ARTICLE REVIEWS

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=fint20

Intelligence and National Security

ISSN: 0268-4527 (Print) 1743-9019 (Online) Journal homepage: http://www.tandfonline.com/loi/fint20

Counterinsurgency agent networks and noncombatant-targeted violence

Britta Stime

To cite this article: Britta Stime (2017) Counterinsurgency agent networks and noncombatant-targeted violence, Intelligence and National Security, 32:1, 107-125, DOI: 10.1080/02684527.2016.1210770

To link to this article: https://doi.org/10.1080/02684527.2016.1210770

Published online: 26 Jul 2016.

Submit your article to this journal

Article views: 1074

View related articles

View Crossmark data

IntellIgence and natIonal SecurIty, 2017 Vol. 32, no. 1, 107–125 http://dx.doi.org/10.1080/02684527.2016.1210770

ARTICLE

Counterinsurgency agent networks and noncombatant-targeted violence

Britta Stime

School of Politics, International Studies and Philosophy, Queen’s university Belfast, Belfast, uK

ABSTRACT In counterinsurgency, agent networks are double-edged swords. They are useful tools for degrading insurgent influence and protecting the population. However, they also endanger the population in some ways, as we have seen with mass executions of suspected agents and agent misdirection of raids. Identifying how/why this occurs is critical for developing intelligence practices to more effectively implement COIN strategies. This exploratory study uses three recent counterinsurgency cases – for which significant secondary open-source agent network documentation is available – to identify, describe, organize and analyse patterns of noncombatant-targeted violence associated with human intelligence networks. Identified cases of noncombatant-targeted violence in Operation Enduring Freedom – Afghanistan, Operation Iraqi Freedom, and the – Israeli–Palestinian conflict are used to develop a theoretical framework that models intelligence-related violence incentivization, which also draws from theoretical literature on human intelligence, civil war violence, police states, community policing, and the sociology of betrayal.

Agent networks are an essential feature of counterinsurgency (COIN) operations. They are extremely important tools for disrupting and degrading insurgent networks and protecting the population. However, they are double-edged swords. Agent networks may also inadvertently motivate corollary violence against noncombatants. For example, intelligence officers lament that, in Afghanistan and Iraq, they lose noncombatants ‘all the time’ whom they recruit to act as agents;1 on 22 August 2014, HAMAS carried out a summary execution of 18 Palestinian noncombatants on suspicions of spying;2 and the Islamic State has repeatedly released videos of beheadings of alleged agents.

Thousands of noncombatants have been severely injured or killed by coercive recruitment tactics, covert and clandestine activities, unauthorized criminal conduct by agents, targeting operations based on poorly validated human intelligence (HUMINT), counterintelligence preventive violence and pun- ishment violence.

Surprisingly, despite the evident dangers of agent networks in COIN and their double-edged effects on securing populations, discussion of agent network-associated violence in the intelligence literature is extremely limited. Most literature on this topic is prescriptive, dealing with how to expand and improve HUMINT collection.3 Normative literature on the subject evaluates human rights dilemmas involving security forces and their agents.4 The literature does not, however, investigate the broader, systemic incentives of agent networks that may inadvertently induce noncombatant casualties.

© 2016 Informa uK limited, trading as taylor & Francis group

CONTACT Britta Stime [email protected]

108 B. STIME

Agent network-associated noncombatant casualties have likely eluded academic investigation because they are often not visible, or viewed as arbitrary, incidental, or necessary evils of wartime intelligence collection that are not directly attributable to the incumbent government. Additionally, the individuals killed tend to be citizens of marginalized groups, such as foreigners and those considered traitors. This depoliticizes and deprioritizes their deaths.

This research shows that agent network-associated noncombatant casualties occur in patterns, cor- relating with systemic incentives induced by agent network recruitment and operation. Agent networks motivate all categories of actors in a conflict (i.e., the incumbent government, agents, noncombatants, and insurgents) to use violence to promote or neutralize agent acquisition5 and handling.6 These incen- tive structures7 for noncombatant-targeted violence8 follow an intelligence war framework as actors compete for access. This theoretical framework is presented in the proceeding sections.

Due to the lack of access to classified data, the assertions herein are based on open sources. This may limit the reliability and generalizability of the conclusions of this research;9 however, the importance of opening this subject to unclassified academic discussion compels the researcher to meet these methodological challenges with research that provides a preliminary examination of agent networks and noncombatant casualties.

Agent networks in COIN

Agent networks

The literature uses a diverse array of terms to refer to agents: informants, informers, sources, human sources, local sources, assets, national security assets, spies, collaborators, contributors, and local national employees, inter alia. These terms are commonly used interchangeably, even though they sometimes refer to different categories of actors. For example, informant and spy are often used inter- changeably; however, spy usually refers to a person who provides information10 to a foreign intelligence service, while informant generally refers to someone who provides information to a domestic intelli- gence/investigative or law enforcement organization. The use of the terms source and asset have grown increasingly common, because of a conscious effort by intelligence and law enforcement agencies to employ more neutral or positive labels. This study uses the word agent, because agent and asset are the most widely-used terms by foreign intelligence services. Agent also denotes one of the most common terms used to describe a broader collection of these individuals: agent network.

