Blurred Lines? Provincial Reconstruction Teams and NGO Insecurity in Afghanistan, 2010-2011
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
stability Mitchell, D F 2015 Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010–2011. Stability: International
Journal of Security & Development, 4(1): 9, pp. 1-18, DOI: http://dx.doi.
org/10.5334/sta.ev
RESEARCH ARTICLE
Blurred Lines? Provincial Reconstruction
Teams and NGO Insecurity in Afghanistan,
2010–2011
David F. Mitchell*
Members of nongovernmental organizations (NGOs) have been critical of the
Provincial Reconstruction Team (PRT) initiative in Afghanistan since its inception,
claiming that the mixture of military and humanitarian operations has resulted
in ‘blurred lines’ that inhibit insurgents from identifying who is and is not a
combatant. Certain organizations have hypothesized that aid workers are more
likely to come under attack as a result of this mixture. Although this claim has
surfaced in multiple outlets over the years, there was a lack of empirical evidence
to support it. This study tests this hypothesis using a panel-corrected standard
error regression model of all 34 Afghan provinces in 2010 and 2011. Preliminary
results show that NGOs were likely to encounter a greater number of security
incidents in provinces with PRTs; however, further analysis reveals this was only
the case in provinces with teams not led by the US. This calls into question the
validity of a general ‘blurred lines’ explanation for decreased aid worker security.
The deliberate targeting of aid workers has to the core humanitarian principles of inde-
increased in recent years, encouraging ana- pendence, impartiality, and neutrality—have
lysts to question the cause of these events. become the deliberate target of combat-
The 2014 Aid Worker Security Report reveals ants. Although some observers attribute
that 251 incidents of major violence against these acts to part of a broader military strat-
aid workers occurred globally in 2013, result- egy employed by recalcitrant insurgencies
ing in 460 workers being killed, kidnapped, or (Hammond 2008), others within the aid com-
seriously wounded.1 Afghanistan alone was munity believe that a 'blurred line' between
home to 81 attacks, the highest number of military engagement and humanitarian
attacks in 2013. It was followed by Syria (43), work has contributed to the violence. For
South Sudan (35), Pakistan (17), and Sudan example, certain nongovernmental organi-
(16) (Stoddard et al. 2014). zations (NGOs) have specifically attributed
As intentional violence is now the leading decreased aid worker security to Provincial
cause of death among aid workers worldwide, Reconstruction Teams (PRTs) operating in
analysts have attempted to decipher why Afghanistan. PRTs were small military units
relief groups—which largely claim to adhere (100–250 troops) with a civilian component
that served three broad functions: reinforce-
* Security Studies, Kansas State University,
ment of the authority and legitimacy of the
Manhattan Afghan government at the provincial and
dfmitch@ksu.edu district levels; development of national,Art. 9, page 2 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
provincial, district, and local governance and and reconstruction (Dziedzic and Seidl
infrastructure; and establishment of eco- 2005: 3). CHLCs were staffed by a dozen
nomic and social stability and security for the Civil Affairs (CA) soldiers in small outposts
people (SIGAR 2010: 75).2 referred to as ‘Chiclets.’ These units were
Those who attribute PRT presence with tasked with small-scale reconstruction pro-
decreased aid worker security believe that jects and the assessment of humanitarian
military involvement in humanitarian needs. They also established relations with
activities ‘blurred the line’ between soldiers the UN Assistance Mission in Afghanistan
and civilians operating in the same realm. (UNAMA) and NGOs already in the field
Therefore, insurgent actors no longer dis- (Perito 2005: 2).
tinguish between combatants and civilians. In November 2002 this concept was
As Fast (2010: 381) has noted, although the further developed in Washington by the
reasoning that militarization of aid increases Department of Defense (DoD), Department
incidents is compelling, ‘the evidentiary basis of State (DoS), and the US Agency for
is not, at present, equally compelling even International Development (USAID). New
though anecdotal evidence does exist.’ Joint Regional Teams (JRTs) were proposed
In accordance with the Enduring Strategic to conduct village assessments, identify and
Partnership Agreement, PRTs began to close coordinate reconstruction projects, and liaise
in the summer of 2012 and were phased out with regional commanders (CHC 2003). The
by the end of 2014. Now that one of the most interim president at that time of the Afghan
significant 21st century experimental under- Transitional Authority, Hamid Karzai, enthu-
takings by military forces is at an end, it is siastically supported the new concept, but
important to analyze the successes and dif- requested the name be changed (Dziedzic
ficulties encountered along the way so that and Seidl 2005: 3). The name was initially
future endeavors are best equipped for suc- altered to Provisional Reconstruction Teams
cess. With that in mind, this study hopes to and later, following the pilot phase, to
offer a small contribution to the discussions Provincial Reconstruction Teams.
