Jump to ContentJump to Main Navigation
Empire's LaborThe Global Army That Supports U.S. Wars$

Adam Moore

Print publication date: 2019

Print ISBN-13: 9781501742170

Published to Cornell Scholarship Online: May 2020

DOI: 10.7591/cornell/9781501742170.001.0001

Show Summary Details
Page of

PRINTED FROM Cornell University Press SCHOLARSHIP ONLINE (www.cornell.universitypressscholarship.com). (c) Copyright University of Cornell University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in Cornell for personal use. date: 28 September 2021

Supplying War

Supplying War

(p.67) 5 Supplying War
Empire's Labor

Adam Moore

Cornell University Press

Abstract and Keywords

This chapter begins by explaining logistics spaces and labor involved in supporting overseas operations in Iraq, Afghanistan, and Africa. It also identifies logistics spaces and labor as two foundational elements of military operations. While logistics spaces receive more attention, it is labor that animates war. Whether drone flights at remote locations in Africa or counterinsurgency campaigns in the Middle East, the U.S. military depends on the beating heart of logistical labor. Due to the increase in contracting, the composition of this labor is increasingly civilian and foreign rather than American and uniformed. This chapter concludes with an analysis of how the U.S. military is now inextricably entangled with the business of transnational labor acquisition.

Keywords:   military overseas operations, Iraq, Afghanistan, Africa, logistics spaces, military labor, U.S. military, transnational labor acquisition

Strategy, like politics, is said to be the art of the possible; but surely what is possible is determined not merely by numerical strengths, doctrine, intelligence, arms and tactics, but in the first place, by the hardest facts of all: those concerning requirements, supplies available and expected, organization and administration, transportation and arteries of communication.

—Martin van Creveld

The U.S. military’s ability to project force across the globe rests on the immense logistical resources it can bring to bear, without which the variety of operations it has carried out since the end of the Cold War would not be possible. Yet it is the rare analysis of warfare—now or in the past—that gives sufficient attention to the import of logistics. One exception is a classic, but little-known, text on the topic written a century ago by Marine Corps colonel George Thorpe, who drew upon the analogy of theater to illustrate the key role that logistics plays:

Strategy is to war what the plot is to the play; Tactics is represented by the role of the players; Logistics furnishes the stage management, accessories, and maintenance. The audience, thrilled by the action of the play and the art of the performers, overlooks all of the cleverly hidden details of stage management. In the conditions now adhering to the drama it would hardly be incorrect to assert that the part played by the stage director, the scene shifter, the property-man, and the lighting expert equals, if it does not exceed in importance, the art of the actor. … ​Logistics is the same degree of parvenu in the science of war that stage management is in the theater.1

As Thorpe perceptively noted, stage management depends on a diverse pool of labor and expertise. Logistics, he argued, is also a multifaceted enterprise with activities ranging from transportation of supplies to care of wounded troops.

Conducting the wars in Iraq and Afghanistan, as well as rapidly expanding counterterrorism operations in Africa, has involved the movement of a tremendous (p.68) amount of goods and people along lengthy and complex supply lines, the construction and maintenance of hundreds of bases—many the size of small cities—in remote and challenging environments, and the provision of a panoply of life support services like food, laundry, showers, and billeting for service members. Consider the remote Arba Minch drone base in Ethiopia that was operational between 2011 and 2015. According to military documents, two medium altitude drones, one MQ-9 Reaper and one MQ-1 Predator, flew from this facility, providing intelligence, surveillance, and reconnaissance (ISR) coverage over Somalia.2

Drone operations from Arba Minch were contingent on extensive logistics networks and the diverse labor of military and civilian workers. If we focus on the people, technologies, and bases that enabled these flights, we would note that the flights were supported by military personnel and facilities across the globe, through a division of labor that the military refers to as “remote split operations.”3 From ground station operators and mechanics at Arba Minch to pilots and sensor operators at Cannon Air Force Base in New Mexico (in 2013 Arba Minch operations were led by the 33rd Special Operations Squadron, based out of Cannon) to teams conducting data processing, exploitation, and dissemination in a variety of locations, in total a single Reaper Combat Air Patrol of four drones requires the work of approximately 170 military personnel.4 Drone operations are also sustained by sophisticated sensor technologies, ground control systems, surveillance and geo-intelligence software, satellite communications, and data relay stations such as the massive Ramstein Air Base in Germany, which serves as the primary conduit for data feeds from African drone bases.5

And then there are the civilian logistics spaces and labors that animated this small outpost of empire. Military personnel at Arba Minch received bimonthly deliveries of food from DLA contractor Seven Seas Shipchandlers, a Dubai firm that shipped containers by sea from Bahrain to Djibouti City’s port, and then hauled them overland to the facility. In addition to food supplies regular fuel deliveries for the drones were provided by the French oil and gas conglomerate, Total S.A., as part of a $51 million DLA contract.6 Contractors also worked with military personnel on site, including drone mechanics provided by the U.S. corporation AECOM.7 The military enrolled local sites and labor as well. A 2015 life support services contract, for instance, stated that the chosen contractor, a large nearby tourist lodge, would provide “131 bed spaces, office space for the Medics, Chaplain, Defense Operations Center, gym, laundry, internet, space to host a closed circuit television (CCTV), as well as NIPRNET [a U.S. military network for unclassified data] access/operations.”8 In addition to staff required to house and feed this contingent, the lodge was expected to supply guards for hotel security. Further afield, it is likely that analysis of drone data was provided by employees (p.69) at one or more of the Pentagon’s favored intelligence contracting firms in the Washington, DC, area.9 In sum, the work required to sustain Arba Minch was remarkably extensive given that the base hosted only two drones.

The above example highlights two central elements in the support of overseas wars. The first is logistics space, which I define as the networked spaces of supply and support, including infrastructure, sites, equipment, information, and technologies that ensure the flow and maintenance of military people and goods.10 Of pivotal importance here is the U.S. “global supply archipelago” of facilities located in dozens of countries across the world, including the various bases constructed and maintained in the Middle East, Afghanistan, and multiple African countries.11 Military operations also depend on civilian logistics infrastructure, from ports and warehouses to rail and road networks to border crossings and airports. Echoing Deborah Cowen’s observation about manufacturing occurring across logistics space rather than at a single site, it is not inaccurate to describe war as taking place along and through a global network of logistics spaces, not just on the battlefield.12

