Article Review 108 on Aqil Shah. “Do U.S. Drone Strikes Cause Blowback? Evidence from Pakistan and Beyond.”17 min read

An MQ-9 Reaper taxis after a mission in Afghanistan.

Beware the grim reaper?

Analyses of drones often generate more heat than light, but Aqil Shah’s article is a welcome change. Shah argues U.S. drone strikes do not cause “blowback” in Pakistan or anywhere else, basing his claims primarily upon field interviews conducted in Pakistan. As he summarizes, “I find no evidence of a significant impact of drone strikes on the recruitment of militants either locally or nationally” (49).

H-Diplo | ISSF Article Review 108

H-Diplo/ISSF Editors: Diane Labrosse and Seth Offenbach
H-Diplo/ISSF Web and Production Editor:  George Fujii

Aqil Shah.  “Do U.S. Drone Strikes Cause Blowback?  Evidence from Pakistan and Beyond.”  International Security 42:4 (Spring 2018):  47-84.  DOI:

Review by Ahsan I. Butt, George Mason University

Published by ISSF on 6 February 2019

Analyses of drones often generate more heat than light, but Aqil Shah’s article is a welcome change. Shah argues U.S. drone strikes do not cause “blowback” in Pakistan or anywhere else, basing his claims primarily upon field interviews conducted in Pakistan. As he summarizes, “I find no evidence of a significant impact of drone strikes on the recruitment of militants either locally or nationally” (49).

Shah is to be commended for intrepidly tackling a question with deep scholarly, policy, and ethical implications. His point that discussions about drones should be grounded in hard empirical rigor is well taken. Most of all, he makes a significant contribution in highlighting pro-drone voices, especially among Pashtuns in Pakistan’s tribal agencies (formerly Federally Administered Tribal Areas), that have generally been unheard in the cantankerous debate about the ethics and efficacy of drones. By showing that drones are more popular than conventional wisdom would have it in former FATA, the region where they are most active, Shah’s intervention is an important corrective that will likely be widely read and cited.

Yet for all its strengths, Shah’s paper suffers from a pair of flaws, one theoretical and the other empirical. First, it does not conceptualize “blowback,” making it more challenging to evaluate the central argument. “Blowback” originally referred to any unintended consequences of covert and clandestine missions. Without being explicit, Shah seems to adopt a higher standard: more terrorists must be created than curtailed for blowback to occur. This distinction is important because while Shah is persuasive in showing that drones can be a very effective counterterrorism tool, it is more problematic to claim that drones do not incite blowback per se. Even while one acknowledges the drone program’s success rate in killing and restricting major militant figures—from Baitullah Mehsud (the leader of the Pakistani Taliban) to Mullah Mansour (the leader of the Afghan Taliban)—it has come at a price, perhaps a worthwhile cost but a cost nonetheless. A subset of those costs can reasonably be described as blowback at each of the levels he analyzes: local, national, and transnational.

Second, Shah repeatedly overstates the difference in quality between his data and that of his opponents. The strong and persistent critiques of the motives and methods of journalistic, polling, and analytical organizations such as the Bureau of Investigative Journalists, the New America Foundation, Pew, and others belie the reality that all data regarding the military and social effects of drones lies on a spectrum of imperfection. Shah’s data itself suffers from important if understandable flaws, and the drone debate would be healthier if Shah extends the same generosity to foils that he expects from his readers.

Shah’s argument, taken on its own terms, is strongest at the local level. There is decreasingly little doubt that drones are quite popular in the areas of Pakistan where they are most active, such as North Waziristan. Perhaps the greatest strength of the paper is that Shah gives voice to tribal Pashtuns who are favorably disposed to drones. Their hitherto absence from the scholarly conversation is deeply problematic, both because it denies them agency and it means that anti-drone views in Pakistan are overrepresented. If a sizable proportion of residents of the areas most afflicted by drones are more favorable towards them than the conventional wisdom would guess (51-53), and more approving of them than their compatriots in the four “settled” provinces, then it is an important endeavor to bring those views to the surface. Shah’s argument is burnished by the impressive number of interviews he conducted (n=167), his fluency in the Pashto language, and the forceful and unambiguous quotes he cites.