regarding humanitarian insecurity generally PRTs combined diplomatic, military, and
and PRT-NGO relations specifically. development components in an effort to
This study uses panel-corrected standard ‘improve stability by building up the capac-
errors of all 34 provinces in Afghanistan in ity of the host nation to govern; enhance eco-
2010 and 2011 to test the hypothesis that nomic viability; and deliver essential public
military involvement in humanitarian activ- services such as security, law and order, jus-
ity has contributed to an increase in violence tice, health care, and education’ (CALL 2011:
against aid workers. The article is divided 2). Teams did not engage in combat opera-
into four sections. Part I provides a brief his- tions, but the integration of military and
torical background of the PRT initiative in humanitarian work enabled civilian repre-
Afghanistan and a review of the existing lit- sentatives to conduct activity with force-pro-
erature. Part II outlines the research design, tection capabilities in hostile regions. All PRTs
followed by empirical analysis in Part III. Part eventually fell under the operational com-
IV provides a few concluding remarks and mand of the International Security Assistance
policy implications. Force (ISAF), while individual nations led the
teams. The official PRT mission statement
A Brief Background of PRTs in as incorporated into the ISAF Operational
Afghanistan Plan states that PRTs ‘will assist The Islamic
Prior to the development of PRTs, Coalition Republic of Afghanistan to extend its author-
Humanitarian Liaison Cells (CHLCs) and ity, in order to facilitate the development of
Afghanistan Civil Affairs Teams (CAT-As) a stable and secure environment in the identi-
supported humanitarian assistance, relief, fied area of operations, and enable SecurityMitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 3 of 18 NGO Insecurity in Afghanistan, 2010-2011 Sector Reform (SSR) and reconstruction PRT to PRT.’ Additionally, staff was subject to efforts’ (ISAF 2009: 3). a high turnover rate; civilians often served The first PRT was established in the city of for twelve months while military members Gardez in Paktia province in November 2002 served for nine. As a result of frequent person- and became fully operational in February nel changes, PRTs would regularly shift their 2003. By the end of 2003, seven more teams direction, objectives, and programs based on were established in Kunduz, Bamyan, Balkh, the expertise of individuals in charge (CALL Parwan, Herat, Nangarhar, and Kandahar. 2011: 15, 28). Initially, US-led PRTs operated in the vola- Following President Obama’s June 2011 tile southern and eastern regions while ISAF announcement that the United States would coalition-led teams remained in the relatively remove 10,000 troops from the country stable northern and western regions. As the by the end of 2011 and 23,000 additional PRT network continued to expand, eleven troops by summer 2012, ISAF and the additional teams were added in 2004. There Afghan Transition Coordination Commission was an increased number of US-led teams in established a plan to guide the transition. the south while ISAF deployed from the north As part of this process, the Afghan govern- to the west in a counterclockwise motion, ment asked the international community eventually extending to the south and east. to begin phasing out PRTs to ensure Afghan By October 2006, all teams came under ISAF institutions would develop the capacity to command (Jalali 2007: 36). ISAF commanded manage development projects. In May 2012, 26 PRTs with 14 separate lead nations; the President Obama and President Karzai signed United States led 12 of these teams. the Enduring Strategic Partnership, outlining US PRTs were structured so that the DoD the US-Afghanistan relationship during tran- was assigned responsibility for improving sition and beyond. As part of the agreement, security while also providing logistical sup- the United States committed to eliminate port and force protection for PRT members. PRTs throughout the country by the end of USAID was tasked with leading reconstruc- 2014. Although all 26 PRTs remained at the tion while the DoS was responsible for politi- start of 2012, the phase-out began that sum- cal oversight, coordination, and reporting. mer with the closure of the first US-operated Examples of PRT activity included political team. At the end of the transition, the func- development (e.g. election support), govern- tions of all PRTs would be handed over to ance (e.g. development of effective legislative the Afghan government, development agen- oversight capabilities), education (e.g. refur- cies, NGOs, or the private sector. In January bishment of schools and provision of mate- 2013, 22 PRTs were still operational, but as rials), health (e.g. medical training), human the phase-out continued only 13 remained rights, economic development (e.g. infra- by that summer (SIGAR 2011: 30; SIGAR structure projects, banking, and enhance- 2012a: 4–5; SIGAR 2012b: 8; SIGAR 2013: ment of agricultural capacity), and rule of 126). The majority of PRTs were eliminated law (e.g. police training, judicial reform, and by mid-2014. development of law school curricula and The civilian humanitarian community has standards) (CALL 2011: 41, 48–50). a different approach to aid facilitation that Guidance for PRT activity came from two unsurprisingly conflicted at times with the sources: ISAF and the member nations lead- PRT concept. Although PRTs engaged in activi- ing the teams. Although ISAF was the pri- ties similar to those of their NGO counterparts, mary source of guidance, individual nations their mission was tied into the overall counter provided the funding for PRTs. The political insurgency strategy of the coalition. NGOs are apparatus of each country often influenced independent, non-profit civilian organizations how each team operated, ‘resulting in differ- that conduct a wide array of humanitarian ent types of operations being conducted from missions. There are both international and
Art. 9, page 4 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
national NGOs with different areas of exper- of views relates not so much to the type of
tise, resources, structures, governing bodies, activity considered to be humanitarian assis-
approaches, and areas of operations; they can tance, but more to the process of delivery
be secular or ‘faith-based.’ Though each is and motivations behind it’ (McHugh and
unique, many of their goals and objectives are Gostelow 2004: 1). A 2009 report compiled
similar. However, there is a divide in the field by eleven NGOs operating in Afghanistan at
between organizations that strictly adhere to the time stated, ‘In the same way that NGOs
a politically impartial distribution and those are not expected to take the lead in the secu-
that embrace political advocacy and struc- rity sector, predominantly military institu-
tural reform. Whereas groups in the former tions should not be expected, or presume, to
camp focus their efforts strictly on providing take a leading role in local development or
relief, those in the latter engage in such activi- governance’ (Waldman 2009: 13).
ties as post-conflict reconstruction, human Many within the aid community have
rights advocacy, democracy promotion, con- suggested that the decline in security for
flict resolution, and peacebuilding.3 All offi- NGOs is the result of a blurring of military
cial organizations are registered under the operations and humanitarianism. The 2009
Ministry of Economy and the Government of NGO report claims that the global increase
Afghanistan’s Law on NGOs (ISAF 2009: 229). in violence against aid workers can be
In 2005, four main NGO coordinating attributed, in part, to military engagement
bodies (ACBAR, ANCM, AWN, and SWABAC) in aid activity, which has ‘blurred the line
developed a code of conduct for NGOs between military and humanitarian actors’
engaged in humanitarian action, reconstruc- and thus ‘adversely affected NGO security
tion, and development in Afghanistan. The [and] endangered the lives of NGO workers.’
mission statement outlined in the code was The report goes on to declare, ‘[W]e do not
defined as follows: believe that PRT engagement in develop-
ment activities is effective or sustainable,
Our general mission as NGOs oper- and strongly oppose the expansion of such
ating in Afghanistan is to address activities or the establishment of new PRTs’
humanitarian, reconstruction and (Waldman 2009: 16).
