Cluster Bombs in Afghanistan
Human Rights Watch Backgrounder,
The United States-led alliance began its
air campaign in Afghanistan on October 7, 2001. While the Pentagon
has been reluctant to talk of specific weapons used in the bombing,
U.S. military sources have told Human Rights Watch that the Air
Force began dropping cluster bombs within a matter of days. During
the first week of the campaign, it is believed that Air Force
B-1 bombers dropped 50 CBU-87 cluster bombs in some five missions.
CBU-87 cluster bomb use has continued after the first week, and
it is believed that other airplanes joined B-1s in dropping cluster
bombs on both fixed and mobile targets.
Human Rights Watch has called for a global
moratorium on use of cluster bombs because they have been shown
to cause unacceptable civilian casualties both during and after
conflict. Cluster bombs have a wide dispersal pattern and cannot
be targeted precisely, making them especially dangerous when used
near civilian areas. Cluster bombs are usually used in very large
numbers and have a high initial failure rate which results in
numerous explosive "duds" that pose the same post-conflict
problem as antipersonnel landmines._ _United Nations officials
have stated that on October 22 U.S. cluster bomb submunitions
landed on the village of Shaker Qala, near the city of Herat in
western Afghanistan, killing nine civilians and injuring fourteen.
The head of the United Nations Mine Action Program in Afghanistan
(U.N. MAPA) noted that villagers are afraid to leave their homes
after encountering the yellow soda can-like objects characteristic
of CBU-87 submunitions that were left scattered in the village
after an air strike on a nearby military camp. He called upon
the United States to provide information on the types of ordnance
dropped on Shaker Qala and elsewhere.
On October 25, the U.S. for the first
time publicly acknowledged using cluster bombs. In response to
a media question, Chairman of the Joint Chiefs of Staff General
Richard Myers said, "Yes, we have used cluster bomb units.
There have not been a great number of them used, but they have
Each CBU-87 cluster bomb contains 202
individual submunitions, also called "bomblets," designated
BLU-97/B. The CBU-87s are formally known as Combined Effects Munitions
(CEM) because each bomblet has an antitank and antipersonnel effect,
as well as an incendiary capability. The bomblets from each CBU-87
are typically distributed over an area roughly 100 x 50 meters.
They can be dropped from virtually any U.S. Air Force, Navy, and
Marine Corps aircraft.
Recent experience in Kosovo, and before
that in the Gulf War, has shown that the exact "footprint,"
or landing area, of the CBU-87's bomblets is difficult to control
and that an initial failure-to-explode rate of some 7 percent
can be expected.
When the bomblets contained inside cluster
bombs fail to explode on contact as intended, they become in effect
antipersonnel landmines-volatile and deadly remnants of war that
can explode from a simple touch. They have proven to be a serious
and long-lasting threat to civilians, soldiers, peacekeepers,
and even clearance experts, because of the high initial failure
rate of the bomblets, because of the large number typically dispersed
over large areas, and because of the difficulty in precisely targeting
A key United Nations clearance expert
has expressed concern about the similarity of the coloring of
the yellow BLU-97/B cluster bomblets and the small yellow food
aid parcels being airdropped in Afghanistan, noting that people
are being encouraged to pick up the food parcels, but that picking
up a bomblet would be lethal. He said, "Our experience in
Kosovo showed us that children and youths were highly susceptible
to the submunitions. It is highly likely that many in Afghanistan
will not know the difference between aerially delivered food aid
and aerially delivered munitions." BBC Worldwide Monitoring
reported that U.S. Psychological Operations units broadcast a
radio message warning Afghan civilians of the similar yellow color
of the cluster bomblets and the food packages, noting that cluster
bombs will not be dropped in areas where food is air-dropped but
stating, "[W]e do not wish to see an innocent civilian mistake
the bombs for food bags and take it away believing that it might
It is noteworthy that during Operation
Deliberate Force in Bosnia in 1995, air combat commander Major
General Michael Ryan (later U.S. Air Force chief of staff) decided
to prohibit the use of cluster bombs, in recognition of the inherent
danger to civilians. "The problem was that the fragmentation
pattern was too large to sufficiently limit collateral damage
and there was also the further problem of potential unexploded
ordnance," says one Air Force-sponsored study. During Operation
Allied Force in Yugoslavia in 1999, the White House prohibited
further use of CBU-87s until technical adjustments could be made,
after a cluster bomb malfunction on May 7 killed many civilians.
Afghanistan is already one of the countries
most severely affected by landmines and unexploded ordnance (UXO).