The second element—which brings the entire supply network to life—is logistics labor. As outlined in chapter 2, this labor is performed by an assemblage of contracting firms employing thousands of people from around the world. This is even the case with the military’s most important logistics entity, U.S. Transportation Command (TRANSCOM). TRANSCOM coordinates the military’s global transportation system, moving a staggering amount of goods and people around the world by sea, air, and land. In just a single year (from October 2011 to September 2012), for instance, it conducted more than 31,000 airlift missions, transporting more than 650,000 short tons of cargo and nearly 1.9 million passengers.13 While airlift is critical, especially for movement of personnel, more than 90 percent of goods are transported by sea. In 2011 TRANSCOM’s sealift branch, Military Sealift Command (MSC), estimated that “since the start of operations in Iraq and Afghanistan, MSC ships have delivered nearly 110 million square feet of combat cargo, enough to fill a supply train stretching from New York City to Los Angeles. MSC ships have also delivered more than 15 billion gallons of fuel—enough to fill a lake 1 mile in diameter and 95 feet deep.”14 Mirroring trends across the armed forces, TRANSCOM now relies heavily on contractors. In 2016 it estimated that commercial entities provided 90 percent of surface transportation (truck and rail), 55 percent of sealift support, 30 percent of airlift cargo, and 80 percent of airlift passenger transport for worldwide contingency operations.15

The wars in the Middle East and Afghanistan, and counterterrorism operations in Africa, have each involved distinctive combinations of logistics spaces and labor, shaped by geopolitics, physical geography, emergent wartime conditions, and preexisting economic relations and infrastructure. This has produced, in turn, (p.70) different supply challenges and operational characteristics. Indeed, while the goal of logisticians is to ensure the smooth flow of people and goods, disturbances and constraints always lurk, especially in the realm of military logistics.16 In Iraq and Afghanistan, for example, insurgents frequently targeted truck convoys and FOBs. In addition to causing supply disruptions, this also compelled, as discussed in the introduction, labor-exporting states to impose bans on citizens traveling to these countries to work for military contractors. These dynamics are absent in counterterrorism operations in Africa, where the greatest challenges involve distance and rudimentary logistics infrastructures. Consequently the military has been much more reliant on airlift to support “small footprint” operations. It is now time to examine the logistics spaces and labors of military operations in Iraq, Afghanistan, and Africa.

Iraq: Building, Maintaining, and Sustaining Baseworld

In 2005 an incredulous New York Times reporter wrote a story about life at Camp Liberty, part of Victory Base Complex surrounding the international airport in Baghdad. Contrary to expectations of basic amenities and oppressive heat the base, he asserted, had “the vague feel of a college campus” where troops lived in air-conditioned trailers, surfed the internet during off hours, worked out in gyms with modern equipment and a variety of exercise classes, and ate at dining halls that offered “a vast selection of food and beverages, ethnic cuisine nights, an ice cream parlor and, occasionally, a live jazz combo.”17 Four years later the paper would feature another story about bases in the country, noting that while a part of the Iraqi landscape they were in many ways “a world apart from Iraq with working lights, proper sanitation, clean streets and … ​thousands of contractors and third-county citizens to keep them running.”18

The scale of the military’s base network in Iraq at the height of operations was enormous. So too was the logistics labor required to construct and maintain it. Due to the dependence of the former on the latter, the geographical distribution of contractors is useful for limning the military’s presence. This is especially the case with LOGCAP workers, whether employed by KBR or one of its many subcontractors. As noted in chapter 2, LOGCAP personnel constituted the largest portion of contractors in Iraq in 2008 (37 percent) when the number of troops in the county reached its peak. The reason for this is that through LOGCAP the military could contract KBR to conduct an incredibly wide range of services (table 5.1). Core LOGCAP tasks involved base support activities such as laundry, food, billeting, morale, MWR, facilities management, waste and sewage disposal, (p.71)

Table 5.1. Logistics services provided by KBR in Iraq through LOGCAP III contract

Base life support services

Supply operations and material management (not procurement)

Other operations and services

Facilities management

Class I: Subsistence (food and water)

Engineering and construction projects

Laundry services

Class II: Clothing, administrative and housekeeping supplies, individual equipment (weapons, tents, tool kits, communications gear, etc)

Transportation (movement control, cargo transfer, port/terminal operations, motor pool operations and maintenance, line haul, etc)

Food services

Class III: Petroleum, oil and lubricants (POL)

Mortuary affairs

Morale, welfare and recreation (MWR)

Class IV: Construction materials

Retrograde operations

Vector and pest management services

Class V: Ammunition

Postal operations

Hazardous material storage

Class VI: Personal demand items (soap, toothpast, snacks, beverages, cigarettes, personal electronics, batteries, etc)

Ice production

Power generation and electrical distribution

Class VII: Major items (missle systems, helicopters, tanks, other vehicles, mobile machine shops, etc)

Medical services


Class VIII: Medical supplies

Test, measurement and diagnostic equipment (TMDE) services

Water production

Class IX: Repair parts

Waste and sewage management

Firefighting and fire protection

Clothing exchange and repair

Personnel support (badging, etc)

pest control, firefighting services, and water and power production. KBR also frequently provided materials management and operations support (but not procurement) for the military’s various supply classes. This involved activities such as tracking materials, operating warehouses, and managing bulk fuel distribution. In addition to this the company performed a number of other services, from engineering and construction to transportation, ice production, and mortuary affairs support.

So what does the military’s baseworld in Iraq look like from the perspective of logistics labor? Drawing on data from the 2nd quarter 2008 contractor census, figure 5.1 shows bases according to the size and composition of the LOGCAP workforce. Concerning the former I have divided the bases into four tiers, based on the number of workers. LOGCAP contingents varied substantially, from just thirteen people at a small FOB called McHenry near the town of Hawija to more than 8,700 workers at Victory Base Complex. The smallest tier of sites with fewer than 100 LOGCAP employees, like McHenry, were often FOBs with 1,000 or (p.72)

Supplying War

Figure 5.1. Distribution and composition of LOGCAP labor in Iraq during peak of military operations in 2008

fewer troops. At the other end of the spectrum the largest tier of bases with more than 2,000 LOGCAP workers were all, with the exception of the International Zone (IZ) complex, situated around airfields used for flight operations.19 The majority of these large bases also served as key logistics hubs (which I discuss below).

At most bases in Iraq TCNs constituted the majority of LOGCAP laborers, a pattern produced in large part by KBR’s reliance on subcontractors from Turkey and Gulf states, whose workers were almost exclusively recruited from South and Southeast Asia. Several thousand additional TCNs—mainly from Bosnia, Macedonia, (p.73) and Kosovo—worked directly for KBR, scattered among bases across the country. Exceptions to the prevalence of foreign labor fell along two lines: 1) small bases where a handful of KBR employees from the U.S. represented the bulk of LOGCAP support personnel (i.e., McHenry, Brassfield Mora, Caldwell, and Paliwoda), and 2) bases located in predominately Shia-populated portions of the country (i.e., Al Hillah, Echo, Delta, and Scania), where KBR subcontracted a significant portion of its work to local companies that used Iraqi LN labor. In contrast, bases located in predominately Sunni-inhabited sections of the country had either only a handful of Iraqi LOGCAP employees or none at all. This does not mean that LNs did not work at these bases, just that those who did were typically contracted directly by the military through the JCC-I/A framework. For instance, in Balad—the second-largest base in the country by number of LOGCAP workers with nearly 8,200—more than 1,100 LNs provided base support and construction services through JCC-I/A contracts. The remaining outlier among LOGCAP sites when it comes to the composition of labor was Harbur Gate, the primary crossing point for goods between Turkey and Iraq, where a small contingent of KBR employees and a local company split duties.