While the term agent networks (also referred to as human source networks and informant networks) is routinely used in intelligence manuals, it is not explicitly defined. Based on the information in these manuals,11 agent networks can be described as a subcategory of collaborators or denouncers12 (general terms referring to people who give information to the incumbent government along a formality contin- uum running from people who volunteer information,13 to people who are recruited, trained and paid by intelligence collectors).14 Intelligence organizations intentionally develop agent networks through multiple contacts with individuals they identify, target, recruit, and handle. While here the primary focus is on incumbent-developed networks, many other actors develop agent networks.15

Political-military utility

Agent networks are critical political-military tools in COIN. They serve as low-cost force multipliers – relative to the average trained, salaried soldier – that mirror insurgent advantages and combat insurgent strategies. A single intelligence officer operating in a new Area of Operation (AO) may expand his/her coverage through networks of agents who, like insurgents, hide among civilian populations; operate agilely, transnationally, in decentralized networks, and with few to no bureaucratic procedures; leverage local knowledge and relationships; and are less restricted by the laws of war.16

These unique capabilities help government security forces build a detailed pattern of life picture, locate high-profile insurgents (referred to as high-value targets), receive alerts of the presence of mines

INTELLIgENCE AND NATIONAL SECURITy 109

or improvised explosive devices and ambushes, monitor local populations, and characterize the oper- ational environment.17

State security forces often use agents as targeting scouts in neighborhoods, captive lineups, and raids.18 The rumored presence of agents has a psychological impact on civilian populations, aiding separation of insurgents from their civilian support base, the first key tenet of COIN.19 The known presence of agents sows suspicion and demoralization within organizations.20 It can also create the perception of popular support for the incumbent government while simultaneously delegitimizing the insurgent’s cause.21

Finally, agents are also used as agents provocateurs to conduct covert and clandestine activities (e.g., torturing suspects, placing homing devices for drone strikes, committing extra-judicial killings, sabo- taging insurgent operations, entrapping conspirators in law enforcement-monitored crimes, provoking internal dissension, and swaying insurgent decisions).22

Noncombatants vs. insurgents vs. agents

For the purposes of this paper, noncombatants refers to law-abiding civilians who are not integrally involved in the insurgent movement. The term insurgents is used to refer to those civilians who are unlawful combatants, actively involved in the insurgent movement. Agents operating on behalf of the incumbent, like insurgents, are also civilians who are unlawful combatants. They may not wield weapons or directly partake in violence, but they actively deal in lethal information. Agents are recruited from within insurgent groups and noncombatant civilian communities. Insurgents and agents use the civilian population as a cover. They hide among and endanger a particular population, group or demographic by exposing them to violence used for intelligence, targeting, and counterintelligence purposes.

It is important to note the clear distinction between recruiting an insurgent versus a noncombatant to work as an agent. The recruitment of an insurgent only exposes an existing combatant (and the insur- gent group that is his/her cover) to additional possibilities of intelligence-, targeting-, and counterintel- ligence-related violence. Conversely, the recruitment of a noncombatant as an agent transforms him/ her into an unlawful combatant, making the individual a legitimate target of violence, and potentially exposing the individual’s noncombatant community or demographic to intelligence, targeting, and counterintelligence-related violence.23 For this reason, while the theory presented in this study focuses only on modeling violence targeted at noncombatants, it also sometimes includes violence targeted at noncombatants-turned-agents, as these targets often overlap.

Incentives for violence

Intelligence war framework

Intelligence is a competition for access, commonly referred to as an intelligence war, in which actors struggle to gather intelligence while preventing their opposition from doing the same through coun- terintelligence strategies and tactics. In the context of COIN HUMINT, this is manifested as the struggle to access the population (who possesses valuable information) through the development and oper- ation of agent networks. Agent network development refers to agent recruitment (acquisition). Agent network operation refers to agent handling (i.e., the use of agents for intelligence collection, insurgent identification and location, covert and clandestine activities, and population control). This HUMINT war framework is illustrated in Figure 1.

This paper focuses on the left side of Figure 1: incumbent HUMINT gathering and insurgent counter- intelligence activities.24 In this context, as the incumbent vies to develop and operate agent networks (illustrated with the light gray downward-pointing arrow on the left), the insurgents and civilian pop- ulations sympathetic to their cause struggle to counter the development and operation of incumbent agent networks (illustrated with the dark gray, upward-pointing arrow).

Actors may use violence against noncombatant populations for intelligence and counterintelligence purposes according to the logic of this intelligence war framework.