sustainable development needs in The eleven signatory NGOs to the report are
Afghanistan, with a special focus on not alone; additional organizations have ech-
the rights of those who are disad- oed similar sentiments. A 2003 policy brief
vantaged and vulnerable. We work in by CARE International argues that extensive
partnership with each other, the gov- coalition engagement in reconstruction may
ernment, donors, and communities. put aid workers at risk and should therefore
(ACBAR 2005: 3) be left to the Afghan government, UN, and
other civilian aid agencies. The brief recom-
The purpose of the code was to promote mended that coalition forces ‘should take all
improved understanding of NGOs; transpar- necessary steps to ensure that communities,
ency, accountability, and good management policy makers and the general public do not
practices; and improved quality of services. confuse military- and civilian- implemented
The code also highlighted the independence, assistance’ (CARE 2003: 6). Additionally,
impartiality, and neutrality of organizations. InterAction—an alliance organization of over
Considering their adherence to this creed, it 180 NGOs—has accused the US military of
is unsurprising that many NGOs were reluc- ‘increasing risks to aid workers and under-
tant to interact with PRTs whose primary mining sustainable development efforts’
objective through relief work was the sup- through the provision of humanitarian assis-
pression of insurgency. According to a 2004 tance generally and PRT activity specifically
report by Save the Children, ‘the divergence (InterAction 2013).Mitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 5 of 18 NGO Insecurity in Afghanistan, 2010-2011 However, the ‘blurred lines’ explana- groups within the population to associate tion for decreasing aid worker security is humanitarian aid workers with the military’ not unanimously accepted. For example, (McHugh and Gostelow 2004: 34–35). Hammond (2008) believes that the attacks One of the most notable instances of con- are deliberate and ‘intended to demonstrate troversy between the military and NGOs the might of the attacker, the weakness of concerned uniforms and clothing during the the victim, and the inability of the oppos- preliminary phase of the PRT initiative. To ing force to prevent such attacks.’ Thus, the minimize the negative psychological impact deliberate targeting of aid workers is part of of his military unit’s interaction with the civil- a broader military strategy and ‘should be ian population, one commander allowed his seen not as the result of extreme confusion troops to ‘wear baseball caps, grow beards, but as deeply conscious acts’ (Hammond and mix with local people in the market’ 2008: 175, 177). (Perito 2005: 8). NGOs expressed concern that Other observers believe that humanitar- their personal safety was at risk, as the popu- ian activity is inherently political in nature lace were unable to differentiate between (Barnett 2011). This view holds that, regard- civilian aid workers and military personnel. less of whether aid workers publicly claim Although events like this strained rela- to adhere to the core principles of inde- tions with NGOs, the military showed ini- pendence, impartiality, and neutrality, their tiative to alter approaches in an effort to work seeks to alter existing structures which build trust with the civilian aid community. can influence local perceptions. In recent In 2004, UNAMA established a civil-military decades, many humanitarian groups have working group to address ongoing concerns. shifted toward human rights advocacy, peace- Following the ‘civilian clothing’ backlash, building, and post-conflict reconstruction military personnel ceased the practice after to include gender equality, religious rights, addressing the grievances expressed by the legal reform, and election monitoring among NGO community. PRTs also altered their a host of other politically-motivated, democ- approach toward NGOs by no longer visiting racy-promoting endeavors. Organizations no unannounced in full ‘battle-rattle.’ The PRT longer solely offer succor to those in need; handbook states that ‘[i]nitial contact with they also seek to restructure environments NGOs is best done through a non-military to address the root causes of suffering and agency (NGO coordinating body, UNAMA or instability. As insurgents are not likely to relevant IRA department). It is requested that embrace the alteration of existing norms in armed PRT personnel, including civilians with favor of Western democratic values, NGOs a military escort, do not visit an NGO office could face increased vulnerability due to the without first obtaining permission from very nature of their activity. the NGO’ (ISAF 2009: 240–241). Additional Nonetheless, the dispute between the two guidelines in the handbook outline how to communities was highlighted in June 2004 properly interact with NGOs, security proto- when Doctors without Borders withdrew cols, and weapons policies. from Afghanistan, claiming that PRT opera- It should be noted that NGOs may have tions in Badghis province contributed to the also helped contribute to ‘blurred lines’ early murder of five of its staff. A spokesman for in the conflict, as they often drove the same the Taliban took responsibility for the attacks, vehicles and used the same equipment as the stating, ‘We killed them because they worked military. Moreover, their bases of operations for the Americans against us using the cover were frequently located near the military. of aid work. We will kill more foreign aid According to sources with deployed experi- workers.’ McHugh and Gostelow of Save ence who were interviewed by Watts (2004), the Children believe this quote ‘highlights these factors made it difficult for locals to the readiness of insurgents and disaffected distinguish between actors (Watts 2004: 11).
Art. 9, page 6 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
PRTs also often operated in violent areas attempting to comprehend attacks. Although
that NGOs tended to avoid. For example, contemporary data are by no means perfect,
when the Taliban killed two German civil- it is now possible to engage in quantitative
ians in Paktia in 2003, most NGOs fled the empirical analysis in areas that have previ-
province for over a year. Similar instances ously posed great challenges.
occurred in Kandahar and Helmand in 2005; Two studies were conducted regarding
however, PRTs remained to carry out recon- NGO security and its possible relationship
struction in the provinces (Malkasian and to military operations. Following a marked
Meyerle 2009: 30).4 increase in the death of civilian aid workers in
Afghanistan in 2004, McHugh and Gostelow
Prior Studies (2004) of Save the Children conducted an
A rather limited number of studies have anecdotal analysis of the possible impact
been conducted regarding NGO security. of PRTs on humanitarian security. The lim-
This is due to multiple factors, one of which ited analysis sought to identify the linkages
is that the movement to target civilians is between PRT activities and possible changes
relatively recent. Additionally, it has his- in humanitarian security. Four areas of PRT
torically been difficult to collect data on the activity were assessed: security, reconstruc-
structure, finances, and activities of humani- tion, strengthening government authority,
tarian organizations. Prior to the 1980s, few and relief operations. The results of the analy-
organizations kept track of their financial sis suggested a positive effect of PRT activities
activities, missions, and relationships with on humanitarian security when the teams
partners. Data collected during this period were involved in security, reconstruction,
often proves unreliable because there was and expanding central authority. However,
no industry-wide standardization (Barnett there were negative effects on humanitarian
and Weiss 2008: 30). Although reporting security when teams were involved in relief
improved during the 1990s, several organiza- activities. The authors of the report conceded
tions either refused to release information or that their analysis was not sufficient to draw
actively misrepresented figures due to safety general conclusions regarding the impact
concerns. As a result, empirical analyses are of PRTs on humanitarian security. However,
often a daunting task, as it is nearly impos- their findings demonstrated it was plausible
sible to determine the number of organiza- that PRTs operating in similar areas as NGOs
tions, their structure and employee makeup, represented ‘the most significant potential
relationships with donors (individuals, states, threat to humanitarian security’ (McHugh
and corporations), and field activity. and Gostelow 2004: 30–35).
