Home Basing of the MQ-9 Marine Unmanned Aerial Vehicle Squadron and KC-130J Marine Aerial Refueler Transport Squadron, Marine Corps Base (MCB) Hawaii Kaneohe Bay.
Aloha Friends, I have just submitted the following comment to the military (deadline has been extended to Sept. 21). Feel free to use all or part of none of it in your own activism. Thank you for all your great work. –Koohan
Draft Environmental Assessment, Home Basing of the MQ-9 Marine Unmanned Aerial Vehicle Squadron and KC-130J Marine Aerial Refueler Transport Squadron, Marine Corps Base (MCB) Hawaii Kaneohe Bay.
COMMENT
Koohan Paik-Mander
August 30, 2022
The Draft EIS to home-base a squadron of six Reaper drones and 15 refueling aircraft exemplifies the tragedy of dystopian America, when public moneys by the tens of billions are being poured into efforts that put all of humanity and life on Earth at risk. It is part of a grand, diabolical experiment to conduct warfare through robots and artificial intelligence (AI). And Hawaii, the land of aloha, is heart-wrenchingly “ground zero” for the Grand Experiment that is on track to pit algorithm against algorithm. Once the battling algorithms escalate into high-stakes stages marked by nanoseconds, the Experiment will have ended, and we will weep for the untold suffering caused, and if we are still alive, swear — once again — to never let history repeat itself.
Oh—did I mention? Each MQ-9 Reaper drone costs the public over $1.6 billion apiece.
We are careening into a free-fall arms race with China and Russia. No one benefits except the weapons dealers like Lockheed Martin and General Atomics, which indeed profit richly and criminally. It’s time to slam on the brakes. Follow the clarion call for AI-arms control by Germany Foreign Minister Heiko Maas. Nations of the world can and must work toward a global treaty to ban lethal autonomous weapons.
Favor the NO ACTION alternative in the Draft EIS.
The awkward and flimsy Draft EIS exists to supposedly ensure democratic process and environmental oversight. But how can democracy be served with a document as non-transparent as this Draft EIS? It gives no hint of the implications of waging networked, AI-driven war that sees Hawaii – and the entire Pacific hemisphere – as a geography in which the U.S. can do whatever it pleases, regardless of impacts to environment and communities.
THE JADC2:
FOSTERING A SYMBIOTIC RELATIONSHIP BETWEEN THE MQ-9 AND THE OTHER COMPONENTS OF THE “KILL CHAIN”
The Draft EIS fails to mention the Joint All Domain Command and Control concept, or “JADC2,” in which the Marines squadron of MQ-9 Reaper Drones will play a key role.
The JADC2, still in development, will be a scalable, AI-driven, networked system of distributed warfare. Its development is of paramount importance to the Pentagon. It is intended to be the foundation for 21st century warfare. It involves much more than overlaying new technologies and hardware over existing force structures; it is a process of far-reaching, disruptive change. The Draft EIS must reflect this.
For the Marines’ part, it is positioning the MQ-9 Reaper drone at the center of its vision to integrate seamlessly with the other forces of the military, by seeking to “develop multi-axis, multi-domain precision fires organic at all echelons, enabled by a federated system of networks to ensure all elements can fight in a degraded command and control environment.” This is according to Force Design 2030, the “vision” to restructure the Marine Corps for the 21st century.
To put it more simply, General Eric Smith, Assistant Commandant of the Marine Corps, has described, “[The MQ-9] is an airborne “quarterback” to pass data, because when we are cut off from the space layer for short periods of time in a maritime environment or any environment, we have to be able to work inside that bubble to pass data back to our navy and joint partners. Back to an Aegis system or back to an Air Force passing fighter.”
Smith continues, “The drone piece, it’s both AI, the ability to use a drone for spotting, though that’s five years ago, Now it’s a matter of using the algorithms that connect what you see to Joint All Domain Command and Control, which is something the Department of Defense works on daily so that every sensor on the battlefield is fused to then provide that target-quality data to the best possible shooter (italics mine), be it a HIMARS launcher, be it a fighter, be it a bomber.”
What Smith explains above is the operative kernel to how the squadron of six MQ-9s are intended to function for the Marines. Yet this essential information has been omitted from the Draft EIS. While the Draft EIS blandly states that the Reaper will conduct “persistent intelligence, surveillance, and reconnaissance,” it neglects to include that it will transmit data of a kind that is less than benign. Actually, its key role is to transmit “target-quality” data at an early and integral stage in the “kill chain.” The network itself is the instrument of lethality, just as an orchestra is the instrument of symphonic music. It is the network of converged weapons that should be assessed, not just the MQ-9, which is only one component of it. To assess only the MQ-9 is deceptive, rendering the current Draft EIS incomplete and nontransparent — yet this is what has been generated for the public. America deserves better.