But even here, at its strongest, Shah’s article suffers from the central conceptual and empirical issues highlighted above. Conceptually, one can concede that drones have not incited locals in former FATA into anti-American militancy, but that signifies only that drones are worth it for the U.S., not that there is no blowback. Locals in former FATA did indeed suffer the consequences of America’s drones:

“As the agency’s operations in Waziristan came to resemble an air war, they changed life on the ground. The strikes spurred militants to try to identify spies who might have betrayed them. Around North Waziristan’s main towns, Miranshah and Mir Ali, which took the brunt of the C.I.A. attacks, paranoia spread. The Taliban blamed local maliks who had long presided over the area’s economy—smuggling, arms dealing, mining, and government contracting. Taliban gunmen seeking control of local rackets executed maliks and their family members in the hundreds.”[1]

The Taliban also tortured such maliks into confessing that they spied on behalf of the C.I.A. By Shah’s definition, the torture and murder of hundreds of maliks and their families is not blowback because no new anti-American terrorists have been created. This is surely a problematic conclusion.

Empirically, Shah’s point that drones are more popular than many would expect in North Waziristan is well taken. But he need not disparage data that show otherwise. It would be more useful to accept those alternate findings, albeit with a grain of salt, as conflicting results in a complicated war.

For instance, Shah criticizes a 2010 New America survey, which found over 70% FATA residents “strongly opposed” drones, for two reasons: (i) it is not representative because “a sizable portion of FATA’s population was internally displaced,” and (ii) the possibility of “social desirability bias” (53). The first is a strange criticism, given that Shah’s study is based entirely on a convenience sample of a displaced population. If non-representative surveys of displaced populations are inherently suspect, why does Shah base his thesis on the same? More importantly, the findings of this poll are broadly consistent with other polls of FATA residents across time.

Study Year Percentage in favor Percentage against
New America 2010 22 76
CAMP[2] 2010 29 59
CAMP 2011 6 63
Yousaf[3] 2017 27 69


As for social desirability bias, this could infect literally any in-person survey, not least Shah’s. Moreover, given his repeated criticism of other polls (e.g. Reprieve’s) for surveying FATA residents outside FATA (54), it would behoove him to be forgiving of whatever issues arise from polling inside FATA. Rather than describing all surveys and studies that contradict his thesis as suspect, it would be fairer for Shah to say that the picture of public opinion on drones in FATA is mixed, and that some areas (e.g. North Waziristan) may be quite pro-drone while others are less so. Any claim stronger than that should be made with care, at least given this data.

At the national level, meanwhile, Shah seeks to debunk the argument that a nationalist backlash, stirred by anger over civilian casualties and violations of Pakistan’s sovereignty, has increased the terrorist threat faced by the United States. To do so, Shah adopts a kitchen-sink strategy, entailing

  • Questioning the extent to which drones kill civilians
  • Questioning how unpopular drones are in Pakistan
  • Questioning whether the unpopularity of drones caused increased disapproval of the U.S.
  • Questioning whether disapproval of the U.S. caused a graver militant threat

In other words, Shah does not concede a single leg of the argument. Notably, however, only on the fourth count does the article have a solid case.

First, Shah writes that “to evaluate the blowback thesis at the national level, it is necessary” to know how many civilians are killed by drones. At the outset, it is difficult to understand why the number of civilian deaths is relevant to the blowback thesis. Are we to infer that a drone strike that kills 32 civilians will elicit more blowback than one that kills 23? If assessing backlash against the U.S., surely the pertinent factor is simply noting that drones do indeed kill civilians, and sometimes dozens of them, as in a 2011 strike at Data Khel that Shah briefly discusses. Even stipulating that blowback is tied to casualty figures, Shah’s warnings against imbibing statistics from organizations such as the Bureau of Investigative Journalism or the New America Foundation ring as over cautionary. The two organizations’ estimates are reasonably close—between 255 and 315 for the latter, and within a range of 424-969 for the former (67)—suggesting that despite all the grey in the data, we can discern that, give or take, there is about one civilian death per drone strike in Pakistan (n=429) (48).

Second, Shah questions just how unpopular drones are. In disputing the national unpopularity of drones, Shah provides no survey data of his own and again is overly critical of alternative data. For instance, he admonishes Pew’s polls for being unrepresentative because Pew’s samples are “disproportionately urban, whereas Pakistan is a predominantly rural country” (68). It is hard to understand this criticism because Shah himself footnotes that Pew weights its results, a standard technique for pollsters. Substantively, Pakistan is no longer a rural country and has not been for at least a decade.