However, recent developments have made it A study by Watts (2004) of the Combating
easier for analysts to collect reliable informa- Terrorism Center (CTC) at West Point ana-
tion for studies on NGO security. Organizations lyzed NGO security incidents from January
such as the Afghanistan NGO Safety Office 2003 to February 2004 at the provincial level
(ANSO) coordinate, track, and report on secu- in Afghanistan. The results of his empirical
rity incidents with the consent and help of study indicated that provinces which shared
NGOs in the field. Furthermore, greater num- a border with Pakistan and those with a
bers of humanitarian groups are now pro- higher percentage of home radios displayed
viding general information on projects and a positive relationship with violence against
expenditures. Nonetheless, the seemingly NGOs. Conversely, provinces with higher lev-
simple task of analyzing the data is quite els of poppy cultivation witnessed decreased
difficult. Fast (2014) persuasively argues levels of violence against NGOs. All other var-
that the intangible internal dynamics of iables—including US military presence, dis-
organizations are of great importance when tance from each provincial capital to Kabul,Mitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 7 of 18
NGO Insecurity in Afghanistan, 2010-2011
warlord control, shared border between a Commission’s Humanitarian Aid and Civil
province and a neighboring country, per- Protection Department (ECHO) and—among
centage of population in need of food aid, other tasks—collects information on the secu-
number of beneficiary population receiving rity environment around the country. ANSO
food aid, cereal allocation, number of learn- uses their security incident reports to dis-
ing spaces, and number of formal/informal seminate information and advice for NGOs
schools—were found to be statistically insig- operating in Afghanistan. ANSO classifies
nificant determinants of aid worker security. ‘incidents’ as fatalities, injuries, abductions,
As a result of his analysis, Watts suggested and criminal activity toward NGO workers.
a reevaluation of US and Afghan policy Provincial level ANSO data were readily avail-
directed at eradicating poppy cultivation able on ReliefWeb for 2010, 2011, and 2012.
(Watts 2004). Data from 2012 were not included in this
Studies by Save the Children and CTC did study because the PRT phase-out began the
not identify aid worker proximity to PRTs summer following the Enduring Strategic
and military presence respectively as signifi- Partnership Agreement.
cant determinants of NGO security. However,
the landscape in Afghanistan has changed Independent Variable
significantly over the years, and an updated PRT presence within a province is the inde-
analysis would be helpful to validate prior pendent variable. Information on PRT loca-
findings. This study uses multiple regression tions and team leads was obtained from the
to test the hypothesis that the presence of Institute for the Study of War (ISW). US-led
PRTs within a province is likely to lead to teams are defined as those operated by the
attacks against NGOs. US, while coalition-led teams are defined
as those operated by ISAF nations other
Research Design than the US. The PRT presence variable is a
This study relies on a panel-corrected stand- dummy and is given the value ‘1’ if present
ard error regression model of all 34 Afghan within a province or ‘0’ if not present.
provinces for the years 2010 and 2011. The
research builds upon Watts’ 2004 study, with Control Variables
the inclusion of additional control variables. The following variables were included in
Watts broadly measured the relationship Watts’ study and have been updated with
between US military presence in a province data from 2010–2011: US military presence,
and NGO security. However, it should be poppy cultivation, border with a neighbor-
noted that only three PRTs were operational ing country, border with Pakistan, distance
by mid-2003, so his inclusion of the vari- of the provincial capital from Kabul, popula-
able was likely not feasible. As mentioned tion, and home radio ownership. Information
previously, the PRT initiative expanded sig- on US military presence was obtained from
nificantly throughout the country shortly ISW (Morgan 2012). The placement of com-
thereafter. bat forces is provided down to the battalion
level and includes the following categories:
Operational Definitions of Variables maneuver (i.e., infantry, armor, and cavalry)
Dependent Variable units, artillery units, military police units,
The dependent variable is the number of NGO most types of engineer and explosive ord-
security incidents by province. Consistent nance disposal units, and ‘white’ special oper-
with Watts (2004), data on security inci- ations forces. Information on ‘black’ special
dents were obtained from ANSO. ANSO operations units is not provided. This variable
is sponsored by the International Rescue is a dummy and is given the value ‘1’ if pre-
Committee (IRC) and funded by the European sent within a province and ‘0’ if not present.Art. 9, page 8 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
The number of ‘formal’ and ‘informal’ The distance of each province’s capital from
schools has been replaced by the number of Kabul was calculated using FreeMapTools.
students enrolled in government-run gen- com, which provides a distance calculator
eral education and private general educa- using Google Maps’ application program
tion schools. As information on food aid—a interface (API). Data on coalition fatalities
variable included by Watts to test economic are from iCasualties.org, while information
prosperity and NGO activity—was not readily on journalist security incidents is from Nai.
available, it has been replaced by the number Nai, an Afghan NGO, supports open media
of NGO projects conducted and total NGO with the assistance of Internews and has
expenditures by province. NGOs often move been tracking threats against media work-
from province to province, and monitoring ers throughout Afghanistan for the past ten
individual agency activity over an extended years. Percentages of radio and mobile device
period of time can be difficult. Therefore, ownership and access to the Internet were
the number of projects and the amount of obtained from Internews, which made availa-
money spent by all organizations within a ble data from Altai Consulting’s comprehen-
province are the best determinants of overall sive Afghan Media in 2010 report that was
NGO activity. commissioned by USAID. Finally, provincial
Watts included the percentage of home border information was derived from maps
radio ownership in his study as a means of published by ISAF. As provinces which share
testing economic prosperity. However, he a border with Pakistan tended to experience
was later informed by a Pakistani military greater overall levels of violence, this variable
officer that the US military provided radios was included to test if the security situation
to locals in dangerous provinces as part of was also more volatile for aid workers. Both
reconstruction efforts and this measure- border variables are dummies.
ment may not be an adequate indicator of
modernization and economic advancement. Empirical Analysis
Therefore, this study includes the percentage Descriptive Statistics
of those who own a mobile device, in addi- Table 1 provides descriptive statistics of the
tion to those who have access to the Internet. variables included in the model by province.