Excluding the JADC2 context from an impacts-assessment of the proposed Reaper Drone squadron is a glaring omission of the Draft EIS. It is like trying to assess impacts of a quarterback’s performance without ever mentioning football. Or like assessing the performance of an orchestral instrument without ever mentioning music. They are inextricably linked and should be assessed as such. Not doing so is segmentation of the full proposal and its impacts.
Because the JADC2 operates on electromagnetic frequencies, bandwidth and electromagnetic frequency concerns should also be included in the EIS. How will civilian bandwidth be affected? What are the sources of electromagnetic frequencies for any given scenario, especially those that will be used and practiced in Hawaii? Please cite studies on the impacts of these frequencies on Hawaii’s birds, insects and other biodiverse wildlife. What method of wireless communication will be used to transmit underwater, to submarines? If these communications networks will impact cetaceans, corals, turtles or other endangered sea creatures, please cite studies.
THE CULTURAL, SOCIAL AND ENVIRONMENTAL IMPACTS OF DISRUPTING HUMANITY
It is often said that the first revolution in warfare was gun powder, the second was nuclear weapons, and the third revolution is the present development of AI-driven, networked warfare. That is how game-changing the metamorphosis, now underway, is.
For any technology this existentially disruptive, the EIS must conduct an exhaustive examination of the MQ-9’s widely diverse functions, as the MQ-9 fulfills one of the most important roles in the operation – hovering overhead for endless lengths of time, and gathering and processing data around the clock while passing along target-quality data to every other fire-able weapon in the system. This is no “normal” aircraft, and its impacts should not be assessed as such. It is all the more urgent that we wrestle with the social, political, environmental and cultural impacts RIGHT NOW, before a horrific incident of destruction takes place, made ever more likely by the experimental nature of the foolhardy pursuit to prevail with robots, machines and AI. Our instrument for this examination is the EIS. A proper EIS must assess impacts in the context of its symbiotic relationship with the other weapons in its networked system, and across the spectrum of possible mission and war-game permutations that the system is called to perform. To do anything less would not be a consideration of the cumulative impacts required by NEPA, and would also be an expression of deceptive segmentation.
For example, during RIMPAC 2022, one MQ-9 supported 63 missions, which included 25 maritime operation missions, seven personnel recovery missions, six opposition forces missions, and six intelligence, surveillance and reconnaissance missions, an amphibious assault scenario, war-at-sea and surface exercises, and sinking a decommissioned warship. They also loaded and launched 16 Hellfire missiles.
An adequate EIS must conduct studies on the environmental, social and cultural impacts of each and every exercise and operation that the MQ-9 supports now, as well as those that it is anticipated to support in future technological generations. Please rectify this glaring omission of Cumulative Impacts.
EXPERIMENTS IN KILLING ARE NOT WELCOME IN HAWAII
A May 6, 2022 article in Marine Corps Times reported that the MQ-9 will “serve as the base for a kind of ‘family of systems’.” Retired Lieutenant General Mark Wise stated in the same article, “The MQ-9 will not be the end state. There will be something after that and something after that.”
Please provide models of systems and missions that are anticipated to emerge from operation of the MQ-9, as well as from the networked warfare with which it is inextricably linked.
General David Berger, who wrote Force Design 2030, a report that describes how the Marines are being restructured, said, “We have made significant progress to date in our force design efforts. While these efforts have undeniably been productive and will inform our divestment and investment decisions going-forward, we should view them as first steps in a longer journey… We simply must have more analysis and evidence, which comes from modeling and experimentation (italics mine).”
It is more than a little disquieting that General Berger admits that Force Design 2030 (to which the Hawaii homing of the MQ-9 Marine Unmanned Aerial Vehicle and the KC-130J Marine Aerial Refueler Transport Squadrons are central) is a giant “experiment.” For many in Hawaii, the statement is downright infuriating, after the ever-forgiving Pacific has already borne the atrocities of other “experiments,” such as the atomic bombs dropped on the Marshall Islands, the battering of numerous islands used for war practice, or hundreds of thousands of injuries and deaths to whales, endangered turtles, migratory birds and other creatures.
We in Hawaii, and all peoples of the Pacific, are not testing grounds and guinea pigs. Our islands and waters are sacred and do not exist for military ravaging. We vehemently oppose the homing of Reaper drones anywhere in Hawaii can call for the NO ACTION alternative.
