“The drones were terrifying. From the ground, it is impossible to determine who or what they are tracking as they circle overhead. The buzz of a distant propeller is a constant reminder of imminent death.” This account of what a drone feels and sounds like from the ground comes from David Rohde, a journalist who was kidnapped and held by the Taliban for seven months in 2008. Yet this kind of report rarely registers in debates in the United States over the use of drones. Instead these debates seem to have reached an impasse. Opponents of drone strikes say they violate international law and have caused unacknowledged civilian deaths. Proponents insist they actually save the lives of both U.S. soldiers, who would otherwise be deployed in dangerous ground operations, and of civilians, because of the drone’s capacity to survey and strike more precisely than combat. If the alternative is a prolonged and messy ground operation, the advantage of drone strikes in terms of casualties is indisputable, and it is not my intention to dispute it here.

But the terms of this debate give a one-sided view of both the larger financial and political costs of drones, as well as the less than lethal but nonetheless chronic and intense harm continuous strikes wage on communities. This myopia restricts our understanding of the full effects of drones; in order to widen our vision, I provide a phenomenology of drone strikes, examining both how the world appears through the lens of a drone camera and the experience of the people on the ground. What is it like to watch a drone’s footage, or to wait below for it to strike? What does the drone’s camera capture, and what does it occlude?

Contemporary drones introduce an unparalleled capacity to see and survey. The use of unmanned aircraft dates back at least to the Vietnam War, and missiles have long been used to target individuals or discrete locations. They started out as fairly basic surveillance aircraft, first used in the mid-1990s in the wars in the former Yugoslavia, but by early 2001, drones were equipped with firepower. Since their first appearances in the war on terror, in the first two years after 9/11, drones have been regarded as a distinctly global resource, the means by which the war on terrorism “goes global.” The first-generation Predator, pressed into service after 9/11 in Afghanistan, was armed with two Hellfire missiles that had a limited range of 400 nautical miles. All this changed in 2007 with the second generation Reaper: equipped with four Hellfire missiles as well as two 500lb laser guided bombs, and able to fly over 16 hours fully armed for up to 3200 nautical miles, this new drone is the “first purpose built hunter-killer UAV.” The Reaper also has an all-weather, day or night radar, linked to a sensor ball that houses image-intensified and infrared cameras. And even this impressive surveillance ability will pale in comparison to the new generation drone called Gorgon Stare, which will increase the single video feed of the Reaper to 12 and eventually to 65 video feeds, so that “a drone which now stares down at a single house or vehicle could keep constant watch on nearly everything that moves within an area of 1.5 square miles. The year after that, the capability will double to 3 square miles.”

In addition to their military prowess, the remote operation of drones from control centers thousands of miles away has made them legendary, the stuff of science fiction. The pilots who command the drone and fire its missiles do so based on a real-time video feed. So no matter how expansive the drone’s vision, or how natural the images it relays seem, all this seeing and killing is based on what can be seen through a camera. Drone strike footage is not a film in any common sense of the term, but it is still video footage, shot from a camera and visible on a screen, and its filmic qualities demand attention.

What the Camera Can See

There is a longstanding intimacy between air power and the visual field. Aerial vision, from the primitive conditions of pilots flying bi-planes over deserts and mountains, to the infrared camera footage from a drone’s eye, belong to what Martin Jay calls the “scopic regimes of modernity.” Paul Virilio’s classic War and Cinema: The Logistics of Perception was written before the advent of drones, but its general argument is more applicable today than ever. For Virilio, the story of twentieth-century war is inseparable from evolving cinema techniques. From telescopic viewfinders atop rifles to sophisticated cybernetic cameras, optical devices mediate and produce the act of taking aim, aligning an imaginary axis from eye to object. As Virilio explains, “the act of taking aim is a geometrification of looking, a way of technically aligning ocular perception along an imaginary axis that used to be known in French as the ‘faith line’ (ligne de foi). . .  to denote the ideal alignment of a look which, starting from the eye, passed through the peep-hole and the sights and on to the target object.” Aerial vision occupies a special place within Virilio’s narrative: “At the turn of the century, cinema and aviation seem to form a single moment. By 1914, aviation was ceasing to be strictly a means of flying and breaking records; it was becoming one way, or perhaps even the ultimate way, of seeing.”