110 B. STIME

Violence against noncombatants in the intelligence war framework

Available evidence suggests that the competition for access to HUMINT may incentivize all categories of actors in an insurgency setting to employ violence to promote and to counter the development and operation of agent networks within the population.

All actors in a COIN conflict – the incumbent, agents, noncombatants, and insurgents – are both instigators and targets of violence. Literature often misinterprets political violence based on the false premise that only the traditional state and insurgent actors actively participate in a conflict, while noncombatants simply get caught in the middle of the fighting. However, Stathis Kalyvas’ theory25 of violence in civil war shows how political violence in civil conflict is a joint venture involving the incum- bent, the insurgents, the noncombatants and the agents alike (which he generalizes as ‘denouncers’).

When we analyse violence against noncombatants associated with the incumbent’s agent networks, we see that, in accordance with Kalyvas’ theory, all the categories of actors in the conflicts use violence (among many other methods of influence) to drive intelligence and counterintelligence. The incumbent uses violence against noncombatants to develop and operate agent networks. The incumbent’s agents use violence while they are being recruited and handled. The noncombatants use violence to counter the development and operation of agents within their communities.26 And, finally, the insurgent group and its agents (groups which are combined, because they are indistinguishable in the literature) employ violence to counter the development and operation of incumbent agent networks.

This research indicates that the development and operation of agent networks by the incumbent motivates six patterns of violence, categorized as: coercion, covert and clandestine activities, unau- thorized crime, misdirected targeting, preventive violence, and punitive violence. Figure 2 provides a theoretical framework for conceptualizing these six patterns of violence. Each pattern correlates with the actor and its incentive (developing and operating agent networks vs. countering the development and operation of agent networks).

These six patterns of violence against noncombatants follow a systemic logic of incentives; each is described and analysed below, using open-source literature on intelligence incidents from the three selected case studies previously mentioned.

Patterns of violence against noncombatants and systemic motivation

Research design and cases

The presented theory of noncombatant-targeted violence associated with COIN agent networks (shown in the previous diagram) was developed through an analysis of three high-profile COIN cases that

Figure 1. Human intelligence war framework.

INTELLIgENCE AND NATIONAL SECURITy 111

exhibit significant documentation of casualties associated with intelligence practices: OEF-A, OIF and the Israeli–Palestinian conflict.

Israeli military and intelligence assets in the Israeli-Occupied Territories and the allied coalition forces involved in Iraq, Afghanistan, and the Federally Administered Tribal Areas (FATA) between Afghanistan and Pakistan have developed extensive agent networks. These agent networks, comprised mostly of noncombatants-turned-agents and minimal insurgents-turned-agents (due to a lack of access to insurgents), have functioned as crucial HUMINT collection resources, insurgent identifiers and locators, proxies for covert and clandestine activities, and social control mechanisms.

In OEF-A and OIF, specially-trained HUMINT collectors (as part of HUMINT collection teams HCTs)27 were forward deployed to the peripheries of hostile AOs to develop agent networks.28 Through daily close contacts with locals and refugees,29 cordon-and-search operations,30 block-by-block neighborhood searches, traffic checkpoints, and official liaison with a few trusted host nation law enforcement and intelligence organizations,31 soldiers identified potential contacts for intelligence officers to recruit as agents.32 In many cases, HUMINT collectors also used trusted local civilians as handlers (or principal agents) of their own sub-agent networks33 to augment their recruitment efforts. Similar to the Israeli internal security service’s (SHABAK) Palestinian agent networks, these agent networks helped the intervening forces collect HUMINT, identify insurgents within lineups of detainees and locate them in neighborhoods, prevent ambushes, and implement neighborhood systems of block control.34

The SHABAK uses elaborate agent networks of Palestinians from all spheres of public life in the occupied territories. These networks originated from mass arrests prior to and during the 1967 Six-Day War. The SHABAK has since greatly expanded these agent networks to include informers within the Palestine Liberation Organization (PLO) and other designated terrorist organizations, but also civilians in universities, unions, councils, and other realms of civil society. The SHABAK uses Palestinian agent networks to gather intelligence on Palestinian underground cells and their activities to predict and thwart potential terrorist attacks, to prevent the Palestinian population from consolidating around the PLO, and to disrupt nation-building processes.35

For each of these three selected cases, instances of violence against noncombatants associated with HUMINT were identified, described, actor-attributed, and traced to their purported purpose. As illustrated in Figure 2, these patterns of violence, across all three cases, have been collated by actor and purpose. In the following six sections, each pattern of violence identified is (1) described, (2) explained according to systemic incentive structures induced by agent network recruitment, and (3) illustrated using select examples from the three analysed cases. Each section is labeled according to the pattern’s typology (Actor, Type, Purpose): the Actor(s) directly implementing the violence; the Type of violence

Figure 2. actors and agent network development/operation violence framework.