Given Berger’s cavalier attitude over matters of existential consequence, it becomes more important than ever to ask, and have answered, the following, and similar, questions:
What are each of the projected permutations of missions that the MQ-9 will serve? What additional hardware and software is anticipated in the realization of each of these projected missions?
ON THE PATH TO AN OBSOLETE HUMANITY
In an interview with the Center for Strategic and International Studies, General Eric Smith, Assistant Commandant of the Marine Corps, described the speeds with which current warfighting takes place:
You have to be able to fire and move immediately. You no longer have six minutes, which is a really well oiled gun crew, artillery. From pulling-the-last-round, to you’re-on-the-move. You know, six, seven minutes [means] you’re pretty well oiled, you’re good. What we have to see now is that there are autonomous, loitering munitions that are looking for that “signature,” and as soon as they see that signature, we call it a POO — a point of origin — they’ve already got lethal authority to strike that. You don’t have six minutes to move. Whereas a HIMARS (High Mobility Artillery Rocket System), you can shoot and be gone in literally seconds, less than a minute.
General Smith gives us a sense of the prized value of weapons that eradicate the limitations of time. Given that humans require time for decision-making, human involvement becomes a deficit in the new way of war. For this reason, hundreds of billions of dollars are now being spent to develop ways to eliminate as much of “cumbersome” humanity from the process as possible. What we would then be left with would be algorithms fighting algorithms, which raises the risk of escalation and its irreversible consequences.
Clearly, the use of robotics and AI in war raises a multitude of profoundly existential ethical issues, all of which must be addressed in the EIS. The first ethical issue to address is that the people of Hawaii have not had adequate public discussion on the fact that our archipelago will be one of two premiere experimentation grounds (the Mariana archipelago being the other) for these comprehensively disruptive systems of killing. Until such public fora take place, the only ethical conclusion is a NO ACTION alternative.
ON THE PATH TO ARMAGAEDDON
Given how our economic model of capitalism is prone to enabling “efficient” algorithms to highjack nearly every aspect of our lives — from booking flights to monitoring what posts gets distributed on Facebook – it is quite easy to see AI at the helm of warfare decisions, especially when human decision-making is now considered too time-consuming. For example, hypersonic interception is being designed so that artificial intelligence will actually do most of the “thinking” required to “pull the trigger.” Because the time between launch and strike of an incoming missile could be as brief as 6 minutes, it is believed that humans would be prone to panic within such a short duration of time, whereas machines would not. The rapid, rational thought processing required during such a moment of urgency is thought to be best handled by machines.
As machine decision-making accelerates warfare, it is plain to see how conflict would easily escalate. Compressed time and space creates the incentive for each side to strike first and strike fast in a perceived crisis. This is a recipe for crisis instability. It’s sort of like a Twitter war, with WMDs instead of words. Even if neither party initially planned to strike first, the accelerated dynamic inherent in an AI-driven scenario forces the likelihood of mutually assured destruction.
In 2010, we saw how an analogous unintended escalation in the financial markets, wiped over a trillion dollars off the stock market in minutes, driven by trading algorithms feeding off each other in a dizzying spiral. Imagine if those algorithms were controlling not digital currency, but instead weapons of mass destruction. How would one mitigate Armagaeddon?
There is only one way to mitigate Armagaeddon: the NO ACTION alternative and an international treaty to ban lethal autonomous weapons. (It is true that the Reaper drone is not fully autonomous, nor have fully autonomous weapons been developed so far. Nonetheless, the distinction between partially autonomous and fully autonomous can be very muddy. In any case, fully autonomous weapons are in development, the Reaper drone is a stepping stone to that development, and they would likely be used as part of the same kill-chain that would also involve a Reaper drone.)
COMPLEXITY = VULNERABILITY TO ERROR
One of the reasons that nuclear weapons are so controversial is because the ghastly, irreversible consequences of a mishap leave little to no wiggle room for error. AI-driven warfare is identically controversial, only more so, because nukes converge into the symbiotic mix of ever-evolving algorithms that up the ante by geometric proportions.
Yet, not a whisper is mentioned in the Draft EIS about the risks of complex machine-driven war and how they foster an ever-evolving symbiosis between all weapons, including nuclear warheads. As writer and retired army colonel Ralph Peters explains, “The more complex any system becomes, the more inherent vulnerabilities it has.”
Please elaborate on the vulnerabilities and risks inherent to AI-driven JADC2 as they involve the MQ-9’s symbiotic relationships with all other weapons with which it is networked.
VULNERABILITY TO ERROR = KILLED CIVILIANS
If the costly 20-year Global War on Terror achieved nothing else, it provided data on vulnerabilities and risks of killer-drone operation, including “collateral damage.” We were able to see first-hand how the glowing promises of this new technology were not foolproof.