We have become too accustomed to seeing from the air, which violates all the familiar geometry and perspective of our mundane, grounded vision. The exhilaration of the bird’s-eye view, or the god’s-eye view, so palpable in early accounts of flying, stems from the possibility of outstripping human limitations. But in another respect, aviation is very much tied to the modern mode of seeing, because from the very beginning it has been linked to photographic and cinematographic representation. Shooting a film, or focusing on a target, are not cheap puns, but reminders of a shared genealogical origin. Indeed, this way of looking is so naturalized that we forget that seeing through an aperture produces a particular and partial visual construction.

Aerial vision at once expands the range of view and hones in on a perceived target. But this focus inwards, this claim of precise aim, is not just one among other ways of looking. Rather, the accuracy of the drone’s eye structures more than vision; it shapes the way we think about, talk about, and evaluate a bombing. We focus in on the target, the moment of impact. We dispute how contained or collateral the damage was, how many civilians died alongside the chosen target. These questions begin to eclipse all other questions about the global military apparatus that makes the strike possible or about civilian injury that goes beyond body counts.

Let us then take a closer look at the visual regime of the drone. Let us see what a drone sees (and what it does not). Here is a representative clip, chosen more or less randomly from the many available online.

While many commentators worry about the “video-game style warfare” of such footage, the comparison is both exaggerated and inapt. Contrary to drone footage, video games offer a deeply immersive environment in which at least the player’s virtual life is at stake. Perhaps what fuels the comparison of drone footage to video games is the aura of detachment they share. The worry is that detachment eases the ability to kill.

In his study On Killing, Dave Grossman, a colonel in the military, argues persuasively for a correlation between distance and the ease of killing.

Like the overhead shot in film, which excludes face-to-face dialogue, policing action both begins and ends with the criminalization of the enemy.

On the one hand, the video feed of drone footage transmitted to a distant location, precisely fits Grossman’s maximum range category: “a range at which the killer is unable to perceive his individual victims without using some form of mechanical assistance—binoculars, radar, periscope, remote TV camera, and so on.” At this distance, Grossman reports, “I have not found a single instance of individuals who have refused to kill the enemy under these circumstances, nor have I found a singe instance of psychiatric trauma associated with this type of killing.” On the other hand, the drone’s ability to zoom in to a sight line just a few hundred feet above the ground produces images of startling intimacy. In the end, we should be less concerned with how the mediation of the drone’s camera increases or decreases the pilot’s willingness to fire—since that decision is dispersed along a complex chain of command, referred to in military circles as the “kill chain”—than with how the purely visual quality reinforces certain conditions of control and asymmetric violence.

Looking at the clip again, one element is obviously missing: sound. Although the pilots can hear ground commands, there is no microphone equivalent to the micro-scopic gaze of the drone’s camera. This mute world of dumb figures moving about on a screen has particular consequences for how we experience the image. As Michel Chion notes in The Voice in Cinema, although sound or voice is easily swallowed up by the image, it nonetheless structures the image: “only the creators of a film’s sound—recordist, sound effects person, mixer, director—know that if you alter or remove these sounds, the image is no longer the same.” In the case of the drone strike footage, the lack of synchronic sound renders it a ghostly world in which the figures seem unalive, even before they are killed. The gaze hovers above in silence. The detachment that critics of drone operations worry about comes partially from the silence of the footage.

The camera angle is always the same: the overhead shot. By definition, the overhead shot excludes the shot/reverse shot, the series of frontal angles and edits that make up face-to-face dialogue. With the overhead shot, there is no possibility of returning the gaze. The overhead shot neither invites nor permits participation in its visual economy. It is the filmic cognate of asymmetric war.

Asymmetric war is typically a conflict between a regular army and a guerilla force, but could describe any conflict in which one side cannot retaliate in kind. The lasting insight of Schmitt’s evaluation of air power in The Nomos of the Earth (1950) is that the technological imbalance inherent in the use of air power transforms conflicts by adding an element of policing. The introduction of air power combined specific spatial transformations within a global nomos with changes in the technology of weaponry. Schmitt saw with prescient clarity that air war would not only create an “intensification of the technical means of destruction” and the “disorientation of space,” but also intensify the problem of unequal sides, and allow the dominant side to re-label enemies as criminals. Schmitt understood that air power would create a world in which those who command the sky could police and punish those who do not. For Schmitt, this widening gap is both the cause and result of a juridification of war, a shift towards conceptualizing war as a policing activity of criminals:

Both sides have a specific relation to the types of weapon. If the weapons are conspicuously unequal, then the mutual concept of war conceived of in terms of an equal plane is lacking. To war on both sides belongs a certain chance of victory. Once that ceases to be the case, the opponent becomes nothing more than an object of violent measures. Then the antithesis between the warring parties is increased exponentially. From the distinction between power and law, the vanquished are displaced into a bellum intestinum (internal war). The victors consider their superiority in weaponry to be an indication of their justa causa and declare the enemy to be a criminal because it no longer is possible to realize the concept of justus hostis.

