VIII. PRIOR U.S. USE OF CLUSTER BOMBS

Operation Enduring Freedom is the most recent of a trio of U.S. air campaigns that have made significant use of cluster bombs. During the past decade, the United States has also dropped these weapons on Iraq and Kuwait during the 1991 Gulf War (Operation Desert Storm) and on the Federal Republic of Yugoslavia during NATO's Operation Allied Force in 1999. In order to determine if the U.S. use of clusters in Afghanistan represents an improvement over its use in the past, it is important to understand their use in these previous conflicts.

The Gulf War
Cluster bombs accounted for about one-quarter of the bombs dropped on Iraq and Kuwait during the Gulf War.²⁷¹ Between January 17 and February 28, 1991, the United States and its allied coalition used a total of 61,000 cluster bombs, releasing twenty million bomblets. About 15 percent of those were CBU-87s, then new to the U.S. arsenal.²⁷² Other, less reliable Vietnam-era types included the Rockeye and CBU-52, CBU-58, and CBU-71. ²⁷³ Although media coverage highlighted the use of precision guided "smart" weapons, "dumb" cluster bombs played a major role in the campaign.²⁷⁴

The coalition used cluster bombs against a range of strategic and tactical targets. Taking advantage of clusters' effectiveness against targets with uncertain locations, the coalition chose them as the best means to attack mobile SCUD missiles.²⁷⁵ They were also used against Iraqi tank and vehicle columns retreating from Kuwait.²⁷⁶ As a result, unexploded bomblets littered roads, culverts, and bridges.²⁷⁷ The coalition also used CBU-87s in urban areas, particularly in southern Iraq.²⁷⁸ The selection of such targets led to the cluster bombing of infrastructure and dual use targets frequented by civilians during and after the war.

The manner of dropping the bombs interfered with their precision. In order to avoid anti-aircraft fire, pilots released most of the clusters from medium to high altitudes. The bombs, however, were designed for use at lower altitudes, and pilots had more experience with delivery closer to the ground. The change in altitude decreased the accuracy of strikes and increased the dispersal pattern of the bomblets.²⁷⁹ The new CBU-87 fared better because of its radar proximity fuze and a spin mechanism that controlled dispersal, but it still had a lower accuracy rate than designed.²⁸⁰ This lack of precision increased the risk of immediate collateral damage from a strike. It also decreased the control pilots had over the location of unexploded bomblets.

Duds caused most of the civilian cluster bomb casualties in the Gulf War. As of February 1993, unexploded bomblets had killed 1,600 civilians and injured 2,500 more.²⁸¹ Post-war research revealed an "excessively high dud rate" due to the high altitude from which they were dropped and the sand and water on which they landed.²⁸² The large quantity of bombs added to the problem; even a 5 percent dud rate would have left 2.2 million unexploded submunitions.²⁸³ The high number of duds, combined with the location of the unexploded bomblets in dual use areas, presented significant risks for the civilian population. The plethora of unexploded bomblets on major roads, for example, put both refugees and foreign relief groups at risk.²⁸⁴ The bomblets particularly endangered children; 60 percent of the victims were under the age of fifteen. In addition to being less cautious in battlefield areas, children were attracted by the colorful bomblets, which one reporter described as resembling "white lawn darts, green baseballs, [and] orange-striped soda cans."²⁸⁵

Unexploded bomblets caused other significant side effects. First, they slowed economic recovery because industrial plants, communication facilities, and neighborhoods had to be cleared before they could be restored. Iraqi authorities said that they removed tens of thousands of bomblets from such areas.²⁸⁶ Submunitions also needed to be cleared before people could extinguish the oil fires in Kuwait.²⁸⁷ Second, during and after the war, unexploded ordnance, including submunitions, represented the "greatest threat" to U.S. troops.²⁸⁸ The General Accounting Office reported, that in some cases, "ground movement came to a halt because units were afraid of encountering unexploded ordnance."²⁸⁹ Bomblets killed or injured more than one hundred American soldiers and killed an additional one hundred clearance workers.²⁹⁰

