Do U.S. Drone Strikes Cause Blowback? Do U.S

Do U.S. Drone Strikes Cause Blowback? Do U.S. Drone Strikes Aqil Shah Cause Blowback? Evidence from Pakistan and Beyond

Since the terrorist attacks of September 11, 2001, the United States has relied almost exclusively on armed unmanned aerial vehicles, or drones, to ªght Islamist militants on un- conventional battleªelds in countries including Pakistan, Somalia, and Yemen. Critics, including analysts, ofªcials from human rights organizations, and former U.S. government and military ofªcials, claim that drone strikes create blowback: rather than reducing the terrorist threat, drone strikes increase it by providing terrorist groups with fresh recruits. For example, Human Rights Watch has warned that “the deaths of numerous civilians” in drone strikes in Yemen have “fueled public anger and frustration...against the United States, handing a recruiting card to Al-Qaeda in the Arabian Peninsula (AQAP).”1 Similarly, retired four-star Marine Corps General and former Vice Chairman of the Joint Chiefs of Staff James Cartwright cautions that “drones cause anger, bitterness and resentment among Muslim populations targeted in the attacks, and will cause blowback against the United States.”2 According to proponents of this thesis, blowback can occur in three theaters: (1) locally, where family members living in the same target zone as a relative killed in a drone strike become revenge-seeking, anti-American militants; (2) nationally, in areas of a country not exposed to drone attacks but where people are motivated to politically or ªnancially support or join militant groups because they view such strikes both as humiliating violations of their country’s sovereignty and as indiscriminate killers of their fellow citizens (in

Aqil Shah is Assistant Professor in the Department of International and Area Studies at the University of Oklahoma.

For insightful comments and suggestions on earlier drafts, the author thanks Bushra Asif, Tricia Bacon, C. Christine Fair, Michael Horowitz, Patrick Johnston, Asfandyar Mir, Jack Serle, participants at the University of Oklahoma’s Middle East Seminar and the International Relations Semi- nar at the University of Texas–Austin, conference participants at the 2017 annual meeting of the International Studies Association, and the anonymous reviewers. Funding from the University of Oklahoma’s Vice President for Research and the College of International Studies is gratefully acknowledged.

1. Human Rights Watch, “Between a Drone and al-Qaeda: The Civilian Cost of U.S. Targeted Kill- ings in Yemen” (New York: Human Right Watch, October 22, 2013), https://www.hrw.org/report/ 2013/10/22/between-drone-and-al-qaeda/civilian-cost-us-targeted-killings-yemen. See also “The Civilian Impact of Drones: Unexamined Costs, Unanswered Questions” (New York: Center for Civilians in Conºict and Columbia Law School Human Rights Clinic, 2012). 2. Quoted in Mark Mazzetti and Scott Shane, “Inºuential Aide to Obama Voices Concern on Drone Strikes,” New York Times, March 21, 2013.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 48

some cases, drone strikes can exert additional national-level effects by motivat- ing allied governments to succumb to public pressure and curtail or sever cooperation with U.S. counterterrorism efforts); and (3), transnationally, when Muslims in other parts of the world are radicalized and/or participate in terrorist activities in response to co-religionists dying in drone strikes. Although scholars have extensively debated the military effectiveness,3 legal status, and ethics of drone warfare,4 there has been no systematic study of drone blowback beyond statements from some important public ªgures, anecdotes about individual terrorists,5 advocacy-driven research, and media commentary.6 In this article, I examine the conventional wisdom on drone blowback and ªnd that most critics of drone warfare assume, rather than demonstrate, the occurrence of blowback, typically pointing to a speciªc theater of supposed blowback or lumping together different theaters, often without specifying the causal mechanisms that connect drone strikes to blowback in each case. The purpose of this article is not to deny the harmful effects of U.S. drone strikes on civilian populations or to ignore that the use of drones in sovereign countries not at war with the United States is a manifestation of American unilateralism—even from the perspective of many U.S. allies. This article uses new data and secondary sources to evaluate variants of the blowback thesis. At the local and national levels, it analyzes the case of Pakistan, where since 2004 the Central Intelligence Agency (CIA) has launched an estimated 429 drone strikes (roughly 75 percent of its known total strikes worldwide).7 Using a diverse convenience sample of interviews with 167 well- informed adults from North Waziristan Agency (NWA), the most heavily tar-

3. See, for instance, Daniel Byman, “Why Drones Work: The Case for America’s Weapon of Choice,” Foreign Affairs, Vol. 92, No. 4 (July/August, 2013), pp. 32–43; David A. Jaeger and Zahra Siddique, “Are Drone Strikes Effective in Afghanistan and Pakistan? On the Dynamics of Violence between the United States and the Taliban” (Bonn, Germany: Forschungsinstitut zur Zukunft der Arbeit, December 2011), http://ftp.iza.org/dp6262.pdf; and Patrick B. Johnston and Anoop K. Sarbahi, “Impact of U.S. Drone Strikes on Terrorism in Pakistan,” International Studies Quarterly, Vol. 60, No. 2 (2016), pp. 203–219, doi:10.1093/isq/sqv004. For the opposing argument, see Audrey Kurth Cronin, “Why Drones Fail: When Tactics Drive Strategy,” Foreign Affairs, Vol. 92, No. 4 (July/August 2013), pp. 44–54; and Megan Smith and James Igoe Walsh, “Do Drone Strikes De- grade Al Qaeda? Evidence from Propaganda Output,” Terrorism and Political Violence, Vol. 25, No. 2 (2013), pp. 311–327, doi:10.1080/09546553.2012.664011. 4. For an accessible treatment of these issues, see Avery Plaw, Matthew S. Fricker, and Carlos R. Colon, The Drone Debate: A Primer on the U.S. Use of Unmanned Aircraft Outside Conventional Bat- tleªelds (Lanham, Md.: Rowman and Littleªeld, 2016); and John Kaag and Sarah Kreps, Drone War- fare (London: Polity, 2014). 5. One such example, discussed in a subsequent section of this article, is the Pakistani-American terrorist Faisal Shehzad, who tried to bomb Times Square in May 2010. 6. One major constraint on examining the link between drone strikes and terrorist recruitment is the dearth of reliable data on militant recruitment. 7. The Pakistan drone program is different from those of some other countries where the United States operates lethal drones. For example, the Pakistan program is run primarily by the Central

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 49

geted district located in Pakistan’s Federally Administered Tribal Areas (FATA), bordering Afghanistan; extensive interviews with respected experts on terrorism; and an ofªcial Pakistani police survey of 500 detained terrorists from southern Sindh Province, I ªnd no evidence of a signiªcant impact of drone strikes on the recruitment of militants either locally or nationally. Instead, my data and secondary sources suggest that militant recruitment is a complex process driven by a variety of factors, some of which scholars and analysts have already identiªed. These include political grievances,8 the Pakistani state’s sponsorship of militancy as a tool of foreign policy,9 state re- pression,10 weak governance,11 and coercive recruitment by militant groups.12 Additionally, my examination of the trial testimony and accounts of terrorists convicted in the United States, together with social science scholarship on Muslim radicalization in the United States and Europe, offers little or no evidence that drone strikes create a systematic pattern of transnational blowback. Although jihadists typically explain their actions as a response to U.S. military interventions in Muslim countries, the main causes of global militant Islamism are not drone strikes but factors such as identity crises suffered by young immigrants, the nature of state integration policies, social networks, and online exposure to extremist ideologies. The article proceeds as follows. In the next three sections, I explicate the the- oretical logic of the local, national, and transnational versions of the blowback theses; discuss the methodology employed to asses each one; and use my interview ªndings and other data to evaluate the blowback hypothesis. I conclude with a discussion of the implications of my ªndings for scholarship and policy.

Local Blowback

Context is essential when examining the blowback thesis. The United States has targeted its lethal drone operations primarily in Pakistan’s northwestern Federally Administered Tribal Areas, a quasi-autonomous frontier region gov-

Intelligence Agency rather than the U.S. military (e.g., Afghanistan) or jointly by the two organizations (e.g., Yemen). 8. Ted Robert Gurr, Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970). 9. See, for example, Hussain Haqqani, Pakistan: Between Mosque and Military (Washington, D.C.: Carnegie Endowment for International Peace, 2005). 10. Jeff Goodwin, No Other Way Out: State and Revolutionary Movements, 1945–1991 (New York: Cambridge University Press, 2001). 11. See International Crisis Group, “Pakistan: Countering Militancy in FATA,” Asia Report No. 178 (Brussels: International Crisis Group, October 21, 2009). 12. Kristine Eck, “Coercion in Rebel Recruitment,” Security Studies, Vol. 23, No. 2 (2014), pp. 364– 398, doi: 10.1080/09636412.2014.905368. See also International Crisis Group, “Pakistan.”

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 50

erned under the 1901 Frontier Crimes Regulation (FCR). There, the Taliban, al-Qaida, and other foreign militant groups found sanctuary after the U.S. invasion of Afghanistan in October 2001.13 FATA has an estimated population of 5 million, most of whom are ethnic Pashtuns. Several of its seven agencies (districts)—in particular, South Waziristan (SWA) and North Waziristan (NWA)—14 have become sanctuaries, training bases, and launching pads for militant attacks in both Afghanistan and Pakistan. Pakistan deployed its military for the ªrst time in FATA in 2002–03 under pressure from the United States to prevent al-Qaida and Taliban militants from mounting cross-border attacks against U.S.-led coalition troops in Afghanistan. Pakistan deployed its military in FATA in 2002–03 under pressure from the United States to prevent al-Qaida and Taliban militants from mounting cross-border attacks against U.S.-led coalition troops in Afghanistan. In 2004, the military launched anti-terrorism operations in SWA. After suffering major losses at the hands of local militants and their foreign allies, the military opted for a policy of appeasement, entering into peace accords with the militants, including one signed in April 2004 in SWA with militant commander Nek Mohammad, who was subsequently killed in a CIA drone strike in June 2004. The short-lived agreement offered the militants amnesty and money in return for a pledge to renounce violence. In NWA, the military sporadically clashed with different militant factions in 2005–06, which ultimately resulted in a similar truce in September 2006 with Taliban commander Haªz Gul Bahadur. The overall effect of these peace deals was to cede more space to the militants and to enhance their political status, which helped them gradually expand their inºuence over the region and establish a parallel administration and court system modeled on Taliban rule in Afghanistan.15 Initially, U.S. drone strikes focused on South Waziristan, the home base of Baitullah Mehsud, head of the Pakistani Taliban, and his al-Qaida-afªliated foreign allies. After the Pakistan Army launched a major offensive in SWA in 2009, the locus of militancy and drone strikes shifted to North Waziristan. In addition to al-Qaida, drone strikes have targeted two main militant groups: the Tehrik-i-Taliban Pakistan (TTP), an umbrella group created by Mehsud in

13. This ªgure includes the strike that killed Afghan Taliban chief Mullah Mansur Akhtar in Paki- stan’s southwestern Balochistan Province in June 2016. 14. It also includes frontier regions or tribal areas adjoining Peshawar, Kohat, Bannu, and Dera Ismail Khan districts of the Khyber Pakhtunkhwa Province. 15. Intikhab Amir, “Waziristan: Whose Writ Is It Anyway?” Herald (Pakistan), March 2006; “Taliban Slap Taxes in Miranshah,” Dawn (Pakistan), October 23, 2006; “Mujahideen Shura Re- solves to Enforce Sharia,” News (Pakistan), November 6, 2006; and International Crisis Group, “Pakistan’s Tribal Areas: Appeasing the Militants,” Asia Report No. 125 (Brussels: International Crisis Group, December 11, 2006).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 51

2007 that has targeted the Pakistani state; and the Haqqani network, a group aligned with the Afghan Taliban that has fought U.S.-led NATO and Afghan security forces across the border.16 According to the existing literature, lethal drone strikes provide militant groups with a steady stream of new recruits seeking revenge for the deaths of relatives.17 The operating mechanism here is familial or tribal revenge. As Audrey Kurth Cronin writes, “Drones are killing operatives who aspire to attack the United States....Butthey are also increasing the likelihood of attacks over the long term by embittering locals and cultivating a desire for ven- geance.”18 According to two other prominent experts, David Kilcullen and Andrew Exum, drones have a negative net effect on terrorism. While violent extremists may be unpopular, for a frightened population they seem less ominous than a faceless enemy that wages war from afar and often kills more civilians than militants. Press reports suggest that over the last three years drone strikes have killed about 14 terrorist leaders. But according to Pakistani sources, they have also killed some 700 civilians. That is 50 civilians for every militant killed, a hit rate of 2 percent—hardly “precision.” American ofªcials vehemently dispute these ªgures....Nevertheless, every one of these dead noncombatants represents an alienated family, a new desire for revenge, and more recruits for a militant movement that has grown exponentially even as drone strikes have increased.19 Although some of those who support terrorist groups or become terrorists themselves may be motivated by the desire to seek revenge for loved ones killed in drone strikes, neither Kilcullen and Exum nor other proponents of the blowback thesis provide empirical data to support their arguments. Indeed,