Coalition fatalities, number of people incar- There was an average of four NGO security
cerated, and number of security incidents incidents per province per year in 2010 and
against journalists have also been included 2011, with a high of 18 incidents in 2010 in
as a means of measuring the overall security Herat (see Figure 1). As Figure 1 indicates,
of each province. no security incidents occurred in 2010–2011
Data on population figures, enrollment in in Badghis, Daykundi, Nimroz, Paktika,
government-run general education programs, Panjshir, and Zabul. Some provinces had
enrollment in private general education incidents in one year only, including Bamyan
schools, number of people incarcerated, NGO (2010), Helmand (2010), Samangan (2011),
projects, and NGO expenditures are from the and Uruzgan (2011). Maps 1 and 2 provide
Afghanistan Central Statistics Organization a comparative visual overview of the security
(ACSO). Information on poppy cultivation situation for both years.
was gathered from the UN Office on Drugs The average number of NGO projects per
and Crime’s (UNODC) 2011 Afghanistan province was 174, with a minimum of 15
Opium Survey. Provinces considered ‘poppy- in 2010 in Zabul and a maximum of 1,250
free’ by UNODC are those estimated to have in 2011 in Kabul. The number of NGO pro-
less than 100 hectares of cultivation, and are jects increased exponentially across all prov-
coded as ‘0’ in the UN survey.5 Poppy cultiva- inces, from 1,165 in 2010 to 5,140 in 2011,
tion and total NGO expenditures were logged an increase of 341 per cent.6 Average annual
to provide a normal distribution. NGO expenditures were $21.29 million, withMitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 9 of 18 NGO Insecurity in Afghanistan, 2010-2011 Variable Observations Mean SD Minimum Maximum NGO Security Incidents 68 4.49 4.88 0.00 18.00 PRT Presence 68 0.76 0.43 0.00 1.00 US PRT Presence 68 0.41 0.50 0.00 1.00 Coalition PRT Presence 68 0.44 0.50 0.00 1.00 US Military Presence 68 0.74 0.44 0.00 1.00 NGO Projects 68 173.76 183.05 15.00 1,250.00 Total NGO Expenditures (in 68 21.29 34.52 0.64 188.84 millions $USD) Coalition Fatalities 68 17.22 41.37 0.00 290.00 Journalist Security Incidents 68 1.44 4.03 0.00 32.00 People Incarcerated 88 651.35 1,261.25 1.00 7,591.00 Hectares of Poppy 68 3,72915 11,860.82 0.00 65,045.00 Cultivation Afghan Border 68 0.56 0.50 0.00 1.00 Border with Pakistan 68 0.32 0.47 0.00 1.00 Distance of Province Capital 68 173.88 122.81 0.00 492.37 from Kabul (in miles) % of Population Own a 68 70.62 21.74 22.00 99.00 Radio % of Population Own a 68 61.26 25.09 8.00 97.00 Mobile Device % of Population with 58 7.32 8.78 0.00 33.00 Internet Access People Enrolled in Govt 68 215.98 207.69 21.11 1,086.82 General Education (in thousands) People Enrolled in Private 68 2,209.60 7,713.19 0.00 49,371.00 General Education Population (in thousands) 68 727.55 645.03 136.30 3,818.70 Table 1: Descriptive Statistics (by Province). a minimum of $640,000 in 2010 in Panjshir of poppy cultivation was 3,729 hectares, and a maximum of $188.835 million in 2011 with a high of 65,045 in 2010 in Helmand. in Kabul. There was an average of 17 coalition fatali- Data Analysis ties per province per year, with a high of 290 Table 2 provides statistical results using fatalities in 2010 in Helmand. The average panel-corrected standard errors of all 34 number of journalist security incidents was provinces in 2010 and 2011.7 Models 1 and 2 one, although Kabul in 2011 experienced a fit the data well, as indicated by their highly high of 32. The average number of people statistically significant chi-squares. Model 1 incarcerated was 651, with a maximum of reveals that all variables except NGO projects, 7,591 in 2011 in Kabul. The average amount coalition fatalities, and the number of people
Art. 9, page 10 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
Figure 1: NGO Security Incidents 2010 vs 2011.
incarcerated were statistically significant at that PRT presence was likely to increase
ten per cent or better.8 Consistent with NGO NGO security incidents by three; however,
complaints, the model suggests that if PRTs US military presence decreased incidents
were present within a province, NGOs were by one, holding other variables constant. A
likely to encounter more security incidents. second model was run in which general PRT
However, the findings also reveal that US presence was replaced with both US- and
military presence was likely to be associated coalition-led PRT presence. This model was
with fewer incidents. The results suggest developed to assess if there was a differenceMitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 11 of 18 NGO Insecurity in Afghanistan, 2010-2011 Map 1: 2010 NGO Security Incidents. between teams with different leads, as NGOs coalition teams were structured differently were less likely to encounter incidents in and engaged the populace in unique ways. areas where US troops were present. Although it is possible that specific methods The results of Model 2 are revealing. utilized by non-US teams may have resulted ‘US-led teams’ was not statistically signifi- in decreased NGO security, this needs to be cant; however, ‘coalition-led teams’ was sig- analyzed in greater detail. nificant at the one per cent level. The model Similar to Watts’ findings, provinces which suggests that if a coalition-led team was pre- share a border with Pakistan and those with sent within a province, NGOs were likely to a higher percentage of home radios were experience three more incidents, holding likely to experience violence against NGOs. all other variables constant. This accounts The AfPak border was largely perceived as a for all incidents indicated by the general generally volatile region for military actors, ‘PRT Presence’ variable in Model 1. Model 2 and both studies suggest these dangers therefore suggests that PRT presence within extended to NGOs. As previously mentioned, a province tended to result in decreased although Watts initially used radio owner- security for NGOs, but only in provinces ship to control for economic prosperity, he with coalition-led teams; US PRTs were insig- was later informed that the US military pro- nificant indicators of NGO security.9 US and vided radios to locals in dangerous provinces
Art. 9, page 12 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
Map 2: 2011 NGO Security Incidents.
as part of reconstruction efforts during the Also consistent with Watts, provinces with
early stages of the conflict. This suggests higher levels of poppy cultivation tended to
higher levels of radio ownership may equate be safer areas for NGOs. Thus, cultivation
to greater volatility (Watts 2004: 16). The may actually have had a stabilizing effect
findings of this study appear to support this on the region. One speculation is that cul-
hypothesis. Additionally, results call into tivation only occurs in regions with exist-
question the effectiveness of expanded radio ing stability, as farmers may not want to
access as a means of quelling violence. risk growing crops in conflict-ridden zones.