Yugoslavia
Cluster bombs played a smaller, but still significant, role in the NATO air campaign against Yugoslavia. Of the 26,000 bombs dropped between March and June 1999, about 1,765, or 7 percent, were cluster bombs.²⁹¹ They released a total of about 295,000 bomblets over 333 strikes. The United States dropped about 1,100 CBU-87s as well as some CBU-99s and CBU-100s, updated versions of the Rockeye. The United Kingdom dropped 500 RBL-755s, and the Netherlands 165 CBU-87s.²⁹² While the numbers were significantly lower than during the Gulf War, the immediate and aftereffects of cluster bombs remained problematic.

In Yugoslavia, the strikes themselves caused the most significant casualties. Between ninety and 150 civilians died during cluster bomb strikes. These figures represent 18 to 30 percent of the total deaths from Operation Allied Force.²⁹³ Targets included airfields, communications sites, military posts, vehicles on roads, troop concentrations, and armored units.²⁹⁴ The most notable case of civilian deaths, which occurred in Nis on May 7, 1999, demonstrated the danger of using cluster bombs in populated areas. A technical failure caused a CBU-87 to open immediately after the plane released it, instead of over the airfield it was targeting. The bomblets fell on an urban area, killing fourteen and wounding twenty-eight civilians. The incident led President Clinton to suspend temporarily U.S. use of cluster bombs in this campaign.²⁹⁵

As in Iraq, bomblets continued to harm the civilian population after the war. The U.N. Mine Action Coordination Center estimated that a dud rate between 7 percent and 11 percent, depending on bomb model, left more than 20,000 bomblets.²⁹⁶ Some bomblets penetrated up to fifty centimeters (twenty inches) deep, making clearance slow and difficult.²⁹⁷ In the year after the war's end, bomblets killed about fifty civilians and injured 101, with children being frequent victims.²⁹⁸ The deaths from duds represent 10 percent of all civilians killed by U.S. and NATO forces during the bombing. Fortunately, only one BLU casualty has been reported since August 2000.²⁹⁹ Bomblets also interfered with the return of refugees and slowed agricultural and economic recovery.³⁰⁰ Although the campaign did not involve a ground component, the BLUs threatened peacekeeping forces that patrolled the region after the bombing campaign ended. Ten days after the KFOR multinational force deployed, two British soldiers were killed while clearing bomblets.³⁰¹

Afghanistan: Lessons Learned and Not Learned
An analysis of cluster bomb use in Afghanistan gives the United States a mixed report card on its lessons learned from previous wars. The United States dropped about 1,149 cluster bombs during Operation Enduring Freedom and an additional eighty-four in Operation Anaconda.³⁰² These figures are smaller than those for either of the previous campaigns and represent about 5 percent of the 26,000 bombs dropped by the United States between October and March. The United States seems to have improved technology and restricted use more than before. Nevertheless, cluster bomb attacks in Afghanistan raised the same basic problems as in the Gulf and Yugoslavia, and the United States repeated several of its past mistakes.

The United States has made efforts to improve cluster bomb technology over the past decade. In Afghanistan, the United States used the CBU-103 with WCMD for the first time in combat.³⁰³ While still not a precision guided munition, this weapon is designed to have increased accuracy and a more contained dispersal pattern. In its list to the United Nations, the United States estimated footprints with a 1,500-foot (457-meter) radius for CBU-87s and a 600-foot (183-meter) radius for CBU-103s.³⁰⁴ These changes have the potential to reduce collateral damage during strikes and to give the military more control over the location of unexploded bomblets. The United States has also modified the BLU-97 in an apparent effort to reduce the dud rate. The newer version has a cap, or spider, that comes off more easily and a slightly different parachute. These modifications could decrease the chance of a bomblet malfunctioning, but there is no technical proof that this is the case. Despite having the technically better option of the CBU-103, however, the United States continued using the older CBU-87, and Human Rights Watch found evidence of both versions of the BLU in Afghanistan. Furthermore the United States still stockpiles the older versions as well as some Vietnam-era submunitions, which could be used in future wars.

