Israel has made few public remarks regarding its general policy toward cluster munitions, despite its long history of use, production, and trade of the weapon. Israel is party to the Convention on Conventional Weapons but did not support a proposal at the CCW Review Conference in November 2006 to begin negotiations on cluster munitions within the CCW.50 Israel was also not among the states that gathered in Oslo in February 2007, Lima in May 2007, and Vienna in December 2007 to commit to negotiating a new cluster munitions treaty outside the CCW.
Israel has not ratified CCW Protocol V on Explosive Remnants of War, although it participated in the development of the protocol in 2003 and has expressed support for it. The protocol has provisions regarding the obligations of the user of weapons that become explosive remnants, including cluster munitions, to assist with the cleanup.
Prior to 2006, Israel had used cluster munitions in Syria in 1973 and in Lebanon in 1978 and 1982.51 During the 1978 and 1982 Lebanon conflicts, the United States placed restrictions on the use of its cluster munitions by Israel, although this appeared to have little impact. Indeed, in response to Israels use of cluster munitions in 1982 and the civilian casualties that they caused, the United States issued a moratorium on the transfer of cluster munitions to Israel. The moratorium was lifted in 1988. Unexploded cluster submunitions from the weapons used more than two decades agothough far less extensive than in 2006continued to affect Lebanon up to the beginning of the 2006 conflict.52
Israel is a major producer and exporter of cluster munitions. Israel Military Industries (IMI), an Israeli government-owned weapons manufacturer, has produced, license-produced, and exported artillery projectiles (105mm, 122mm, 130mm, 152mm, 155mm, 175mm, and 203mm), mortar bombs (120mm), and rockets (TCS, EXTRA, GRADLAR, and LAR-160) with submunitions.53
Most notably, it has produced artillery projectiles and ground rockets containing the M85 Dual Purpose Improved Conventional Munition (DPICM) submunition equipped with a back-up pyrotechnic self-destruct fuze. Experts have touted the M85 as among the most reliable and sophisticated submunitions in existence, but as discussed in the Civilian Harm chapter below, it performed poorly in Lebanon in 2006.54 IMI reported that by 2002 it had produced more than 60 million M85 DPICM submunitions.55 IMI concluded licensing agreements in 2004 with companies in India (Indian Ordnance Factories) and the United States (Alliant Techsystems) to produce M85 DPICMs. Companies in Argentina (CITEFA), Germany (Rheinmetall), Romania (Romtechnica), and Switzerland (RAUG Armasuisse) have assembled or produced these submunitions under license to Israel.56
Israel also produces several types of air-dropped cluster munitions. The Rafael Corporation is credited with producing the ATAP-300, ATAP-500, ATAP-1000 RAM, TAL-1, and TAL-2 cluster bombs, as well as the BARAD Helicopter Submunition Dispenser.57
Israel has imported M26 rockets with 644 DPICMs each from the United States for its MLRS launchers. Experienced Israeli non-commissioned officers leading platoons with an MLRS unit told Human Rights Watch that prior to the 2006 conflict, the IDFs stockpile of M26 rockets totaled approximately 18,000. These weapons would contain about 11.6 million submunitions.58 Israel has also imported from the United States M483A1 155mm artillery projectiles with 88 or 72 DPICMs each, Rockeye cluster bombs with 247 Mk 118 bomblets each, and CBU-58B cluster bombs with 650 BLU-63 bomblets each.59
In addition to the US-supplied M26 rockets, IMI has produced a new MLRS rocket called the Trajectory Correction System (TCS). Dubbed Destroyer by the IDF, Israeli media first reported its use in Lebanon on July 17, 2006.60 According to IMI, the TCS improves the accuracy of free flight artillery rockets to that of conventional tube artillery . By providing in-flight trajectory correction, the system simultaneously controls up to 12 rockets in the air, increasing engagement potential while reducing the number of rockets needed per target.61 The TCS underwent operational testing in April 2006 and reportedly reduces the circular error probable (the radius of the area in which half of rockets can be expected to fall) for rocket impact to less than 50 meters at a maximum range of 40 kilometers.62 The number of M85 dual-purpose submunitions contained in each TCS rocket is not publicly known. The US company Lockheed Martin won a contract in 1998 to produce 1,974 rocket motors for integration with the TCS warhead.63
In the 2006 conflict in Lebanon, Israel used cluster munitions delivered by artillery projectiles, ground rockets, and aircraft bombs carrying five main types of submunitions: M42, M46, M77, M85 (with and without self-destruct devices), and BLU-63. These submunition types are unguided weapons that pose grave danger to civilians because of their inaccuracy, wide dispersal pattern, and high dud rates. Human Rights Watch researchers documented each of the five types lying unexploded in villages and surrounding fields in south Lebanon.