112 B. STIME

targeting noncombatants; and the Purpose of the violence (agent network development, operation, counter-development, or counter-operation).

Incumbent coercion in network development

In order to develop agent networks, the incumbent sometimes uses coercion: violence, the threat of violence, deception, and the promotion of conditions that restrict access to basic needs.36 States use coercion to recruit noncombatants and insurgents who would not otherwise voluntarily agree to act as agents;37 coercion changes the individuals’ circumstances and decision calculus in favor of spying. The incumbent may also use coercion to compel an already-recruited agent to perform certain espionage and covert or clandestine operations that he/she would not otherwise agree to do.38 Through coercion, the incumbent converts an individual into an unlawful combatant (or into a more involved unlawful combatant) against his/her will. This study includes deception in the coercion definition, because this, too, turns a noncombatant into an unwitting or unwilling agent, thus exposing the individual to punitive violence for an action to which he/she had not knowingly consented. For example, the case of Alan Bani Odey in the Israeli–Palestinian conflict: the SHABAK tricked him into murdering his cousin; he was subsequently sentenced to death.39 Finally, while this study does not discuss morality in spying, it is important to note a moral distinction between the coercive recruitment of a noncombatant versus an insurgent; transforming a noncombatant into an agent and potential target of violence against his/her will is of greater moral significance than exposing an insurgent combatant to additional possibilities of violence as an agent.

As one would expect, volunteer agents are widely considered more reliable than agents who are coerced into their role.40 However, voluntary agents can be scarce, especially when insurgent organi- zations adeptly deter spying with extremely violent methods. The highly-volatile nature of COIN and high strategic utility of agents incentivizes the incumbent to use coercive recruitment methods to build its networks in the absence of other viable options.41

As of this date, there are no publicly-available data to estimate the frequency of such coercive practices in COIN. However, this study found many confirmed instances of incumbent use of coercive methods to recruit and run agents, similar to the cases described above. This is consistent with findings of earlier studies, such as Marx’s 1974 survey of US police agent recruitment, which is very similar to COIN agent recruitment. Marx showed that six of 15 agents (40%) were recruited through coercive methods (only legally-acceptable coercive methods – threat of imprisonment and deception through agent provocateurs – not violence or threat of violence, as this is prohibited in liberal Western democracies).42

There is significant evidence suggesting that incumbents in OEF-A, OIF, and the Israeli–Palestinian conflict have used coercion to recruit and run agents. Israel has used and threatened recruitment targets and/or the targets’ families with violence, imprisonment, torture, blackmail, misleading rumors con- demning individuals as agents, and the confiscation of travel permits required for work and medical care.

According to Morad Jadalah, a researcher with ADDAMEER, a prisoners’ rights organization in the West Bank, Palestinians are coerced through arrests, at checkpoints, and during raids. Cook reports:

Long jail terms and the use of administrative detention – imprisonment on secret charges – are the most obvious threats, but there are other ways to pressure Palestinians in detention, says Jadalah ... The interrogators may beat them, or threaten to beat or rape their mother or sister, or arrest a close relative ... They may even threaten to spread rumours that the family are already acting as agents.43

There have also been reports of Israeli military and intelligence assets blackmailing homosexual Palestinians into working as agents, since this revelation could result in their murder.44 In addition, Jadalah and co-authors Cohen and Dudai describe how the SHABAK has leveraged access to work permits and medical care to recruit agents. Jadalah explains:

They usually already know something about the family, so they can threaten, for example, to revoke the father’s work permit ... In other cases, the Israeli security services may offer inducements. Israel controls most people’s lives, including their ability to work and move around. Between 30 and 40 per cent of adults are unemployed. That gives Israel the leverage it needs to recruit collaborators.45

INTELLIgENCE AND NATIONAL SECURITy 113

Coalition intelligence assets in OEF-A and OIF are reported to have used similar forms of coercive recruitment. Highly-valued potential agents (both insurgent and noncombatant) were threatened with detention in guantanamo Bay or with transfer to foreign security services with a history of human rights abuses. These individuals were sent to these locations to be further coerced if they initially declined to inform.46

Incumbent covert and clandestine activities in network operation

State security forces sometimes use agents for covert and clandestine activities: un-attributable, unpop- ular, or illegal violent acts, colloquially referred to as ‘dirty tricks’ or ‘dirty work’. Here, the focus is on the use of agents for covert and clandestine paramilitary operations and lethal action, although covert and clandestine operations encompass all operations meant to be plausibly deniable or conducted in secret. A few examples of the use of agents for covert and clandestine activities involving violence include sabotaging insurgent operations, provoking internal dissension, swaying insurgent decisions, torture, extra-judicial killings, and placing homing devices for drone strikes.47 There is significant evidence of Israeli, Afghan- and Iraqi-directed covert and clandestine operations by agents, especially for the place- ment of electronic homing devices for drone strikes and for torturing insurgent suspects and sources.