Aerial bombing of those who have no chance to retaliate is not a war but an unequal exchange, which by its very nature accelerates the process through which war becomes a policing action and the adversary becomes a criminal or a mere object of violent reprisal. Policing action both begins and ends with the criminalization of the enemy. The overhead shot, coeval with air power itself, both produces and solidifies asymmetry and criminalization, which in turn produces a moral and legal justification of the violence.

The public availability of the drone’s footage deserves as much attention as its diegetic world. What is the video of supposedly classified strikes doing on YouTube? Who put it up? These clips are released by the Department of Defense through its Defense Video and Imagery Distribution System (DVIDS) Web site. With over ten million hits online, the clips are consumed voraciously, and attract a community of viewers (judging from the comment profiles, mostly men) who comment on what they portray and inform each other of new postings. Given the distinct action in these clips and the obsessive interest in them, some commentators have called the phenomenon “drone porn.” This offensive moniker does not so much equate the subject matter with that of a snuff film as offer a clue to the structure of the videos. Just as pornography caters to masculine desire, and the so-called money shot or male orgasm structures the film and retrospectively casts the action leading up to it as anticipation, so the experience of watching the drone strike footage is characterized by anticipation of the coming explosion, the moment of the strike. But while discussions of drone strikes in the United States focus on the precision of impact, the experience of drone strikes from the ground cannot be understood as a singular moment but as a structuring reality. And it is to that reality that I now turn.

What The Camera Can’t See

If drone operators can see but not hear the world below them, the exact oppositeis true for people on the ground. Because drones are able to hover at or above 30 thousand feet, they are mostly invisible to the people below them. But they can be heard. Many people from the tribal areas of Pakistan (FATA) describe the sound as a low-grade, perpetual buzzing, a signal that a strike could occur at any time. The locals call the drones machar, mosquitos. Because the drone can surveil the area for hours at a time, and because each round of surveillance may or may not result in a strike, the fear and anxiety among civilians is diffuse and chronic.

While drone strike footage has entered our culture as fantasy, drones have entered these regions as psychological trauma. In interviews, humanitarian workers, doctors and psychologists all attest to widespread occurrence of PTSD and anticipatory anxiety. Recent studies go beyond the disputation of casualty counts to a more thorough examination of life under the constant threat of drone strikes, offering ample evidence of a severely traumatized population, living under constant fear of the next strike. “Living Under Drones,” the comprehensive Stanford/NYU study of the impact of drone strikes in the tribal areas of Pakistan, includes first-hand interviews with many witnesses and survivors of the strikes. The words of one interviewee reveal an almost textbook definition of anticipatory trauma: “God knows whether they’ll strike again or not. But they’re always surveying us, they’re always over us, and you never know when they’re going to strike and attack.” While in law, the term “imminent” is frequently used for justification, here “imminent” takes on an altogether different and terrifying meaning, one distinguished by sound: “one man described the reaction to the sound of the drones as ‘a wave of terror’ coming over the community.” In another testimony, Hisham Abrar states, “when children hear the drones, they get really scared, and they can hear them all the time.”

The same is true in the other countries subject to repeated strikes. The resulting consequences have been devastating: parents refusing to risk sending their children to school, tribal councils wary of meeting, and shops shuttered into bankruptcy. In April of 2011, a young Yemeni author who spent time as an exchange student in the United States testified before the Senate, described his first-hand experience of a drone strike: “They [the locals] were moving erratically and frantically pointing toward the sky. Based on their past experiences with drone strikes, they told us that the thing hovering above us—out of sight and making a strange humming noise—was an American drone. My heart sank. I was helpless.”

Emphasizing the diffuse but chronic deterioration of life offers a thicker definition of civilian harm, and dismantles the visual regime of the drone, its will to omniscience and precision.