I, personally, had a glimpse of the terror that was inflicted in the middle east when I attended the quadrennial World Congress of the International Union for the Conservation of Nature in 2012, in South Korea. Ten thousand members of the international scientific community were in attendance. There, I was approached by an elderly woman who beseeched me to help her country. She had tears in her eyes and a thick accent. I could hardly understand what she was saying. Then, it became clear, that this traumatized woman was telling me, an American, to help her stop drone attacks in her home country of Pakistan. I later learned that the number of civilians killed during the War on Terror were significant.
In the Summer 2018 issue of The Independent Review, Christopher J. Coyne and Abigail R. Hall co-wrote “The Drone Paradox: Fighting Terrorism with Mechanized Terror”. An excerpt is included at the end of this comment as an Addendum. It gives a thorough accounting of various assessments of civilian deaths by drone attacks during the Global War on Terror.
Though Coyne and Hall’s statistics reflect early-style drone strikes, rather than “multi-axis, multi-domain precision fires organic at all echelons” that are characteristic of JADC2 warfare, there are commonalities between the two generations of unmanned warfare. For example, both versions of warfare prioritize “signature” strikes, which means strikes that are based on machine determinations of who to kill. The MQ-9’s surveillancefunction is integral to this stage in the kill chain. Coyne and Hall explain how signature strikes work, and how they increase the likelihood of killed civilians:
The likelihood that innocent civilians will be harmed by drone strikes is exacerbated by the U.S. government’s commitment to using “signature strikes” against targets (1). Instead of relying on a preidentified target, signature strikes involve targeting a person or group of people based on their geographic location and broad patterns of behavior that are determined to be suspicious. This means that the government cannot be sure exactly who is being killed by drone strikes. Intended targets may be killed by signature strikes, but so, too, might innocent civilians. There is no way to obtain concrete numbers for these two categories because of the lack of specific reporting in areas where drone strikes take place, the methodology of counting enemy combatants, and the general secrecy surrounding the government’s drone program.
FOOTNOTE:
(1) De Luce, Dan, and Paul McLeary. 2016. Obama’s Most Dangerous Drone Tactic Is Here to Stay. Foreign Policy, April 5. At http://foreignpolicy.com/2016/04/05/obamas-most-dangerous- drone-tactic-is-here-to-stay/.
Numerous, credible studies (described in the Addendum to this comment) prove that drone strikes cause death and injury to innocent civilians. This flies in the face of repeated government claims, such as that from CIA director John Brennan, who stated that drones have “surgical precision—the ability with laser-like focus to eliminate the cancerous tumor called al Qa’ida, while limiting the damage to the tissue around it.” At a minimum, evidence suggests that drones lack the scalpel-like precision that their proponents often claim as a defining feature of this technology.
Proponents may maintain that the numbers of civilian deaths cited in Coyne and Hall’s report reflect drone technology from earlier generations, and therefore, the JADC2’s systems are not comparable. Actually, the earlier studies remain more relevant than ever, because what has not changed at all is the reliance on mere machines to determine who should live and who should die. That fact locks in a certain margin of error (civilian deaths), regardless. In fact, it can be deduced that the margin of error will actually increase over early drone strikes with the greater complexity of the JACD2, since greater complexity leads to greater margins of error. The early studies therefore serve as baseline numbers from which to generate new models of risk that do extrapolate JADC2 projections over the much larger Pacific theater.
Please provide such numbers of anticipated civilian deaths that would occur during JADC2 operations that involve the MQ-9 Reaper drone.
PENTAGON ASSURANCES VERSUS REALITY
Without studying how the numbers of dead civilians betray the Pentagon’s promises that drones kill with scalpel-like precision, people in Hawaii have experienced their own cognitive dissonance with the glowing promises bestowed by the U.S. military. For example, the military has been promising for decades that the Red Hill fuel tanks were no threat to Oahu’s most important aquifer; today, military and civilian families alike have lost faith in the Navy after thousands of people, as well as pets, were sickened from drinking the tainted water last year.
To add insult to injury, the Navy will not commit to draining the tanks asap, which would thereby remove the serious health and environmental hazard. In fact, a provision in the National Defense Authorization Act says that drainage depends upon the military’s ability to provide fuel for war by alternative means. In other words, the purity of our drinking water is not as important as the Pentagon’s assessment of warfighting capabilities.The U.S. military behaves as if Hawaii and surrounding waters exist only to serve the U.S.’s war economy, whether as a giant range complex to practice death games, or a command from which to project firepower across a hemisphere toward China.