Incumbents use agent networks for covert and clandestine operations because they are plausi- bly deniable. The incumbent can ‘obfuscate responsibility of the state for certain types of operations’ through agent networks.48 It is important to note that covert and clandestine activities include forms of violence with a greater potential to generate noncombatant casualties, because the incumbent is initiating operations that it would not be inclined to – or would be unable to – perform overtly for fear of political and/or legal ramifications.

While government agencies do not confirm their use, there is significant reported evidence of intelli- gence agencies employing agents to place electronic homing beacons, nicknamed patrai in the Middle East, to direct predator drone strikes on insurgent targets.49 Increased reliance on HUMINT for targeting will likely increased in use due to insurgent measures to counter signals intelligence (SIgINT)-driven targeting.

Israeli military and intelligence assets have allegedly used Palestinian informers to torture terrorist subjects; Cohen and Dudai write:

A common practice in Israeli police, military and SHABAK interrogation facilities is using informers, known as ‘birds.’ These informers introduce themselves as activists and force them [the terrorist subjects] to prove that they are not collaborators by telling who they are, what they did and who were the people that sent them on their missions ... ‘[B]irds’ also torture the detainees that refuse to disclose information.50

The use of agents to torture is evidenced in a 2013 case involving Palestinian prisoner Arafat Jaradat. Mr. Jaradat died while in detention at Megiddo prison in central Israel. His relatives claim his death resulted from harsh interrogation tactics carried out by the SHABAK and by Palestinian prisoners work- ing as agents.51

The SHABAK has also used agents to carry out extra-judicial killings. Returning to an example used in the Coercion section, the SHABAK booby-trapped the car of Palestinian agent Bani Odeh and requested that he lend it to his cousin. Unsuspecting, Odeh did so, and his cousin was killed in the explosion.52 By employing an agent for this operation, the SHABAK was attempting to portray it as an apolitical murder, thus obfuscating its political nature and sidestepping Israel’s legal and political responsibility.

Agent unauthorized crime in network development

The incumbent sometimes must protect agents from sanction for their criminal activities in order to keep the agents operating in the field. This indirectly precipitates casualties by subjecting noncombatants (as well as insurgents on occasion) to violence that might have been prevented had the perpetrator not been informing for the incumbent and thus been detained according to legal, law enforcement procedures upon his/her first criminal offense. Agents sometimes commit three general categories

114 B. STIME

of crimes: (1) low-level authorized crimes legally deemed essential for the maintenance of their cover as participating informers,53 (2) unauthorized crimes that are permitted by the handlers through the obstruction of justice, and (3) unauthorized crimes that are not permitted by their handlers.54

Around the world, agents have committed extrajudicial killings and torture without the consent or directive of their handlers. This is often mislabeled in the news media as collusion (which is a component of violence through covert and clandestine activities), because unauthorized crime can mix with covert and clandestine operations on a continuum of security force awareness of and complicity with such acts (i.e., when they further national political objectives).

The available sources did not indicate evidence of unauthorized crimes in OEF-A, OIF, and the Israeli–Palestinian conflict. However, the existence of this pattern of violence against noncombatants is probable, because the use of immunity is a commonly-practiced and generally-accepted recruitment technique in these cases. People with privileged access who have committed a crime are targeted, turned, and released into the public with the possibility of – and, often, tendency to – commit further crimes. Documentation of unauthorized crime in the selected cases is scarce because these intelli- gence wars are recent and ongoing; thus intelligence and legal documentation is still closely guarded to protect sources and methods. Unauthorized crimes, however, proved to be extremely prevalent in the Northern Ireland Troubles, a case with similar actors and HUMINT collection practices.55 Further research on this pattern of violence is warranted.

Agent-misdirected targeting in network operation

Agents sometimes provide false information about their targets. Conspiratorially and unwittingly, they misidentify individuals as targets who, in actuality, are not. As a result, the incumbent carries out tar- geted violence against people whom it would not have necessarily targeted had the information been validated.

Most, though not all,56 intelligence acquired from agents is cross-checked against other intelligence sources before the incumbent acts upon it. Intelligence packages for kinetic57 targeting are usually composed of at least a combination of SIgINT and HUMINT; satellite imagery, drone surveillance, and communications matrices (such as the National Security Agency’s (NSA) geo Cell program)58 inform and validate HUMINT collection and agent operations (agent identification of insurgents, descriptions of relationships, and location of targets).59 However, HUMINT is very difficult to verify at times, and SIgINT can falsely appear to confirm agents’ claims.60 Marx explains this in theoretical terms: ‘[the secrecy of the agent’s role] gives the [agent] an advantage over his employer, who in any given instance may find it hard to assess the accuracy of the information he gets’.61 This of course, must take into account intelligence agencies’ use of source and content ratings that, over time, help handlers assess the validity of each source’s information.