The sprawling baseworld in Iraq did not come ready-made. Every facility, from small FOBs to large air bases with tens of thousands of troops, was the product of massive construction investments and labor. As Tom Englehart observed in 2009, the country was “a Pentagon construction site.”20 One journalist working for the DoD’s own newspaper, Stars & Stripes, reported that by 2010 the military had spent more than $2.1 billion dollars on base construction projects in Iraq since 2004, with plans for an additional $323 million in projects to be completed before withdrawing at the end of 2011.21 As astounding as this figure sounds, it is likely on the low side. For instance, at just a single base—Balad—construction projects worth more than $240 million were completed, initiated, or allocated funds between the beginning of 2004 and September 2005 according to a now declassified, but heavily redacted, base master plan.22 This included $11.8 million for a wastewater treatment plant, $12.6 million for an aviation maintenance facility, $23.8 million for hospital construction and a class VIII warehouse, $7.4 million for a postal distribution center, $25 million for a fixed-wing hanger, $2.3 million for a new Army and Air Force Exchange Service (AAFES) shopping center … ​and the list goes on. In total there were nearly thirty projects in this less-than-two-year period. Two years later the base continued to be a hive of activity, a “giant construction project, with new roads, sidewalks, and structures going up” everywhere.23

In addition to constructing a new baseworld, by the time the number of uniformed personnel in Iraq reached its apex in 2008 the military had developed an extensive supply network to sustain operations. The linchpin holding this together (p.74) was Kuwait. How important was the country for both Iraq and wider operations in the region? One indication is that over 80 percent of U.S. military forces transited Kuwait while rotating in and out of CENTCOM, with roughly 1,750,000 troops passing through in 2008 alone.24 Similarly, the majority of supplies—from equipment to food to fuel—entered and exited Iraq from Kuwait. A remarkable 2009 DoS cable titled “A Big Footprint in the Sand: The U.S. Presence in Kuwait” details the multifaceted role this “indispensable ally” played.25 The document begins by noting that the U.S. received over $1.2 billion annually in benefits such as “free access to bases, waived port and air support fees, customs waivers, subsidized fuel and other services.” Bases that Kuwait offered “essentially open access” to the military included Ali Al Salem Air Base (the primary airport for moving U.S. forces to forward deployed sites across CENTCOM), Camp Buehring and the surrounding Udairi Range facility (used for “spin-up” or predeployment training before heading to Iraq), Camp Virginia (the main staging site for military convoys to Iraq), and Camp Arifjan, (the largest surface logistics center in the country and home to nearly 5,000 contractors). The military also had access to Kuwait Navy Base and Shuaiba, a large industrial port south of Kuwait City, while the DLA’s prime food delivery contractor for Iraq, the Kuwaiti firm PWC (renamed Agility in 2006), utilized the country’s largest commercial port, Shuwaikh (figure 5.2).

One of the more extraordinary elements of Kuwaiti logistical support involved border crossings, where the U.S. was given nearly unlimited control over the flow of people and supplies. One example of this, as discussed in the introduction, was Kuwait’s decision not to enforce travel bans to Iraq imposed by labor-exporting countries in 2004. Another is the development of border crossings exclusively dedicated to the transit of goods and equipment by the U.S. military, coalition partners, and military contractors. The first of these, Navistar, was built next to Al-Abdali, the primary civilian border facility between Iraq and Kuwait. In 2005 Kuwait signed a memorandum of understanding (MOU) that acknowledged the “trust placed in the United States by Kuwait for day-to-day management of the Coalition Forces Crossing [Navistar].”26 The MOU also announced plans to build a new dedicated military crossing in the desert expanse several dozen kilometers to the west, which began operations in 2007.

This new facility, called Khabari or K-Crossing, had several advantages from the military’s perspective. First, all northbound military and contractor convoys were allowed to stage at Camp Virginia and other bases in Kuwait and then pass through Khabari without processing by Kuwaiti border authorities.27 Second, there was no Iraqi government presence in the vicinity of Khabari, further facilitating the unimpeded flow of supplies.28 Third, the new route was shorter and safer. The Navistar crossing and main supply route (MSR) arcing through southern (p.75)

Supplying War

Figure 5.2. Iraq-Kuwait logistics network (key fuel and food supply hubs and facilities in italics)

Iraq passed through several towns prior to reaching the first two logistics hubs south of Nasiriyah (Camp Cedar II and Tallil Air Base), increasing the risk of improvised explosive device (IED) attacks and hijackings. In contrast, the new route from Khabari to these bases was a relatively straight shot through desert with “virtually no habitation.”29

Another important aspect of Kuwaiti support involved fuel. From late 2002 to March 2005, the country supplied aviation fuel to the military free of charge and from 2005 to the end of 2008 it sold fuel at below-market rates.30 It also approved the military’s prewar construction of a fuel pipeline that ran from Mina (p.76) Abdullah refinery in the southeast corner of the country to Camp Virginia, and then forward to a pump station at the border in the northwest of the country, not far from where Khabari would eventually be built. Just hours after the invasion commenced, military engineers began extending the pipeline, called the Inland Petroleum Distribution System (IPDS), through the Iraqi desert. IPDS consisted of thousands of nineteen-foot-long, six-inch-diameter sections of aluminum pipe, joined by coupling clamps. Due to aluminum’s high thermal reactivity, changes in temperature could cause the pipeline to shrink or expand by two feet for every fifty sections, thus expansion loops were required at regular intervals, as were smaller pump stations every twenty kilometers. In a remarkable display of engineering capability, by late April the completed pipeline—with a throughput capacity of 720,000 gallons of fuel daily—extended to Tallil and Cedar II, 360 kilometers away from Mina Abdullah.31

As the military settled into its occupation of Iraq, it required a more robust and flexible fuel distribution system than IPDS, which was a tactical solution designed for use during the outset of operations. By early 2009 nearly 1,500 DLA-contracted trucks a day were traversing the region delivering gasoline, diesel, and aviation fuel from Kuwait, Jordan, and Turkey to a series of large bases located along the military’s MSR network that served as primary fuel storage sites (figure 5.2). From these facilities fuel was then distributed to the rest of the bases in the country. In total the military consumed more than 1.5 million gallons of fuel a day, the vast majority of this being jet fuel.32 More than 60 percent came from suppliers in Kuwait. DLA also shipped fuel to ports in Jordan (20 percent) and Turkey (13 percent), where it was then loaded onto trucks bound for Iraq.33 Paradoxically only a small fraction of the military’s fuel needs was provided by suppliers in Iraq, even though the country possesses some of the largest oil reserves in the world. In one of history’s many ironies, U.S. dependence on fuel imports to sustain operations stemmed from the collapse of Iraq’s refining capabilities due to infrastructure damage during Operation Desert Storm and the imposition of sanctions in the decade following.