COOPERATION, NOT WAR, WILL SOLVE OUR PROBLEMS
The present anticipation of war with China began in 2011 when Obama first announced the “Pacific Pivot.” At that time, the spectre of climate catastrophe was still an unknown variable in a vague distant future. Even at the COP 21 conference in 2015, Parisians said that climate was a far less urgent matter than the migrant and refugee crisis.
Who would have guessed that now, in 2022, only seven years hence, the major rivers of the northern hemisphere would be simultaneously evaporating off the face of the Earth? The Yangtze, the Seine, the Thames, the Colorado, the Euphrates, the Danube… and the list goes on. It’s as if someone put the whole planet in a giant hot air fryer. Out-of-control fires plague China, the U.S., Europe, Siberia, Canada and more. And then once the planet’s moisture rises into the atmosphere, it swells to bursting in the form of Biblical-scale rain bombs that sweep away cars, homes, highways, livestock and people in 500-year-floods (that now seem to be the “new normal”) because the ground is too parched to absorb the water. One third of the nation of Pakistan is under water at the time of this writing.
Videos from China show us a giant economy that has been hobbled by climate catastrophe. Temperatures of 120 degrees have been recorded in many places. The Yangtze River has been reduced to a dessicated riverbed. Hydropower plants are producing only half their normal output. Sichuan province has imposed rolling black-outs across factories and international companies have ground to a halt, even with coal-fire plants operating at full capacity. In Dazhou, power has been cut off to communities for 6-7 hours per day. The price of commodities such as silicon metal has risen due to the power restrictions, and there are growing concerns about a shortage of automobile parts in Shanghai for companies including the Shanghai Automotive Industry Corporation and Tesla. The droughts have also been causing problems for farmers, with a shortage of drinking water among nearly 200,000 livestock across farms in Sichuan. About 433,000 hectares (1,069,966 acres) of crops have been affected by the water shortages, with the resulting direct economic loss amounting to 3.5bn yuan, according to data released by Sichuan’s emergency management authorities.
A “rising China” may still be perceived by American military planners as a threat, but in actuality, Mother Earth has slammed China’s economy into grave crisis. We no longer need to throw hundreds of billions of dollars at weapons and war infrastructure to take down China, an action which pulls us ever further from climate equilibrium. China is already down. To continue kicking them smacks of anti-Asian hatred.
In the context of the current climate reality, it is morally (and fiscally) unsound to cling to the outdated “Great Power/Peer Competition” approach cited in the Draft EIS. Again, no one benefits except the weapons manufacturers. Their idea of an ersatz “clash of the Titans” was the fantasy of a bygone era trying to revive the Cold War that once proved so lucrative. Those days are gone. China is hurting and so are we, facing the shared foe of climate catastrophe. Our chances of survival as a human species on our shared planet are wholly contingent upon cooperation – not competition — with China. Come on, Pentagon! Get REAL!!!
One Reaper drone could pay for 275 half-million-dollar homes to help house those who have lost their homes to fire or flood. Or provide 27,516 60,000-year jobs to Americans suffering from record-level inflation. Or provide healthcare.
I, myself, quit working a full-time job in 2020 to be a full-time caregiver for my husband who suffers from several serious health conditions. We both live on his social security check, which is grossly insufficient. When I occasionally fall ill due to the stress of full-time caregiving, there is no government safety net. It raises my ire to see tens of billions of dollars a pop go toward drones and other unmanned, AI-based warfare, which is scandalously costly. Policy is written to benefit the weapons industry, not the needs of everyday people like me, or those who have lost their homes as climate refugees, or to joint cooperative projects with China to find ways toward a livable future in the face of climate catastrophe. Americans are fed up.
It doesn’t take a rocket scientist to see the injustice and inequity. In the zero-sum game of federal funding, costly high-tech warfare takes from Americans the basic-needs support needed to pursue life, liberty and happiness. As such, it is a violation of the Constitution. It is a double insult when the most dire needs in human history – combatting climate catastrophe – go largely ignored. Ronald Reagan’s disarmament partner, Mikhail Gorbachev, who died the day of this writing, wisely queried, “Is it not clear by now that wars and the arms race cannot solve today’s global problems?” The elder statesman added, “War is a sign of defeat, a failure of politics.”
Once weapons were manufactured to fight wars. Now wars are manufactured to sell weapons. – Arundhati Roy.