Informing is a component of noncombatant ‘denunciation’, which Kalyvas theorizes is often done ‘maliciously’ (self-servingly) and not just ‘politically’ (pursuing only the same political purposes as the political actor to which the information is supplied).62 In his theory of political violence and noncom- batant involvement as information sources, Kalyvas theorizes: ‘denunciation turns the production of selective violence into an outcome jointly produced by political actors and noncombatants’.63 He also notes that false denunciations are common: ‘individuals are tempted to settle private and local conflicts’ by supplying the incumbent with false or misleading information.64

Agents misdirect targeting by supplying misinformation (unintentionally false information) and disinformation (intentionally false, misleading information) to the incumbent, which informs kinetic targeting. Often, agents just want to collect their compensation without supplying real or certain infor- mation, or they have personal, local tribal, or political motives to intentionally mislead the incumbent to attack their own targets.

Agents intentionally mislead their handlers for political motives driven by the goals of the insurgent or by local, tribal power competitions. Sometimes, agents act as triple and re-doubled agents. Agents that have been doubled pretend to act as agents for the incumbent when, in fact, they are actually

INTELLIgENCE AND NATIONAL SECURITy 115

working for the insurgent. They supply information meant to deceive the incumbent in order to afford the insurgent a political or strategic advantage.

Agents also act on behalf of their tribes and tribal leaders, falsely accusing members of rival tribes as ‘insurgents’. Intelligence practitioners in OEF-A and OIF have expressed frustration that agents (tribal leaders included) are commonly using this tactic. For example, in Mosul, Iraq, American forces asked their Shia agent to identify insurgents from a lineup; all chosen were Sunni.65 Finer quotes two American soldiers in Afghanistan who acknowledge this form of disinformation: ‘they want to get everyone who’s not in their tribe’ and ‘we almost never get anything good from them [the agents] … I think they just pick people from another tribe or people who owe them money or something’. As the second soldier notes, agents also supply disinformation in order to carry out violence through the incumbent that would fulfill a personal vendetta.

Intelligence personnel have expressed concern about agents providing false information about potential targets simply to reap the monetary rewards from the incumbent and/or to direct incumbent violence at personal enemies. As Callam observes: ‘the lack of multiple intelligence sources also inhibits the ability of drones to accurately identify targets. Local agents are notoriously unreliable and can exploit the attacks for personal gain by pointing drone attacks toward personal rivals’.66 Malicious information supplied by agents is not always false; it may also be true, even though it was divulged with ulterior motives. For example, a 13-year-old Iraqi, whom American soldiers nicknamed ‘Steve-O’, informed on his father who had been abusing him and became an American war ‘hero’.67

Evidence suggests that disinformation is pervasive in the OEF-A, OIF, and Israeli–Palestinian conflict cases. Said Riegle, who runs a private intelligence firm that worked in Iraq and Afghanistan, observed that only ‘about half’ of the information from agents was reliable.68 Israel also reports extremely low reliability of Palestinian agent tips, attributed to HAMAS’ strategy of flooding the system with disinfor- mation to obscure true information and provoke the government to commit politically costly errors.

Mayer writes that Israel’s unmanned airstrikes in the Palestinian territories are ‘more exacting’ than the US’ in Iraq and Afghanistan. Despite this, an Israeli news agency notes that Israeli airstrikes are often misguided by faulty intelligence:

In the past, Israel has blamed HAMAS for civilian casualties, saying the group uses schools, mosques and residen- tial areas for cover while carrying out attacks. But cases of mistaken identity or faulty intelligence have also been known to result in civilian casualties.69

Human Rights Watch also reports70 that attacks included bombings of noncombatant locations (e.g., homes and farm groves). For example, a strike on 21 November 2013 killed a 48-year-old farmer, Talal al-Asaly and two of his children, Ayman, 19, and Abeer, 11, in northern gaza while working in their garden. Another drone strike two days earlier killed Ibrahim al-Astal, 48, and his nephew, Omar al-Astal, 14, in the southern area of Khan younis. In each case, residents said they did not observe Palestinian militants trying to fire missiles nearby, the usual target of such Israeli strikes.71 On 22 August 2014, an Israeli airstrike aimed at the chief of HAMAS’ military wing killed the leader’s wife and two of his children. Analysts assert that, based on the accuracy of this airstrike, Israel’s military would almost certainly have needed HUMINT to direct it.72

There is a significant number of accounts suggesting that incumbent raids and airstrikes in OEF-A and OIF have been executed using poorly validated HUMINT. Writing on OIF, Kahl observes:

Too often, raids have been based on dubious intelligence gleaned from agents seeking to settle scores. And when soldiers and marines have encountered armed resistance from families seeking to defend their homes, they have sometimes responded [in hindsight] with disproportionate force.73