The M42, M46, M77, and M85 submunitions are DPICMs whose purpose is to injure persons and pierce armor. The majority of submunitions found in Lebanon have been DPICMs. These submunitions are cylinder shaped; civilians often describe them as resembling batteries. Connected to the top of each of these submunitions is a white ribbon that unfurls when the submunition is released. The ribbon both releases the firing pin, thus arming the submunition, and orients the submunition so that it falls with its shaped charge facing downward.64 The shaped charge is a concave copper cone inside a DPICM designed to explode and pierce armor when it hits perpendicular to its target. A metal fragmentation cylinder is designed to explode and kill people.
M42 and M46 submunitions are delivered by M483A1 155mm artillery projectiles. Each projectile carries 88 M42 and M46 submunitions. Both the submunitions and the projectiles were made in the United States. The submunitions are able to penetrate more than 2.5 inches of armor.65 The test condition failure rate of these two submunitions is between 3 and 14 percent.66 As of January 2008, clearance personnel have destroyed 46,082 unexploded M42 and M46 submunitions, 33 percent of the total number of duds destroyed.67
Israel also widely used the M77 in Lebanon. M77 submunitions are delivered by M26 MLRS rockets. The launchers, rockets, and submunitions were produced in the United States. Each rocket contains 644 M77 submunitions, and each MLRS can fire up to 12 rockets at once. A typical volley of six rockets would release 3,864 submunitions over an area with a one-kilometer radius. Called Steel Rain by Gulf War soldiers, the submunitions can pierce up to four inches of armor.68 The M77, visually distinguishable from the M42 and M46 by its white stripe, has a reported test condition failure rate of 5 to 23 percent.69 The US use of M26 rockets in Iraq in 2003 caused hundreds of civilian casualties.70 Deminers in Lebanon have cleared more M77s than any other type of submuntion57,271 submunitions, which represent 41 percent of the total.71
M85 submunitions are delivered by M395 and M396 155mm artillery projectiles, which contain 63 and 49 M85 submunitions, respectively. The submunitions and the projectiles were made in Israel. Israel has produced at least two versions of the M85 submunition, an older model similar to the M42, M46, and M77, and a newer model with a self-destruct device. Many military experts consider the newer version to be one of the most reliable and sophisticated submunitions in existence.72 The
submunitions reported failure rate is 1.3 to 2.3 percent under test conditions.73 Based on a study of strike locations where the self-destruct models landed, however, both weapons experts and MACC SL estimate that the self-destruct M85s had an actual failure rate 10 percent or higher.74 (See below for more information.) Clearance groups have destroyed 6,892 M85s with and without self-destruct systems, 5 percent of the total number of submunitions found in Lebanon.75
Israel used only limited numbers of its new Trajectory Correction System MLRS rockets with M85 submunitions. Israeli soldiers told Human Rights Watch that the IDF fired a total of 130 TCS rockets and used them exclusively in the earlier stages of the conflict.76 An IDF reserve officer told a reporter that his battalion used only a small number of RAMAM rockets (the Hebrew acronym for TCS) and just in the first days of the war.77 The control unit for TCS, inside an armored vehicle, required level ground for proper guidance operation. TCS fire missions involved shooting one to three rockets at a target, in contrast to the mass firing of M26 rockets in later weeks. Soldiers in the battalion received little, but contradictory, feedback on the performance of TCS.78
Israel also used aerially delivered CBU-58B cluster bombs with BLU-63 submunitions, both made and supplied by the United States. Each CBU-58B contains 650 BLU-63 bomblets, which are ball-shaped, weigh roughly one pound, and measure three inches in diameter.79 The bombs and bomblets are Vietnam war-era weapons developed in the early 1960s. While fewer BLU-63s were used than DPICMs, deminers have still found 28,136 duds from 2006 throughout Lebanon, 20 percent of their total clearance numbers.80 MACC SL officials blame the submunitions high dud rate on the fact that it is an ancient weapon.81 The United States last used this cluster bomb in the 1991 Gulf War and no longer has it in its inventory.
In Nabatiyah, Zawtar al-Gharbiyeh, and Beit Yahoun, among other places, Human Rights Watch researchers examining sites in the immediate aftermath of the 2006 conflict saw CBU-58B canisters stamped with load dates of September 1973, meaning that their original contents were loaded in 1973. Most of the CBU/BLUs found by deminers have been from the 1970s, particularly the years 1973, 1976, and 1978.82 Deminers have also encountered several CBU-58B catastrophic failures, where the weapon completely failed to function and none of the submunitions dispersed or exploded.83
In addition, demining groups have found 1,207 Chinese-made MZD-2 submunitions in Lebanon, 1 percent of the total submunitions cleared so far.84 Human Rights Watch saw one unexploded MZD-2 on the side of a road in Beit Yahoun. Human Rights Watch documented that Hezbollah fired Type-81 122mm cluster munition rockets containing MZD-2 (also called Type-90) submunitions into Israel during the conflict.85 Since Israel is not known to have this Chinese-made weapon in its arsenal, it is most likely that the MZD-2s found in Lebanon belonged to Hezbollah and not Israel, and either misfired, fell short, or were left behind following Israeli strikes on the weapons.86
50 Israel also was not among the dozens of CCW states parties that provided information regarding their views on IHL and explosive remnants of war, including cluster munitions, as part of the work of the CCW Group of Governmental Experts in 2005 and 2006.