ADDENDUM
EXCERPT FROM “The Drone Paradox: Fighting Terrorism with Mechanized Terror” by Christopher J. Coyne and Abigail R. Hall
“Collateral Damage,” or the Maiming and Killing of Innocents
Nearly every argument for the expansion of the U.S. government’s use of drones stems from the idea that they are believed to be a more efficient means of achieving the government’s foreign-policy goals relative to the alternatives (see Hall 2015). Drones minimize the potential harm to members of the U.S. military, it is argued, while accurately targeting terrorists. When in office, President Barack Obama explicitly stated that drones are better at targeting and killing foreign adversaries. Drones “are effective,” he said. “Dozens of highly skilled al Qaeda commanders, trainers, bomb makers and operatives have been taken off the battlefield. . . . [T]he primary alternative to [drones] would be the use of conventional military options. . . . Conventional airpower or missiles are far less precise than drones” (Obama 2013). As this quote illustrates, the use of drones is typically compared and contrasted with conventional bombings assumed to be the relevant alternative.
A related argument is that drones reduce the costs of conflict in terms of reduced civilian casualties or “collateral damage.” CIA director John Brennan, for example, stated that drones have “surgical precision—the ability with laser-like focus to eliminate the cancerous tumor called al Qa’ida, while limiting the damage to the tissue around it” (Brennan 2012). Harold Koh, the former legal adviser of the State Department, stated that “[b]ecause drone technology is highly precise, if properly controlled, it could be more lawful and more consistent with human rights and humanitarian law than the alternatives” (quoted in The Economist 2015). Other commentators have made similar claims, stating that “drones kill fewer civilians . . . than any other weapon” (Saletan 2013) and that “[drones are] actually the most humane form of warfare” (Lewis 2013). In 2011, Brennan, at the time counterterrorism adviser to the president, stated, “[T]here hasn’t been a single collateral death [in a year] because of the exceptional proficiency, precision of the [drone] capabilities we’ve been able to develop” (quoted in Shane 2011).
As these statements suggest, the overarching idea is that the U.S. government can intervene in other societies and exterminate confirmed threats with precision while avoiding harming innocent civilians.(1) Moreover, it is claimed that drones are more effective than alternatives, with conventional bombing typically cited as the relevant substitute. The standard rhetoric and claims about drones raise a range of important issues.
For one, if we take the claim that drones are more accurate than conventional bombing as the appropriate comparison, it is not clear, ex ante, that the adoption of drones will result in fewer total deaths of innocent people. The economic logic underlying this claim is that drones reduce the price of an attack, which allows the military to move down the demand curve, increasing the quantity of drone strikes demanded. The result is that although the use of drones might reduce deaths in any single strike by substituting for another, more deadly alternative (conventional bombing), this reduction might be offset by an increase in the total death of innocents due to an increase in the overall number of drone attacks due to the lower relative price of employing drone technology to strike targets. (2)
In addition, presenting conventional bombing as an alternative to drone bombing is an artificially narrow dichotomy. If the U.S. government’s foreign-policy goal is to eliminate individual enemy targets, then it isn’t clear that conventional bombing should be presented as the appropriate alternative to drone bombing. The appropriate alternative should instead be something akin to special-operations missions against specific targets.(3) Drone strikes also raise a host of issues related to international law and state sovereignty, the ethics of robotic warfare, and the international precedent being created by U.S. drone policy. But even if these (significant) issues are put aside, existing evidence calls into question the precision of drones in striking the desired target while avoiding the imposition of significant harms on innocent human beings.
To date, efforts to quantify the number of civilian casualties from U.S. drone strikes have led to different estimates. This variation is due in part to differing methodologies and definitions as well as to alternative sources of data regarding drone strikes and casualties (see Singh 2013). Another confounding factor is the secrecy of the U.S. government’s drone program. The fact that these missions are so covert makes tracking drone strikes and their outcomes extremely difficult if not impossible. Nonetheless, existing estimates provide some range of civilian casualties from drone strikes and, more importantly, highlight the human cost of the use of drones despite rhetoric to the contrary by U.S. government officials.
The New America Foundation (NAF) collects from credible news sources data on U.S. drone strikes in Pakistan (2004–present), Yemen (2002–present), and Somalia (2003–present).(4) The Pakistan data include only U.S. drone strikes, but the Yemen data are broader and includes all U.S. air strikes, drone and non-drone. The Somalia data include air strikes and ground operations by special-operations forces. This means that the data for Yemen and Somalia capture the effects of drone strikes but also of other types of military operations.
