I. SUMMARY AND RECOMMENDATIONS

On October 7, 2001, the United States launched an air campaign in Afghanistan that represented the beginning of its worldwide war on terrorism. The campaign sought to destroy al-Qaeda, the international terrorist organization responsible for the September 11 attacks in New York and Washington, D.C., and the Taliban, Afghanistan's oppressive fundamentalist regime which had sheltered al-Qaeda. In addition to precision guided munitions and traditional unguided "dumb" bombs, the U.S. arsenal included cluster bombs, large weapons that release hundreds of smaller submunitions, or bomblets. While cluster bombs have military value because they can destroy broad or moving targets, they also have serious civilian side effects. Human Rights Watch and others have criticized the bombs' large and imprecise "footprints" (the areas over which bomblets disperse) as well as the fact that they leave large numbers of unexploded submunitions that become de facto landmines. Their use in Afghanistan renewed this debate.

The United States dropped about 1,228 cluster bombs containing 248,056 bomblets between October 2001 and March 2002.¹ Cluster bombs represented about 5 percent of the 26,000 U.S. bombs dropped during that time period. The United States primarily used two aerially delivered models, the CBU-87, a veteran of the Gulf War and the NATO bombing campaign in Yugoslavia, and the new, "wind corrected" CBU-103. A small number of Navy CBU-99s, CBU-100s, and JSOW-As were also used. In 232 cluster strikes, the United States hit targets across Afghanistan, including military bases, frontlines, villages where Taliban and al-Qaeda troops were hiding, and cave complexes. Reports that one bomb went astray and killed at least nine people near Herat outraged critics, who remembered the civilian casualties cluster bombs had caused in past wars. In response, the Pentagon defended cluster bombs as important area and antipersonnel weapons and said that it used them with care.

In a three-and-a-half week mission to Afghanistan in March 2002, Human Rights Watch found ample evidence that cluster bombs caused civilian harm. It confirmed that at least twenty-five civilians died and many more were injured during cluster strikes in or near populated areas. Holes in the walls and roofs of numerous homes were still visible. The casualty figures do not represent the total for the country because some deaths and injuries go unreported; furthermore, the Human Rights Watch team focused on determining potentially disturbing patterns and incidents in the bombing rather than identifying every civilian casualty.

Cluster bombs also left unexploded bomblets, or live duds, which continue to injure and kill innocent civilians long after the attacks. The precise dud, or initial failure, rate of cluster bombs used in Afghanistan, i.e., the percentage of bomblets that did not explode on impact, is not known. Even using a conservative estimate of 5 percent, however, the cluster bombs dropped by the United States likely left more than 12,400 explosive duds that threaten civilians and require clearance. From October 2001 to November 2002, at least 127 civilians as well as two deminers were killed or injured by cluster bomblets. Common post-strike victims in Afghanistan include shepherds grazing their flocks, farmers plowing their fields, and children gathering wood. Duds have also interfered with the economic recovery of the country. Clearance has proceeded with impressive speed, but deminers have had to overcome several obstacles, including shortages in information, equipment, funding, staff, and time.

The use of cluster bombs in Afghanistan, as elsewhere, raises concerns under international humanitarian law (IHL). This body of law, which governs conduct during armed conflict, requires belligerents to distinguish between combatants and non-combatants and prohibits as "indiscriminate" any attacks that fail to do so. Some kinds of cluster bomb attacks consistently rise to the level of being indiscriminate. Particularly troublesome are strikes in or near populated areas, which regularly cause civilian casualties both during strikes, due to the difficulty in precisely targeting cluster bomblets, and after strikes, due to the large number of explosive duds inevitably left by cluster bombs. The aftereffects of unexploded bomblets, especially when they litter an area that will be frequented by civilians or when the dud rate is high, are also problematic. The United States did not intentionally target civilians in Afghanistan, but in some of its cluster bomb attacks, it used means and methods of attack that could be interpreted as indiscriminate. Furthermore, given the foreseeable dangers of using cluster bombs in certain circumstances, it failed in some cases to "take all feasible precautions" to avoid civilian harm as required under IHL.

The use of cluster bombs in Afghanistan did not cause as much humanitarian harm as in some other conflicts, but the case illuminates common and recurrent problems with these weapons. The United States and its allies dropped 61,000 cluster bombs in the Gulf War, and unexploded bomblets killed 1,600 civilians. In Yugoslavia, the NATO alliance used 1,765 clusters, which killed between ninety and 150 civilians during strikes and another fifty after the conflict. The smaller number of cluster bomb casualties in Afghanistan is due in part to the smaller number of cluster bombs used and the extensive pre-existing demining infrastructure, which speeded clearance of unexploded bomblets. The United States also learned lessons from its previous cluster bomb mistakes, making improvements in targeting and technology. Some lessons remain to be learned, however. The United States ignored the major targeting lesson of Yugoslavia when it used cluster bombs in populated areas, and it still has not solved the problem of unexploded bomblets by lowering the dud rate to an acceptable level.