Prior to October 7, 2001, the known contaminated area was estimated
at 724 million square meters, including 344 million square meters
classified as high priority land for clearance. From 1990 through
2000, more than 225,000 landmines and 1.3 million pieces of unexploded
ordnance (including submunition duds) were detected and destroyed.
The Taliban and the United Front (Northern Alliance) have used
surface-delivered cluster munitions, fired from BM-21 122mm multiple
According to information received by Human
Rights Watch, the U.S. inventory alone contains more than one
billion individual submunitions. The United States has more than
forty different types of air and surface-delivered cluster bombs
and submunitions. It is thought that at least eighteen nations
produce cluster munitions and more than four dozen have stockpiles
of the weapons.
What Are Cluster Bombs?
Modern cluster bombs are of two main types-those
delivered by air and those delivered by surface artillery or rockets
(including artillery projectiles and multiple rocket launchers).
The bombs are designed to disperse submunitions (often called
"grenades" in surface-delivered weapons and "bomblets"
in air-delivered weapons) over a large area, thereby increasing
the radius of destructive effect over a target. Typical targets
for cluster bombs would include troop concentrations, airfields,
and air defense units.
The large number delivered increases the
density of explosives in the target area, with submunitions designed
to strike every few feet or so. They saturate an area with explosives
and tiny flying shards of steel. Depending on the type, bomblets
can be dispersed to areas as large as the size of several football
fields. An air attack typically disperses thousands of submunitions
within a small space; a common target area for a single weapon
under optimal conditions covers an area of roughly 100 x 50 meters.
Air-delivered cluster bombs are
composed of a large dispenser with attached fins (called the tactical
munitions dispenser, or TMD, in the newest systems); fuzes and
electronic devices to control, spin, and direct the weapon during
fall; and submunitions or bomblets. The bomblets themselves are
of a variety of designs and shapes. Once released, cluster bomb
units (CBUs) fall for a specified amount of time or distance before
the dispenser opens and dispenses the submunitions, allowing them
to cover a wide-area target. Depending on the type, the submunitions
are activated by an internal fuze, and can detonate above ground,
at impact, or in a delayed mode. Existing versions of submunitions
do not incorporate self-destruct or self-deactivating mechanisms.
Modern air-delivered cluster bombs
can be set to determine height of burst and the dispersal pattern.
As the aircraft drops the TMD, tail fins open and stabilize the
bomb body. At the selected time or altitude, the dispenser begins
to spin, the spin rate determining the dispersal pattern. As the
bomblets fall and disperse, they arm in different ways depending
on their design.
The U.S. CBU-87 Combined Effects Munition
is one of the newest standard air-delivered cluster bomb units
(CBUs) in the U.S. arsenal. It has been in the U.S. Air Force
inventory since 1986 (and in production since 1984), and has replaced
aging and less effective Vietnam-era cluster bomb units and antitank
mines. A myriad of delivery settings (high and low altitudes,
extremely high speeds, and various toss modes) makes it a significant
advance over older bombs. Combining light antiarmor capabilities
with antipersonnel and incendiary effects, it was the first weapon
in the U.S. inventory to include all three "kill mechanisms."
The 1,000-lb. class cluster bomb
is compatible with virtually all current Western-produced tactical
fighter aircraft, U.S. and foreign. The dispenser has been approved
for use on the A-10, AV-8B, F-15, F-16, F/A-18, B-1, and B-52
aircraft. International aircraft currently certified for the CBU-87
include NATO F-16s, British Hawk and Harrier, French Mirage V,
German Alpha Jet, Japanese F-1 and FX, and multinational Tornado
and Jaguar. It is manufactured by Alliant Techsystems of Minnesota.
The weapon contains 202 BLU-97B bomblets (officially called "combined
effects bomblets," CEBs), which are seven inches long, with
a two-and-a-half inch diameter and a weight of 3.41 pounds._
In contrast with earlier cluster bombs, the ground pattern
size and shape of the bomblet dispersal can be determined in the
CBU-87 by setting the spin rate of the dispenser and the height
of burst. A single CBU-87 set at a low spin rate (e.g., 500 rpm)
can disperse bomblets to an area 120 by 200 feet, with bomblets
scattered an average of nine feet apart. A range of impact patterns
from 70 by 70 feet to 150 by 450 feet can be achieved depending
on altitude. In general, the bomblets cover an area of 800 by
400 feet, given medium- to high-altitude delivery.