16. On the Pakistani Taliban, see Shehzad H. Qazi, “Rebels of the Frontier: Origins, Organization, and Recruitment of the Pakistani Taliban,” Small Wars & Insurgencies, Vol. 22, No. 4 (2011), pp. 574– 602, doi:10.1080/09592318.2011.601865; and Hasan Abbas, “A Proªle of the Tehrik-i-Taliban Paki- stan,” CTC Sentinel, Vol. 1, No. 2 (January 2008), pp. 1–3, http://www.dtic.mil/dtic/tr/fulltext/ u2/a539361.pdf. On the Haqqani network, see Vahid Brown and Don Rassler, Fountainhead of Jihad: The Haqqani Nexus, 1973–2012 (Oxford: Oxford University Press, 2013). 17. See, for instance, Brian Glyn Williams, Predators: The CIA’s Drone War on al-Qaida (Washington, D.C.: Potomac, 2013), pp. 207–217; Michael J. Boyle, “Costs and Consequences of Drone Warfare,” International Affairs, Vol. 89, No. 1 (January 2013), pp. 1–29, doi:10.1111/1468-2346.12002; and Da- vid Kilcullen and Andrew McDonald Exum, “Death from Above, Outrage Below,” New York Times, May 16, 2009. 18. Cronin, “Why Drones Fail.” See also Stanley McChrystal, “Special Address to the International Institute for Strategic Studies,” London, England, October 1, 2009, p. 2, https://www.iwp.edu/ docLib/20131125_GenMcCrystalAddress1012009.pdf. 19. Kilcullen and Exum, “Death from Above, Outrage Below.” Kilcullen also made this claim in David Kilcullen, The Accidental Guerrilla: Fighting Small Wars in the Midst of a Big One (New York: Oxford University Press, 2009), pp. 230–232. For a similar argument, see Leila Hudson, Colin S. Owens, and Matt Flannes, “Drone Warfare: Blowback from the New American Way of War,” Middle East Policy, Vol. 18, No. 2 (Fall 2011), http://www.mepc.org/journal/middle-east-policy- archives/drone-warfare-blowback-new-american-way-war.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 52

Kilcullen and Exum based their claims about blowback on dubious ofªcial Pakistani ªgures that appeared in a local press report.20 But even if one accepts Kilcullen and Exum’s hit rate as valid, it demonstrates only that drones kill more civilians than terrorists, not that they create more terrorists than they kill. John Feffer has argued that blowback would exist even if only 0.1 percent of FATA’s 4 million residents joined the Taliban out of anger over drone strikes.21 Feffer is right to count even the smallest number of drone-spawned militants as blowback, but in the absence of actual data, his argument is mere speculation. As mentioned earlier, studies by human rights advocacy groups also argue that lethal drone strikes create militants. These include a widely cited report conducted by the Law School Clinics of Stanford and New York Universities.22 The report’s ªndings, based on eyewitness testimony, are undermined, however, by a serious conºict of interest: the London-based charity, Reprieve, known for its anti-drone advocacy, commissioned the report. Moreover, the study’s participants were paid for their travel to Islamabad and Peshawar to sit for interviews by Reprieve and its Pakistani partner, the Foundation for Fundamental Rights, which the report describes as “the most prominent legal advocate for drone victims in Pakistan.”23 In addition, similar to other advocacy research,24 the report paints a biased picture because it bases its conclu- sions on three drone strikes that reportedly killed a larger number of civilians than a typical drone strike. The report also suffers from other empirical and methodological problems.25 Admittedly, the human costs of drone strikes should not be minimized, and documenting these costs is both necessary and desirable. Nevertheless, the charge that drones cause excessive casualties, which has a radicalizing effect, does not withstand scrutiny. In fact, drones ap-

20. Amir Mir, “60 Drone Hits Kill 14 Al-Qaeda Men, 687 Civilians,” News (Pakistan), April 10, 2009. 21. John Feffer, “The Coming Drone Blowback,” Foreign Policy in Focus, May 25, 2016, http:// fpif.org/coming-drone-blowback/. 22. See Stanford Law School and New York University (NYU) School of Law, Living under Drones: Death, Injury, and Trauma to Civilians from U.S. Drone Practices in Pakistan (Stanford, Calif. and New York: Stanford Law School and NYU School of Law, 2012), pp. 133–135. 23. Ibid., p. 3. 24. For instance, Amnesty International based its investigation of U.S. drone strikes in Pakistan on 9 of the 45 strikes that occurred in NWA between January 2012 and August 2013. See Amnesty In- ternational, “Will I Be Next?’ U.S. Drone Strikes in Pakistan” (London: Amnesty International, Oc- tober 22, 2013), https://www.amnestyusa.org/ªles/asa330132013en.pdf. 25. See C. Christine Fair, “Ethical and Methodological Issues in Assessing Drones’ Civilian Impacts in Pakistan,” Monkey Cage blog, Washington Post, October 6, 2014, https://www .washingtonpost.com/news/monkey-cage/wp/2014/10/06/ethical-and-methodological-issues- in-assessing-drones-civilian-impacts-in-pakistan/?utm_termϭ.706209dbd5f6.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 53

pear to cause unintended deaths at a rate lower than their alternatives, such as Pakistani military campaigns involving F-16 jets.26 Some analysts have cited opinion polls to substantiate their claims about local blowback. For example, a 2010 poll based on a sample of 1,000 residents in all seven FATA agencies showed that 76 percent of respondents opposed drone strikes; only 16 percent thought that such strikes accurately target insurgents; and 48 percent believed that the strikes largely kill civilians.27 Moreover, 60 percent believed that suicide attacks against the United States were “sometimes” or “always” justiªed. Peter Bergen and his collaborators, who commissioned the survey, have used these ªgures to claim that suicide attackers are widely popular across FATA, and that the main motivation for anti-American radicalization “stems from anger at CIA-directed drone strikes at militants living in the area.”28 Flaws in the poll, however, undermine its ªndings. The ªrst ºaw is the presence of a “social desirability bias” (i.e., responses seen as so- cially acceptable or safe), which may be enhanced in conºict situations where respondents fear reprisals from the government or insurgents. In the words of Nizam Dawar, head of a network of local development organizations in FATA, “There was no space to do independent research in FATA at the time. My own family in North Waziristan was threatened by militants after I spoke about drone attacks to a foreign media outlet in 2010. Survival required a policy of see no evil, hear no evil, say no evil.”29 Second, the polling sample is not representative of the population. In 2010, a sizable portion of FATA’s population was internally displaced. For instance, one-third of SWA’s estimated total population of 600,000 had left the area prior to the Pakistani military’s October 2009 offensive.30

methods for assessing local blowback To assess the occurrence of local blowback, I relied primarily on interviews with inhabitants of North Waziristan. NWA offers distinct advantages in this

26. See Avery Plaw and Matthew S. Fricker, “Tracking the Predators: Evaluating the U.S. Drone Campaign in Pakistan,” International Studies Perspectives, Vol. 13, No. 4 (November 2012), pp. 351– 354, doi:10.1111/j.1528-3585.2012.00465.x. 27. New America Foundation and Terror Free Tomorrow, “Public Opinion in Pakistan’s Tribal Re- gions” (Washington, D.C.: New America Foundation and Terror Free Tomorrow, September 2010). 28. Ken Ballen, Peter Bergen, and Patrick Doherty, “U.S.-Led Drone War Is Self-Defeating,” CNN, September, 10, 2010, http://www.cnn.com/2010/OPINION/09/30/ballen.bergen.pakistan/index .html. 29. Author telephone interview, January 2017. 30. United Nations Ofªce for the Coordination of Humanitarian Affairs, “Pakistan—South Waziristan Displacement,” situation report No. 4 (New York: United Nations Ofªce for the Coor- dination of Humanitarian Affairs, November 19, 2009) https://reliefweb.int/sites/reliefweb.int/ ªles/resources/655DAC32B6BBE197C1257674003AD402-Full_Report.pdf.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 54

regard. As mentioned earlier, 322 of the CIA’s 429 drone strikes in Pakistan (75 percent) have occurred in the agency since the ªrst recorded strike in SWA in 2004. The CIA’s drone campaign heavily targeted several subdistricts that had become hubs of militant activity, including Datta Khel, Mir Ali, and Miranshah. Therefore, the probability of ªnding evidence of local blowback should be higher in these locales than in less frequently targeted or nontargeted FATA districts, such as Mohmand or Orakzai. NWA thus consti- tutes a “most-likely case” that must closely conform to the predictions of a theory if one is to have any conªdence in its validity.31 At the same time, state and armed nonstate actors constrain the ability to do research in FATA. Ironically, my access to research subjects was made possible by the Pakistan military’s June 2014 offensive against the Taliban in NWA, which displaced virtually all its estimated 400,000 residents to the nearby districts of Bannu, Hangu, and Peshawar, as well as to Afghanistan. Although I conducted my interviews outside NWA,32 building trust was still essential for convincing participants to discuss a topic as sensitive as drones.33 Because random sampling was impossible under the circumstances, I used a non-random snowball sample to build a diverse population of interviewees. Snowball sampling is a method in which initial contacts provide additional contacts.34 This process proved useful because referrals from a trusted or reliable source increased the likelihood that others would speak with me.35 The sample (n ϭ 167) includes legislators, masharan (elders), maliks (head- men), clerics, local ofªcials, reporters, lawyers, traders, shop owners, human rights activists, teachers, university students, and local leaders and activists from seven political parties—including the Islamist Jamiat Ulema-e-Islam and Jamaat-e-Islami—covering the full spectrum of party ideologies in Pakistan. Admittedly, snowball sampling carries a risk of selection bias, because respon-

31. Harry Eckstein, “Case Study and Theory in Political Science,” in Fred I. Greenstein and Nelson S. Polsby, eds., Handbook of Political Science, Vol. 7: Strategies of Inquiry (Reading, Mass.: Addison Wesley, 1975), pp. 79–137; and John Gerring, “Is There a (Viable) Crucial-Case Method?” Compara- tive Political Studies, Vol. 40, No. 3 (March 2007), pp. 231–253. 32. I conducted these interviews during two ªeld trips to Pakistan, in May–July 2015 and Decem- ber 2015–January 2016. 33. Most survey respondents requested anonymity for reasons of personal safety. 34. On snowball sampling, see David L. Morgan, “Snowball Sampling,” in Lisa M. Given, ed., The SAGE Encyclopedia of Qualitative Research Methods, Vol. 2 (London: Sage 2008), pp. 816–817. 35. Being a native of northwestern Pakistan, I was relatively well positioned to conduct in-depth interviews in the respondents’ native Pashtu language, which was also crucial for establishing trust and reliably interpreting cultural cues and local idioms. On the advantages and disadvan- tages of being an indigenous researcher, see Jenny Wüstenberg, “When the Field Is Home: Con- ducting Research in One’s Country of Origins,” APSA-CP Newsletter, Vol. 19, No. 2 (Summer 2008), pp. 19–23. See also Robert G. Burgess, In the Field: An Introduction to Field Research (New York: Routledge, 2004).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 55

dents are likely to recommend people with similar characteristics or similar views on the topic of the interview.36 Hence, the snowball sample may have produced additional interviewees who shared a generally favorable view of drone strikes. Although I tried to reduce this bias by choosing the initial set of respondents from an array of social and professional backgrounds, collect- ing a convenience sample free of bias is impossible. I conducted interviews using open-ended questions, such as: “What is your opinion of drone strikes in NWA?” These were followed by probes to capture respondents’ speciªcs views of and/or experiences with drones. I used thematic content analysis to analyze the interview data (see the online appendix).37 I began by carefully reading each interview transcript. Using existing works on drone blowback, I identiªed key themes, such as civilian casualties, military operations, mental distress, displacement, militant recruitment, and militancy, as initial coding categories. Then, I highlighted text that appeared to describe an opinion about these themes, as well as new themes that emerged from the interviews, such as militant brutality and property damage. Once I coded all the transcripts in accordance with these themes, I examined the data within each code to record the incidence/variance and direction (e.g., supportive/opposed) of responses, which I descriptively report as percent- ages for the whole sample. To further illustrate my ªndings, I chose quotations typical of an opinion or sentiment expressed in the respondents’ accounts. My sample is superior to existing samples for at least two reasons. First, although the sample size falls short of a quantitative sample, it has the same order of magnitude as the previously mentioned Stanford/NYU sample, but without any conºict of interest. Second, unlike most existing advocacy reports, my study does not rely on third parties, such as translators or interpreters, thus eliminating translation-related biases that can affect the reliability of the research ªndings. Although the interview sample is large compared with those of all similar studies, one caveat is in order: there is no way of knowing how representative the interviewees are of the entire population of NWA. The interviews do, however, present a different “sample point” by capturing