Afghanistan: The Geopolitics of Supplying a Logistics Island

Perhaps the defining difference between the Iraq and Afghanistan campaigns concerns the spaces and geopolitics of logistical support.34 In the former an extensive road network facilitated the flow of goods, and preexisting Iraqi military installations—most notably large air bases situated throughout the country—could be developed to serve as logistics hubs. Even more important was Kuwait’s (p.77) willingness to serve as the primary staging area for personnel and material, enabling use of its modern port facilities and airfields, spin-up training at desert bases, U.S. control over border operations, and reliable provision of refined fuel products. In addition to Kuwait, goods to Iraq could be routed through Turkey and Jordan, both U.S. allies. The military’s supply chain for operations in Iraq, in sum, was relatively short and robust.

Afghanistan, in contrast, is more akin, as Pierre Belanger and Alexander Arroyo have observed, to a “logistics island.”35 To begin it is a landlocked country. One of the closest ocean-going ports, Chabahar in Iran, is more than 900 kilometers from the nearest Iran-Afghanistan border crossing. For geopolitical reasons, Iran is not a viable option for transiting U.S. or NATO coalition military supplies into Afghanistan. Instead, for much of the war the military’s primary ground line of communication (GLOC) has run through Pakistan. This is an incredibly long supply route with two branches (figure 5.3). The first extends from the port of Karachi, to Quetta and the Chaman border crossing, then on to the massive base at Kandahar Airfield (KAF), the military’s main logistics depot in southern Afghanistan roughly 900 kilometers away. The second, longer—approximately 1,700 to 2,000 kilometers depending on the route taken through Pakistan—and more perilous branch to Bagram Air Base, the primary logistics center in the north of the country, runs from Karachi to Peshawar, through the Torkham border crossing, and then over Khyber Pass into Afghanistan. Transit of goods from Karachi to Kandahar or Bagram typically takes one to three weeks. But accidents, strikes, and delays at the border crossings have often produced significant delays, forcing the military “to budget months for travel that should take days.”36

In addition to long supply lines and frequent delays, utilizing Pakistan as the primary logistical conduit to Afghanistan has presented several other limitations and deficiencies for the U.S. military compared with operations in Iraq. First, due to Pakistani government restrictions, troops, weapons, and ammunition must be flown to bases in Afghanistan, whereas in Iraq the military had the option of staging troops and equipment in Kuwait and then traveling overland to bases. Second, the GLOC through Pakistan is much less secure than routes through Kuwait, Turkey, or Jordan, with supply operations in the former plagued by pilferage and attacks on trucks, bridges, and staging areas. Part of the problem is the inability of U.S. personnel to oversee the flow of goods. As AMC’s deputy commander acknowledged in 2010, “Once the piece of equipment gets off the boat at Karachi, no American [soldier] touches it—it is all contract [labor] because of the political situation in Pakistan.”37 Consequently the military has increasingly turned to remote technologies such as radio-frequency identification tags, shipping container intrusion monitoring devices, and satellite tracking to combat the (p.78)

Supplying War

Figure 5.3. Main supply lines and logistics sites for Afghanistan operations (lines are approximate and do not indicate exact routes taken)

(p.79) problem of en route theft.38 A bigger issue has been the inability of Pakistan to prevent militants from carrying out attacks within its territory, a problem that was especially acute around Peshawar and the Torkham crossing in 2008–9. In December 2008, for instance, around 300 cargo trucks and military vehicles were destroyed in a series of attacks on staging yards in Peshawar.39

The most significant challenge, however, has been the fraught geopolitical relationship between Pakistan and the U.S. Though nominally allies, contradictory interests and mutual distrust pervade the relationship, a tension colorfully captured by one senior U.S. diplomat’s characterization of the two countries as “frenemies.”40 Indeed, despite public praise by U.S. officials calling Pakistan a crucial partner for counterterrorism and counterinsurgency campaigns in the years after 9/11, behind the scenes there is deep concern that the country supports, or at least tolerates, the activities of various extremist groups within its borders who in turn carry out attacks against U.S. forces in Afghanistan. Following the 2011 raid on Osama bin Laden’s compound in Abbottabad—and reports that he had been hiding out in this military town located only fifty kilometers from Pakistan’s capital for years—such suspicions swelled. Pakistani political and military authorities in turn have their own concerns, including fear that the U.S. is increasingly orienting itself toward their main rival, India.41 Another irritant from Pakistan’s point of view are drone strikes—and occasional cross-border raids—conducted primarily in the country’s northwest borderlands, the Federally Administered Tribal Areas (FATA). Drone strikes began in 2004, peaked in 2010, and continue to the present day. In total more than 400 strikes have been conducted killing up to 4,000 people, nearly a quarter of them civilians.42

In 2008 the U.S. began to put into motion long-standing plans to develop the Northern Distribution Network (NDN) for Afghanistan, with the goal of reducing reliance on the Pakistani supply route. This was not a completely new concept. As far back as 2005 DLA had begun sourcing fuel from Azerbaijan and Kazakhstan, with roughly 30 percent coming from these two countries by the end of 2007. Additionally, prior to 2005 TRANSCOM shipped some food and construction materials across Europe to Afghanistan by rail.43 It also used air bases in Uzbekistan (Karshi-Khanabad, known as K2, 2001–5) and Kyrgyzstan (Manas, 2001–2014) as transit centers for flying personnel into Afghanistan.44 In addition to insecurity in Pakistan the NDN initiative was motivated by two further calculations. First, troop levels in Afghanistan were growing—even before the Afghanistan surge implemented by President Obama in 2009—which led to worries about a lack of surplus capacity along the Pakistan route.45 Second, U.S. officials were concerned that Pakistan might threaten to close the border as geopolitical leverage, or in response to cross-border operations. This was not a (p.80) theoretical concern as Pakistan closed the border for several days in 2008 after the bombing of a military outpost in FATA.46