As the CBU-87's soda can-sized bomblets
fall, a "spider" cup is stripped off the body, releasing
a spring which pushes out a nylon "parachute" (called
the decelerator), which inflates and then stabilizes and arms
the bomblet. The bomblets orient perpendicular to the ground for
optimal top attack, and the descent is slowed to approximately
125 feet per second. On impact the primary firing mechanism detonates
the bomblet. A secondary firing system is included to detonate
if the bomblet impacts other than straight on, or if the bomblet
lands in soft terrain or water.
The BLU-97's parachute-like decelerator,
firing system, fuze, and downward-firing shaped charge are all
packed in a steel case with a fire-starting (incendiary) zirconium
ring. The case is the main part, made of scored steel designed
to break into approximately 300 preformed thirty-grain fragments
upon detonation of the internal explosive. The fragments then
travel at extremely high velocities in all directions. This is
the primary antipersonnel effect of the weapon. An explosively
shaped charge (a formed molten copper jet slug) is the primary
antiarmor effect. If the bomblet has been properly oriented, the
downward-firing charge travels at 2,570 feet per second, able
to penetrate most armored vehicles. The zirconium ring provides
for fuel and other fires by spreading small incendiary fragments.
The shaped charge has the ability to penetrate five inches of
armor on contact. The tiny steel case fragments are also powerful
enough to damage light armor and trucks at fifty feet, and to
cause human injury at 500 feet. The incendiary ring can start
fires in any combustible environment.
Use of CBU-87s and other Cluster Bombs
U.S., British and Dutch aircraft dropped
more than 1,765 cluster bombs containing more than 295,000 cluster
bomblets during the NATO air campaign in Yugoslavia from March
to June 1999. During Operation Allied Force, the U.S. dropped
about 1,100 CBU-87s (each containing 202 submunitions), the United
Kingdom dropped about 500 RBL-755 cluster bombs (each containing
147 submunitions), and the Netherlands dropped 165 CBU-87s.
The U.N. Mine Action Coordination Center
(MACC) reported that NATO provided the locations of 333 cluster
bomb strike areas. On the basis of the clearance rate by March
2001 of unexploded cluster bomblets, the MACC estimated that around
seven percent of the CBU-87's bomblets and eleven percent of the
BL-755's failed to explode on impact. According to the MACC, more
than 20,000 bomblets remained after the war, and the bomblets
"are in a highly sensitive state, and can explode as a result
of being moved or picked up. This volatile condition means that
NATO-dropped CBU are a major part of the mine/UXO problem in Kosovo."
In its June-September 2000 quarterly report, MACC wrote, "During
the previous quarter, it had been recognized that CBU were a major
contributing factor to incidents involving civilians. In particular,
CBU incidents generally involved groups of younger people, often
with very tragic results."
The MACC has also noted, "The CBU
problem is exacerbated by the fact that many bomblets have penetrated
the ground and some have been found up to 50 centimeters below
the surface. This means that CBU strike areas must be subjected
to sub-surface clearance using detection equipment before the
area can be declared free of UXO."
Human Rights Watch criticized NATO for
use of cluster bombs in Kosovo, particularly in or near populated
areas. Human Rights Watch believes there were nine to fourteen
cluster bomb attacks resulting in civilian casualties during the
conflict, causing an estimated ninety to150 civilian deaths, or
18 to 30 percent of all civilian deaths, even though cluster bombs
represented just 6 percent of weapons expended in the air war.
A NATO air strike involving cluster bombs on an airfield in Nis
on May 7 went off target, hitting a hospital complex and adjoining
civilian areas. On April 24, five boys were reported to have been
killed and two injured when what was evidently a cluster bomb
submunition exploded near the village of Daganovic.
The civilian toll due to cluster bombs
was even greater following the end of the conflict. According
to the International Committee of the Red Cross, from June 1999
through May 2000, there were at least 151 casualties due to cluster
bomblets, including fifty dead and 101 injured. 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.
The MACC has reported that in the year 2000, there were twenty-four
CBU casualties, and that as of late October 2001, successful clearance
operations resulted in just one cluster bomb incident since August
Use of Cluster Bombs in the Gulf War
More than 1,600 Kuwaiti and Iraqi civilians
have been killed, and another 2,500 injured, by the estimated
1.2 million explosive cluster bomb duds left following the 1991
Persian Gulf War, which saw the most extensive use of cluster
bombs in history. Some 62,000 air-delivered cluster bombs, 100,000
DPICM artillery shells, and 10,000 MLRS rockets were used, containing
a total of 24 to 30 million submunitions.
The United States has continued to use
cluster bombs in Iraq. While great press attention was paid to
President George W. Bush's decision to bomb Iraqi targets on February
16, 2001, there was scant recognition that some U.S. jets used
cluster bombs, those formally known as the Joint Stand-Off Weapon
(JSOW). JSOW was first used in combat in Iraq on January 25, 1999.