The presence of problems with targeting, aftereffects, and clearance, even in a less egregious case like Afghanistan, suggests that cluster bombs have fundamental flaws. Tactical, technical, and legal steps should be taken to minimize future humanitarian harm. The United States should consider whether the cluster bomb, while effective in some circumstances, is still necessary to its arsenal. Airdropped cluster bombs appear to be of diminishing importance to the U.S. military, given the prevalence of less expensive precision guided munitions and the existing and emerging alternatives to cluster bombs. The international community should formally regulate cluster bombs as it has other problematic weapons, such as antivehicle landmines and incendiary weapons. At least fifty-six countries, many of which have less developed technology and may employ less careful targeting than the United States, stockpile cluster munitions. At least nine states have used them in conflict, most notably the United States in seven countries.² Specific new international law could clarify and strengthen existing IHL restrictions on cluster bombs.

The war in Afghanistan coincided with and spurred an international discussion about the best way to regulate these weapons. In December 2001, the Second Review Conference of the Convention on Conventional Weapons (CCW) appointed a group of governmental experts to evaluate ways to deal with explosive remnants of war (ERW), including the possibility of negotiating a new protocol. Cluster bombs fall under the experts' mandate because the bombs leave unexploded bomblets. Reports of civilian cluster casualties in Afghanistan added impetus to the argument that these weapons should be controlled through a CCW protocol. The experts, who will meet again in December 2002 to present recommendations for next year's activities, have a valuable opportunity to help reduce the humanitarian impact of cluster bombs.

The following report provides a detailed assessment of cluster bomb use in Afghanistan while offering recommendations for minimizing their civilian effects in the future. Human Rights Watch has studied these weapons for years and was the first group to call for a moratorium on their use.³ This report presents detailed findings about the most recent use of cluster bombs based on a post-bombing field mission to Afghanistan. It places these findings in the context of the general cluster bomb debate, evaluating the legality of the weapons under IHL and analyzing how cluster bomb use has evolved over the past decade. The report concludes that while the United States has made efforts to improve the targeting and technology of cluster bombs, the weapon has fundamental flaws that require additional changes and new international regulation.

Methodology
As the air war in Afghanistan moved from populated areas to more isolated caves, Human Rights Watch prepared to send a team of researchers to investigate the conduct of the war and the extent of civilian casualties. One researcher was assigned to focus on cluster bombs. While unitary bombs produce primarily immediate effects, cluster bombs also present special post-conflict threats and clearance obstacles. The civilian impact of both kinds of weapons merited study.

The Arms Division of Human Rights Watch performed an extensive bomb damage assessment in Afghanistan from March 9 through April 3, 2002. Its three-person team traveled around the country to evaluate the impact of the U.S. bombing campaign Operation Enduring Freedom. The team visited more than 250 sites and most of the country's major urban areas, including Kabul, Kunduz, Mazar-i-Sharif, Pul-i Khomri, Kandahar, and Herat. Security concerns made travel to Jalalabad and the far east of Afghanistan impossible.

The Human Rights Watch team laid the groundwork for its mission in the months before its departure. With regard to cluster bombs, it compiled press accounts of strikes and civilian casualties that resulted. It also acquired the list of cluster strikes that the United States had given to the United Nations to help with clearance. This information was processed and inputted into a database harmonized with ArcView geographic information system software, a computer-mapping program. It helped researchers determine which regions and which individual sites to visit.

Once on the ground, team members visited cluster bomb sites in and around the major urban areas listed above. At each site, they took global positioning system (GPS) coordinates, a measure of a location's latitude and longitude, so that they could plot the site on maps and satellite photos. They looked at both the remains of weapons and the types and patterns of destruction they caused. Cluster bombs leave much debris, which provides clues to their model and use. Team members documented what they saw with numerous photographs. They also interviewed witnesses who provided the time and place of strikes, the names of casualties, and information about the long-term effects of unexploded bomblets; such testimony revealed the human significance of the events.

Demining groups around the country provided invaluable assistance to the Human Rights Watch mission. These groups, including the U.N. Mine Action Program for Afghanistan (MAPA), HALO Trust, the Demining Agency for Afghanistan (DAFA), and the Organization for Mine Awareness and Afghan Rehabilitation (OMAR), shared statistics on clearance efforts and civilian casualties. They guided the team to both cleared and uncleared sites, many of which had not been mentioned in press reports or U.S. government documents. Deminers also explained the steps they had taken to clear bomblets and increase awareness of their danger.

Work continued after the mission. In addition to processing its data, Human Rights Watch obtained updated lists of casualties from the various organizations recording that information on the ground. It also spoke with officials from the U.S. State Department, Department of Defense, Navy, and Air Force, from whom it received new information on cluster bomb use. These post-mission interviews clarified the U.S. government's understanding of the events and allowed Human Rights Watch to paint a more complete picture of what happened.

Outline of Report
The report that follows comprises three parts: an overview of the cluster bomb debate, details on cluster bomb use in Afghanistan, and a comparison of use in this conflict and past ones. As a whole, the report illustrates the serious humanitarian problems presented by this weapon and suggests guidelines for future regulation.