The United States learned some targeting lessons from its experience in the Gulf and Yugoslavia, but it continued to make costly cluster bomb strikes on populated areas. Many CBUs fell on uncontroversial military targets. Unlike in Iraq, the United States did not target roads or bridges with either unitary or cluster bombs. This decision preserved civilian infrastructure and protected returning refugees from some unexploded bomblets. The U.S. military also did not use cluster bombs on industrial plants, communication sites, or other potentially dual-use facilities that would need to be cleared for recovery. Despite these positive changes in targeting practice, the United States ignored the lesson of Yugoslavia and continued to drop cluster bombs in or near populated areas. The thirteen deaths from a stray bomb in Qala Shater are reminiscent of the fourteen deaths from a stray bomb in Nis. While Afghan villages are smaller than Yugoslavian cities, such targets accounted for most, if not all, civilian casualties during cluster bomb strikes in Afghanistan.

The aftereffects of cluster bombs have been less severe in Afghanistan, but that is largely because of the pre-existing demining infrastructure and the fact that fewer bomblets were released. Clearance should still be improved, but it has moved remarkably rapidly in Afghanistan. The dud rate, however, continues to cause problems. Although fewer bombs were dropped, there is no evidence that the rate of unexploded bomblets decreased. Orchard trees that snagged parachutes rather than sand that softened the bomblets' landing caused high dud rates for certain strikes. The unexploded submunitions have led to scores of civilian casualties, including at least twenty-nine deaths, and have endangered coalition troops. The bomblets also continue to slow economic recovery. Afghanistan is a largely agricultural nation so the fact that clusters were not dropped on industrial facilities is relatively insignificant. They did litter farmland, orchards, and grazing areas, which provide Afghans sustenance. While the United States learned some lessons from past wars, it did not always adapt them to the unique circumstances of the current one.

Conclusion: Lessons for the Future
In the future, the United States and other countries using cluster bombs must not only weigh the humanitarian consequences of these troublesome weapons against their military effectiveness, but also learn the lessons of Afghanistan. Some lessons of Afghanistan are clear: cluster bombs should not be used in or near populated areas, the dud rate must be reduced, and clearance efforts should be improved. Even many of the least controversial strikes resulted in civilian casualties, however. The CBUs dropped on large military bases caused casualties because of their proximity to urban areas vulnerable to stray munitions and because people gathered wood or grazed sheep within their borders. Cluster bombs dropped on cave complexes endangered the lives of U.S. troops with their duds. Such effects raise the question of military necessity. Unitary bombs, like those used at Kandahar East Barracks, would have caused the same damage to military facilities with fewer risks. Cluster bombs were most effective against frontlines, but an antipersonnel unitary weapon might have been equally effective. Perhaps the most important lesson of Afghanistan is that despite efforts to correct the errors of past wars, the fundamental problems of cluster bombs remain. Cluster bombs continue to endanger civilians during strikes, leave a lethal legacy after conflicts, and require an international effort for clearance.

²⁷¹ Human Rights Watch, "U.S. Cluster Bombs for Turkey?" p. 15. For more information on cluster bomb use in the Gulf War, see Human Rights Watch, "Ticking Time Bombs," pp. 8-12; Human Rights Watch, "Cluster Bombs in Afghanistan," pp. 8-9.

²⁷² Human Rights Watch, "U.S. Cluster Bombs for Turkey?" pp. 15-16.

²⁷³ Ibid., p. 17.

²⁷⁴ Ibid., p. 15.

²⁷⁵ Ibid., p. 18 (citing U.S. Air Force, Gulf War Air Power Survey, vol. IV, pt. I (1993), p. 290 [hereinafter GWAPS]).

²⁷⁶ Ibid. (citing GWAPS, vol. IV, pt. I, p. 231).

²⁷⁷ Human Rights Watch, "Cluster Bombs for Turkey?" p. 18 (citing GWAPS, vol. IV, pt. I, pp. 43, 48).