Finer also records an instance in which American forces rounded up hundreds of villagers identified as insurgents by a Shia agent in Mosul, Iraq. He quotes a student who was among those detained: ‘your source is not good, these are all innocent men ... we are all Sunnis. That is why he [the agent] chose us. He is Shia’.74 In an interview with Semple, a member of the US Army’s Tactical HUMINT Team in Iraq, acknowledged that they needed to ‘rein in’ their agents, because they were ‘fingering, like, 25, 30 at a time ... they want to get everyone who’s not their tribe’.75 Declassified US Army records also describe

116 B. STIME

a case in 2006, in which US forces killed an Iraqi noncombatant when they raided the wrong house identified by two agents as the home of a member of an explosively formed penetrator (EFP) cell.76

Airstrikes in OEF-A have been similarly misguided. Mayer explains: ‘the strikes are only as accurate as the intelligence that goes into them. Tips from agents on the ground are subject to error, as is the interpretation of video images … local agents, who also serve as confirming witnesses for the airstrikes, are notoriously unreliable’.77 This claim is echoed by Callam78 and demonstrated in a number of unofficial ‘close calls’ and botched cases. A former CIA officer recalls a time when he barely held off an airstrike directed by his superiors and based on the HUMINT supplied by his Afghan agent:

They scrambled together an élite team ... We caught hell from headquarters. They said ‘Why aren’t you moving on it?’ when we insisted on checking it out first. It turned out to be an intentionally false lead.79

Mayer also records cases in which airstrikes on invalid targets were actually carried out: In September [2009], a NATO air strike in Afghanistan killed between seventy and a hundred and twenty-five peo- ple, many of them civilians, who were taking fuel from two stranded oil trucks; they had been mistaken for Taliban insurgents. (The incident is being investigated by NATO.) According to a reporter for the guardian, the bomb strike ... left such a tangle of body parts that village elders resorted to handing out pieces of unidentifiable corpses to the grieving families, so that they could have something to bury. One Afghan villager told the newspaper, ‘I took a piece of flesh with me home and I called it my son’.80

In addition, it is estimated that the campaign to kill Baitullah Mehsud took 16 missile strikes, resulting in 207–321 additional deaths over 14 months.81 Amnesty International also records a number of cases in which only noncombatants were killed in drone strikes.82 Based on the organization’s investigation, it is unclear what source(s) of intelligence led to these cases, although it is highly probable that HUMINT was a significant element.

On the other hand, it should be noted that a recent survey of FATA residents found that approximately 64 per cent of locals living with drone strikes believe that they accurately target militants, and greater than 66 per cent believe that most of the non-militant civilians who do die in drone strikes are militant sympathizers or radicalized collaborators.83

Insurgent and noncombatant-led preventive violence for network counter-development

Insurgents and noncombatant civilian communities publicly maim and kill suspected agents with the expressed purpose of deterring others from informing. The primary target of counterintelligence vio- lence is a particular demographic population (usually a civilian community or the insurgent’s own per- sonnel) that identifies with the secondary target, the violently punished suspected agent. This strategy incorporates especially brutal tactics, as it inflicts suffering for a broader psychological effect to advance counterintelligence objectives. It sends a clear message to the primary audience: do not inform.

Incumbent agent network development motivates counter-network violence by noncombatant civilian communities and insurgents. It is extremely difficult to identify agents once they are turned. Thus, it is strategically favorable to deter people from becoming agents in the first place. According to Kalyvas’ theory of the political economy of denunciation, the supply of denunciations is constrained by: ‘the likelihood of retaliation faced either by the denouncer or by the local committee that vets denunciations’.84 It is in the interest of the actors that would be threatened by the presence of agents to publicize the high likelihood of retaliation against agents. They employ excessively violent measures to discourage would-be informers. The victims of counterintelligence violence often have nothing to do with incumbent agent networks. Not only does this not detract from the strategy’s effectiveness, it actually enhances it. People are deterred from providing any sort of information to any incumbent representative for fear of being used as a violent propaganda example against spying. This strategy deters spying, makes it easier to identify agents talking to authorities, and further separates the pop- ulation from the incumbent.

Preventive violence analysed in the three cases indicates that both insurgents and local communities deter informing. In the literature, preventive violence is often labeled ‘terrorism’ when performed by

INTELLIgENCE AND NATIONAL SECURITy 117

insurgents and ‘public executions’ or ‘demonstrations’ when performed by noncombatant communities. Why is it in the interest of communities to deter informing? They, too, often consider agents a threat for largely three reasons: (1) the community may be sympathetic to the insurgent influence, (2) the presence of agents may be viewed as a liability that exposes the community to misdirected incumbent targeting driven by agents’ personal motives, and (3) agent presence may be feared to provoke insur- gent counterintelligence preventive and punishment violence against innocent community members misidentified as agents.

