munition use and were caused by cluster munition remnants, including explosive submunitions, which failed to detonate during strikes. Data on casualties due to cluster munition strikes is more difficult to systematically collect and is often not included in casualty reporting. The other 2,141 casualties were recorded from cluster munition strikes. Casualties at time of use are grossly under-reported; therefore the actual number of casualties, both known and estimated, is massively under-represented.
In cases where status was recorded, civilians accounted for the majority (94%) of casualties (10,447 of the total of 17,194 cluster munition casualties), while humanitarian deminers (clearance personnel) accounted for 3%, and security forces (military, police, and other security forces) accounted for another 3%. However, for 6,076 cluster munition casualties (more than a third of all recorded casualties), the civilian status was not indicated or recorded.
Last updated on: 10 December 2013
The 2008 Convention on Cluster Munitions prohibits use, development, production, stockpiling, and transfer of cluster munitions, as defined.+ More
Protocol I additional to the 1949 Geneva Conventions seeks to protect the victims of international armed conflicts. It regulates the conduct of hostilities and sets out basic rules on the use of weapons, means and methods of warfare.+ More