36. See Erik Bleich and Robert Pekkanen, “How to Report Interview Data,” in Layna Mosley, ed., Interview Research in Political Science (Ithaca, N.Y.: Cornell University Press, 2013), p. 87. See also Anthony Seldon and Joanna Pappworth, By Word of Mouth: Elite Oral History (London: Methuen, 1988). 37. See Philipp Mayring, “Qualitative Content Analysis Forum,” Qualitative Social Research, Vol. 1, No. 2 (June 2000), http://www.qualitative-research.net/index.php/fqs/article/viewArticle/ 1089/2385; Robert Philip Weber, Basic Content Analysis (London: Sage, 1990), pp. 117–124; and Mi- chael Quinn Patton, Qualitative Research and Evaluation Methods (Thousand Oaks, Calif.: Sage, 2002), pp. 452–457. The online appendix is available at doi:10.7910/DVN/AGCWUR.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 56

the voices of a diverse, informed section of a population directly affected by drone strikes but generally disregarded in the larger debate.

assessment of local blowback My interview ªndings contradict the local blowback thesis. When asked whether drone strikes create militants, 71 percent of respondents disagreed, 11 percent agreed, and 18 percent were unsure or had no opinion. I followed this question with a speciªc probe about whether respondents knew of individuals in their family, clan, or village who had become militants and whether these militants had relatives who died or were injured in drone strikes. Al- though most respondents claimed to personally know or be aware of someone in their clan or village who had been involved in militant activity or who had been indirectly linked to militants, none believed that the reason was the loss of a relative in a drone strike. The minority who agreed with the blowback thesis spoke in general terms and could not identify speciªc cases of drone- inspired militants. The opinions of those I interviewed do not deªnitively prove or disprove the existence of local blowback. But again, context matters. Virtually every family in NWA has been affected by the conºict, whether through the death of a relative, the destruction of property, or displacement resulting from a military offensive. Yet these people have largely been left out of debates about the threats they face and the impact of these threats on their lives. More- over, the inhabitants of FATA identify themselves as members of a particular Pashtun qabail or qaum (tribe) divided into khels (sub-tribes), each of which consists of extended clans or families.38 Most inhabitants are therefore enmeshed in dense social networks, which makes them uniquely informed about the effects of drone strikes on their community. My interviews conªrm anecdotal accounts describing the CIA’s intensiªed drone campaign in 2009–11 as having forced al-Qaida, Taliban, and other militants to limit their movements, curtail communications, and devote time and resources to survival rather than planning attacks—thus undermining both their operational capabilities and their ability to take advantage of the hypo- thetical stream of recruits generated by drone strikes.39 A well-connected Wazir tribal elder from Miranshah conªrms this ªnding: “Forget about new

38. “Tribe” is a problematic category typically associated with colonial ethnography. I use it here in the sense that local inhabitants use it: as a marker of their social identity and spatial location in the different agencies of FATA. 39. See Jane Perlez and Pir Zubair Shah, “Drones Batter Al Qaeda and Its Allies within Pakistan,” New York Times, April 4, 2010; and Pir Zubair Shah, “My Drone War,” Foreign Policy, February 27, 2012, http://foreignpolicy.com/2012/02/27/my-drone-war/.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 57

members. The drone strikes either killed or drove the foreign and local militants underground. They could no longer roam freely, train openly, and force or entice the locals to join them. They had to worry about their own safety.”40 Even if militants wanted to replenish their ranks, they had to worry that new recruits could be CIA or Pakistani spies.41 As David Rhode, a New York Times journalist held captive by the Haqqani network for several months in NWA in 2009, writes, drones were a “terrifying presence” for the militants, “hindered their operations,” and created “a paranoia among the Taliban” about spies in their midst. 42 A letter possibly written by Osama bin Laden acknowledges the negative impact of drones on al-Qaida’s operations: “Over the last two years, the spying aircrafts beneªted the enemy greatly and led to the killing of many jihadi cadres, leaders, and others. This is something that is concerning us and exhausting us.”43 Another letter, this one written by bin Laden’s deputy, Atiyah Abd al-Rahman, states: “We are facing difªculties due to the grave shortages in personnel in some cadres and the abundance of spies operating in our areas.”44

taliban, tanks, and drones Several studies have suggested that in addition to physical destruction, drone strikes produce mental distress, which also fuels anger against the United States and incites terrorism.45 Even though it is hard to separate the psychological impact of drone strikes from those of other forms of violence prevalent in NWA, such as Taliban suicide attacks or Pakistani airstrikes, my interviews conªrm that drones do indeed create stress and anxiety, especially among children. According to a former British Broadcasting Corporation correspondent with extensive experience reporting about militancy in FATA, the psychological effects of drones are linked to the fear that “the drones are always hovering, watching us and can strike any time.”46 But as Mohsin Dawar, a lawyer and human rights activist who hails from NWA, points out: “Local

40. Author interview with respondent 19, Peshawar, June 2015. 41. Asfandyar Mir and Dylan Moore, “Drones, Surveillance, and Violence: Evidence and Theory from a U.S. Drone Program,” University of Chicago and Stanford University, 2017. 42. David Rhode, “A Drone Strike and a Dwindling Hope,” New York Times, October 20, 2009. 43. “Summary on Situation in Afghanistan and Pakistan” (Washington, D.C.: Ofªce of the Direc- tor of National Intelligence, n.d.), https://www.dni.gov/ªles/documents/ubl/english/Summary %20on%20situation%20in%20Afghanistan%20and%20Pakistan.pdf. 44. United States of America vs. Abid Naseer, Motion in Limine to Admit Bin Laden Media Docs at the Trial, case 1:10-cr-00019-RJD, United States District Court, Eastern District of New York, Feb- ruary 15, 2015, http://kronosadvisory.com/Abid.Naseer.Trial_Abbottabad.Documents_Exhibits .403.404.405.420thru433.pdf. 45. See, for example, Stanford Law School and NYU School of Law, Living under Drones. 46. Author interview with Dilawar Wazir, Peshawar, December 2015.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 58

people were confronted with a dilemma: military offensive or drone strikes. The military solution [bombardment] was sometimes worse than the disease. Drones were noisy but necessary.” Dawar and most of my interviewees noted, however, that no evidence exists to support the claim that psychological distress from drone strikes increases support or sympathy for the Taliban, let alone produces new militant recruits.47 With regard to the destruction of property, 84 percent of my interviewees seemed to dread drone strikes less than they did the Pakistani military’s ground and aerial offensives, which have reportedly caused much more extensive damage to civilian life and property, as well as the displacement of roughly 1 million FATA residents.48 An automobile dealer from Miranshah, whose shop was destroyed during the Pakistan military’s June 2014 offensive against the Taliban, offered the following account: “Drones did not destroy our property. Drones did not displace us. Drones did not humiliate us. The Pakistan Army did. If anything, our youth should be motivated to pick up arms against them Punjabi soldiers.”49 The owner of a cement business in Miranshah’s main market who witnessed the aftermath of the offensive, com- mented: “Drones are precise in most cases. After the military comes, nothing is left. In the 2014 operation, they [the Pakistani military] destroyed everything: markets razed to the ground, houses reduced to rubble by heavy bombings. Talk to anyone from NWA and you will know why people are upset and angry.”50 These sentiments may be overrepresented in the sample, because the interviewees or their families suffered hardships related to their displacement by the Pakistani military. They are consistent, however, with the spirit of the “Peshawar Declaration,” a peace plan put forth by a broad-based coalition of political parties and civil-society groups from the tribal areas during the intense drone activity of 2009 and prior to the military operation in NWA. The declaration concludes that drones are the only counterterrorism instrument that “the people of the war-affected areas are satisªed with.”51 Further evidence can be gleaned from the FATA-based Pashtun Tahafuz Movement, which mobilized thousands of protesters in Islamabad in February 2018 to de- mand accountability for the extra-judicial killings and disappearances of

47. Email communication to author, January 24, 2018. 48. Umar Farooq, “Civilians Bear Brunt of Pakistan’s War in the Northwest,” Foreign Policy, Febru- ary 11, 2013, http://foreignpolicy.com/2013/02/11/civilians-bear-brunt-of-pakistans-war-in-the- northwest/; and International Crisis Group, “Drones: Myth and Reality in Pakistan,” Asia Report No. 247 (Brussels: International Crisis Group, May 21, 2013). 49. Author interview with respondent 61, Rawalpindi, June 2015. 50. Author interview with respondent 67, Bannu, December 2015. 51. Cited in Farhat Taj, The Taliban and the Anti-Taliban (Newcastle upon Tyne, U.K.: Cambridge Scholars Publishing, 2011), pp. 198–212.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 59

Pashtuns at the hands of the security forces. The movement’s grievances are targeted entirely against the Pakistani state, not drone strikes. According to one prominent member, Ayaz Wazir, “The most interesting aspect of protest was the complete absence of drone attacks from the discourse emanating both from the stage and the audience during the sit-in in Islamabad. The reason: drones were never a problem for FATA residents.”52 These pragmatic perspectives on the utility of drones are inextricably linked to at least two factors. The ªrst factor consists of the coercive tactics the Taliban used to terrorize and control the local population, especially after the peace deal of September 2006.53 These include taxation and harsh religious penalties for minor offenses, such as petty theft, and for violations of the ban on shaving beards, listening to music, and girls going to school. As a government contractor from the town of Mir Ali, who closely witnessed the operation of this parallel justice system, described it: “The badshakla [ugly] Taliban made our lives miserable. We had no choice but to obey them or suffer harm. We had to pay compulsory donations for their ‘jihad’ or protection money to save our liveli- hoods.”54 As part of this campaign, the militants also attacked and sought to subvert alternative centers of social authority. According to a local activist from Miranshah who was afªliated with the moderate left-of-center Pakistan People’s Party, “They killed our elders and attacked jirgas [deliberative coun- cils made up of maliks] while the government looked away. We were not safe anywhere.”55 “We suffered on many counts,” said an accountant employed in the NWA political administration who lost a relative in a suicide bombing. “We were trapped between the tanks and the Taliban, between bombardment from [ªghter] jets and Taliban terror. The Americans killed the militants for their own interests. But it restored some normalcy to our lives.”56 The second factor relates to local resentments rooted in a sense of betrayal by the Pakistani government. Whereas the rest of Pakistan is governed under the 1973 constitution enshrining Pakistanis’ fundamental rights, FATA is still governed under the 1901 Frontier Crimes Regulation, designed by British colonial authorities. Under the FCR, a government ofªcial known as the “political agent” administers each agency through maliks, who depend on monthly government allowances and enjoy virtually unlimited executive, ªnancial, and magisterial powers. Although jirgas can resolve local disputes, the political au-

52. Email communication to author, February 10, 2018. 53. Zeeshan Haider, “Taliban-Style Militants Roam in North Waziristan after Pact,” Reuters, Janu- ary 18, 2007. 54. Author interview with respondent 89, Bannu, January 2016. 55. Author interview with respondent 72, Hangu, June 2015. 56. Author interview with respondent 154, Bannu, December 2015.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 60

thority can overturn their decisions. Moreover, the FCR prescribes harsh penalties for crimes without the right of appeal in a court of law,57 including collective punishments for offenses committed by one or more members of a tribe or those committed by anyone in its area. During its operations, the Pakistani military has used collective punishments, including razing of resi- dential and business properties, economic blockades, and asset seizures, to punish families or clans whose members they suspected of harboring foreign militants. The state’s application of these draconian, often indiscriminate mea- sures against the local population, and its appeasement of militants through peace agreements, has only compounded the sense of alienation stemming from counterinsurgency operations that, some studies show, beneªt insurgents.58 But in North Waziristan, many locals are alienated from both the Pakistani state and the militants. Given the dense social and kinship ties of NWA inhabitants, ªnding evidence of blowback from well-informed locals should be relatively easy. As one elder from Dande Darpa Khel, a base of the Haqqani network in tehsil Miranshah and the target of more than a dozen reported drone strikes since 2008, explained: “We hear rumors that this or that man joined the Taliban or al-Qaida because of anger over drone strikes. It is possible. But much of it is propaganda by the militants and intelligence agencies. I know almost every family in my area, and I do not know of a case where a local man or boy joined the Taliban as the direct result of death or injury to a close relative in a drone strike. In fact, most of the Taliban ªghters were already radicalized, or inclined toward militancy for various reasons.”59 I conªrmed this ªnding with another tribal elder from Dande Darpa Khel, as well as with more than four dozen other interviewees from the drone-targeted tehsils of Datta Khel, Ghulam Khan, and Guryam.60 According to a Wazir elder loyal to the Pashtun nationalist party, the Awami National Party, who survived a Taliban suicide attack in 2009 in Miranshah, “I have seen many people recruited by militants, including people I knew growing up. But their radicalization preceded the drone campaign. Many of the Pakistani Taliban leaders attended madrassas [religious schools] in and around the tribal areas or even in the settled districts established during the Afghan jihad. Some fought alongside the mujahideen or Afghan Taliban. They