For the 2004–16 (through June) period, the NAF calculates 403 total strikes in Pakistan. These strikes killed between 1,853 and 3,032 militants as well as between 255 and 315 civilians, with between 176 and 278 uncategorized deaths (NAF n.d.a). In Yemen, it estimated 156 strikes over the 2002–16 (through June) period. These strikes killed an estimated 895–1,129 militants and 87–93 civilians, with an estimated 33–52 uncategorized deaths (NAF n.d.c). Finally, in Somalia the NAF estimated that 36 strikes over the 2003–16 (through June) period killed 299–343 militants and 28–40 civilians, with an estimated 0–19 uncategorized deaths (NAF n.d.b).
The Bureau of Investigative Journalism (BIJ) tracks drone strikes in Yemen (2002–present), Pakistan (2004–present), Somalia (2007–present), and Afghanistan (2015–present). The data are collected from a variety of sources, including news sources, publicly available information (e.g., lawsuits), and field investigations. In addition to estimating total deaths, the BIJ presents estimates of civilian deaths as well as a separate death toll for children (up to seventeen years old) killed by drone strikes. It also presents an estimate of the number of people injured by drone strikes.
In Yemen, the BIJ identifies 120–40 confirmed drone strikes through June 2016 (all data from BIJ 2017). These strikes killed 535–782 people, including 65–101 civilians and 8–9 children. An additional 96–227 people are estimated to have been injured by these strikes. There were 424 drone strikes in Pakistan through June 2016, resulting in an estimated 2,499–4,001 total deaths. Of this total, it is estimated that civilian deaths range from 425 to 967, including 172–207 children. The estimated number of people injured by the drone strikes in Pakistan range from 1,161 to 1,744. In Somalia, the BIJ identifies 26–30 drone strikes through June 2016, resulting in 219–383 total deaths. It estimates that 3–10 civilians, including 0–2 children, have been killed by the strikes, with another 0–2 civilians injured. Finally, in Afghanistan, the BIJ estimates that the U.S. government has carried out 332–37 drone strikes (through June 2016), resulting in 1,610–2,123 deaths. Estimated civilian deaths range from 75 to 106, including 4–18 children. An estimated 163–69 people have been injured by the drone strikes.
Other efforts have been made to estimate the number of civilian casualties from drone strikes. For example, a report by the Human Rights Clinic (2012) at Columbia University draws on the data from the aforementioned two independent sources to compile estimates of civilian deaths. Other reports have studied the effects of a specific sample of drone strikes. For example, a report by Human Rights Watch (2013) reviews the effects of six targeted killings via drone strikes by the U.S. government in Yemen over the 2009–12 period. The report concludes: “Two [of] these attacks were in clear violation of international humanitarian law—the laws of war—because they struck only civilians or used indiscriminate weapons. The other four cases may have violated the laws of war because the individual attacked was not a lawful military target or the attack caused disproportionate civilian harm, determinations that require further investigation. In several of these cases the US also did not take all feasible precautions to minimize harm to civilians, as the laws of war require” (2013, 1). A report by Amnesty International (2013) reviews nine drone strikes in Pakistan during the January 2012–August 2013 period. The report details each strike and traces some of the costs incurred by innocent civilians, ranging from injury to death.
Further insight into the harm caused to civilians by drone strikes is provided by recently released government documents on Operation Haymaker, which targeted members of the Taliban and al-Qaeda along Afghanistan’s northeastern border with Pakistan (see Scahill 2016, 154–76). Haymaker involved a combination of special- operations forces and other members of the intelligence community on the ground with drone strikes from above to carry out targeted killings. Among other things, the government documents reveal that “during a five-month stretch of the campaign, nearly nine out of ten people who died in airstrikes were not the Americans’ direct target” (Scahill 2016, 156). Further, the documents include “a chart revealing that airstrikes killed 219 people over a fourteen-month period in 2012 and 2013, resulting in at least thirty-five jackpots [the killing of intended targets]” (Scahill 2016, 169). This means that the 184 other casualties—84 percent of the total people killed during this period—were not the intended targets of the U.S. airstrikes.
As this review of the existing, public evidence indicates, there is a lack of consensus on the specific number of civilian injuries and deaths caused by drone strikes. In some cases, the estimated number of civilian deaths and injuries falls within a wide range. For our purposes, this variance is irrelevant. What does matter is that there is evidence from numerous credible sources of drone strikes causing death and injury to innocent civilians. (5) (boldface mine) At a minimum, this evidence suggests that drones lack the scalpel-like precision that their proponents often claim as a defining feature of this technology.
FOOTNOTES:
(1) Even where drone strikes successfully kill the target, there is a lack of consensus as to whether this success actually contributes to achieving the U.S. government’s foreign-policy goals (see Trofimov 2016). Evidence indicates that drones and military strikes in general are not the best method for eliminating terrorists (see Jones and Libicki 2008, 18–19). Further, there is evidence that terrorist organizations can use drone strikes as a recruiting tool (see Kilcullen and Exum 2009), an issue discussed further later in this article.