Sight on one side and sound on the other. Focus on one side and diffusion on the other. It is precisely this distribution of senses that produces the assertion of pinpoint accuracy and the disavowal of widespread harm. As I suggested earlier, the visual regime of the drone’s camera extends well beyond the video it produces, structuring larger discussions of legitimacy and efficacy. For instance, in testimony before the U.S. Senate reported by the Times, Retired Colonel Martha McSully insists “drones offered more oversight and precision because they could hover over a target, with the ability to abort a strike until the last second, and with the chain of command and lawyers watching.” The same prolonged hovering that produces the terrifying buzzing here adds oversight to sight, combining surveillance with legal scrutiny. But the layers of supervision effectively evacuate the world of sound and the interpersonal reality that sound produces; to argue about how precise or imprecise, focused or unfocused, such strikes are is to remain within a visual economy. These arguments, preoccupied with casualty figures, tend to dominate critical commentary in the West. While outlandish claims by the administration that there hasn’t been “a single collateral death” in Pakistan from the August 2010 to July 2011 have been repudiated by various organizations, disputes over the exact number of civilian deaths remain. Emphasizing instead the diffuse but chronic deterioration of life offers a thicker definition of civilian harm, and dismantles, in part, the visual regime of the drone, its will to omniscience and precision. It also begins to dispel the fantasy of air power in general.

But another major contributor to the dominance of air power in general is the idea that such power provides a neat alternative to the messiness of ground forces: either because aircraft can operate in areas where, for whatever reason, it is not possible to send troops, or because air power provides an alternative in areas where, for financial and strategic reasons, the United States wishes to scale back its presence (for instance, in the debate over the troop surge in Afghanistan during the summer of 2010, the Biden plan called for strategic air strikes throughout the country in lieu of additional troops). In this vision, as with previous hopes for air power, selective drone attacks are seen as the substitute for ground troops from the horn of Africa to the mountains of central Asia.

In reality, drone strikes often entail an increase in the infrastructure of ground support, in terms of personnel, the need for landing strips and bases, and an intelligence network for locating targets. Indeed, the fact that the actual drone is unmanned is misleading. The drones themselves do not have pilots (they are officially called Unmanned Aerial Vehicles), but it takes between 180 to 200 personnel to support a single combat air patrol. Moreover, even though drones may be operated from bases thousands of miles away, drones still need to take off and land somewhere near the theater of active conflict. U.S. Air Force drones that take off from the base at Kandahar are operated from Creech Air Force Base in Southern Nevada, and CIA drones are operated from Langley, Virginia, but it is only once the drones are airborne that the controls handed over to operators at Creech or Langley. This in turn, demands a new network of bases and airstrips around the world.

Consider the 2002 strike against Al-Harethi in Yemen. In that case, the drone took off from the state of Djibouti, on the Horn of Africa, where U.S. presence dramatically increased following the events of 9/11. By 2002, a thousand soldiers were stationed at Camp Lemonier, once occupied by the French Foreign Legion, and in a visit in December of 2002, Donald Rumsfeld predicted that the facilities would still be running for years to come. Since 2002, the camp has grown, renewing its leases and planning more permanent housing and facilities. Not coincidentally, the increasing military presence has been accompanied by the need to police the waters off the Horn of Africa against pirates. In addition to the base in Djibouti, there is a counter-piracy base in Bahrain, and surveillance drones are stationed in the Republic of Seychelles. As with the Al-Harethi targeted killing, personnel in all these locations are suspected of being involved in the targeted killing of Saleh Ali Saleh Nabhan, south of Mogadishu in September of 2009.

And this vast “constellation of secret drone bases,” as the Washington Post put it, is developing quickly. The debate over the government’s decision to kill an American citizen, Anwar al-Awlaki, in 2011, failed to note that the strike involved a new drone base in an undisclosed Middle East location, designed to make Yemeni drone strikes easier.

All of this comes at a financial and political cost. In 2004, Chalmers Johnson wrote about the unintended consequences of U.S. global militarism: “this vast network of American bases in every continent except Antarctica actually constitutes a new form of empire—an empire of bases with its own geography not likely to be taught in any high school geography class.” And as Ian Shaw notes, “the Droneworld is the evolution of the Baseworld.” Now instead of large garrison bases, there are more small facilities to house drones and their local operators. To those, such as political scientist Andrew Bacevich, who argue that U.S. national security requires a much smaller global military footprint, drones are no real alternative. Like so many earlier dreams of air power, they are no more than a flight of fancy; to civilians traumatized by their very sound, they are the transport of terror.