Drones Gone Wild!

The recent vaporization of Taliban leader Mullah Mansur brings up a quiz or two...

Mansur’s potential death provides a real-world, real-time ability to test two hypotheses about the policy of killing terrorist leaders. These are based upon the objectives of the strike, according to the Pentagon press release, as well as subsequent statements by President Barack Obama and Secretary of State John Kerry.

Hypothesis one: Mansur’s death will reduce Taliban attacks and fatalities against Afghanistan national security forces, U.S. and coalition troops, and Afghan civilians.

Hypothesis two: Mansur’s replacement will be more likely to participate in the long-stalled peace and reconciliation negotiations with the Afghan government.

There has been a tremendous amount of social science research on these challenging policy puzzles. These policy-evaluative publications have reached somewhat conflicting conclusions, and are often contested by U.S. military and intelligence staffers who I speak with. However, those staffers never publish their research findings for public scrutiny, and are unable—given they would be referring to classified information—to clearly articulate their problems with the existing research.

On whether killing terrorists leaders and lower-level militants reduces violence, Max Abrahms and Phillip Potter assessed that when leaders of militant groups are killed or targeted, lower-level members have to assume tactical responsibility, and they increase the proportion of the group’s violence against civilian targets. Patrick Johnston and Anoop Sarbahi determined, “We find no statistically significant evidence of a positive relationship between drone strikes and terrorism.”  Meanwhile, Vincent Bauer, Keven Ruby, and Robert Pape found that “drone strikes are only marginally effective at reducing militant violence in the short term, and that the effect dissipates over time.”

On leadership targeting and the strength and durability of terrorist groups: In 2009, Jenna Jordan examined 298 leadership targeting incidents from 1945 through 2004, and concluded that “decapitation is not an effective counterterrorism strategy,” and oftentimes prolongs the life of a terrorist group. On the other hand, Bryan C. Price concluded, by analyzing the effect of leadership decapitation on 207 terrorist groups from 1970 to 2008, the killing or capturing leaders significantly increases the mortality rate of the group. In 2014, Jordan reviewed the impact of 109 attacks on Al Qaeda leadership from 2001 to 2011, and did not find a “significant degradation of organizational capacity or a marked disruption in al-Qaida’s activities,” measured in the number of attacks and their lethality.

There is also a CIA “Best Practices in Counterinsurgency” report from July 2009 examined nine cases of high-value targeting and found that five failed outright, two succeeded, and two had mixed results. The report specifically warned: “The Taliban’s military structure blends a top-down command system with an egalitarian Afghan tribal structure that rules by consensus, making the group more able to withstand HVT operations, according to clandestine and U.S. military reporting.”

How might someone determine if hypothesis one has been achieved? United Nations Assistance Mission in Afghanistan (UNAMA) reports on the protection of civilians, which have been produced since 2007. Fortunately, for determining causal effect, UNAMA began releasing its reports bi-annually in 2009, and just last month releasing them quarterly. So, when the third quarter UNAMA report is released in November, look for an increase or decrease in attacks by “anti-government elements,” meaning the Taliban. (It is worth noting that Taliban attacks are decreasing relative to other perpetrators: In 2015, the group was responsible for 62 percent of all civilian fatalities, a decrease from 78 percent in 2013.)

There is also the Global Terrorism Database, which produces its excellent summary of terrorist attacks for all countries by date, perpetrator group, fatalities or casualties, and target type. The 2016 data for Afghanistan will probably be posted online sometime in mid-2017.

There has been no new data for total attacks on U.S. or coalition forces since 2013, but U.S. troop fatalities are constantly updated at the Pentagon’s casualty status website, and military contractors working for the Department of Defense at a Department of Labor website. As for Afghanistan security forces, the Ministries of Defense and Interior apparently prepare an annual total of military fatalities, which has previously been provided to western journalists.

How might one determine if hypothesis two has been achieved? This simply requires determining if Mansur’s replacement, or a council of recognized Taliban leaders, decide to negotiate directly and faithfully with the government of Afghanistan. One member of the government-appointed High Peace Council stated, “Mansour’s death doesn’t necessarily mean that peace is closer than it was yesterday.”

We will soon find out if this is true, and if targeting Taliban leadership succeeds at achieving the objectives as articulated by the administration.