The 1,000 pound, fourteen-foot-long weapon carries 145 anti-armor
and antipersonnel incendiary bomblets that disperse over an area
that is approximately 100 feet long and 200 feet wide.
Department of Defense Justifications
The Pentagon primarily justifies
use of cluster bombs based on the perceived effectiveness of the
weapon. In fact, the utility, reliability, and effectiveness of
different types of cluster bombs and submunitions varies tremendously.
But the Pentagon has also offered up rebuttals to some of the
other criticisms that have been made regarding cluster bombs._
_Initial Failure Rates - The Department of Defense will sometimes
claim that the initial failure rate of most cluster munitions
is not much different from other munitions, such as gravity bombs,
mortar rounds, or artillery shells. Human Rights Watch is unaware
of a serious or comprehensive study of this matter. Regardless,
however, the initial failure rate of cluster munitions, whether
equal to or greater than other munitions, is a special problem
because of the large number of submunitions used, and their particular
volatility. When each bomb contains hundreds of bomblets, and
hundreds of thousands of bomblets are used in a campaign (as in
Kosovo) or even millions of bomblets (as in the Gulf War), even
a small initial failure rate can quickly translate into a major
Moreover, the current initial failure
rate for even advanced U.S. cluster munitions, like the CBU-87,
is clearly far too high to be acceptable. Estimates of the initial
failure rate range from 2 percent to 30 percent or more, depending
on conditions. The best data on this has been gathered in Kosovo,
where, as noted above, U.N. clearance experts estimate a 7 percent
initial failure rate for the CBU-87 bomblets.
While not saying so directly, the U.S.
military has acknowledged that the initial failure rate on cluster
bombs is too high. On January 10, 2001, then-U.S. Secretary of
Defense William Cohen issued a memorandum stating:
It is the policy of the DoD [Department
of Defense] to reduce overall UXO [unexploded ordnance] through
a process of improvement in submunition system reliability - the
desire is to field future submunitions with a 99% or higher functioning
rate. Submunition functioning rates may be lower under operational
conditions due to environmental factors such as terrain and weather.
_The Services may retain "legacy" submunitions until
employed or superseded by replacement systems in accordance with
the above policy. The designation "legacy" would apply
to submunition weapon acquisition programs reaching Milestone
III prior to the First Quarter of Fiscal Year 2005._The Services
shall evaluate "legacy" submunition weapons undergoing
reprocurement, product improvement, or block upgrades to determine
whether modifications should be made to bring them into compliance
with the above policy._The Services shall design and procure all
future submunition weapons in compliance with the above policy.
A "future" submunition weapon is one that will reach
Milestone III in FY 2005 and beyond._[Secretary of Defense William
Cohen, Memorandum for the Secretaries of the Military Departments,
Subject: DoD Policy on Submunition Reliability (U), January 10,
The volatility of cluster bomb
duds also makes them more dangerous than many other types of unexploded
ordnance. Again, Human Rights Watch is not aware of specific studies
comparing UXO volatility, but the testimony of clearance personnel,
such as that cited above regarding Kosovo, gives an indication
of the special dangers posed by cluster bomblets.
CBUs vs. Other Weapons - Department
of Defense and other U.S. officials will sometimes claim that
cluster bombs pose less danger to civilians than alternative weapons
that might be used, noting for example that the explosive power
of unitary munitions (such as bombs and artillery shells) could
cause far more collateral damage. However, this argument ignores
the documented inability to ensure cluster bomblets stay within
the confines of the intended target area, and does not take into
account the ever-greater precision (and smaller warheads) of unitary
munitions, which results in less and less civilian impact.
Cluster Bombs and Landmines
While cluster bomblet duds undeniably
function like antipersonnel mines, they are not covered under
the 1997 Convention on the Prohibition of the Use, Stockpiling,
Production, and Transfer of Anti-Personnel Landmines and on their
Destruction, also known as the Mine Ban Treaty, which entered
into force on March 1, 1999. There are 122 States Parties and
an additional 20 signatories. The United Kingdom is a State Party;
the United States has not signed.
The treaty defines antipersonnel mine
as a munition "designed to be exploded by the presence, proximity
or contact of a person." That is, the mine must be designed
to be victim-activated. Cluster bombs are not designed to be victim-activated;
they become so when they fail to function as designed. In this
way, cluster bombs are not different than other types of unexploded
ordnance (failed mortar rounds, artillery shells, grenades, etc.),
none of which are covered by the Mine Ban Treaty.