(2) We thank Robert Whaples for bringing this point to our attention.
(3) Even the standard dichotomy between drone bombing and conventional bombing is not as clear-cut as its proponents make it seem. According to one estimate, drone strikes conducted in Afghanistan from mid- 2010 to mid-2011 were ten times more deadly for civilians than air strikes carried out by fighter jets (Ackerman 2013; Zenko and Wolf 2016). In writing about life in the Palestinian territories during the Israeli summer offensive of 2014, Atef Abu Saif (2015) describes, among other things, how the Palestinian population experienced different forms of bombing and how drones were no less terrifying than other forms of attack. This suggests that it isn’t the technology itself that is the determining factor of precision but rather the manner and context in which the technology is employed.
(4) For the NAF’s general report on drones, see NAF n.d.d.
REFERENCES:
Amnesty International. 2013. “Will I Be Next?” US Drone Strikes in Pakistan. New York: Amnesty International. At http://www.amnestyusa.org/sites/default/files/asa330132013en.pdf.
Brennan, John. 2012. The Efficacy and Ethics of U.S. Counterterrorism Strategy. At http://www.wilsoncenter.org/event/the-efficacy-and-ethics-us-counterterrorism-strategy.
Bureau of Investigative Journalism (BIJ). 2017. Drone Wars: The Full Data. London: BIJ. At https://www.thebureauinvestigates.com/category/projects/drones/drones-graphs/.
The Economist. 2015. Drone Strikes and International Law: Fallout Reaches the Ivory Tower. April 22. At http://www.economist.com/blogs/democracyinamerica/2015/04/drone- strikes-and-international-law.
Hall, Abigail R. 2015. Drones: Public Interest, Public Choice, and the Expansion of Unmanned Aerial Vehicles. Peace Economics, Peace Science, and Public Policy 21, no. 2: 273–300.
Human Rights Clinic. 2012. Counting Drone Strike Deaths. October. At http:// web.law.columbia.edu/sites/default/files/microsites/human-rights-institute/files/ COLUMBIACountingDronesFinal.pdf.
Human Rights Watch. 2013. “Between a Drone and Al-Qaeda”: The Civilian Cost of US Targeted Killings in Yemen. New York: Human Rights Watch. At https://www.hrw.org/sites/ default/files/reports/yemen1013_ForUpload.pdf.
Lewis, Michael W. 2013. Drones: Actually the Most Humane Form of Warfare Ever: How Better Targeting and Surveillance Can Reduce the Number of Civilian Casualties. Atlantic, August 21. At http://www.theatlantic.com/international/archive/2013/08/drones-actually-the- most-humane-form-of-warfare-ever/278746/.
New America Foundation (NAF). n.d.a. Drone Strikes: Pakistan. In Depth: America’s Counterterrorism Wars. At http://securitydata.newamerica.net/drones/pakistan- analysis.html.
———. n.d.b. Drone Strikes: Somalia. In Depth: America’s Counterterrorism Wars. At http:// securitydata.newamerica.net/drones/somalia-analysis.html.
———. n.d.c. Drone Strikes: Yemen. In Depth: America’s Counterterrorism Wars. At http:// securitydata.newamerica.net/drones/yemen-analysis.html.
———. n.d.d. World of Drones. Washington, D.C.: NAF. At http://securitydata.newamerica.net/world-drones.html.
Obama, Barack. 2013. Speech on U.S. Drone and Counterterror Policy. New York Times, May 24. At http://www.nytimes.com/2013/05/24/us/politics/transcript-of-obamas-speech- on-drone-policy.html?pagewanted5all&_r50.
Saletan, William. 2013. In Defense of Drones: They’re the Worst Form of War, except for All the Others. Slate, February 19. At http://www.slate.com/articles/health_and_science/human_ nature/2013/02/drones_war_and_civilian_casualties_how_unmanned_aircraft_reduce_ collateral.html.
Scahill, Jeremy. 2016. The Assassination Complex: Inside the Government’s Secret Drone Warfare Program. New York: Simon and Schuster.
Shane, Scott. 2011. C.I.A. Is Disputed on Civilian Toll in Drone Strikes. New York Times, August 12. At http://www.nytimes.com/2011/08/12/world/asia/12drones.html?_r50.
Singh, Ritika. 2013. Drone Strikes Kill Innocent People: Why Is It so Hard to Know How Many? New Republic, October 25. At https://newrepublic.com/article/115353/civilian-casualties- drone-strikes-why-we-know-so-little.