The NDN was originally constituted by two distinct routes, each with their own variants (figure 5.3). The first “Russian” route began at the Latvian port of Riga (with operations later expanded to include Tallinn and Klaipeda in Estonia and Lithuania) where cargo was loaded onto trains. After traversing Russia, Kazakhstan, and Uzbekistan by rail, goods were then unloaded at the Uzbek border town of Termez and then driven to their final destinations in Afghanistan. Alternatively cargo would be unloaded in Kazakhstan and then hauled by truck though Kyrgyzstan and Tajikistan and then into Afghanistan at the Nizhny Pyandzh border crossing. In 2015 Russia rescinded transit permission across its territory, thereby closing off this northern circuit. The second “Caucasus” route begins at the Georgian port of Poti. After crossing Georgia and Azerbaijan by rail, cargo is then ferried across the Caspian to Kazakhstan, where it is then reloaded onto trains that end at Termez. As with the Russian route a variant of this approach utilizes line haul across Kazakhstan, Kyrgyzstan, and Tajikistan into Afghanistan.47 In 2010 the military also began shipping goods through the port of Mersin in Turkey, even using overland transport from DLA warehouses in Germersheim, Germany, to feed into this second supply line.48

The first NDN shipments began in 2009, and by June 2011 nearly 40 percent of supplies to Afghanistan were being delivered via the network despite the fact that this was nearly three times more expensive than routing cargo through Pakistan.49 U.S. reliance on the NDN routes increased dramatically later that year in the wake of the decision by Pakistan to close its borders to Afghanistan-bound military supplies for eight months following a U.S. military attack on two border posts that resulted in the deaths of twenty-four Pakistani soldiers in November. The military also relied upon the NDN to facilitate retrograde—military speak for removal—of equipment during troop withdrawals beginning in 2012, while at the same time increasingly utilizing airlift to move equipment to sites in the Middle East and Europe, where it is subsequently loaded onto ships for delivery back to the U.S.50

Any analysis of the NDN needs to go beyond a narrow economic calculation of shipping costs as its development has also affected politics and human rights in the region. The linchpin through which the vast majority of cargo enters Afghanistan, for instance, is Uzbekistan, an authoritarian state with one of the worst human rights records in the world.51 Or as it was put more delicately by the U.S. embassy in Tashkent in 2009 when it was cultivating Uzbek support for the NDN scheme, “A non-democratic regime with a troublesome human rights record in the center of a strategically important, but unstable region.”52 It is also a country that is extremely sensitive to criticism along these lines. Beginning in late 2001 (p.81) Uzbekistan granted the U.S. use of an air base at K2 in the south of the country. In exchange the U.S. provided more than $200 million in military hardware and surveillance equipment the following year.53 Uzbekistan also received support for its campaign against the Islamic Movement of Uzbekistan, a group of radical militants that were also fighting alongside the Taliban in Afghanistan.

This geopolitical quid pro quo was not to last. In May 2005 Uzbek security forces killed several hundred protestors in the southern city of Andijan, leading to calls for an international investigation by human rights groups, DoS, and a number of U.S. senators. Of particular concern for critics were reports that military hardware provided by the U.S. was used in the attack.54 Angered by these criticisms and fearful that Andijan might be used as a pretext for a “color revolution”—especially following the collapse of the ruling regime in neighboring Kyrgyzstan just months before in the Tulip Revolution—Uzbekistan’s president, Islam Karimov, moved quickly to evict the U.S. from K2. The realpolitik lesson U.S. officials learned was “to not push Central Asian regimes too hard on democracy and human rights issues, especially when important security cooperation and basing rights were at stake.”55 It was a lesson they would not forget during subsequent NDN operations, leading to accusations that they were “whitewashing … ​abuses” of the Karimov regime and other states in the region.56

Arguably the most striking difference between Iraq and Afghanistan concerns the practice and geopolitics of logistics operations within each country. Consider the distribution process for food supplies to bases in the two countries in 2008, which was supervised by DLA. In the case of Iraq military escorts would meet truck convoys at the Kuwaiti border (Khabari) and then travel with them to one of the primary logistics centers in Iraq (figure 5.3). From there truckers would unload their goods or pick up new escorts that would travel with them to their final destination. After deliveries were completed the trucks were then escorted back to Kuwait, where the process would begin again. In Afghanistan, in contrast, trucks carrying DLA foodstuffs were not provided with military escorts, whether hauling supplies to the two primary logistics hubs in the country (Bagram and Kandahar) or delivering goods directly to FOBs. Also, rather than entering immediately trucks were required to stage outside a FOB for at least twenty-four hours in a “cooling yard” where they were inspected for IEDs by contractors or Afghan National Army personnel. Upon completion of delivery, truckers then returned to supply warehouses, again unescorted.57 As with food, DLA deliveries of fuel in Iraq were accompanied by military escorts at all times while military escorts were generally not provided for fuel trucks in Afghanistan.58 A 2009 Host Nation Trucking (HNT) contract that simplified supply operations by awarding contracts to six prime trucking contractors in Afghanistan codified this practice by stating that the “contractor is responsible for all security” and convoys should (p.82) be conducted “independently, without military escorts, unless otherwise determined by the USG [U.S. government] at its sole direction.”59

Unprotected logistics supply lines, of course, are inviting targets for any enemy. Truck convoys in Afghanistan have been frequently attacked—and with deadly results. According to one military briefing, attacks on convoys resulted in nearly 100 fatalities between December 2005 and February 2008.60 As supply operations expanded the following year in response to the troop surge, attacks and casualties mounted. Trucking companies reacted by turning to private security companies to protect their convoys, a strategy that was mandated by the 2009 HNT contract. In practice, however, private security companies are often little more than thinly disguised fronts for local warlords who run what amounts to a protection racket, demanding bribes in exchange for refraining from attacking trucks that transit territories they control.61

Even more concerning is ample evidence that a significant portion of the fees earned by Afghan security companies have ended up in the pockets of the Taliban and other insurgents, perhaps as much as 10 percent of logistics contracts—amounting to hundreds of millions of dollars—according to an explosive report by The Nation’s Aram Roston in 2009.62 As one U.S. contractor in this article observed, “the Army is basically paying the Taliban not to shoot at them.” Pakistan’s subsequent closure of the border would demonstrate that this quote was not an exaggeration—and that the practice of paying the Taliban not to attack truck convoys is not limited to Afghanistan. Following the reopening of the border in July 2012, a series of news accounts suggested that the Taliban was more adversely affected by the suspension of the Pakistan GLOC than the U.S. military. According to one Taliban commander, “The NATO supply [route] is very important for us,” and in fact, “stopping these supplies caused us real trouble” as “earnings dropped down pretty badly. Therefore the rebellion [in past months] was not as strong as we had planned.”63

Two years later John Sopko, head of the Special Inspector General for Afghanistan Reconstruction, the U.S. government watchdog for military and civilian reconstruction efforts in Afghanistan, sharply castigated military officials for their inaction in relation to this problem:

As I have pointed out in our last six quarterly reports, the Army’s refusal to suspend or debar supporters of the insurgency from receiving government contracts because the information supporting these recommendations is classified is not only legally wrong, but contrary to sound policy and national-security goals. I remain troubled by the fact that our government can and does use classified information to arrest, detain, and even kill individuals linked to the insurgency in Afghanistan, but apparently (p.83) refuses to use the same classified information to deny those same individuals their right to obtain contracts with the U.S. government. There is no logic to this continuing disparity.64

Actually there was a logic. It was just an insidious one. Several Afghan security contractors suspected of funneling protection payments to the Taliban were politically connected and ostensibly coalition allies. Among those highlighted in Roston’s article were Watan Risk Management, which was run by two cousins of Afghanistan’s then-president, Hamid Karzai, and NCL Holdings, run by the son of the then-defense minister, Abdul Rahim Wardak.