51 There are unconfirmed reports of Israeli use of cluster munitions in Lebanon in 1996 and 2005. See Handicap International, Circle of Impact: Footprint of Cluster Munitions on People and Communities, May 2007, http://www.handicap-international.org.uk/page_709.php (accessed September 3, 2007).
52 For details on past use and on the US restrictions and moratorium, see Landmine Action, Cluster Munitions in Lebanon, November 2005, http://www.landmineaction.org/resources/resource.asp?resID=1009 (accessed September 3, 2007).
53 Information on surface-launched cluster munitions produced and possessed by Israel is taken primarily from Israel Military Industries Ltd. (IMI), http://www.imi-israel.com/Homepage.aspx?FolderID=11 (accessed September 3, 2007). It has been supplemented with information from Janes Ammunition Handbook , Terry J. Gander and Charles Q. Cutshaw, eds. (Surrey, UK: Janes Information Group Limited, 2001) and US Defense Intelligence Agency, Improved Conventional Munitions and Selected Controlled-Fragmentation Munitions (Current and Projected) DST-1160S-020-90, June 8, 1990, partially declassified and made available to Human Rights Watch under a Freedom of Information Act request.
54 Military experts from numerous countries that stockpile the M85 or variants of it have made this claim in discussions with Human Rights Watch during sessions of the CCW in recent years.
55 Mike Hiebel, Alliant TechSystems, and Ilan Glickman, Israeli Military Industries, Self-Destruct Fuze for M864 Projectiles and MLRS Rockets, presentation to the 48th Annual Fuze Conference, Charlotte, North Carolina, April 27-28, 2004, slide 9, http://www.dtic.mil/ndia/2004fuze/hiebel.pdf (accessed November 28, 2006).
56 See Israel Military Industries Ltd. (IMI), http://www.imi-israel.com/Homepage.aspx?FolderID=11 (accessed September 3, 2007); Janes Ammunition Handbook; US Defense Intelligence Agency, Improved Conventional Munitions and Selected Controlled-Fragmentation Munitions.
57 Janes Air Launched Weapons, Robert Hewson, ed. (Surrey, UK: Janes Information Group Limited, 2004), pp. 370-380.
58 Human Rights Watch interviews with IDF reservists (names withheld), Tel Aviv and Jerusalem, Israel, October 2006.
59 All of these US-produced cluster munitions, except the Rockeye bombs, were used by Israel in Lebanon. The details of the transfers are not known.
60 Hanan Greenberg, IDF: Our Rockets More Dangerous, Ynetnews.com, July 17, 2006, http://www.ynetnews.com/Ext/Comp/ArticleLayout/CdaArticlePrintPreview/1,2506,L-3277034,00.html (accessed October 26, 2006).
61 Israel Military Industries Ltd. (IMI), TCSTrajectory Corrected System, http://www.imi-israel.com/Business/ProductsFamily/Product.aspx?FolderID=36&docID=311 (accessed October 26, 2006).
63 Lockheed Martin Vought Systems Receives Contract for Israeli MLRS Trajectory Correction System, PRNewswire, November 2, 1998.
64 Database of Demining Incidents and Victims, Ribbon Oriented Dual Purpose Submunition, http://www.ddasonline.com/SubsKB1-M42.htm (accessed November 29, 2006).
65 Globalsecurity.org, Dual-Purpose Improved Conventional Weapons, http://www.globalsecurity.org/military/systems/munitions/dpicm.htm (accessed November 29, 2006).
66 The 3 percent figure is contained in Office of the Under Secretary of Defense (Acquisition, Technology and Logistics), Report to Congress: Cluster Munitions, October 2004. The 14 percent figure is from US Army Defense Ammunition Center, Technical Center for Explosives Safety, Study of Ammunition Dud and Low Order Detonation Rates, July 2000, p. 9, and Office of the Under Secretary of Defense for Acquisition, Technology and Logistics, Unexploded Ordnance Report, undated, but transmitted to the US Congress on February 29, 2000, table 2-3, p. 5.