57. Parliament recently passed legislation that will extend the writ of Pakistan’s courts to FATA. See “National Assembly Extends SC, High Court’s Jurisdiction to FATA,” Pakistan Today, January 12, 2017. 58. See, for instance, Jason Lyall, Graeme Blair, and Kosuke Imai, “Explaining Support for Com- batants during Wartime: A Survey Experiment in Afghanistan,” American Political Science Review, Vol. 107, No. 4 (November 2013), pp. 679–705, doi:10.1017/S0003055413000403. 59. Author interview with respondent 17, Peshawar, December 2015. 60. Author interviews with respondents 13, 14, 18, and 27, Bannu and Peshawar, June 2015.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 61

were able to mobilize the madrassa network across the region, as well as use their links to Sunni sectarian groups like Lashkar-e-Jhangvi.”61 According to Sailab Mehsud, a well-regarded journalist from SWA who has interviewed key Taliban commanders and covered extremism and violence in FATA for more than a decade: “Talibanization followed the U.S. invasion of Afghanistan, and the Pakistani military’s operations and peace deals with Taliban militants, which happened years before the drone campaign picked up momentum. How can we say that drones are driving radicalization and militancy among local youth?”62 Even local leaders of Islamist and other right-wing parties acknowledged that the ability of drone strikes to spawn militants is exaggerated. According to a local member of the Jamiat Ulema-e-Islam (JUI), a party with long-standing ties to the Afghan Taliban, Pakistani Taliban, and other Deobandi militant organizations, “The people know drones kill the widely despised militants. Our society is based on close family and community relations. So, it is not difªcult to determine who dies in drone strikes and who becomes a militant.”63 Simi- larly, a former JUI member of the national parliament from NWA, who personally knows the families of several militants killed in drone strikes in Mir Ali and Miranshah, stated that “drones did not fuel militancy. I think most of the tribal people know that drones are precise. Those who died in drone attacks were already militants or al-Qaida and Taliban leaders.”64 In fact, members of the most vehemently anti-American party, the Pakistan Tehrik-e-Insaf (Movement for Justice party), characterized its national leadership’s claims about blowback from drone strikes in terms of political expediency. According to one local PTI ofªcial, “Our leader, Imran Khan, is cynically using drones for political gain. He has never been to NWA or SWA. The gunehgar [sinful] militants, not drones, were blowing up innocent people to pieces. They are our real enemies.”65 Similarly, two members of Jamaat-e-Islami contradicted the claim of their emir (party chief), Syed Munawar Hasan, that drone strikes “are killing nearly 100 percent innocent civilians.”66 According to one, “Drones kill in- nocents sometimes. But the [Pakistan] military and militants have more innocent blood on their hands. The death and destruction caused by them boils our blood.”67 Others with deep knowledge of FATA believe that the roots of militancy also

61. Author interview with respondent 53, Bannu, June 2015. 62. Author interview with Dera Ismail Khan, May 2015. 63. Author interview with respondent 32, Peshawar, June 2015. 64. Author interview with respondent 33, Bannu, December 2015. 65. Author interview with respondent 37, Peshawar, December 2015. 66. Author interviews with respondents 43 and 44, Peshawar, December 2015. Quoted in Abbot, “New Light on Drone War Death Toll.” 67. Author interview with respondent 43, Peshawar, December 2015.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 62

lie in social and political conditions that have plagued the tribal belt since at least the 1979 CIA-led anti-Soviet jihad in Afghanistan, when the region was used by Pakistan’s Inter-Services Intelligence agency for training and arming the mujahideen.68 Aided by the Pakistani state, the erosion of the region’s tribal social structure during that war at the hands of Islamist militants, com- bined with Pakistan’s consistent failure to provide FATA with meaningful levels of economic development and political representation, created the struc- tural conditions necessary for an insurgency. In addition, the Pakistani military’s long-standing policy of using jihadi proxies to wrest Kashmir from archrival India or to deny New Delhi inºuence in Afghanistan has backªred. Since the September 11 attacks, the Pakistani military has selectively sided with the United States in the “war on terror,” ªghting groups such as the TTP, which has launched attacks against the Pakistani state, while continuing to support other groups, such as the Haqqani network, that commit violence in Afghanistan.

tribal revenge One of the principal claims in the literature is that casualties in drone strikes compel family members of the killed or injured to seek revenge, as prescribed by their unwritten honor code of Pashtunwali (or the way of the Pashtun). In the rush to make sense of the Pakistan-Afghanistan border region in the aftermath of the September 11 attacks and the subsequent U.S. invasion of Afghanistan, many journalists, scholars, and policymakers fell back on British colonial stereotypes and myths about Pashtuns.69 The most pervasive and re- silient of these concerns the primordial identity and customs of tribal Pashtuns as the key motivator of their social and political behavior. As late as 2006, the Economist wrote, “His honour besmirched...aPashtun is obliged to have his revenge, or badal,” and “the wildest Pashtun places, including Waziristan,” are likely to remain trapped in “their simmeringly murderous tribal ways” in the foreseeable future.70 Thomas Johnson and Chris Mason describe Pashtuns as a “revenge-oriented people,” with “zero tolerance for insult,” who take

68. Author interview with columnist and former senator Afrasiab Khattak, Peshawar, January 2015. 69. Much of the contemporary policy discourse on FATA has its roots in the writings of British colonial ofªcials and observers, who presented it as a remote lawless borderland populated by hill tribes who needed special forms of control and criminal justice presumably true to their native tra- ditions. See, for instance, Olaf Caroe, The Pathans: 500 – 1957 (London: Kegan Paul Interna- tional, 1958). For a critique of this colonial model of frontier management, see Benjamin D. Hopkins, “The Frontier Crimes Regulation and Frontier Governmentality,” Journal of Asian Studies, Vol. 74, No. 2 (March 2015), pp. 369–389, doi:10.1017/S0021911815000030. 70. “Honor among Them,” Economist, December 9, 2006, http://www.economist.com/node/ 8345531.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 63

up arms against outsiders because of their “peculiar cultural” mores.71 Johnson and Mason write that, of all the Pashtun tribes, “the Waziris of greater Waziristan are reputed to be the most conservative and irascible.”72 At the heart of these stereotypes lies a culturally deterministic interpretation of Pashtunwali as a normative straitjacket governing Pashtun life, whereby each member strictly conforms to its expectations. This view, however, ignores that the tribe is only one marker of Pashtuns’ many identities, and there is no reason to conclude that it is the most important one. As Patrick Porter argues, orientalist tracts assume “that Pashtuns [are] bound by their honor code regardless of rationality or material considerations [which is] symptomatic of a broader tendency in popular and academic discourse to reduce people with multiple interests and identities to one-dimensional beings.”73 And even if some Pashtun groups organize themselves in tribes, individual members may choose not to follow group customs. As Makulika Banerjee writes, “Notions of Pashtunness are not static but rather the subject of negotiation and innovation.”74 Badal (exchange or reciprocity) is generally considered part of Pashtunwali. The historian Robert Nichols contends that “badal has been reduced by colonial orientalists and western writers to the exchange of violence or tribal revenge. Dismissing local behavior and responses as tribal and barbaric can be used to justify unaccountable state violence and power.”75 Tribal norms of revenge are, in other words, part of an “invented tradition” designed to render special forms of colonial and, later, postcolonial governance possible.76 This so-called revenge thesis ignores that badal is meant to be exacted against individuals in the context of family disputes, not against the civilian population or security forces of a state, as Islamist militants have done in FATA, the rest of Pakistan, and Afghanistan. Moreover, these disputes can be ended amicably through a process known as rogha (mediated peace/truce), which typically involves maaª (forgiveness), recompense of victims, or both. In fact, the more than one dozen respondents in my sample who self-identiªed as

71. Thomas H. Johnson, and M. Chris Mason, “No Sign until the Burst of Fire: Understanding the Pakistan-Afghanistan Frontier,” International Security, Vol. 32, No. 4 (Spring 2008), p. 50, doi:10 .1162/isec.2008.32.4.41. 72. Ibid. 73. Patrick Porter, Military Orientalism: Eastern War through Western Eyes (New York: Oxford Uni- versity Press, 2009), p. 145. 74. Makulika Banerjee, The Pathan Unarmed: Opposition and Memory in the Northwest Frontier (New Delhi: Oxford University Press, 2001), p. 15. See also Robert Nichols, Settling the Frontier: Land, Law, and Society in the Peshawar Valley, 1500–1900 (Ithaca, N.Y.: Cornell University Press, 1982), pp. 6–7. 75. Email communication to author, January 2017. 76. Terence Ranger, “The Invention of Tradition in Colonial Africa,” in Eric Hobsbawm and Ranger, eds., The Invention of Tradition (New York: Cambridge University Press, 1983), pp. 211–262.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 64

relatives of drone victims were seeking redress from Pakistani or U.S. authorities, not revenge.77 The revenge thesis suffers from other misinterpretations as well. First, as several well-informed tribal elders and experts emphasized in my interviews, Taliban ªghters and other militants are motivated by a jihadi ideology, not family honor. In fact, many Taliban are not ethnic Pashtuns but Punjabis— hence, the moniker “Punjabi Taliban.” Second, the revenge thesis fails to account for Taliban suicide attacks on jirgas, markets, sports grounds, and schools in FATA that invariably kill local noncombatants (i.e., the very reason for the presumed drone blowback).78 Finally, the Taliban have assassinated hundreds of local inhabitants of NWA and other FATA agencies on suspicion of their spying for the United States or sympathizing with the Pakistani military.79 Yet, aggrieved family members have not resorted to retributive attacks against Taliban militants.80

National Blowback

A subset of the literature on the blowback thesis examines the indirect effects of U.S. drone strikes at the national level. According to this literature, drone strikes impinge on a country’s territorial integrity and sovereignty, enraging a broad swathe of its population and providing anti-U.S. militant groups with a sizable reservoir of sympathizers and potential recruits. Civilian deaths in such strikes amplify the blowback effect. Thus, the operating mechanism here is nationalism. According to a Stimson Center Task Force on U.S. drone warfare, “Even where strikes kill only legitimate targets, the perceived insult to sovereignty—in places like Pakistan and Yemen—sparks bitterness, feelings of nationalism...hostile to the U.S.” In addition, “civilian casualties, even if relatively few, can anger whole communities, increase anti-U.S. sentiment and become a potent recruiting tool for terrorist organizations.”81 According to Kilcullen and Exum, the backlash against U.S. drone strikes extends beyond FATA to “visceral opposition across a broad spectrum of Pakistani opinion in

77. Author interviews, Bannu and Peshawar, December 2015. 78. See data on suicide attacks in Waziristan at the South Asian Terrorism Portal, Institute for Conºict Management, New Delhi, India, http://www.satp.org/satporgtp/countries/pakistan/ waziristan/datasheet/suicideattack.htm. 79. In 2005–06 alone, militants reportedly murdered 200 Wazir tribal elders who dared to oppose their reign of terror. See Carlotta Gall and Ismail Khan, “Taliban and Allies Tighten Grip in North of Pakistan,” New York Times, December 11, 2006. 80. Interviews with multiple respondents, December 2015. 81. John P. Abizaid and Rosa Brooks, “Report and Recommendations of the Task Force on U.S. Drone Policy” (Washington, D.C.: Stimson Center, 2015), pp. 29–30.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 65

Punjab and Sindh, the nation’s two most populous provinces.”82 The Pakistani opposition leader and former cricketer Imran Khan, who revived his dormant political career in part by loudly criticizing the U.S. drone campaign, has claimed that drone strikes carried out in Pakistan kill many civilians and “are turning young men into angry jihadis.”83

methods for assessing national-level blowback Assuming that the national blowback thesis is true and that the backlash from drone strikes is consequential, one should observe signiªcant numbers of Pakistanis becoming radicalized or motivated to join militant organizations because drone strikes feed their nationalist impulses. To test this argument, I conducted interviews with terrorism experts and counterterrorism ofªcials from Punjab and Sindh, which proponents of the blowback thesis claim are the prime sites of drone blowback. I included journalists and scholars in my sample, as well as police and intelligence ofªcials from counterterror units in the two provinces, given their ªrsthand experience of dealing with terror suspects.84 In addition to the interviews, I analyzed the ªndings of the Sindh Counterterrorism Department’s (CTD’s) 2017 survey of 500 detained terrorists.85 The CTD sample is not representative of the universe of Islamist militants in the province, but it does include active members of the key militant groups operating there, such as al-Qaida, Harkatul Mujahideen al-Almi, the Sunni sectarian Lashkar-e-Jhangvi, and the TPP.86

assessment of national blowback To evaluate the blowback thesis at the national level, it is necessary to examine whether civilian casualties from drone strikes amplify the backlash against drone strikes in nontargeted areas of a country. Unfortunately, no one can reliably know the number of civilian deaths caused by drone strikes, in good