Although radio ownership holds a positive Another possibility is that many of the pro-
relationship, mobile device ownership and jects NGOs engage in are beneficial to culti-
access to the Internet were found to be nega- vators. These organizations not only provide
tive. These variables were incorporated to test the necessary materials and equipment to
for economic advancement within a province. farmers, but also engage in infrastructure
Given that provinces with greater access to projects such as roads and bridges which
these technologies experienced decreased facilitate crop transportation within rural
attacks on aid workers, perhaps more econom- areas of the province.
ically developed areas were less likely to either Moreover, the findings reveal that, although
produce or host actors who targeted NGOs. the number of NGO projects conducted wasMitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 13 of 18
NGO Insecurity in Afghanistan, 2010-2011
Model 1 Model 2
Variable Coefficients SE Coefficients SE
PRT Presence 3.3328 ***
0.6421 -- --
US PRT Presence -- -- 1.6548 1.8711
Coalition PRT Presence -- -- 2.8056*** 0.2588
US Military Presence -1.2755 **
0.7266 -0.7159 1.5382
NGO Projects -0.0022 0.0064 -0.0011 0.0057
LOG Total NGO 1.3680 *
0.9258 1.2718 *
0.9517
Expenditures (in mil-
lions $USD)
Coalition Fatalities -0.0039 0.0032 -0.0094*** 0.0028
Journalist Security 0.2062*** 0.0537 0.1150* 0.0732
Incidents
People Incarcerated -0.0006 0.0008 -0.0005 0.0007
LOG Hectares of Poppy -0.4652 ***
0.1187 -0.4136***
0.1217
Cultivation
Afghan Border -3.1086*** 0.5739 -2.4404*** 0.5491
Border with Pakistan 2.7003 *
1.7939 3.0242 **
1.4465
Distance of Province 0.0264*** 0.0054 0.0243*** 0.0044
Capital from Kabul (in
miles)
% of Population Own a 0.0987*** 0.0068 0.1006*** 0.0181
Radio
% of Population Own a -0.0224* 0.0168 -0.0266*** 0.0115
Mobile Device
% of Population with -0.0584*** 0.0173 -0.0599*** 0.0270
Internet Access
People Enrolled in Govt 0.0298*** 0.0033 0.0285*** 0.0033
General Education (in
thousands)
People Enrolled -0.0003** 0.0002 -0.0003* 0.0002
in Private General
Education
Constant -12.8101*** 1.3384 -12.3337*** 1.2981
R2 0.58 R2 0.58
Wald Chi2 76.99*** Wald Chi2 57.78***
N = 68 N = 68
p < 0.01; **p < 0.05; *p < 0.10, one-tailed test.
***
Table 2: Panel-Corrected Standard Error Models of NGO Security Incidents by Province.Art. 9, page 14 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
statistically insignificant, total expenditures for causing ‘blurred lines’ between military
held a statistically significant relationship and aid activity. Although certain NGOs have
with increased incidents. One speculation accused PRTs of decreasing the overall security
is that NGOs may have been more likely to situation for aid workers within a province,
encounter attacks while engaged in a ‘large there was a lack of empirical evidence to sup-
scale’ activity that utilized a greater number port this claim. A report that often surfaces
of resources. When NGOs spend significant in this debate is Clinton Watts’ 2004 study,
amounts of time and money in one area, which analyzed NGO security incidents in
insurgents may be able to monitor daily Afghanistan that occurred from 1 January
activity closely, thus rendering aid workers 2003 to 15 February 2004. Although Watts’
more vulnerable to attacks. findings revealed that US military presence
Although the number of coalition fatalities was not a significant variable concerning NGO
and the number of people incarcerated were security, PRTs were not included in his analysis.
statistically insignificant, journalist security To build upon Watts’ findings, this study
incidents held a statistically significant rela- ran a panel-corrected standard error regres-
tionship with NGO security incidents. This sion model with the inclusion of PRT pres-
suggests that provinces where NGOs encoun- ence as an independent variable. Results
tered attacks may have been hostile areas indicated that if PRTs were present within a
for other civilian actors as well. Information province, NGOs were likely to encounter a
on Afghan civilian casualties should be greater number of security incidents; how-
included in a future analysis to further test ever, if the US military was present, NGOs
this hypothesis. After accounting for the dif- were likely to experience fewer incidents.
ference in team leads in Model 2, coalition Given that the findings revealed a posi-
fatalities became statistically significant at tive relationship between increased NGO
the one per cent level. incidents and PRT presence, yet a negative
Finally, educational enrollment was also relationship with US military presence, a sec-
shown to be a significant determinant of ond model was run substituting PRTs with
incidents. Interestingly, provinces with a high US- and coalition-led PRTs more specifically.
number of citizens enrolled in government- Model 2 results indicated that US PRTs were
run education programs held a positive rela- insignificant indicators of NGO security;
tionship with NGO attacks, while provinces however, coalition PRTs were found to be
with a high number of people enrolled in highly significant. Given these results, it is
private education programs held a negative difficult to conclude that PRTs in general are
relationship. Similar to mobile phone own- culpable for lax NGO security.
ership and Internet access, perhaps a greater As the ‘blurred line’ explanation is not
number of people enrolled in private edu- supported in this context, policymakers
cation indicates higher levels of economic seeking remedies for aid worker insecurity
prosperity in a region and therefore results should examine other possible causes of the
in less incentive to turn to violence. increased attacks. The results of this study
concur with Watts that a reevaluation of
Conclusion anti-drug policies directed towards eradicat-
In an effort to curtail the insurgency in ing poppy cultivation is needed. Additionally,
Afghanistan, the US and coalition forces have provinces with higher levels of private school
blended military and humanitarian opera- enrollment, mobile phone ownership, and
tions, much to the dismay of many within the access to the Internet all held a negative rela-
NGO community. One of the major debates tionship with NGO incidents, suggesting that
surrounding this effort concerns the PRT greater economic advancement may result in
initiative, which many NGOs have faulted less incentive to turn to violence. However,Mitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 15 of 18
NGO Insecurity in Afghanistan, 2010-2011
the paradox faced by aid workers is that they analyses has recently been suggested by Fast
tend to operate in less developed regions of (2014). Aid workers—both as individuals and
the country by design. within organizations—approach their work
Additionally, this study’s findings reveal a in diverse ways, and not all are uniformly
positive relationship between higher levels perceived as favorable or unfavorable by the
of NGO expenditures and security incidents. communities in which they operate. NGOs
This suggests that aid workers may have expe- can contribute to these perceptions through
rienced greater attacks in regions where they their individual interactions and organiza-
were most active (although total NGOs projects tional practices, and internal dynamics must
was statistically insignificant). Nonetheless, be analyzed for a more complete understand-
without access to the number of aid workers ing of the cause of these attacks.
operating in each province, it is difficult to
test for incident proportionality. If data from Replication Data
the projects that NGOs were engaged in when The data set and STATA output for this study
they were attacked—including length of time are available at www.davidfmitchell.com/data.
in the field, manpower, and public visibility—
is made available in the future, this should be Acknowledgements
analyzed in greater detail. An earlier version of this article was presented
Given the results of this study, future at the 2014 Annual Meeting of the Midwest
research should seek to explain the different International Studies Association in St. Louis,
approaches of lead nations in greater detail. MO. I am grateful to Emizet Kisangani, COL.