²⁷⁸ Ibid., p. 16.

²⁷⁹ Ibid. (citing GWAPS, vol. II, pt. II, pp. 162-63).

²⁸⁰ Ibid. (citing Murray Hamrick, "Aerial Views: USAF Air-to-Air Combat," International Defense Review (July 1991), p. 743).

²⁸¹ Ibid., p. 18.

²⁸² Human Rights Watch, "U.S. Cluster Bombs for Turkey?" p. 16 (quoting GWAPS, vol. II, pt. I, p. 261).

²⁸³ Ibid., p. 17.

²⁸⁴ Ibid., p. 18.

²⁸⁵ Ibid., p. 17 (quoting James Vincent Brady, "Kuwaitis Dying from Old Menace: Unexploded Bombs," Forth Worth Star-Telegram, January 12, 1992, p. 1).

²⁸⁶ Ibid., p. 19.

²⁸⁷ Human Rights Watch, "U.S. Cluster Bombs for Turkey?" p. 19.

²⁸⁸ Ibid.

²⁸⁹ Ibid., p. 20 (citing U.S. General Accounting Office, "Operation Desert Storm: Casualties Caused by Improper Handling of Unexploded U.S. Submunitions," GAO/NSIAD-93-212, August 1993, p. 9).

²⁹⁰ Ibid. (citing Patrick J. Sloyan, "U.S. Bomblets Killed 14 Americans in Gulf War," Newsday, September 19, 1991, p. 4; Brady).

²⁹¹ Human Rights Watch, "Civilian Deaths in the NATO Air Campaign," p. 12; Human Rights Watch, "Cluster Bombs in Afghanistan," p. 7.

²⁹² Human Rights Watch, "Cluster Bombs in Afghanistan," p. 7.

²⁹³ Human Rights Watch, "Civilian Deaths in the NATO Air Campaign," pp. 8, 2.

²⁹⁴ Human Rights Watch, "Ticking Time Bombs," p. 4.

²⁹⁵ Human Rights Watch, "Civilian Deaths in the NATO Air Campaign," pp. 27-28.

²⁹⁶ Human Rights Watch, "Cluster Bombs in Afghanistan," p. 7 (citing ICBL, Landmine Monitor Report 2001, p. 952); "U.N. Mine Clearance Center Expects To Finish Kosovo Mine Clearance by End of 2001," Agence France Press, July 11, 2001).

²⁹⁷ Ibid., p. 8 (quoting ICBL, Landmine Monitor 2001, p. 952).

²⁹⁸ Ibid. (citing ICRC, "Cluster Bombs and Landmines in Kosovo," August 2000, p. 12. The ICRC notes that the actual number of CBU casualties is likely higher because there were 108 incidents in which the cause of injury was unknown.).

²⁹⁹ Ibid. (citing email from John Flanagan, Program Manager, UNMIK Mine Action Program, Kosovo, October 25, 2001).

³⁰⁰ Human Rights Watch, "Ticking Time Bombs," p. 6.

³⁰¹ Human Rights Watch, Cluster Bombs: Memorandum for Convention on Conventional Weapons (CCW) Delegates, December 16, 1999.

³⁰² U.S. Cluster Bomb List--March.

³⁰³ For an explanation of the WCMD, see footnote 8-10 and accompanying text. The Air Force has contracted with Lockheed Martin for 6,000 WCMDs and plans to procure 40,000 WCMDs, including 30,000 for cluster bombs. Each costs $10,000, significantly less than the target cost of $25,000. Tuttle. "WCMD became one of the greatest success stories in the history of Defense Department munitions acquisition, for which it received this year's Schriever award for outstanding program management," said Col. Ken Merchant, director of Eglin Air Force Base's Area Attack System Program Office. Jake Swinson, "New Munitions Dispenser Greatly Improves Accuracy," Air Force Link, December 14, 2001, at http://www.af.mil/news/Dec2001/n20011214_1799.shtml (last visited November 22, 2002).

³⁰⁴ U.S. Cluster Bomb List-January.