The irony of supply operations in Afghanistan providing a key source of funding for insurgents waging war against U.S. forces is that the decision not to provide military escorts was motivated in part by a concern for casualties that this would entail. In Iraq, attacks on truck convoys in 2004 compelled the U.S. to bolster military escorts, in part to head off travel bans by labor-exporting states like India and the Philippines. Though ultimately unsuccessful in achieving this goal, the practice continued throughout the occupation of that country. Providing military escorts, however, carries significant risks for U.S. troops. In early 2011 Steven Anderson, a senior military logistician who was in Iraq in 2006–7, estimated that approximately 1,000 U.S. personnel had been killed while on “fuel-related missions in Iraq and Afghanistan,” with the bulk of casualties occurring in the former theater.65 Remarkably, this represented nearly one-quarter of all battlefield deaths suffered in the two conflicts to that point. Another particularly evocative article in Armed Forces Journal the following year referred to the “direct link between fuel [demand] and casualties” as “logistical fratricide.”66

As noted above, logistical operations in Iraq were relatively straightforward compared to those in Afghanistan. Therefore the decision not to provide military escorts for truck convoys in the latter—which was made in late 2003 or early 2004—made sense, initially.67 In the end, however, attempts to avoid logistical fratricide have just displaced the problem and ultimately provided the monetary fuel for insurgent operations across the country. Logistics contracting in Afghanistan amply illustrates, in others words, Derek Gregory’s argument that “the business of supplying war produces volatile and violent spaces in which—and through which—the geopolitical and geo-economic are still locked in a deadly embrace.”68

(p.84) Africa: Developing the Sinews of Support for Counterterrorism Operations

In contrast to bases in the Middle East and Afghanistan facilities in Africa tend to be small and austere. At one level this reflects the military’s preference for a small footprint approach to deployment in Africa Command (AFRICOM) that eschews large bases in favor of a network of small, relatively unobtrusive “lily pads” that facilitate force projection across the continent.69 But as the following vignette from a Navy SEAL who served as commander of several dozen operators at Camp Simba in Manda Bay, Kenya, in 2005 illustrates, this also reflects the difficulty of supplying personnel in Africa: “We had no fresh fruit or vegetables at Manda Bay. Our supply officers in Djibouti tried to get us fresh fruit, but it was difficult to transport an orange from Europe to Djibouti, from Djibouti to Mombasa, and from Mombasa up to Lamu. We ate peaches soaked in syrup packaged in MRE bags.”70 Military presentations, reports, articles, and theses on Africa almost inevitably include comments about the logistical challenges that operations pose due to the “tyranny of distance” and underdeveloped transportation infrastructure on the continent.71 Consequently, in recent years the U.S. military has focused attention on the development of more robust logistics networks, with an assemblage of contractors playing a key role in providing services.

Contracting facilitates two related operational priorities on the continent. First, it allows the military to maintain a relatively low profile, even as it operates from dozens of facilities, including several drone bases and SOF compounds (figure 5.4).72 Due to Africa’s vast size, contracted air transportation is key, and is of particular importance for SOF teams who conduct operations across a wide swath of the Sahel, Maghreb, and Central and East Africa, and who rely on U.S.-based flight contractors. In 2016, for instance, Special Operations Command, Africa (SOCAFRICA) issued a solicitation for two helicopters based out of a previously unknown base in Arlit, Niger, to provide support for military operations in the “North and West Africa Area of Operations,” which includes the countries of Mali, Algeria, Libya, Tunisia, Chad, Cameroon, Nigeria, Benin, Mauritania, Senegal, and Burkina Faso. Notably the contract language specified that “aircraft shall not be painted in a color that is close to military colors and paint schemes. A conservative, predominately white, civilian-style paint scheme is preferred.”73 Contracting documents indicate that the main hubs for fixed-and rotary-wing air transportation for SOF operations include Entebbe in Uganda, and Niamey in Niger.

Airlift works well for small, mobile SOF teams. But as the military expands its presence across the continent, the cost and limitations of transporting goods and equipment by air has precipitated efforts to develop “adaptive” logistics networks that utilize international and African surface transportation firms. In 2011 one (p.85)

Supplying War

Figure 5.4. Drone bases, CSLs, Marine staging bases, and SOF sites supporting operations in Africa

of the first such initiatives, dubbed “the new spice route,” combined sealift with line haul by local truck companies to move goods between several bases and temporary forward operating locations (FOLs) in Kenya, Uganda, Ethiopia, and Djibouti.74 Following the successful completion of this operation the aforementioned Dubai firm Seven Seas Shipchandlers was awarded a two-year contract by the DLA to make twice-monthly refrigerated and dry goods truck deliveries to three facilities in the “Ethiopia Operational Deployment Zone”: Camp Gilbert at (p.86) Dire Dawa, the drone base at Arba Minch, and an unknown facility in Negele.75 This was followed by another two-year contract in 2013. Nothing is known about Negele, though it is possible to infer from the contract that a similar number of personnel (~100+) were located there as there were at Arba Minch, which suggests that it was a key SOF base for missions in Somalia. Seven Seas Shipchandlers and its subcontractors also regularly deliver supplies to Camp Simba, which, according to 2016 contracting documents, has expanded into a facility that hosts a steady state population of “approximately 325 military personnel with potential surges up to 510 personnel.”76 In recent years AFRICOM has developed a surface distribution contracting network that extends across the entire continent, beginning with a 2014 award for contracts worth up to $10 million each to five different companies “to perform surface transport and distribution of general cargo within all fifty five (55) nations of the AFRICOM AOR and Egypt.”77

The use of civilian contractors to the reduce the visibility of military operations in Africa extends beyond logistics to ISR as well, the best examples being two previous manned surveillance operations—codenamed Creeksand and Tuskersand—in West and Central Africa, respectively.78 As with air and ground transportation, contracted ISR operations are by design intended to be as low-profile as possible to reassure “host nations” that are “uncomfortable with U.S. military platforms.”79 Tender documents and contracts often include specific language on the number of personnel, the flight and surveillance equipment to be used, and aircraft appearance, such as a solicitation from 2010 that states that operations must “present a relatively inconspicuous presentation, including but not limited to: (i) no distinctive US or military markings, other than the required US registration number and placards, and (ii) no ‘one of a kind’ platform which would invite attention. Aircraft should have a ‘slick’ appearance with little to no external variation (i.e., antenna arrays, baggage pods, fuel pods).”80 Small, unmarked, civilian planes favored by contractors, such as the Pilatus PC-12 and Beechcraft King Air series, also have the added benefit of requiring only a handful of people to operate and being capable of flying out of remote and rudimentary airfields if necessary.

