After the Pentagon scuttled a longstanding pledge to destroy its existing cluster-munitions stockpile, the Army is moving ahead with renewed vigor to acquire at least three new foreign-made versions of the weapons for its artillery. Late last year, the Trump administration canceled a Defense Department policy that limited the military’s ability to use cluster munitions, which, at a conference on Friday in Arlington, Va., Deputy Defense Secretary Patrick Shanahan attributed specifically to the threat posed by North Korea.
Cluster munitions are a type of weapon that distributes smaller weapons, called submunitions or bomblets, over a targeted area. They have been condemned by lawmakers and arms-control groups for causing indiscriminate harm to civilians even decades after conflicts end. The now-abandoned policy, drafted in 2008 under Robert Gates, the defense secretary at the time, required any submunitions used after 2018 to have a dud rate, or the percentage of submunitions that don’t detonate when they are supposed to, of 1 percent or less — a standard the Pentagon appeared unable to meet, even a decade after the policy was put in place. The 2017 policy change stated that newly produced cluster munitions must have a dud rate of 1 percent or less, but left open the use of older cluster munitions with higher dud rates, which allowed the United States to maintain its large cluster-munitions stockpile. The Army has since ramped up its efforts to seek newer cluster-munition models.
According to military officials, the new generation of weapons the Pentagon is looking to buy is designed to selectively destroy armored vehicles and either self-destruct or deactivate itself if it does not find an appropriate target — making it theoretically less likely than older versions of the weapon to harm civilians. But bomb technicians, who are trained to always treat an unexploded submunition as armed and capable of exploding, say such safety features are inherently unreliable, because mechanical and electronic systems built into the submunitions can and will fail.
The United States military has used submunitions in many conflict zones, including Vietnam, Laos, the Persian Gulf and the former Yugoslavia. They were also used early on in the wars in Afghanistan and Iraq, as well as in naval missile strikes in Yemen. Many of these weapons failed to explode and littered the ground with “de facto land mines,” as one Army field manual described them, that remained dangerous long after those conflicts ended. The weapons have also harmed American forces in almost every war they have used them in. In 1991, during the first Persian Gulf war, the Army fired more than 17,000 cluster-munition rockets into Iraq and Kuwait, each containing 644 submunitions. American forces also dropped tens of thousands more cluster bombs on and fired tens of thousands of submunition artillery rounds at Iraqi forces. In 2002, the Government Accountability Office reported that 22 American service members were killed and 58 were wounded during and after Desert Storm by unexploded cluster bomblets dropped by the United States. “The department determined that there are specific deliberate targeting situations in which the lives of American service members can be saved through the use of cluster munitions while mitigating risk to civilian populations,” Lt. Col. Joseph Buccino, a Defense Department spokesman, said in a statement to The Times regarding the potential for similar instances of fratricide with new systems. “The department will continue to invest in technology which reduces the risk of collateral damage.”
In late October, a spokeswoman at Picatinny Arsenal, home to the Army’s weapons and ammunition development, confirmed that the service plans to buy 3,100 Swedish-made Bonus 155-millimeter artillery projectiles from BAE Systems. It is also reviewing a similar projectile called SMArt 155, a German munition marketed by General Dynamics, according to a representative of the company and an announcement on the Army’s website. The two weapons are similar in form and function. They are fired from a howitzer toward concentrations of military vehicles and eject into the air two submunitions that use sensors to locate vehicles below. If the sensors select a target, the submunitions then explode, sending a slug of metal, known as an explosively formed penetrator, or E.F.P., down onto a targeted vehicle.
Army budgeting documents indicate that the service plans to test a new antipersonnel cluster weapon called the M999, which is offered by IMI Systems, an Israeli company. At a recent arms-trade exposition, an official at the company told The New York Times that the shell contains nine submunitions, each roughly the shape and size of a soda can and each with a fuse that sends out radio waves to measure how far it is above the ground and to tell the fuse when to detonate. The IMI Systems representative also said the submunitions contain self-destruct features.
Those features are potentially important to the Pentagon, which has faced criticism for the high dud rates of many of its submunitions and for its refusal to join an international treaty that more than 100 other nations have signed to ban the manufacture, stockpiling and use of weapons that cannot detect and destroy a single target, contain more than nine submunitions and do not have a self-destruct feature. The Army declined to release data on the dud rates of the weapons it is pursuing and also declined to say how much it plans to spend on the Bonus rounds. It did say that the Bonus munitions are compliant with the Defense Department’s policy, but it will not perform any additional tests to verify that.
Recent arms-development history provides a basis for skepticism on both dud rates and costs. In the 1980s and 1990s, the Army developed the Search and Destroy Armor Munition, or SADARM, which ejected two E.F.P. bomblets that searched for targets. In 2001, after spending $266 million and building only 348 serviceable rounds of the 47,000 that it had budgeted for, the Army canceled the program because of the weapon’s poor reliability. The Pentagon had also made bold claims about the reliability of its airdropped CBU-97 Sensor Fuzed Weapon, each of which contains 40 submunitions; in 2006, it claimed that 99 percent of its submunitions would either find their targets and explode or would self-destruct. But photos emerged on social media from Yemen in 2015 of dud submunitions that had been provided by the United States and dropped by Saudi warplanes, suggesting that the Air Force’s claim did not hold up in combat use. The manufacturer, Textron Systems, announced in 2016 that it would no longer make the weapons. SADARM and CBU-97 each had “sensor fused” submunitions that are very similar to those used in the Bonus 155-millimeter artillery projectiles recently bought by the Army and in the SMArt 155 weapons that the Army is evaluating.
One arms-trade analyst and proponent of the cluster-munition ban, Jeff Abramson of the Landmine and Cluster Munition Monitor, said the Pentagon should not field any new submunitions at all. “Military planners can envision nearly any scenario, no matter how unlikely, to find a rationale for using or developing a weapon,” he said. “The global community has agreed that indiscriminate weapons such as land mines and cluster munitions should never be used.”