Although it is widely understood that coali- J. Bryan Mullins, and the anonymous review-
tion forces approached their PRT mandate in ers for their helpful comments.
unique ways, an in-depth comparative study
would be beneficial for researchers. As the Notes
scholarly literature on PRTs is limited, a pro- 1
Kidnappings have been the most com-
ject of this sort may best be served by exten- mon form of attack against aid workers
sive interviews with those who have served worldwide since 2009, and have steadily
on teams in the field. increased from seven in 2003 to 134 in
If information becomes available, future 2013.
studies on NGO insecurity should include the 2
It should be noted that PRTs in Iraq had
nature and affiliation of the assailant, whether different compositions and missions
the NGO was directly targeted, if the NGO was from their counterparts in Afghanistan.
local or international, what type of activity This article strictly addresses the latter.
or project the NGO was engaged in (includ- 3
Barnett (2011) refers to these two camps
ing length of time in the field, manpower, as emergency humanitarianism and
and public visibility), and where the incident alchemical humanitarianism. The first
occurred in proximity to PRT operations. In largely focuses on symptoms, while the
smaller provinces, PRTs may have been opera- latter seeks to remove the root causes of
tional in most areas; however, in much larger human suffering.
provinces, an NGO may come under attack in 4
An important caveat is that a lack of aid
an area with no history of PRT activity. worker incidents in a region may not be
In addition to external variables, it is due to decreased violence or vulnerabil-
equally important to analyze the internal ity; rather, it could reflect the absence of
dynamics of NGOs operating in the field. NGOs operating in that area due to secu-
This is a critical factor which is often omitted rity concerns.
from studies concerning aid worker security. 5
‘Poppy-free’ provinces were given a value
The inclusion of internal vulnerabilities in of one in the data set.Art. 9, page 16 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
6
Whether this is truly representative or Afghan%20Media%20in%202010.pdf
simply the result of improved reporting [Last accessed 9 January 2015].
is debatable. Barnett, M 2011 Empire of humanity: A his-
7
‘Significance’ relates to the probability tory of humanitarianism. Ithaca and Lon-
that a given variable ‘x’ has no effect on don: Cornell University Press.
the dependent variable ‘y.’ Thus, the three Barnett, M and Weiss, T G 2008 Humani-
levels of significance imply that the prob- tarianism: A brief history of the present.
ability that variable ‘x’ has no impact on In: Barnett, M and Weiss, T Humanitari-
variable ‘y’ is 10%, 5%, or 1%. anism in Question: Politics, Power, Ethics.
8
A variance inflation factor (VIF) was Ithaca and London: Cornell University
used to test for multicollinearity. Results Press. pp. 1–48.
showed that the coefficient of the ‘popu- CARE International 2003 A New Year’s
lation’ variable was above the critical level resolution to keep: Secure a lasting peace
of 30. Once the variable was removed, the in Afghanistan. Policy Brief (January).
coefficients of the remaining variables all London: CARE International. Available at
fell below 16, with a mean of 5.51. http://reliefweb.int/sites/reliefweb.int/
9
US military presence became statistically files/resources/169E8405579D7C32492
insignificant after controlling for the dif- 56CB60026AF14-care-afg-13jan.pdf [Last
ference in team leads. accessed 9 January 2015]: CARE.
Center for Army Lessons Learned 2011
References Afghanistan provincial reconstruction
Afghanistan Central Statistics Organi- team: observations, insights, and lessons.
zation 2011 Statistical yearbook, 2010– Available at http://cgsc.contentdm.
2011. Available at http://cso.gov.af/en/ oclc.org/utils/getdownloaditem/collec-
page/1500/4722/4725 [Last accessed 9 tion/p15040coll4/id/31/filename/32.
January 2015]: ACSO. pdf/mapsto/pdf/type/singleitem [Last
Afghanistan Central Statistics Organi- accessed 9 January 2015]: CALL.
zation 2012 Statistical yearbook, 2011– Center for Humanitarian Cooperation
2012. Available at http://cso.gov.af/en/ 2003 The provincial reconstruction team
page/1500/4722/7108 [Last accessed 9 (PRT) in Afghanistan and its role in recon-
January 2015]: ACSO. struction. Available at http://reliefweb.
Afghanistan NGO Safety Office 2011 int/report/afghanistan/provincial-recon-
ANSO quarterly data report q.4 2011. struction-team-prt-afghanistan-and-
Available at http://reliefweb.int/ its-role-reconstruction [Last accessed 9
sites/reliefweb.int/files/resources/ January 2015]: CHC.
ANSO%2520Q4%25202011.pdf [Last Dziedzic, M J and Seidl, M K 2005 Provin-
accessed 9 January 2015]: ANSO. cial reconstruction teams and military
Agency Coordinating Body for Afghan relations with international and nongov-
Relief 2005 Code of conduct for NGOs ernmental organizations in Afghanistan.
engaged in humanitarian action, reconstruc- Washington, DC: United States Institute
tion, and development in Afghanistan. of Peace. Available at http://www.usip.