The indiscriminate dangers posed to civilians
by explosive cluster bomb duds, however, are essentially no different
than antipersonnel mines. They will in most cases explode upon
contact, whether being picked up or kicked or otherwise touched.
One difference is that antipersonnel mines are generally designed
more to maim than to kill, with a relatively small amount of explosive,
while cluster bomblets (at least in the case of CBU-87s) are more
likely to kill their victims. As noted above, experience in Kosovo
has also shown that cluster bombs can pose special problems and
dangers for the clearance professionals as well.
Explosive Remnants of War and the Convention
on Conventional Weapons
One avenue for attempting to deal
with the humanitarian problems associated with use of cluster
bombs is the 1980 Convention on Prohibitions or Restrictions on
the Use of Certain Conventional Weapons Which May Be Deemed To
Be Excessively Injurious or To Have Indiscriminate Effects, also
known as the Convention on Conventional Weapons (CCW). The CCW
has four Protocols (and one amended protocol) that contain restrictions
on the use of landmines, blinding lasers, incendiary weapons,
and weapons causing injury by non-detectable fragments. The U.S.
is a party to the CCW and all of its protocols._ _The second review
conference of the CCW will be held in Geneva in December 2001,
and a series of preparatory meetings have been held during the
past year. Governments have had extensive discussions on what
is being called the "Explosive Remnants of War" issue,
growing out of an initiative by the International Committee of
the Red Cross. The initiative began largely in response to the
increased sensitivity of the international community to the use
of cluster bombs. It is aimed at reducing the number of explosive
remnants of war (whether caused by cluster bombs or other weapons)
and establishing responsibility for clearing the detritus of war.
Different governments are embracing the issue with widely varying
degrees of enthusiasm, but it appears that a likely outcome of
the December review conference will be the establishment of a
working group with a mandate to do further work on the issue.
The United States and all other combatants
should immediately halt the use of cluster bombs in Afghanistan.
At a minimum, cluster bombs should not be used in or near populated
The U.S. and others should in a timely
manner provide relevant information regarding types and locations
of ordnance used to appropriate clearance organizations, as requested
by United Nations officials.
There should be a global moratorium on
use of cluster bombs until effective measures can be put in place
to lessen their impact on civilian populations. Governments should
examine possible technical solutions, as well as options related
to use and targeting; they should research and analyze past use
of cluster bombs, including military usefulness, civilian impact,
safety and overall effects; and, they should conduct a legal review
of cluster bombs and their consistency with international humanitarian
Governments should urgently address the
cluster bomb problem as part of the review process of the Convention
on Conventional Weapons. At the December 2001 CCW Review Conference,
an expert group should be formed to look at the Explosive Remnants
of War issue, with a focus on problems caused by cluster bombs
and submunitions. The group's mandate should be broad, and allow
for consideration not just of technical factors, but also those
related to use and targeting. The group should aim to conclude
its work in no more than one year. This work should pave the way
for immediate negotiations aimed at a new protocol to the CCW
to be concluded in a similar period of time.
Human Rights Watch believes that any long-term
approach to ameliorating the negative humanitarian impact of cluster
bombs must include the following:
Reduction of the initial failure rate
of cluster bombs to a tolerable level from a humanitarian perspective.
That level should be determined by humanitarian and military experts,
but should certainly be less than 1 percent.
A prohibition on use in or near populated
or urban areas.
A requirement for accurate recording and
mapping of cluster bomb use to assist in post-conflict clearance
and awareness efforts.
Post-use requirements such as marking,
warnings to civilians, clearance, and timely sharing of relevant
information with appropriate clearance and awareness organizations
In examining possible solutions to the
cluster bomb problem, there needs to be:
A realistic assessment of the initial
failure rate of existing cluster bombs, and a determination of
whether cluster bombs have a higher initial failure rate than
other weapons that become unexploded ordnance, and whether unexploded
cluster bomblets are more volatile or dangerous than other UXO.
An examination of options to improve the
reliability and safety of cluster bombs, and of the factors that
affect cluster bomb reliability and safety (including fuzes and
secondary fuzes, number of bomblets, area coverage, dispersal
pattern, method and circumstances of delivery, special features
like incendiary rings, characteristics of the target area, age
An assessment of the feasibility and effectiveness
of putting self-destruct, self-neutralizing and/or self-deactivating
mechanisms on all cluster bombs, both new production and existing
An assessment of the feasibility and effectiveness of various
other potential restrictions on use aimed at avoiding collateral
damage, including target limitations and high-altitude delivery.
Landmine & Cluster Bomb watch