67 Email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 18, 2008.
68 Globalsecurity.org, Dual-Purpose Improved Conventional Weapons.
69 A 5 percent failure rate was reported in US Office of the Under Secretary of Defense (Acquisition, Technology and Logistics), Report to Congress: Cluster Munitions, pp. 2-6. A 16 percent failure rate was reported in US Office of the Under Secretary of Defense for Acquisition, Technology and Logistics, Unexploded Ordnance Report, table 2-3, p. 5. A 23 percent failure rate for some newly produced lots was reported in US General Accounting Office, GAO/NSIAD-92-212: Operation Desert Storm: Casualties Caused by Improper Handling of Unexploded US Submunitions, August 1993, pp. 5-6. UK testing has indicated a 5 to 10 percent failure rate, which is largely dependent on ground conditions and range. DLO Secretariat, DLO Andover, Response to Landmine Action Question, Reference 06-02-2006-145827-009, March 27, 2006.
70 Human Rights Watch, Off Target: The Conduct of the War and Civilian Casualties in Iraq (New York: Human Rights Watch, 2003), http://www.hrw.org/reports/2003/usa1203/.
71 Email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 18, 2008.
72 Military experts from numerous countries that stockpile the M85 or variants of it have made this claim in discussions with Human Rights Watch during sessions of the CCW in recent years.
73 In tests carried out in Norway in September and October 2005 of the Norwegian stockpile of cluster munitions as well as of identical UK-owned DPICM projectiles, submunition failure rates of 2.3 percent, 2 percent and 1.3 percent were achieved. Some UK test results have also been made available: The manufacturers firing trials indicated that 97% of armed grenades will have a successfully functioning self-destruct mechanism. The results of the acceptance proofs for lots 1 to 3 for which 60 shells (2,940 bomblets) were fired with 22 bomblet failures represent[s] a failure rate of 0.74%. Of these failures, only 6 of the bomblets had armed . In Sep 05 the first in-service safety and performance test was carried out at Hjerkinn Range, Dombass, Norway. During the test 175 shells were fired of which none failed, 8,575 bomblets deployed of which 197 failed, giving a bomblet failure rate of 2.3%. DLO Secretariat, DLO Andover, Response to Landmine Action Question.
74 For a detailed discussion of the M85 with self-destruct device and its failure in Lebanon, see C. King Associates, Ltd., Norwegian Defence Research Establishment, and Norwegian Peoples Aid, M85: An Analysis of Reliability (Norway: Norwegian Peoples Aid, 2007). See also information provided by Ove Dullum, Chief Scientist, Norwegian Defence Research Establishment, April 19, 2007; Chris Clark, program manager, MACC SL, Unexploded Cluster Bombs and Submunitions in South Lebanon: Reliability from a Field Perspective, paper presented at ICRC Expert Meeting, Montreux, Switzerland, April 18-20, 2007, http://www.icrc.org/web/eng/siteeng0.nsf/htmlall/cluster-munition-montreux-310507 (accessed April 30, 2007); email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 16, 2008.
75 Email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 18, 2008.
76 Human Rights Watch interviews with IDF reservists (names withheld), Tel Aviv and Jerusalem, Israel, October 2006. The TCS was used only during the second week of operations according to one soldier serving in the reserve MLRS battalion.
77 Meron Rapoport, A Barrage of Accusations, Haaretz, December 8, 2006.
78 Human Rights Watch interviews with IDF reservists (names withheld), Tel Aviv and Jerusalem, Israel, October 2006.
79 Designation-Systems.net, BAK to BSU/BSGEquipment Listing, http://www.designation-systems.net/usmilav/asetds/u-b.html (accessed September 3, 2007).
80 Email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 18, 2008.
81 Human Rights Watch interview with Chris Clark, program manager, MACC SL, Tyre, October 21, 2006.
82 Human Rights Watch interview with Allan Poston, chief technical advisor, National Demining Office, UNDP, Beirut, November 29, 2006.
83 Chris Clark, program manager MACC SL, presentation to CCW Delegates, Geneva, August 30, 2006 (notes by Human Rights Watch).
84 Email communication from Dalya Farran, media and post clearance officer, MACC SL, to Human Rights Watch, January 18, 2008.
85 Lebanon/Israel: Hezbollah Hit Israel with Cluster Munitions During Conflict, Human Rights Watch news release, October 19, 2006, http://hrw.org/english/docs/2006/10/18/lebano14412.htm.
86 Human Rights Watch interview with Andy Gleeson, program manager and technical operations manager, Mines Advisory Group, Kfar Joz, October 25, 2006. The speculation is that the submunitions were dropped or abandoned by Hezbollah, or dislodged by an Israeli strike. For more information on Hezbollahs cluster munition attacks, see Human Rights Watch, Civilians under Assault, pp. 44-48.