In addition to an unobtrusive presence logistics contracting also facilitates AF-RICOM’s stated goal of organizing force posture “to maximize operational flexibility and agility.”81 Contractors provide base operations and life support services to the growing number of U.S. military sites and operations in Africa. In some cases—such as SOF facilities and drone operations in Niger, Cameroon, and Somalia; small bases in Uganda, Central Africa Republic, Democratic Republic of the Congo, and South Sudan that were used in counter-Lord’s Revolutionary Army operations; and OUA—these contracts are awarded through established channels such as LOGCAP.82 In other instances ad hoc solicitations or no-bid contracts (p.87) are used due to the small size of the base, temporary duration of operations, or difficulty in identifying qualified providers. In 2015, for instance, the Marine Corps solicited bids to provide base support services for up to four months for twenty-four troops conducting training exercises with the Ugandan military at Camp Singo in Uganda in the fall.83 The U.S. frequently uses Camp Singo—which is located approximately seventy kilometers northwest of Kampala—for training exercises with Ugandan and other African military contingents, and has even established a small fenced compound with buildings, tents, water tanks, and generators.84 But rather than permanently stationing troops there, it rotates them in as desired, relying on short-term contracts for base and life support.

The development of contracting capabilities in Africa has been accompanied by the emergence of an increasingly dense network of sites that facilitate the movement of U.S. personnel and equipment across the continent. Foremost among these are Cooperative Security Locations (CSLs), which the DoD characterizes as facilities “with little or no permanent U.S. presence, maintained with periodic Service, contractor, or host nation support.”85 CSLs are typically located at large airports and are valued because they provide “a foothold for conducting the full range of military options, forced entry, humanitarian relief, NEO [noncombatant evacuation operation], peacemaking, peace keeping, and other stabilization operations.”86 Following the September 2012 attack on the U.S. consulate in Benghazi, Libya, that resulted in the death of U.S. ambassador Christopher Stevens, AFRICOM began upgrading several CSLs into staging bases for use by Special Purpose Marine Air-Ground Task Force, Crisis Response (SPMAGTF-CR) teams as part of a strategic shift that military documents and online resumes refer to as the “new normal” or Operation New Normal.87 These expanded CSLs are, according to one military article, capable of hosting “within hours … ​nearly 200 troops for as long as they need to stay.”88 News accounts and contracting documents suggest that SPMAGTF-CR bases exist or are being set up in Ghana, Gabon, Senegal, Niger, and Uganda, with SPMAGTF-CR units also operating out of larger military bases in Djibouti, Spain, and Italy (figure 5.4).89 In addition to CSLs and Marine staging bases, the Navy also utilizes a number of African ports for fuel bunkering. Moreover, all of these sites are supported by an extensive network of logistics nodes across Europe and the Middle East, and existing strategic airlift and sealift channels and sites maintained by TRANSCOM.90

AFRICOM’s growing logistics network, in conjunction with an increasingly robust assemblage of contractors, ranging from small African trucking firms to massive multinational corporations, supports more than 1,700 SOF troops on the continent.91 It also facilitates a remarkably large number of military actions, from joint training exercises with dozens of African and European militaries (12 in 2015), to security cooperation activities (400 in 2015), to military operations (p.88) (75 in 2015), all while promoting flexibility and a low-on-the-ground profile that belies this activity.92 This in turn has deepened political and military entanglements between the U.S. and African governments, especially in the realm of counterterrorism activities.

Similar to the case of securing logistical support in Central Asia for operations in Afghanistan, the political effects of these entanglements have often been deleterious. Four of the countries that AFRICOM has established the closest counterterrorism partnerships with have experienced successful or attempted military coups in recent years: Mauritania (2005 and 2008), Niger (2010), Mali (2012) and Burkina Faso (2014 and 2015). In the case of Mali, the coup was led by Captain Amadou Haya Sanogo, a participant in “several” U.S. military training programs, while the leader of the most recent coup attempt in Burkina Faso, General Gilbert Diendere, was the country’s “point person on the U.S. Trans-Sahara Counter Terrorism Partnership.”93 Chad, another key partner, has seen several attempted coups against an authoritarian government led by Idriss Deby, who himself came to power though a coup in 1990. As the Oxford Research Group remarked in a 2014 report, “The pursuit of counterterrorism operations and basing or logistics infrastructure across the Sahel-Sahara is dependent on maintaining relationships and status of forces agreements with national governments,” with the result being that these states have become “largely immune from pressure to improve their repressive treatment of citizens and political opponents” due their status as reliable partners in the “war on terror.”94 In other words, for political and military elites in the Sahel, binding themselves to AFRICOM’s counterterrorism assemblage can be useful for better securing their own authority and privileges against potential challengers. This dynamic is not limited to Africa—or Central Asia—as the U.S. has “repeatedly collaborated with murderous, antidemocratic regimes and ignored widespread evidence of human rights abuses” in countries that it relies upon for overseas bases of operation.95


(7.) This information comes from online contractor résumés. Following Trevor Paglen (2009), I refer to this as “résumé intelligence” or RESUMINT. Online contractor and military personnel résumés offer especially rich veins of information on the various operations and activities conducted by the U.S. military around the world over the past decade. This said, I have decided not to provide links to individual résumés in the footnotes. There are two reasons for this. First, they can be easily altered. Second, the information they reveal constitutes breaches of operational security on the part of contractors and military personnel, therefore creating the risk of personal repercussions. However, full webpage PDFs of all pertinent résumés have been created and copies remain on file with the author. For more on RESUMINT, see Paglen 2009, 70–74.

(10.) This definition draws from Deborah Cowen’s (2013, 8–9) discussion of logistics space.

(11.) Belanger and Arroyo 2012. On US military infrastructural investments in the Arabian Peninsula in the 20th century, see Khalili 2018.

(16.) For more geographical analyses of the inherent conflicts and tensions involved in logistics and the circulation of goods around the world, see Chua et al. 2018.

(18.) Santora 2009. As Lair Meredith (2001, chap. 1) demonstrates, luxuriously appointed wartime bases first appeared in the Vietnam War, with contractors also playing a key role in their construction and operations.

(19.) In six cases I have agglomerated data from bases or sites that were separated out in the raw data tables, but were adjoined and/or overlapping on the ground, and thus (p.208) functioned more or less as a single base. These are (1) Diamondback and Marez (surrounding Mosul Airfield); (2) Victory Base Complex surrounding Baghdad International Airport, including camps Victory, Liberty, Radwaniyah Palace, Mayberry, Cropper, and Slayer; (3) Kirkurk and Warrior (surrounding Kirkuk Air Base); (4) IZ sites, including the IZ complex, Tigris, and Freedom Rest; (5) Basra and Harper (surrounding Basra Airfield); and (6) a military training ground east of Baghdad that cycled through a variety of names during the occupation, including Shakoosh, Butler Range, Besmaya Range, and Hammer.