82. Kilcullen and Exum, “Death from Above, Outrage Below.” See also Nathanial Fick, testimony before the Senate Committee on Homeland Security and Governmental Affairs, July 7, 2009, 111th Cong., 1st sess., p. 2, www.hsgac.senate.gov/download/ªck-testimony; Akbar Ahmed, The Thistle and the Drone: How America’s War on Terror Became a Global War on Tribal Islam (Washington, D.C.: Brookings Institution Press, 2013), p. 83; and Boyle, “Costs and Consequences of Drone Warfare,” p. 15. 83. Rob Crilly, “Imran Khan’s Anti-Drone Protest Falls Short of Destination,” Telegraph, October 7, 2012. 84. The sample includes two inspectors general (the country’s highest police rank), four deputy inspectors general, and two senior ofªcials of the Federal Investigation Agency, the Pakistani equivalent of the Federal Bureau of Investigation. 85. Its ªndings have been partially reported in the Pakistani media. See Imtiaz Ali, “Most Mili- tants Surveyed in Jails Suffering from ‘Psychological Problems’: CTD,” Dawn, April 18, 2017, https://www.dawn.com/news/1327614. 86. Email communication to author, August 6, 2017.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 66

measure because the program is shrouded in secrecy. For many years, the George W. Bush and Barack Obama administrations neither conªrmed nor de- nied its existence. Even after acknowledging the program’s existence in 2012, top U.S. ofªcials continued to withhold detailed information, citing national security concerns. Under mounting criticism from the media and human rights organizations, President Obama admitted in a May 2013 speech that drone strikes had resulted in civilian deaths. He went on to reassure critics and the U.S. public that “before any strike is taken, there must be near-certainty that no civilians will be killed or injured—the highest standard we can set.” ) Only in July 2016 did he reveal an ofªcial estimate of the number of civilian casualties (64–116) resulting from U.S. drone strikes in Pakistan, Yemen, and Africa since he had taken ofªce in 2009.87 As critics note, this deceptively low number is the result of how the Obama administration deªned combatants: “all military-age males in a strike zone...unless there is explicit intelligence posthumously proving them innocent.”88 Another obstacle to determining the number of civilian deaths from drone strikes is that most such strikes occur in areas with a heavy presence of Taliban or Taliban-associated militant groups. Local inhabitants complain that militants close down roads and cordon off strike sites.89 Because independent media have no access to these sites, it is usually the militants who provide the casualty ªgures reported in the Pakistani press. The Crisis Group contends that civilian casualty ªgures are also misreported or unreliable because they are based on information provided by unnamed Pakistani or U.S. ofªcials who have an interest in exaggerating or underreporting the civilian death toll.90 Particularly problematic are so-called signature strikes, which target groups of men based on their behavioral patterns rather than on their identity and which are prone to causing high fatalities. One such attack in Datta Khel killed 40 people on March 7, 2011.91 Catastrophic strikes not only anger the victims’ relatives, but can arguably inºame Pakistani public opinion within and be-

87. “Obama Reveals How Many Died in Drone Strikes,” Associated Press, July 1, 2016. 88. See Jo Becker and Scott Shane, “Secret ‘Kill List’ Proves a Test of Obama’s Principles and Will,” New York Times, May 29, 2012, http://www.nytimes.com/2012/05/29/world/obamas-leadership- in-war-on-al-qaeda.html. 89. “Cables Show U.S. Special Operations in Pakistan,” Al Jazeera, May 21, 2011, http://www .aljazeera.com/news/asia/2011/05/201152175836483568.html. See also “Drone Attacks: Chal- lenging Some Fabrications,” Dawn, January 2, 2010. 90. International Crisis Group, “Drones,” p. 25. 91. Among the dead were a local commander afªliated with Haªz Gul Bahadur, khassadars (tribal police), and maliks who were gathered for a jirga meeting called to settle a dispute over chromite mining. See Pir Zubair Shah, “Analysis: U.S. Drone Strikes in Pakistan’s Tribal Areas Create Back- lash,” Global Post, October 10, 2012, https://www.pri.org/stories/2012-10-10/analysis-us-drone- strikes-pakistans-tribal-areas-create-backlash.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 67

yond FATA. According to a reporter who visited the strike site in Datta Khel and has interviewed family members of the victims, “I ªnd drones to be useful in ªghting the Taliban and other militants. But such callously misguided strikes are hard to swallow and make the local people question their utility.”92 The two main databases on drone strikes provide divergent numbers on civilian casualties, which underscores the difªculty of accurate estimation. The London-based Bureau of Investigative Journalism maintains that drone strikes have caused 2,514–4,023 deaths in Pakistan, including 424–969 civilians.93 Ac- cording to New America, the total number of people killed in drone attacks in Pakistan is 2,281–3,672, of which 255–315 were civilians.94 These ªgures are based on Pakistani and international media reports. But because the international media have no access to the tribal areas, their data are drawn from often unreliable reports, some of which have been planted by Pakistani intelligence in the Pakistani media.95 Despite considerably divergent ªgures in the two databases, both show that civilian casualties have declined drastically since 2011,96 a ªnding consistent with improvements in drone technology and modi- ªed targeting protocols designed to minimize casualties.97 The Donald Trump administration’s reported relaxation of guidelines intended to minimize civilian casualties in U.S. airstrikes could result in more civilian deaths, however. Regardless of the veracity of claims about civilian casualties, the belief that U.S. drone strikes kill many innocent people could still produce a backlash outside FATA. Scholars and analysts critical of drone strikes typically use opinion polls conducted by the Pew Research Center to support their claims.98 These polls demonstrate high levels of opposition to drone strikes in Pakistan.

92. Author interview with reporter, Peshawar, December 2015. 93. Bureau of Investigative Journalism, “Drone Wars: The Full Data” (London: Bureau of Investi- gative Journalism, n.d.), https://www.thebureauinvestigates.com/projects/drone-war/pakistan. 94. See Peter Bergen and Katherine Tiedemann, “Drone Strikes: Pakistan” (Washington, D.C.: New America, January 2018), https://www.newamerica.org/in-depth/americas-counterterrorism -wars/pakistan/. 95. C. Christine Fair, “The Problems with Studying Civilian Casualties from Drone Usage in Paki- stan: What We Can’t Know,” Monkey Cage blog, August 17, 2011. 96. Alice Ross and Jack Serle, “Drone Strikes in Pakistan Attack Houses” (London: Bureau of In- vestigative Journalism, May 23, 2014), https://www.thebureauinvestigates.com/2014/05/23/ most-us-drone-strikes-in-pakistan-attack-houses/. 97. For instance, the CIA shifted from targeting houses to targeting vehicles, because the latter pose a lower risk of civilian casualties. See Joby Warrick and Peter Finn, “Amidst Outrage over Ci- vilian Deaths in Pakistan, CIA Turns to Smaller Missiles,” Washington Post, April 26, 2010. See also Steve Coll, “Obama’s Drone War in Pakistan,” New Yorker, November 20, 2014, https://www .newyorker.com/magazine/2014/11/24/unblinking-stare. 98. See, for example, Boyle “Costs and Consequences of Drone Warfare”; Alexander B. Downes, “The Truth about Obama’s Drone War: It’s about Attrition, Not Decapitation,” Hufªngton Post, Jan- uary 16, 2013, https://www.hufªngtonpost.com/alexander-b-downes/obama-drones_b_2427030 .html.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 68

Pew’s survey from 2014, the last year the organization ªelded questions about drones, showed that 67 percent of Pakistani respondents who had heard of the drone program opposed drone attacks because they kill “too many innocent people.”99 U.S. drone strikes are globally unpopular, however, and many countries have higher levels of opposition to U.S. drone strikes than does Pakistan; these include Venezuela (92 percent), Greece (89 percent), and Brazil (87 percent).100 Yet, blowback is not a global phenomenon. In other words, popular disapproval of drones cannot stand in as evidence of blowback. As scholars have recently shown, Pew polling data on Pakistan is not representative for two reasons. First, Pew’s samples are disproportionately urban (55 percent), whereas Pakistan is a predominantly rural country with only 36 percent of the population living in urban areas.101 Second, the polls have a very high frequency of “don’t know” and “did not answer” responses,102 which indicates a public largely uninformed about the drone program and unwilling to express opinions about a sensitive issue. As such, Pew polls typically capture the opinion of a minority of the country’s population about drones.103

expert views on national blowback My interviews with experts on terrorism show little support for the drone blowback thesis at the national level. Of the twenty-four experts interviewed, only two endorsed the national blowback thesis. Nevertheless, even they did not think it was the most salient factor in fostering Islamist militancy, which they blamed instead on the conducive environment created by Pakistan’s sponsorship of militancy as a foreign policy instrument against India.

99. “A Less Gloomy Mood in Pakistan” (Washington, D.C.: Pew Research Center, August 27, 2014), http://www.pewglobal.org/2014/08/27/a-less-gloomy-mood-in-Pakistan; “Pakistani Pub- lic Opinion Ever More Critical of U.S.” (Washington, D.C.: Pew Research Center, June 27, 2012), http://www.pewglobal.org/2012/06/27/pakistani-public-opinion-ever-more-critical-of-u-s/; “On the Eve of Elections, a Dismal Public Mood in Pakistan” (Washington, D.C.: Pew Research Center, May 7, 2013), http://www.pewglobal.org/2013/05/07/on-eve-of-elections-a-dismal- public-mood-in-pakistan/; and Pew Research Center Global Attitudes Project, “Spring 2010 Sur- vey Data” (Washington, D.C.: Pew Research Center, 2010), http://www.pewglobal.org/category/ datasets/?downloadϭ15858. 100. See “Widespread Opposition to Drones” (Washington, D.C.: Pew Research Center, July 11, 2014), http://www.pewglobal.org/2014/07/14/global-opposition-to-u-s-surveillance- and-drones-but-limited-harm-to-americas-image/pg-2014-07-14-balance-of-power-0-01/. 101. Pew claims to adjust for this urban bias by using weights. 102. For example, in 2010, 43 percent of the respondents had not heard about the drone program despite a dramatic increase in drone strikes during the previous two years, and 22 percent did not respond. See “Little Knowledge of Drone Strikes in Pakistan” (Washington, D.C.: Pew Research Center, August 12, 2010), http://www.pewresearch.org/fact-tank/2010/08/12/little-knowledge- of-drone-strikes-in-pakistan/; and Pew Research Center Global Attitudes Project, “Spring 2010 Survey Data.” 103. C. Christine Fair, Karl Kaltenthaler, and William J. Miller, “Pakistani Political Communication and Public Opinion on U.S. Drone Attacks,” Journal of Strategic Studies, Vol. 38, No. 6 (2015), pp. 852–872, doi:10.1080/01402390.2014.997932.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 69

Experts in my sample agreed that drone strikes can be a useful recruiting tool. Beyond that, however, their utility can vary depending on the agenda and objectives of the militant organization in question. Tariq Parvez, founding national coordinator of Pakistan’s National Counterterrorism Authority, argues that organizations with a clear anti-U.S. narrative such as al-Qaida can exploit drone strikes for propaganda purposes, but that “drones are a distant threat for people in Punjab or Sindh, for which you can express indignation but not be really be threatened or motivated by it to become militants.”104 According to Ahmed Rashid, one of the world’s foremost authorities on the Taliban and terrorism in Afghanistan and Pakistan, “There is no evidence to show that Punjabi or other terrorist groups in Pakistan use drone strikes as an important recruitment tool or that these attacks create terrorists en masse. Even drone strikes outside of FATA, such as the one which killed the Afghan Taliban leader Mullah Mansur in Balochistan, did not lead to any anger against drones.”105 Arif Jamal, journalist and author of an important account of the origins of Lashkar-e-Taiba (LeT), the Pakistan-based militant group that enjoys ofªcial backing and operates in the disputed Kashmir region and other parts of India,106 similarly rejects the blowback thesis: “Haªz Saeed [leader of LeT] has criticized U.S. drone strikes in his recent speeches, banding together the U.S. and India as enemies of Islam. But demonizing the ‘inªdels’ [the United States, Israel, and India) has long been part of their organizational rhetoric. Their main strategy is to exploit sympathy for fellow Muslims in Kashmir to mobilize support for jihad against India.”107 A study based on 917 biogra- phies of deceased LeT militants, coauthored by Jamal, concluded: “In general ªghters viewed association with the group as a means to live a more meaningful or purposeful life. Some were speciªcally motivated by corruption in their societies. Others by what they saw as moral depravity. Some articulated an ob- ligation to help fellow Muslims who experienced oppression and death at the hands of non-Muslims in Indian Kashmir.”108 Expressing a view prevalent in my interviews, the analyst Ayesha Siddiqa, who has conducted extensive ªeld research on the rise of militant groups in South Punjab, argues that “the Punjabi jihadists have an expansionist agenda which varies from liberating Indian Kashmir to establishing an Islamic state in Pakistan which they use to