He further claims that “Pew polls typically captured the opinion of a minority of the country’s population about drones” because of a “very high frequency of ‘don’t know’ and ‘did not answer’ responses” (68). This is incorrect. There have been four Pew polls that have questioned Pakistanis nationally on drones: 2010, 2012, 2013, and 2014. Only in the first of these could it be claimed that Pew captured just the minority’s views. By 2012, the percentage of people that had heard “a lot” or “a little” about drones was 55%. Of that 55%, a whopping 97% considered them “a bad thing” and 94% agreed that drones “kill too many innocent people.”[4] In 2013, the “don’t know/refused to answer” proportion fell further still, to 27% of respondents, with 68% of respondents overall (93% of those that answered) disapproving of drones.[5] 2014’s figures were remarkably similar to the previous year’s: a no response rate of 30%, an overall disapproval of 66%, and approval of only 3%.[6] Drones were and are deeply unpopular in Pakistan.

Third, Shah questions the connection between the unpopularity of the drone war and disapproval of the U.S. Referring to polls from 2002, the article misleadingly suggests that there was a consistent level of anti-Americanism from 9/11 to President Barack Obama’s escalation of the drone wars (72). To the contrary, Pakistani (dis)approval of the U.S. has ebbed and flowed, with three post-9/11 inflection points: 2006/7, 2010/11, and 2013/4.

Year “Very favorable” or “somewhat favorable” “Very unfavorable” or “somewhat unfavorable”
2004 21 60
2005 23 60
2006 27 56
2007 15 68
2008 19 63
2009 16 68
2010 17 68
2011 12 73
2012 12 80
2013 11 72
2014 14 59
2015 22 62


There is a strong association between escalation of the drone wars and (dis)approval of the U.S. For instance, there was a sharp fall in U.S. popularity following the drone program becoming widely publicized in 2006, thanks to a horrific strike in Bajaur that, according to the New America Foundation, killed between 79 and 81 civilians. 2010 saw Obama more than double the previous year’s drone strikes; like clockwork, by the next year (2011), U.S. approval suffered. In 2013, Obama drastically reversed course, launching only 26 drone strikes, half of his administration’s previous low. Sure enough, the next year America was less unpopular.[7] Correlation is not causation, but these numbers reaffirm that drones were a key driver of anti-Americanism between 2006 and 2015.

Finally, Shah uses expert testimony, especially from security officials, to debunk the connection between drone-inspired anti-Americanism and the creation of local militants. Shah is correct to note that sectarian, ethnic, class, and local fissures have been the common dimensions along which recruits have been motivated to join militant groups, rather than anti-Americanism. But setting aside the conceptual issue of blowback being conflated with the creation of terrorists—as discussed above, this is a narrow definition that avoids accounting for national developments such as the shooting popularity of strident drone opponent and recently elected Imran Khan—another problem arises here, that of the known unknown. We cannot infer how many militants have been created at time t by counting acts of militancy at t. The CIA’s sponsorship of the Afghan jihad in the 1980s did not witness the manifestation of blowback until 2001, but blowback had set in much earlier—we just did not see it. Similarly, it may be that one or a number of those Pakistanis strongly disapproving of both the U.S. and its drones have been radicalized to the point of militancy, and we are just waiting for the smoking gun. The deep unpopularity of the U.S. in Pakistan, spurred in part by its laissez-faire approach to drones, means that one cannot discount such a possibility.

Shah’s final arrow is aimed at the transnational blowback thesis, which states that Muslims in the west “may become radicalized or retaliate violently” thanks to America’s drone wars (75).

There are two main empirical issues here. First, if Shah’s findings of drones being of little importance for Somalian militant recruitment withstand President Donald Trump’s dramatic escalation in that country, that will be interesting. At this juncture, however, Somalia is an easy case. After all, Somalia has seen the fewest drone strikes of any country. While the U.S. has conducted over 400 drone strikes in Pakistan and almost 300 in Yemen,[8] Presidents George W. Bush and Obama only ordered 41 strikes between them in Somalia.[9]

Second, and more importantly, the article is extremely ungenerous to alternative explanations. Even when there is a clear-cut and documented case of a terrorist citing drones as his (partial) motivation, that of Faisal Shehzad in 2009, Shah muddies the waters. At his arraignment, Shahzad said that “I want to plead guilty and I’m going to plead guilty 100 times forward because until the hour the US pulls its forces from Iraq and Afghanistan and stops the drone strikes in Somalia and Yemen and in Pakistan and stops the occupation of Muslim lands and stops Somalia and Yemen and in Pakistan, and stops the occupation of Muslim lands, and stops killing the Muslims and stops reporting the Muslims to its government, we will be attacking US, and I plead guilty to that.”[10] In a later exchange, when the prosecutor interrupted Shehzad and asked him to consider “the people who were walking in Times Square that night…Did you look around to see who ‘they’ were?” Shehzad’s response was instructive.

“Well, the people select the government. We consider them all the same. The drones, when they hit…”

“Including the children?” the judge interrupted Shahzad once again.

There was a long pause.