The first part of the report discusses the benefits and costs of cluster bombs. Chapter two provides an explanation of the workings of cluster bombs and outlines their military purposes, focusing on the models used in Afghanistan. Chapter three describes the major humanitarian criticisms of these weapons and analyzes how these factors raise serious concerns under IHL.

The heart of the report deals with U.S. use of cluster bombs in Afghanistan. Chapter four provides a brief overview of the air war in Afghanistan and the role cluster bombs played. It then summarizes the public debate that raged around the world during the conflict. The next three chapters present the findings of the Human Rights Watch bomb damage assessment mission to Afghanistan. Chapter five, which focuses on targeting, or immediate effects, uses three case studies to highlight the frequent civilian harm caused by dropping cluster bombs in or near populated areas. Chapter six on aftereffects explains the dangers of unexploded bomblets. Chapter seven describes clearance efforts and the obstacles deminers face in ridding Afghanistan of cluster bomblets.

Chapter eight concludes the report by comparing U.S. use of cluster bombs in Afghanistan to their use by NATO in Yugoslavia in 1999 and the allied coalition against Iraq during the 1991 Gulf War.

An appendix at the end of this report provides more detailed information about the incidents that Human Rights Watch investigated.

Recommendations
While U.S. modifications in targeting and technology appear to have reduced the adverse humanitarian side effects of the cluster bombs used in Afghanistan to some degree, the weapon still poses a danger to civilians in future conflicts because of its broad footprint, lack of accuracy, and high number of explosive duds left behind. Given the proliferation of cluster bombs around the world, a new international instrument is needed to control the fundamental problems inherent in cluster bomb use. The information and recommendations in this report could be used as a starting point for discussion of both additional changes in targeting and technology and future regulation.

Human Rights Watch recommends:

To minimize the humanitarian harm of cluster bombs during strikes

· Cluster bombs should not be used in or near populated or urban areas. The definition of a populated area should include inhabited towns and villages as well as cities.
· The United States and other users of cluster bombs should investigate the causes of cluster bombs that go astray and address them before any future use.
· The United States and other users of cluster bombs should increase transparency about targeting choices, at least post-conflict, so that the effectiveness and dangers of cluster bombs can be better evaluated.

To minimize the aftereffects of cluster bombs

· Cluster bomb use should be suspended until the initial dud rate can be reduced dramatically. By order of the Secretary of Defense, future U.S. submunitions are to have a failure rate of less than 1 percent.
· Whatever the dud rate, militaries should consider the long-term effects of cluster bombs when choosing targets. They should be particularly careful about using cluster bombs in areas to which civilians may return or in environments that increase the dud rate, such as soft terrain.
· Records should be kept of cluster bomb casualties and clearance efforts to allow for future analysis of the aftereffects of this weapon.

To improve clearance

· The United States and other users of cluster bombs should keep accurate records of strikes and report them to the United Nations.
· The United States and other users of cluster bombs should provide clearance assistance, at least in the form of funding, training, and equipment.
· The United Nations and international community should expedite the dissemination of information on cluster bomb locations.
· Cluster clearance groups should develop a clear and consistent system for demarcating areas containing unexploded cluster bomblets and aggressively publicize it in the local civilian community.

To develop better cluster bomb controls for the future

· The United States and others should continue efforts to improve the accuracy and reliability of cluster bombs and submunitions. They should also examine the military necessity of these weapons in modern warfare and consider if other weapons with fewer humanitarian side effects can replace them.
· The international community, especially CCW States Parties, should negotiate a new protocol that addresses the targeting, technology, and clearance of cluster bombs and regulates use of this troubling weapon.

¹ Additional cluster bombs may have been used since March in the mountains of eastern Afghanistan, but the exact number is unknown and likely to be small.

² Human Rights Watch, Memorandum to Convention on Conventional Weapons (CCW) Delegates: A Global Overview of Explosive Submunitions, May 2002 [hereinafter Human Rights Watch, A Global Overview of Explosive Submunitions]. For more information on cluster bomb use by countries other than the United States, see International Committee of the Red Cross (ICRC), Submunitions and Other Unexploded Ordnance (Geneva: ICRC, 2000); Landmine Action, Explosive Remnants of War: Unexploded Ordnance and Post-Conflict Communities (London: Landmine Action, 2002).

³ Human Rights Watch, Cluster Bombs: Memorandum for Convention on Conventional Weapons (CCW) Delegates, December 16, 1999. See also Human Rights Watch, "U.S. Cluster Bombs for Turkey?" A Human Rights Watch Report, vol. 6, no. 19, December 1994; Human Rights Watch, "Ticking Time Bombs: NATO's Use of Cluster Munitions in Yugoslavia," A Human Rights Watch Report, vol. 11, no. 6 (D), May 1999; Human Rights Watch, "Cluster Bombs in Afghanistan," A Human Rights Watch Backgrounder, October 2001. For a complete list of Human Rights Watch documents on cluster bombs, see http://www.hrw.org/arms/clusterbombs.htm.