104. Email communication to author from Tariq Parvez, January 4, 2018. 105. Email communication to author from Ahmed Rashid, June 28, 2017. 106. Arif Jamal, Call for Transnational Lashkar-e-Taiba (LeT): 1985–2014 (Portland, Or.: Avant-Garde, 2014). 107. Email communication to author from Arif Jamal, June 27, 2017. 108. Don Rassler et al., “The Fighters of Lashkar-e-Taiba: Recruitment, Training, Deployment, and Death” (West Point, N.Y.: Combating Terrorism Center at West Point, 2013), p. 9.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 70

recruit and motivate their cadres. Their rhetoric and discourse has little con- nection to FATA or drone strikes.”109 I derived additional data for testing the national blowback thesis from interviews with sixteen senior active-duty police ofªcers, six from Sindh and ten from Punjab.110 Although this sample is not representative of the universe of police ofªcers in these two provinces, the mean length of the respondents’ ser- vice in counterterrorism-related operations and investigations is thirteen years. Among these ofªcers, there was a consensus that drone strikes in FATA were not a crucial factor in militant recruitment in the two provinces. A senior intelligence ofªcial in charge of counterterrorism put it plainly, “We receive terrorism-related reports from our own people, and other law enforcement agencies. We have not seen evidence to suggest that drone strikes play a signiªcant role in fomenting militant recruitment in Punjab or Sindh.”111 According to an ofªcial in the Punjab Counterterrorism Department, “Mili- tants in our province are motivated by a mix of factors which cannot be directly linked to drones. In our investigations, we routinely encounter two types of hardened militants: the Lashkar-e-Jhangvi types who use violence against religious minorities they consider to be inªdels, and the Taliban-type militants trying to destabilize the state.”112 Expressing an opinion shared by eight of the ten Punjab ofªcers mentioned above, Senior Superintendent of Police Sohail Habib Tajik, who was additional director of the Federal Investi- gation Agency’s Counterterrorism Wing from 2008 to 2010, when the U.S. drone campaign was as it peak, stated: “Drone strikes did not generate militancy, militancy generated drone strikes.”113 Terrorism investigators from Sindh conªrmed the views of their Punjab counterparts. To quote a senior superintendent of police from Karachi, “We have interrogated militants from al-Qaida, TTP, Jundullah, and sectarian outªts. They all profess a standard hatred of some enemy: Shias, Iran, the Pakistan Army, India, Hindus, Jews, the United States, or the West in general. Our investigations reveal that criminality, economic problems, inºuence from someone close to them often has a strong impact. The notion that drone strikes fuel militancy is quite simplistic.”114 In response to a question about the effects of drones on militancy in Karachi, an ofªcer with extensive experience of inter- rogating Taliban militants stated that “many militants have moved to Karachi from FATA. What we ªnd is that drones are the reason for their ºight. Not

109. Author interview with Ayesha Siddiqa, London, October 16, 2016. 110. Kilcullen and Exum, “Death from Above, Outrage Below.” 111. Email communication to author from ofªcer D, September 14, 2017. 112. Email communication to author from ofªcer B, July 12, 2017. 113. Email communication to author from Suhail Habib Tajik, January 4, 2018. 114. Author telephone interview with ofªcer C, July 14, 2017.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 71

their anti-state activities.”115 According to a senior intelligence ofªcer whose responsibilities include monitoring militant groups in the interior (rural) parts of Sindh Province, “Interior Sindh districts, such as Sukkur and Jacobabad, that border the militant heartland of southern Punjab are mainly a recruitment ground for Sunni sectarian outªts who socialize vulnerable youth into a narrative of us and them, them being Shias and other religious minorities, who must be eradicated. The [militant] threat engulªng this area is more domestic in nature, having arisen from local social conditions rather than the war on terror.”116 The CTD survey conªrms these observations and shows that the main motivations for joining terrorist groups in Sindh Province are economic and religious/sectarian, rather than anti-Americanism generally or opposition to U.S. drone strikes speciªcally, although grievances against the West appear to play an important role in kindling initial interest in militancy. The ªelded questionnaire asked participants a host of demographic questions related to education, employment, and marital status, as well as level of religiosity. It also asked a series of questions about the sources of the captured militants’ motivations. In response to the ªrst question, “Why did you start thinking that violence in the name of Islam was justiªed?” a plurality cited perceived injus- tices of the West against Islam (41.4 percent), followed by lack of justice in society (19.4 percent), personal experiences (19.8 percent), and other (19.4 percent). The follow-up question was: “How were you inºuenced in developing extremist views?” Of the options listed, 40 percent chose religious cleric, 25.6 percent peers, 15 percent family or community members, and almost 20 percent were in the “other” category. In response to the ªnal and most important question, “What ultimately drew you to join a terrorist/banned outªt?” 41 percent cited unemployment or economic concerns, 40 percent religious concerns, and 16 percent psychological issues. A CTD ofªcer involved in supervising the survey explained that those who answered “other” in this series of questions did not always identify the reasons for their participation in terror groups. He could not recall, however, an example of revenge or anger at drone strikes as a possible motivation.117 According to another CTD ofªcial, “The drone issue is highly overrated as a motivation for militancy. In my experience [of more than twelve years) in police and counterterrorism work, I have seen no evidence of drone-driven militants in Sindh Province.”118

115. Email communication to author from ofªcer K, August 2, 2017. 116. Author telephone interview with ofªcer O, August 5, 2017. 117. Author telephone interview with ofªcer N, August 3, 2017. 118. Email communication to author from ofªcer L, August 2, 2017.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 72

drones and anti-americanism in pakistan Even if one accepts that drone strikes antagonize signiªcant segments of the Pakistani population outside FATA, other issues likely feed and reinforce resentment of U.S. policies, including a carefully curated ofªcial narrative of the United States as an unreliable ally, which dates to at least the 1960s and 1970s.119 According to former Pakistani Ambassador to the United States Husain Haqqani, Pakistani government ofªcials have long invoked anti- American public opinion to extract concessions and aid from Washington.120 More recently, the Pakistani government, the military, and right-wing politi- cians such as Imran Khan have sought to leverage media accounts of alleged civilian deaths in drone strikes to fan anti-Americanism. Although there is almost no reliable reporting from areas where drone strikes occur, many pro- military talk show hosts and journalists have played a key role in inºaming public opinion,121 often by making wild claims about “thousands” of civilian casualties resulting from drone strikes. According to Hassan Abbas, terrorists and their sympathizers “expose and market civilian casualties...quite effec- tively.”122 It is probable that sections of a country’s population not directly exposed to drone strikes may be more vulnerable to media manipulation and political propaganda because they are not well informed about the impact of these strikes. However strident these propaganda efforts may be, their impact on Pakistani public opinion toward the United States does not appear to have signiªcant independent effects. For instance, post–September 11 Pew polls show that unfavorable opinions of the United States have remained static over time. In fact, 69 percent of Pakistanis held a negative view of the United States in 2002 (two years before the ªrst drone strike), and an almost similar percentage (68 percent) in 2009 and 2010 (the ªrst two years of the Obama administration’s intensiªed drone campaign).123 The polls do show an upward trend in anti-U.S. opinion in 2011 and 2012, but this spike may not be exclusively correlated to drone strikes. Instead, it can be more reasonably linked to the events of 2011, the worst year for bilateral relations since Pakistan formally joined the U.S.-led war on terror in 2001. The

119. Shafqat Hussain Naghmi, “Pakistan Public’s Attitudes toward the United States,” Journal of Conºict Resolution, Vol. 26, No. 3 (September 1982), pp. 507–523. 120. See Husain Haqqani, Magniªcent Delusions: Pakistan, the United States, and an Epic History of Misunderstanding (New York: PublicAffairs, 2013), pp. 11–14, 42, 82. 121. Saba Imtiaz, “What Do Pakistanis Really Think about Drones,” in Peter L. Bergen and Daniel Rothenberg, eds., Drone Wars: Transforming Conºict, Law, and Policy (New York: Cambridge Univer- sity Press), p. 102. 122. Hassan Abbas, “How Drones Create More Terrorists Than They Kill,” Atlantic,Au- gust 31, 2013, https://www.theatlantic.com/international/archive/2013/08/how-drones-create- more-terrorists/278743/. 123. Data retrieved from Pew Global Attitudes Database, http://www.pewglobal.org/question- search/?qidϭ844&cntIDsϭ@[email protected]&stdIDsϭPew.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 73

percentage of Pakistanis who see the United States as an “enemy” already stood at 60 percent in 2008, despite the sporadic nature of U.S. drone strikes until 2007. It was the highest (74 percent) in 2012, following the steep plunge in bilateral relations in 2011, and dropped close to the 2008 level (64 percent) in 2013.124 Three events in 2011 seriously damaged U.S.-Pakistan relations. First, in January Pakistani police arrested Raymond Davis, a CIA contractor, who fa- tally shot two reportedly armed Pakistanis in Lahore who he claimed were threatening him. The Obama administration’s position that Davis enjoyed diplomatic immunity, which was contested by the Pakistani government, appeared to conªrm for many Pakistanis that the U.S. government was unconcerned about Pakistani lives or laws. Second, in May U.S. Special Forces killed Osama bin Laden in Abbottabad, sending a signal that the United States had no regard for Pakistani sovereignty. And just as the two states were beginning to mend their relationship, the U.S.-led NATO attack on the Pakistani military’s border post in Salala, in November, killed twenty- four Pakistani soldiers, plunging the relationship to a new low after the Obama administration refused to apologize. In reaction, Pakistan blocked the use of its ground lines of communication used by NATO forces for trans- porting non-lethal supplies to Afghanistan and demanded the CIA vacate the Shamsi Air Base in Balochistan.

drones, sovereignty, and counterterrorism cooperation Some scholars have argued that public anger caused by drone strikes creates pressure on allied governments to end or drastically curtail their cooperation with U.S. counterterrorism efforts. Leila Hudson, Colin Owens, and Matt Flannes contend that “the less direct effect of steady drone attacks...isa smoldering dissatisfaction with the dead-end policy...which results in anti- government agitation and anti-American sentiment, which may force sudden policy adjustments by political and military actors.”125 Similarly, Abbas claims that “anti-U.S. feelings in Pakistan increased substantially as a result of this strategy [drone strikes], weakening the U.S.-Pakistan counterterrorism cooperation.”126 The indisputable unpopularity of drones in Pakistan conditions government responses to drone strikes. Ofªcials in Pakistan’s ministry of foreign affairs

124. Data retrieved from Pew Global Attitudes Database, http://www.pewglobal.org/question- search/?qidϭ793&cntIDsϭ@[email protected]&stdIDsϭ. 125. Hudson, Owens, and Flannes, “Drone Warfare.” 126. Hassan Abbas, “Are Drone Strikes Killing Terrorists or Creating Them?” Atlantic, March 31, 2013, https://www.theatlantic.com/international/archive/2013/03/are-drone-strikes- killing-terrorists-or-creating-them/274499/.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 74

have repeatedly claimed that these strikes are illegal and violate the country’s sovereignty.127 In an October 2012 meeting with President Obama, Prime Minister Nawaz Sharif reportedly emphasized the need to end the program because it was “a continued violation of our territorial sovereignty.”128 This ofªcial disapproval notwithstanding, the United States could not have oper- ated drones in the country’s airspace without the consent of Pakistani authorities, as the military’s air defense weapons systems can easily shoot them down. A former director general of Pakistan’s Inter-Services Intelligence agency, Lt. Gen. Shuja Pasha, told a Pakistani commission inquiring into bin Laden’s death about an “understanding” with the United States regarding its drone operations in the country. In 2013, Pervez Musharraf disclosed that, as president of Pakistan, he had granted permission to the United States to launch drone strikes in FATA.129 (The two sides had reportedly entered into a covert agreement in 2004 that designated zones, or “ºight boxes,” for U.S. drone operations in the Waziristans.130) The CIA was allowed to use air force bases in Jacobabad (Sindh Province) and Shamsi (Balochistan Province) for its drone ºeet. Even proponents of the blowback thesis claim that “Pakistan’s government (both military and civilian) was fully on board . . . which means the sovereignty issue was not relevant in many cases.”131 Khaled Ahmed, one of Pakistan’s most proliªc analysts of Islamist ideologies and sectarian militancy, argues that in “a state without internal control...[where] ter- ritory is controlled by terrorists and insurgents,” opposition to drones is meant to keep up pretenses of external sovereignty.132 Evidence suggests that, despite their public protestations, civilian leaders have also acquiesced in drone strikes as a strategy to defeat terrorism. For instance, Prime Minister Yusaf Raza Gillani publicly criticized the United States for its “illegal and counterproductive” use of drones, arguing that it fuels the Taliban insurgency against the government.133 According to leaked U.S. diplomatic cables, however, Gillani, in a meeting in August 2008

127. “Drones Violating Pakistani Sovereignty, Hina Tells Olson,” Nation, January 23, 2013; and Jethro Mullen, “Rights Groups Challenge U.S. on Drone Strikes in Pakistan, Yemen,” CNN.com, October 22, 2013, http://www.cnn.com/2013/10/22/world/asia/us-drone-strikes-reports/index .html. 128. Mullen, “Rights Groups Challenge U.S. on Drone Strikes in Pakistan, Yemen.” 129. Nic Robertson and Greg Botelho, “Ex-Pakistani President Musharraf Admits Secret Deal with U.S. on Drone Strikes,” CNN.com, April 12, 2013, http://www.cnn.com/2013/04/11/world/asia/ pakistan-musharraf-drones/index.html. 130. Eric Schmitt, “U.S. Prepares for a Curtailed Relationship with Pakistan,” New York Times,De- cember 25, 2011. 131. Abbas, “Are Drone Strikes Killing Terrorists or Creating Them?” 132. Khaled Ahmed, “Two Blowbacks from Afghanistan,” Express Tribune (Pakistan), May 5, 2012. 133. “Gillani Bemoans ‘Trust Deªcit’ with the U.S.,” Dawn, June 28, 2012.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 75

with U.S. Ambassador to Pakistan Anne Patterson, consented to drone strikes “as long as they killed the right people.” He added, “We’ll protest in the National Assembly and then ignore it.”134 Earlier that year, Army Chief of Staff Gen. Ashfaq Pervez Kayani asked the United States for “continuous predator coverage” to aid the Pakistani army’s operations in South Waziristan.135 Senior Pakistani military ofªcials reportedly even muted their public criticism after a drone strike killed TTP leader Baitullah Mehsud in the summer of 2009.136 According to classiªed CIA documents and Pakistani diplomatic memos ob- tained by the Washington Post, U.S. intelligence ofªcials routinely briefed their Pakistani counterparts on speciªc drone strikes and casualty counts. From 2008 to 2010, the United States also targeted individual Taliban militants at the request of the Pakistani authorities.137 Public criticism by Pakistani leaders to assuage public opinion aside, even the most devastating U.S. drone strikes in Pakistan have not severed or signiªcantly weakened bilateral counterterror cooperation, especially the CIA’s use of Pakistani airspace to launch these strikes. And despite Islamabad’s strong diplomatic rebuke when the Trump administration decided in January 2018 to cut off all security-related assistance to Pakistan because of its selective counterterrorism efforts, U.S. drone strikes in Pakistan continue.