Available at http://afghantranslation. org/sites/default/files/sr147.pdf [Last
checchiconsulting.com/documents/reg- accessed 9 January 2015].
ulation/Code_of_Conduct_for_NGOs- Fast, L A 2010 Mind the gap: Documenting
-from_Acbar-org_English_ET.pdf [Last and explaining violence against aid work-
accessed 9 January 2015]: ACBAR. ers. European Journal of International
Altai 2010 Afghan media in 2010: Syn- Relations, 16(3): 365–389. DOI: http://
thesis report. Available at http://www. dx.doi.org/10.1177/135406610935
altaiconsulting.com/docs/media/2010/ 0048Mitchell: Blurred Lines? Provincial Reconstruction Teams and Art. 9, page 17 of 18 NGO Insecurity in Afghanistan, 2010-2011 Fast, L A 2014 Aid in danger: The perils and tarian-military relations in Afghanistan. promise of humanitarianism. Philadel- London: Save the Children. Available phia: University of Pennsylvania Press. at http://www.savethechildren.org.uk/ Free Map Tools 2014 How far is it between. sites/default/files/docs/Provincial_ Available at http://www.freemaptools. Reconstruction_Teams_and_Humani- com/how-far-is-it-between.htm [Last tarian-Military_Relations_in_Afghani- accessed 9 January 2015]. stan_2004_09_1.pdf [Last accessed 9 Hammond, L 2008 The power of holding January 2015]. humanitarianism hostage and the myth Morgan, W 2012 Afghanistan order of battle: of protective principles. In: Barnett, M and February 2009–December 2012. Washing- Weiss, T Humanitarianism in Question: ton, DC: Institute for the Study of War. Politics, Power, Ethics. Ithaca and London: Available at http://www.understanding- Cornell University Press. pp. 172–195. war.org/sites/default/files/afORBATs_ iCasualties 2014 Coalition fatalities by year Feb2009-Dec2012.pdf [Last accessed 9 and province. Available at http://icasu- January 2015]. alties.org/OEF/ByProvince.aspx [Last Nai 2013 Violence against journalists in accessed 9 January 2015] Afghanistan 2001–2013. Available at Institute for the Study of War Provincial http://data.nai.org.af/ [Last accessed 9 reconstruction teams (PRTs). Available at January 2015]. http://www.understandingwar.org/pro- Perito, R M 2005 The US experience with pro- vincial-reconstruction-teams-prts [Last vincial reconstruction teams in Afghani- accessed 9 January 2015]: ISW. stan: Lessons identified. Washington, DC: InterAction 2013 The U.S. military’s expand- United States Institute of Peace. Available at ing role in foreign assistance. Available at http://www.usip.org/sites/default/files/ http://www.interaction.org/files/FABB%20 sr152.pdf [Last accessed 9 January 2015]. 2013_Sec16_NGOAndMilitaryRelations.pdf Special Inspector General for Afghani- [Last accessed 9 January 2015]. stan Reconstruction 2010 Quarterly International Security Assistance Force report to the United States Congress (April). 2009 Provincial reconstruction team (PRT) Available at http://www.sigar.mil/pdf/ handbook. 4th ed. Available at http://info. quarterlyreports/2010-04-30qr.pdf [Last publicintelligence.net/ISAF-PRThand- accessed 9 January 2015]: SIGAR. book.pdf [Last accessed 9 January 2015]: Special Inspector General for Afghani- ISAF. stan Reconstruction 2011 Quarterly Internews 2010 Media landscape in Afghan- report to the United States Congress (July). istan. Available at http://data.internews. Available at http://www.sigar.mil/pdf/ org/af-media/ [Last accessed 9 January quarterlyreports/2011-07-30qr.pdf [Last 2015]. accessed 9 January 2015]: SIGAR. Jalali, A 2007 The legacy of war and the chal- Special Inspector General for Afghani- lenge of peace building. In: Rotberg, R stan Reconstruction 2012a Quarterly Building a New Afghanistan. Cambridge: report to the United States Congress (July). World Peace Foundation. pp. 22–55. Available at http://www.sigar.mil/pdf/ Malkasian, C and Meyerle, G 2009 Provincial quarterlyreports/2012-07-30qr.pdf [Last reconstruction teams: How do we know they accessed 9 January 2015]: SIGAR. work? Carlisle, PA: Strategic Studies Insti- Special Inspector General for Afghani- tute. Available at http://www.strategics- stan Reconstruction 2012b Quarterly tudiesinstitute.army.mil/pdffiles/PUB911. report to the United States Congress (Octo- pdf [Last accessed 9 January 2015]. ber). Available at http://www.sigar.mil/ McHugh, G and Gostelow, L 2004 Provin- pdf/quarterlyreports/2012-10-30qr.pdf cial reconstruction teams and humani- [Last accessed 9 January 2015]: SIGAR.
Art. 9, page 18 of 18 Mitchell: Blurred Lines? Provincial Reconstruction Teams and
NGO Insecurity in Afghanistan, 2010-2011
Special Inspector General for Afghani- Available at http://www.unodc.org/doc-
stan Reconstruction 2013 Quarterly uments/crop-monitoring/Afghanistan/
report to the United States Congress Afghanistan_opium_survey_2011_web.
(July). Available at http://www.sigar. pdf [Last accessed 9 January 2015]:
mil/pdf/quarterlyreports/2013-07- UNODC.
30qr.pdf [Last accessed 9 January 2015]: Waldman, M 2009 Caught in the conflict:
SIGAR. Civilians and the international security
Stoddard, A, Harmer, A, and Ryou, K 2014 strategy in Afghanistan. Available at
Aid worker security report 2014: Unsafe http://www.oxfam.org/sites/www.
passage: Road attacks and their impact oxfam.org/files/civilians-caught-in-the-
on humanitarian operations. Humanitar- conflict-afghanistan.pdf [Last accessed 9
ian Outcomes (August). Washington, DC: January 2015].
USAID. Available at https://www.human- Watts, C 2004 Indicators of NGO security in
itarianoutcomes.org/sites/default/ Afghanistan. West Point: United States
files/Aid%20Worker%20Security%20 Military Academy, The Combating Terror-
Report%202014.pdf [Last accessed 9 Jan- ism Center. Available at http://www.dtic.
uary 2015]. mil/cgi-bin/GetTRDoc?Location=U2&do
UN Office on Drugs and Crime 2011 c=GetTRDoc.pdf&AD=ADA467616 [Last
UNODC Afghanistan opium survey 2011. accessed 9 January 2015].
How to cite this article: Mitchell, D F 2015 Blurred Lines? Provincial Reconstruction Teams and NGO
Insecurity in Afghanistan, 2010–2011. Stability: International Journal of Security & Development, 4(1):
9, pp. 1-18, DOI: http://dx.doi.org/10.5334/sta.ev
Published: 04 March 2015
Copyright: © 2015 The Author(s). This is an open-access article distributed under the terms of the
Creative Commons Attribution 3.0 Unported License (CC-BY 3.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original author and source are credited.
See http://creativecommons.org/licenses/by/3.0/.
Stability: International Journal of Security & Development is a
peer-reviewed open access journal published by Ubiquity Press
OPEN ACCESSYou can also read