(24.) Embassy Kuwait 2007d, 2009.

(25.) Embassy Kuwait 2009. As extensive as U.S. military use of Kuwaiti facilities was in 2009, it appears to have paled in comparison to the beginning of the war according to a 2003 cable that details a much more extensive presence, including the estimate that Kuwait had “set aside approximately 70 percent of its total land area for U.S. military training and bed-down” that year. See Embassy Kuwait 2003.

(26.) Embassy Kuwait 2005. Prior to the MOU a series of ad hoc agreements had governed border-crossing procedures. Due to a handful of disagreements leading to temporary border closures in the previous two years, negotiating the MOU was a priority for the military and DoS.

(27.) Embassy Kuwait 2007b. This cable, which was written before operations began at Khabari, stated that contractor conveys would still be subject to inspections. But this is contradicted by a 2009 military logistics article which states that “the Khabari Crossing, unlike Navistar, would operate as a throughput for convoys, not a staging yard. Staging would take place at other bases before heading for Khabari Crossing. At the new crossing, the previous convoy receptions, inspections, and consent procedures would no longer be used. Instead, civilian transporters would be issued a coalition crossing card—a plastic photo identification card with a bar code containing information linked to the Kuwaiti immigration and customs databases.” See Walker 2009.

(32.) Jet fuel (JP8) is also used by the military as fuel for M1 Abrams tanks, and for cooking, heating, etc.

(33.) McNulty 2009, slide 15.

(41.) For a detailed analysis of U.S.-Pakistan relations, see Kronstadt 2011.

(42.) The most comprehensive and updated data on drone strikes in Pakistan are produced by the Bureau of Investigative Journalism. See https://docs.google.com/spreadsheets/d/1NAfjFonM-Tn7fziqiv33HlGt09wgLZDSCP-BQaux51w/edit#gid=1000652376. For a detailed history of drone strikes in Pakistan from the early years to the peak of operations (p.209) in 2010, see B. Williams 2010. On the exceptional status of FATA in Pakistan, see I. Shaw and Akhter 2012.

(44.) Following the closure of Manas by Kyrgyzstan in 2014—in part due to pressure from Russia—the military began using a Romanian air base, referred to as MK, near the port of Constanta as its primary transit center for troops entering and exiting Afghanistan. See Nickel 2014.

(47.) For earlier instantiations of NDN routes, see Kuchins et al. 2009, 9–10; Cooley 2012, 44. On Russia’s decision to close off the northern line, see Daly 2015.

(49.) Whitlock 2011. Data on shipment costs come from a TRANSCOM document, dated June 21, 2011, that responded to questions submitted by Whitlock in advance of the above story. This document is part of a large batch of NDN-related material made public following FOIA requests that can be accessed at the command’s FOIA reading room. See http://www.ustranscom.mil/foia/reading_room_arc.cfm#hideD.

(51.) At the peak of operations in 2011–12, roughly 70 percent of NDN cargo entered Afghanistan through Uzbekistan. See Kuchins and Sharan 2015, 105.

(57.) On this distribution process, see Rackuaskas 2008, 14 and 17. Multiple Bosnian contractors I interviewed mentioned the cooling yards in Afghanistan. A description of this process is also provided by Task Force Currahee 2014, 41.

(58.) McNulty 2009, slides 15 and 17.

(59.) Contract language cited in Tierney 2010, 10.

(62.) Roston 2009. In addition to Tierney’s (2010) congressional report, Roston’s reporting was subsequently substantiated by an internal U.S. Army investigation. See DeYoung 2011.

(67.) On the decision to avoid supplying military escorts for Afghan truckers, see McDonnell and Novack 2004.

(69.) For a detailed account of the lily pad strategy in Africa and other parts of the world, see Vine 2015, chap. 16.

(71.) See, for example, King, Moss, and Pittman 2014.

(72.) In addition to U.S. military drone bases on the continent, the CIA also operates at least one drone facility, in Dirkou, Niger. See Penney et al 2018.

(p.210) (75.) Moore and Walker 2016, 697. Camp Gilbert was reportedly used as a staging site for SOF missions in Somalia. See Schmitt and Mazzetti 2008.

(78.) From 2007 to 2012 Creeksand flights flew from Burkina Faso and Mauritania, and also provided ISR coverage over Mali and Niger. The Tuskersand operation (beginning in 2009) was part of a multinational campaign against the Lord’s Resistance Army. Tuskersand was based in Uganda, with flights providing ISR coverage over parts of South Sudan, Central African Republic, and the Democratic Republic of the Congo. See Moore and Walker 2016.

(82.) For more on LOGCAP contracting and the use of LOGCAP contractors, see Moore 2017.

(83.) A copy of the solicitation is on file with the author.

(84.) On training at Camp Singo, see Whitlock 2012b.

(86.) Cornella et al. 2005b, viii. The number of CSLs in Africa mushroomed from a total of four in 2005 (Senegal, Ghana, Gabon, and Uganda) to thirteen in 2011 (Senegal, Gabon, Uganda, Ghana, Algeria, Botswana, Kenya, Mali, Namibia, Sao Tome and Principe, Sierra Leone, Tunisia, and Zambia). See Cornella et al 2005b and Ploch 2011. An unclassified 2018 AFRICOM briefing obtained by Nick Turse (2018) suggests that the number is now similar (twelve) but their composition has evolved. However, in my view the AFRI-COM briefing functions more as a disinformation device than a true accounting of U.S. military presence on the continent. The three largest U.S. drone bases on the continent (Niamey, Chabelly and Agedez), for example, are labeled as CSLs, while two others (Garoua and Bizerte) are absent from the map. Moreover, a key SOF base in Kenya (Manda Bay) is listed as CSL even though its runway is too short to accommodate most large military and civilian transport planes. Finally, in responding to Turse’s inquiries the Pentagon refused to acknowledge whether or not its tally is exhaustive—which it clearly isn’t. Consequently figure 5.4. highlights only sites that have been confirmed as CSLs through previous sources.

(87.) See Dickey 2013. According to a 2015 military presentation, at least two of the Operation New Normal Marine staging bases (Libreville and Accra) are supported by LOGCAP contractors. See U.S. Army 2015, 32, 33, and 65.

(89.) On Ghana, Senegal, Gabon, Niger, and Spain, see Seck 2015; on Uganda and Djibouti, see Reif 2014; on Italy, see Vandiver 2014.

(90.) On strategic airlift channels, see Moore and Walker 2016, 698.

(93.) On Mali, see Whitlock 2012a; on Burkina Faso, see Campbell 2015.