Transnational Blowback

Some scholars and analysts have argued that Muslims living in, for example, the United States or Europe may become radicalized or retaliate violently against the United States when their religious kin are killed or injured in drone strikes in Afghanistan, Pakistan, Somalia, or Yemen.138 Hence, the operative mechanism here is “long-distance” or diasporic nationalism.139 According to the Middle East scholar Fawaz Gerges, “[D]rone strikes fuel the ªres of homegrown [Muslim] radicalization in Western societies. This is a rising phenome-

134. “U.S. Embassy Cables: Pakistan Backs U.S. Drone Strikes in Tribal Areas,” Guardian, Novem- ber 30, 2010, https://www.theguardian.com/world/us-embassy-cables-documents/167125. 135. “Kayani Asked for ‘Continuous Predator Coverage,’” Dawn, May 19, 2011. 136. Daniel S. Markey, No Exit from Pakistan: America’s Tortured Relationship with Islamabad (New York: Cambridge University Press, 2013), p. 157. 137. Greg Miller and Bob Woodward, “Secret Memos Reveal Explicit Nature of U.S.-Pakistan Drone Agreement,” Washington Post, October 24, 2013. 138. Stanford Law School and NYU School of Law, Living under Drones, pp. 136–137. See also Chase Madar, “Drone Attacks Undermine National Security,” Al Jazeera America, July 27, 2014, http://america.aljazeera.com/opinions/2014/7/drone-blowback.html. 139. Benedict Anderson, “Long-Distance Nationalism: World Capitalism and the Rise of Identity Politics,” Wertheim lecture given at the Center for Asian Studies, Amsterdam, the Netherlands, 1992.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 76

non that has not been seriously debated, despite a string of high-proªle attacks....[T]he Woolwich attack in London and the Boston Marathon bombings are suspected to be the latest cases in point.”140 Michael Boyle claims that “drones have replaced Guantánamo Bay as the number one [global] recruiting tool for Al-Qaeda today.”141

methods for assessing transnational blowback Gathering conclusive proof of terrorist motivations is notoriously difªcult. To evaluate the transnational blowback thesis, I rely on three main sources. The ªrst source consists of trial testimony and public statements by U.S.-based captured terrorists believed to have been motivated by drone strikes. Second are court documents (i.e., criminal prosecution complaints and testimony) and secondary sources relating to Somali-American men charged with terror- related offenses in the United States. Since 2007, the al-Qaida afªliate al-Shabaab has recruited more than two dozen men from the Minneapolis– St. Paul area (and several others from the Somali diaspora in Europe) to join the insurgency against the United Nations–backed transitional government in Somalia.142 Al-Shabaab, designated by the United States as a foreign terrorist organization (FTO) in 2008, has been the target of some sixty U.S. drone strikes in Somalia since 2008. Hence, an assessment of the motives behind violent jihad among Somali immigrant youth can provide at least a sample for evaluating the transnational blowback thesis. The third source of evidence com- prises scholarly accounts of the motivations for militant Islamism in Europe, one of the main hubs for its transnational spread.

assessment of transnational blowback It is important to note that only a handful of the approximately 100 jihadist- linked or inspired terror plots (successful, failed, or foiled) in the United States since September 11 were reportedly hatched by individuals who claimed that drone strikes were one, if not the only, source of their motivation.143 The

140. Fawaz Gerges, “Why Drone Strikes Are Real Enemy in ‘War on Terror,’” CNN.com, June 21, 2013, http://www.cnn.com/2013/06/21/opinion/terrorism-gerges/index.html. It is un- clear whether the two attacks cited by Gerges involved militants inspired speciªcally by drone strikes. 141. Boyle, “Costs and Consequences of Drone Warfare,” p. 13. See also Glen Greenwald, “An- drew Sullivan, Terrorism, and the Art of Distortion,” Guardian, May 25, 2013. 142. Whereas al-Shabaab recruiting has dwindled since 2010, presumably because of a community backlash against the organization, the Islamic State has become the primary terrorist organization recruiting from the Somali diaspora. See Jack Healey, “For Jihad Recruits, a Pipeline from Minne- sota to Militancy,” New York Times, September 6, 2014, http://www.nytimes.com/2014/09/07/us/ for-Jihad-recruits-a-pipeline-from-Minnesota-to-militancy.html. 143. Martha Crenshaw and Gary LaFree, Countering Terrorism: No Simple Solutions (Washington, D.C.: Brookings Institution Press, 2017), p. 69.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 77

ªrst and most commonly cited case is that of the Pakistani-American terrorist Faisal Shehzad, who tried but failed to bomb Times Square in May 2010. Media accounts show that his radicalization preceded the U.S. drone campaign in Pakistan and Afghanistan.144 Shehzad was reportedly angered by the West’s “crusade against Islam,” exempliªed (in his view) by U.S. foreign policy in the Middle East, Israel’s treatment of the Palestinians, and the publi- cation of satirical cartoons of the Prophet Muhammad in Denmark. In 2009, Shehzad abandoned his family in the United States and traveled to Pakistan, where he was trained in bomb making by the TTP.145 In court testimony, Shehzad justiªed his actions as revenge for a range of U.S. policies, including the U.S. occupation of Iraq and Afghanistan, as well as drone strikes in Pakistan, Somalia, and Yemen.146 In a video recorded prior to the failed attack, he justiªed jihad as a fundamental pillar of Islam and vowed to avenge “the killing of Baitullah Mehsud and Abu Musab Zarqawi [the Jordanian militant who led al-Qaida in Iraq], and for all the weak and oppressed and martyred among the Muslims.”147 Baitullah was, indeed, killed in a drone attack in Pakistan, but Zarqawi died in a U.S. airstrike in Iraq. In other words, Shehzad’s turn to terrorism was driven not by U.S. drone or airstrikes, but by the U.S. government’s more general policy of assassinating foreign adversar- ies. The United States is unlikely to forgo this policy, at least in the near future, even if it were to prioritize the use of political and diplomatic tools in ªghting the “war on terror.” Hence, stopping the use of drones would not have stopped Shehzad from trying to carry out his plot. And from a counterfactual perspective, Shehzad would most likely have become a terrorist even in a world without drones. The second case involved Najibullah Zazi, an Afghan immigrant arrested by the Federal Bureau of Investigation in September 2009 for conspiring to bomb the New York subway system.148 According to his court testimony, Zazi went to Pakistan with his co-conspirators in September 2008 with the ultimate goal of traveling to Afghanistan to join the Taliban in their ªght against U.S- led NATO forces. While they were transiting through the Pakistani city of Peshawar, an al-Qaida cell recruited them for a “martyrdom operation” in the

144. Andrea Elliott, Sabrina Tavernise and Anne Barnard, “For Times Sq. Bomber, Long Roots of Discontent,” New York Times, May 15, 2010. 145. David Tan, “Case 34: Times Square,” in John Mueller, ed., Terrorism since 9/11: The American Cases (Columbus: Mershon Center, Ohio State University, 2011), p. 515. 146. See United States of America vs. Faisal Shehzad, Defendant Plea before U.S. District Court, case 10-CR-541 (MGC), Southern District of New York, June 21, 2010, p. 8, http://online.wsj.com/ public/resources/documents/06l1shap.pdf. 147. The video, released by the Al-Arabiya news channel, is available at https:// www.youtube.com/watch?vϭMang4w4SbZo. 148. Daniel Klaidman, Kill or Capture: The War on Terror and the Soul of the Obama Presidency (New York: Houghton Mifºin Harcourt, 2012), p. 119.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 78

United States. Zazi told the judge at his plea hearing that he was ready to kill himself “to bring attention to what the United States military was doing to civilians in Afghanistan.”149 Neither Zazi’s court statement nor secondary accounts of his path to terrorism make any mention of drones.150 Some scholars and analysts argue, however, that drone strikes in Afghanistan did take him down this path.151 But while they could have been a factor, they were not the most important one. To begin, Zazi’s radicalization was reportedly inspired by YouTube videos of the pro-al-Qaida Indian Muslim televangelist Zakir Naik, whose call for jihad does not involve grievances linked to drones or other targeted killings.152 Second, Zazi had already started conspiring with his partners before the start of the United States’ extensive deployment of armed drones in Afghanistan in 2008.153 Although these two cases may suggest the existence of a wider theater of blowback, there is no credible evidence of systematic transnational blowback in response to U.S. drone strikes. Further, these home-grown militants were not driven by the same grievances. Indeed, one of Zazi’s co-conspirators claimed in court that he wanted to bomb New York because of the “Zionist Jews who, I believe, run a permanent government in the United States.”154 What he and those described above share is an aversion to U.S. foreign policy, including the United States’ occupation of Muslim countries such as Iraq and Afghanistan or its backing of Israel.155 For example, one of al-Qaida’s original objectives was to wage jihad against U.S. troops in bin Laden’s native Saudi Arabia. As a declassiªed 2011 intelligence assessment by the Federal Bureau of Investigation concludes, “A broadening U.S. military presence overseas” was among the factors behind the growing number of U.S.-based Islamic extremist plots between 2001 and 2010. The report pointed to the “perception that the United States is at war with Islam and jihad is the correct and obligatory re-

149. United States of America vs. Najibullah Zazi, Transcript of Criminal Case for Pleading, case 1:09-cr-00663, United States District Court, Eastern District of New York, February 22, 2010, p. 28, https://www.cbsnews.com/htdocs/pdf/USvZazi.pdf. 150. See, for instance, Peter Bergen, United States of Jihad: Investigating America’s Homegrown Terror- ists (New York: Crown, 2016), pp. 113–118. 151. Gerges, “Why Drone Strikes Are Real Enemy in ‘War on Terror.’” 152. Michael Wilson, “Najibullah Zazi: From Smiling Coffee Vendor to Terror Suspect,” New York Times, September 25, 2009. 153. Drones were introduced into combat in Afghanistan at the start of Operation Enduring Free- dom in October 2001, but were rarely used until 2008. See “If Drone Strikes Continue in Afghani- stan, Lack of Transparency Must Not” (London: Bureau of Investigative Journalism, October 16, 2014), https://www.thebureauinvestigates.com/opinion/2014-10-16/comment-if-drone-strikes- continue-in-afghanistan-the-lack-of-transparency-must-not. 154. John Marzulli, “Najibullah Zazi Cohort Zarein Ahmedzay Admits Plot Role,” NY Daily News, April 23, 2010. 155. “Terrorism in America after 9/11” (Washington, D.C.: New America Foundation, n.d.).