“Well, the drone hits in Afghanistan and Iraq,” he finally said, “they don’t see children, they don’t see anybody. They kill women, children, they kill everybody. It’s a war, and in war, they kill people. They’re killing all Muslims.”

Here is a terrorist repeatedly citing drones as his motivation, albeit as part of a panoply of grievances against the United States. Yet Shah argues that: “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” (77).

If the article’s standards for confirming evidence of the blowback thesis are so exacting, it is no wonder that no instances are found. For Shah, it is not enough for terrorists to claim they were motivated by drones. It has to be shown that drones specifically, rather than U.S. policy more generally, led to someone taking up arms against the U.S. But one cannot isolate drones from U.S. militarized behavior in the Muslim world. Drones are one of many American methods of delivering death, and it would be similarly fanciful to find airtight evidence of a terrorist being motivated by a Howitzer gun or F-16.

Interestingly, Shah concludes by arguing that “even if drone strikes do play a part in generating anti-US resentment among some Muslims around the world, their motivations can vary and it is not easy to pinpoint the source of the blowback” (82). If terrorists’ motivations are so manifold and varied that isolating any one cause is a fool’s errand, it would suggest that the question in Shah’s title—Do U.S. drones cause blowback?—can elicit only one answer.

Overall, Shah has written an important article that deserves to be read widely by scholars, analysts, and policymakers concerned with counterterrorism, especially in the Pakistani context. The article has much to recommend it, especially in giving voice to pro-drone Pashtuns in North Waziristan. It would be easy to wholeheartedly endorse the article if Shah’s argument was merely that drones have been a highly effective tool in the fight against Islamist militants in Pakistan’s northwest—itself a reasonably controversial and important claim.

But Shah does more. He sets the bar for “blowback” too high—a net uptick in numbers of terrorists—thus downplaying the costs of the U.S. drone program. Even when a terrorist openly and repeatedly claims to be motivated by drones, Shah does not consider it blowback.  He suggests that the impact of drones cannot be separated from (perceptions of) US intervention and war in the Muslim world more generally. This is a fair enough argument. But if it is not possible for social scientists to disentangle the effects of drones from other instruments of the American war machine, then why circumscribe a question about blowback to drones in the first place? Regardless, the drone debate can only be settled with more and better data, and thanks to his field interviews Shah’s article is an important step towards establishing such a consensus.


Ahsan I. Butt is an Associate Professor at the Schar School of Policy and Government at George Mason University and a nonresident fellow at the Stimson Center. He is the author of Secession and Security: Explaining State Strategy Against Separatists (Cornell University Press, 2017) as well as journal articles that have appeared or are forthcoming in Journal of Global Security Studies, Journal of Strategic Studies, International Organization, Politics and Religion, and Security Studies.

©2018 The Authors | Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License



[1] Steve Coll, Directorate S: The C.I.A. and America’s Secret Wars in Afghanistan and Pakistan (New York: Penguin Press, 2018), 534.

[2] Naveed Ahmad Shinwari, Understanding FATA: Attitudes Towards Governance, Religion & Society in Pakistan’s Federally Administered Tribal Areas Volume V (Islamabad: CAMP, 2012), 88,

[3] Farooq Yousaf, “CIA Drone Strikes in Pakistan: History, Perception, and Future,” CRSS (2017), 1 December 2017. DOI:

[4] Pew Research Center, “Pakistani Public Opinion Ever More Critical of U.S.—Chapter 1. Views of the U.S. and American Foreign Policy,” 27 June 2012,

[5] Pew Research Center, “America’s Global Image Remains More Positive than China’s: But Many See China Becoming World’s Leading Power,” 18 July 2013,

[6] Pew Research Center, “Global Opposition to U.S. Surveillance and Drones, but Limited Harm to America’s Image: Many in Asia Worry about Conflict with China,” 14 July 2014,

[7] Pew Global Attitudes & Trends Question Database, “Question: Please tell me if you have a very favorable, somewhat favorable, somewhat unfavorable, or very unfavorable opinion of…the United States [844],”[email protected]@36.5-&stdIDs_Pew, accessed 24 January 2019.

[8] Peter Bergen, David Sterman, Melissa Salyk-Virk, Alyssa Sims, and Albert Ford, “Drone Strikes: Yemen,” New America,, accessed 18 September2018.

[9] Bergen, Sterman, Salyk-Virk, Sims, and Ford, “Drone Strikes: Somalia,” New America,, accessed 18 September 2018.

[10] Lorraine Adams and Ayesha Nasir, “Inside the mind of the Times Square bomber,” The Observer, 18 September 2010,