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 79

sponse.”156 Ultimately, as Martha Crenshaw and Gary LaFree argue, “individuals follow different pathways to the development of violent intentions and plans for action,” and “there are few clear patterns other than that most perpe- trators are young men who are American citizens or residents ...fromdiverse immigrant family backgrounds.”157

somali americans, radicalization, and terrorism My examination of secondary accounts of the pathways to radicalization of more than thirty Somali Americans charged or convicted in terror-related offenses in the United States provides little or no reason to accept the argument that their decision to materially support or engage in armed violence was the result of U.S. drone strikes or other forms of targeted killings.158 The case of militant recruitment in the Somali-American community in Minnesota, the largest in the United States, sheds light on whether the backlash against drone strikes has transnational reach. Having escaped a civil war in their country in the 1990s, Somali Americans face particular challenges of assimilation compared to most other U.S. Muslim immigrant communities.159 Studies of Somali-American youth in Minnesota show that the decision to become a violent extremist is shaped by a combination of factors, such as an “identity crisis” born out of problems associated with assimilation that can be worsened by the reportedly aggressive and intrusive actions of U.S. authorities, as well as the physical or social media presence of terrorist recruiters who can exploit these vulnerabilities.160 For instance, al-Shabaab recruiters typically rallied young Somali-American men by appealing to their Somali origins and urging them to defend their original homeland by ªghting against U.S.-backed Ethiopian troops.161 A group of Somali men successfully prosecuted in the United States for providing material support to al-Shabaab explained their decision to join the group as a way

156. See “Decade in Review: Self-Selecting U.S. Persons Drive Post-2006 Increase in Anti-U.S. Plotting,” intelligence assessment (Los Angeles: Federal Bureau of Investigation, March 7, 2011), https://www.brennancenter.org/sites/default/ªles/SPolice%20Gen16031414460.pdf. 157. Crenshaw and LaFree, Countering Terrorism,p.95. 158. See “Al-Shabaab’s American Recruits” (New York: Anti-Defamation League, Febru- ary 15, 2015), https://www.adl.org/sites/default/ªles/documents/assets/pdf/combating-hate/ al-shabaabs-american-recruits.pdf. 159. Jessica Stern, “Muslims in America,” National Interest, May/June 2011, p. 42. 160. See Justin Heinz and Errol Southers, “Foreign Fighters: Terrorist Recruitment and Countering Violent Extremism (CVE) Programs in Minneapolis-St. Paul” (Los Angeles: National Center of Ex- cellence for Risk and Economic Analysis of Terrorism Events, University of Southern California, 2015); and Arvin Bhatt and Mitchell D. Silber, “Radicalization in the West: The Homegrown Threat” (New York: New York City Police Department, 2007). 161. Ethiopia intervened in Somalia in 2007 to defend the Somali transitional government against the Islamist insurgency.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 80

of helping rebuild war-torn Somalia. One of them, identiªed in court documents as Ahmed, claimed that his decision to “join al-Shabaab was rooted in [a] sense of responsibility and a desire to do something to help.” Another, Yusuf, stated that he was not a “violent person” or a “fanatic,” but that he was simply “motivated by an abiding belief that the people of Somalia have the right to political self-determination.” “His efforts were directed exclusively toward his homeland, Somalia, out of a profound concern for the safety, security and future of the Somalian people.”162 It is possible that the defendants made these statements in hopes of being given leniency, and that they were motivated primarily by the desire to “kill non-believers and apostates” as argued by the prosecution.163 The federal judge hearing the case, however, turned down the prosecution’s plea to sentence each of them to ªfteen years in prison (the maximum sentence for providing material support to an FTO), and reduced their sentences on the grounds that the men had joined al-Shabaab only to ªght in Somalia against Ethiopian troops.164 In addition, there seems to be no positive correlation between the U.S. drone campaign in Somalia, which was negligible until 2015 (on average, there were two U.S. drone strikes per year from 2011 to 2014), and al-Shabaab’s recruitment in the United States, which peaked in 2008–09 and has since declined sharply. While al-Shabaab is a sworn enemy of the United States and may even be able to draw some recruits by inciting or exploiting anger against U.S. drone or air strikes in Somalia, there is no concrete evidence to suggest that the group’s transnational recruitment is signiªcantly driven by them. More recently, Islamic State (ISIS) recruiters have recruited young Somali Americans to ªght in Syria and Iraq. To date, the largest ISIS conspiracy case in the United States involved nine Somali-American men, three of whom were sentenced to life in prison in November 2016 for providing material support to an FTO by plotting to ªght on its behalf in Syria. Their court testimony reveals that they were motivated to join ISIS because of personal factors, such as peer inºuence and conºicted identities exacerbated by racism or prejudice, which were duly leveraged by ISIS social media campaigns. One of the men who pleaded guilty, Zacharia Abdur Rahman, cited three factors leading to his transformation into an ISIS supporter: a racist incident when he was spat on, inspiration from a friend who had traveled to Syria to join ISIS, and his subse-

162. United States of America vs. Ahmed et al., document 339, case 1:12-cr-00661-JG-LB, United States District Court, Eastern District of New York, January 8, 2016, https://www .investigativeproject.org/documents/case_docs/2919.pdf. 163. Ibid., p. 13. 164. “Two Swedish Citizens Get 11 Years in U.S. Prison for al-Shabaab Support,” Reuters, January 15, 2016.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 81

quent moral outrage over the conºict in Syria.165 In other words, Rahman and some of his co-conspirators ªrst experienced what Quintan Wiktorowicz calls a “cognitive opening,” which can be triggered by a variety of factors, including job loss, discrimination, humiliation, or personal loss, that renders a person amenable to ideas that would otherwise be ignored or rejected. This opening can prompt a search for a “satisfactory system of meaning,” often through social ties and networks involving friends and relatives.166 Ultimately, Rahman’s transition from passively supporting ISIS to joining the organization was sealed with what he called a “hug” from ISIS recruiters on social media.167

muslims in europe Research on the causes of Muslim radicalization and terrorist recruitment in Europe underscores the importance of factors similar to those discussed above,168 including the appeal of a transnational Islamic identity to second- and third-generation Muslims that helps them overcome their conºicted identities,169 state policies on immigration and integration that marginalize Muslim communities in countries such as France,170 the role of radical imams (mosque preachers),171 and more recently, the inºuence of social media both as a facilita- tor of terror recruitment and as a source for self-radicalization.172 Still, context matters in evaluating motivations for violent jihad: factors that are important in a U.S. or European setting may not be applicable in North Africa or South Asia. And even within Europe, the conditions that promote radicalization can vary from country to country, depending on the national background of the Muslim community in question (e.g., Pakistanis in Britain, Turks in Germany, North Africans in France), which emphasize different roles

165. Mukhtar M. Ibrahim, “ISIS Trial in Minnesota: What You Need to Know,” Minnesota Public Radio, May 6, 2016, https://www.mprnews.org/story/2016/05/06/isis-trial-minnesota-faq. 166. Quintan Wiktorowicz, Radical Islam Rising: Muslim Extremism in the West (Lanham, Md.: Rowman and Littleªeld, 2005), pp. 5, 20–22. 167. Laura Yuen, “3 out of 9 Sentenced in ISIS Conspiracy Trial,” Minnesota Public Radio, Novem- ber 14, 2016, https://www.mprnews.org/story/2016/11/14/ªrst-day-of-sentencing-isis-trial. 168. For a concise review of the literature, see Anja Dalgaard-Nielsen, “Violent Radicalization in Europe: What We Know and What We Do Not Know,” Studies in Conºict and Terrorism,” Vol. 33, No. 9 (2010), pp. 797–814, doi:10.1080/1057610X.2010.501423. 169. See, for example, Olivier Roy, Globalized Islam: The Search for a New Ummah (New York: Co- lumbia University Press, 2004), pp. 303–309; and Farhad Khosrokhavar, Suicide Bombers: Allah’s New Martyrs (London: Pluto, 2005). 170. Peter R. Neumann and Brooke Rogers, “Recruitment and Mobilization for the Islamist Mili- tant Movement in Europe” (London: International Center for the Study of Radicalization and Po- litical Violence, King’s College London, 2007). 171. Gilles Keppel, The War for Muslim Minds: Islam and the West (Cambridge, Mass.: Belknap Press of Harvard University Press, 2004). 172. Neumann and Rogers, “Recruitment and Mobilization for the Islamist Militant Movement in Europe,” p. 83.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 82

for religion in political and social life; the history of immigration (recent immigrants in Southern Europe vs. second- and third-generation immigrants in the United Kingdom and France); and variation in national policies on immigration and integration.173 Even if drone strikes do play a part in generating anti- U.S. resentment among some Muslims around the world, their motivations can vary and it is not easy to pinpoint the source of blowback. Moreover, transnational terror groups can ªnd plenty of material for recruitment in U.S. national security policies. For instance, a 2016 al-Shabaab propaganda video exploited the Trump administration’s “Muslim” travel ban to rally support for its cause in the United States.174 As Daniel Byman notes, “The problem is that the list [of grievances] is long, and it is hard to tell if one grievance would simply be replaced by another in the mind of an angry, idealistic, and excitable young volunteer. Instead of hating America or the West for ten reasons, they now hate the West for nine.”175

Conclusion

I have evaluated the validity of the drone blowback thesis, which posits that drone strikes generate a backlash effect by creating more terrorists than they eliminate at the local, national, and international levels. At the local level, the logic of the blowback thesis is that family members of those killed or wounded in drone strikes will mechanically follow tribal norms to avenge their relatives by joining anti-American militant groups such as the Pakistani Taliban. Based on my interviews with a diverse group of respondents from North Waziristan, I argue that the notion that drone strikes turn aggrieved relatives into blood- thirsty militants is deeply problematic because it essentializes an entire ethnic group and reduces their choices to primordial urges, mores, and customs. In my interviews, experts with deep knowledge of the Federally Administered Tribal Areas, especially North Waziristan, in addition to well-informed local elders and others deeply embedded in social networks, reject the claim that drones provide a recruitment card for militant organizations. Even though the sample is not scientiªcally drawn, at a minimum, it reºects the position of many of those who have been directly affected by drone strikes yet whose voices have gone unheard in the debate on drones. Clearly, this article does not

173. Ibid., pp. 27–28. 174. “Al-Shabaab Releases Recruitment Video Featuring Donald Trump,” ABC News, January 1, 2016, https://www.youtube.com/watch?vϭZ9Rr3kD75uM. 175. Daniel Byman, “Thoughts on Counterterrorism and Blowback,” Lawfare blog, Novem- ber 30, 2014, https://www.lawfareblog.com/foreign-policy-essay-thoughts-counterterrorism-and- blowback.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 Do U.S. Drone Strikes Cause Blowback? 83

offer the ªnal word on the effects of drone strikes. It does, however, offer a cor- rective to the literature that has essentially ignored those who should be the most well informed about the impact of drone strikes. Proponents also claim that blowback occurs at the national level in areas not directly targeted by U.S. drone strikes. This is because the people in these areas have not directly witnessed or experienced the effects of drone strikes. Hence, they may be more susceptible to believe anti-U.S. state or militant propaganda in the media that portrays the ºying of U.S. drones as an infringement on national sovereignty and/or as the cause of large numbers of civilian deaths. My interviews with scholars, journalists, and counterterrorism ofªcials from Pakistan, as well as an assessment of a Pakistani police survey of 500 detained terrorists, contradict this argument. Most experts I interviewed maintain that militant groups outside FATA mobilize support and attract recruits by blaming or demonizing a range of domestic and international enemies, including India/Hindus, the West, the United States, and, in some cases, minority Muslims such as Shias. Even the few experts who argue that militant groups have used drones as a propaganda tool claim that drone strikes are not the most important factor in driving recruitment, which instead beneªts from state policies that nurture select militant groups. Although U.S. drone strikes in Muslim countries might play a role in the radicalization or violent extremism of co-religionists around the world, the available evidence does not support the assertion that drones are “fueling the ªres of homegrown radicalization” in Western societies or that these unmanned aerial vehicles are the new Guantánamo. Like the local and national blowback arguments, this argument is based primarily on anecdotal evidence and statements of individual terrorists. Transnational terrorist groups such as al-Qaida have sought to use U.S. drone strikes to incite violence against civilians in, for example, the United States and Europe. Islamist terrorists convicted in the United States do not, however, seem to be solely or even mainly motivated by drone strikes. Instead, they typically express their main grievances in terms of a broader antagonism toward the United States’ foreign policy in the Middle East and its occupation or military presence in Muslim- majority countries such as Iraq or Afghanistan. In other words, militants may use drone strikes as evidence of the West’s “war against Islam,” but it is not their main motivation. The radicalization of individuals in the Somali diaspora, especially in the state of Minnesota, illustrates the role of U.S. policies targeting Muslim communities at home; personal factors, such as conºicted identities among young Somali Americans; and their nationalist desire to evict foreign troops from Somalia, a desire that recruiters from the al-Qaida afªliate al-Shabaab

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021 International Security 42:4 84

have been able to exploit. My brief examination of the social science literature on the drivers of Islamist militancy among Muslims in European countries similarly points to domestic factors such as an identity crisis among some young Muslims, state policies of marginalization and discrimination, and the role of radical preachers and terrorist recruiters who leverage these vulnerabilities for recruitment. In conclusion, the blowback thesis offers a simplistic, monocausal explana- tion of a complex process of radicalization and jihadist recruitment at the local, national, and transnational levels. To develop a better understanding of these activities, scholars and other analysts should begin by focusing their efforts not on anecdotal accounts, but on data-driven empirical evidence. In addition, the argument and ªndings of this study should not be viewed as a license for future U.S. drone strikes. President Trump’s reported decision to expand the powers of the CIA for carrying out drone strikes and to relax Obama-era re- strictions on counterterrorism operations by both the CIA and the military is likely to increase the risk of civilian casualties in conºict zones. Beyond the ob- vious need for transparency when conducting drone strikes, the United States should do its utmost to protect civilians from harm in accordance with international legal and normative standards.

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/isec_a_00312 by guest on 01 October 2021