Numerous examples of preventive violence (by both civilian communities and insurgent groups) exist in each case study. For example, on 2 December 2012, HAMAS killed six suspected civilian collab- orators, in addition to the convicted agent Fadel Shalouf, and paraded their bodies through the streets of gaza City.85 More recently, on 22 August 2014, HAMAS executed at least 18 Palestinians suspected of informing for Israel. They were executed in public with warnings to the public scrawled on the walls to the effect that these people were ‘traitors’ who had ‘given information to the enemy’.86 HAMAS also posted graphic photos of their executions and the aftermath online.

Counterintelligence units in North Waziristan, called the Haqqani Khurasan, cover their faces with balaclavas and wear black tunics with Khorasan Mujahedin in Urdu scrawled across the back. They travel in slow-moving caravans of cars with tinted windows weeding out suspected agents.87 Reports indicate that they abduct and torture suspected agents, often with scalding, boiling water and iron rods; bodies are usually dumped on roadsides with scraps of paper that warn civilians against the consequences of spying for Western forces.88 Between 2009 and 2011, the Khurasan have carried out at least 250 public executions of Pakistani civilians declared agent suspects.89 This number is a minimum estimate, because evidence suggests that civilians are so scared of the Khurasan that everyone refrains from reporting public executions to state authorities. For example, no one reported a daylight execution of three men in a bazaar in the village of Maktab, in which one man was shot 10 times in the face and chest and a Khurasan operative yelled to the crowd, ‘Whoever tries to help the Americans and spies for them will face this!’90 State authorities became aware of this incident four months later when they captured a videotape of it.

The Khurasan broadcast warnings to Pakistani civilians, telling them that they will fight agents ‘before we fight the Americans’.91 A Western reporter described videos of sham trials and executions of sus- pected agents:

Accused tribesmen are abducted from homes and workplaces at gunpoint and tortured. A sham religious court hears their case, usually declaring them guilty. Then they are forced to speak into a video camera … [O]ne accused agent says into a video camera: ‘I am a spy and I took part in four attacks’, said Sidinkay, a young tribesman who said he was paid US$350 to help direct CIA drones to their targets in Pakistan’s tribal belt. Sweat glistened on his forehead; he rocked nervously as he spoke. ‘Stay away from the Americans’, he said in an imploring voice. ‘Stay away from their dollars’. The videos end in execution by hanging, beheading or firing squad … ‘Spies, your days are numbered because we are carrying out raids’, chants the video soundtrack.92

Al Qaeda’s sister organization in Afghanistan, Ittehad-e-Mujahedeen Khorasan,93 has also filmed and published these sorts of sham trials online, in which the accused are found guilty and beheaded.94

Civilian communities also use methods of intimidation to prevent locals from informing. For exam- ple, Shadid reports on a case in Iraq in 2003, in which a village required a father to publicly execute his son for being an informant.95 His son’s information had led to the detainment of 400 residents by US forces. The boy, Kurbul, had a bag over his head while he pointed out guilty parties in a line-up for American forces, but the villagers recognized him by his hand and sandals. Shadid writes that even local children chanted rhymes about him: ‘masked man, your face is the face of the devil’. Another example from the National Public Radio (NPR)96 describes how HAMAS had widespread community support for removing Khulud Badawi’s husband, a suspected agent, from prison in gaza City before his final court appearance that day to appeal the spying charges. Badawi, along with five other suspected spies, were somehow taken from prison without a trial, shot in broad daylight, and dragged through the street behind motorcycles.

118 B. STIME

Insurgent and noncombatant-led punitive violence for network counter-operation

Insurgents and noncombatant civilian communities counter agent operations through punitive vio- lence. Punitive violence and preventive violence (from the previous section) often overlap, since punishment of suspected agents (no matter how inaccurate) usually also serves to deter those who witness the event.97 As in preventive violence, both noncombatants and insurgents perform punitive violence.

Frequently, agents have been killed for their role as betrayers, an especially stigmatized activity.98 In nearly every culture, death is considered the proper punishment for informing, like treason. For exam- ple, a senior Pakistani Taliban commander explains: ‘in every civilized society, the penalty for spying is death’.99 Similarly, an Iraqi noncombatant asserts: ‘It’s justice … in my opinion, he deserves worse than death’.100 In the Israeli-occupied territories, Cohen and Dudai observe that agents are targets of commu- nity anger and punishment ‘because they are a serious threat to collective social values and interests, binding the community and establishing a sense of sovereignty’.101 Palestinian agents are often killed in mob beatings, their families are usually forced to disown them, and communities shun their families.102

US military intelligence personnel in OEF-A and OIF have reported extremely high rates of agent mortality. ‘We are losing agents all the time’, lamented one anonymous senior Western intelligence offi- cial, speaking with a reporter.103 Just how frequently agents have been killed in these theatres has not been disclosed, however, not only because of the sensitivity of the information, but (disconcertingly) also because security services often do not know how many have been killed.104

According to one US intelligence officer, in 2010 the Taliban discovered that a group of recent high school graduates in Afghanistan’s ghazni province had been feeding information to the Americans. The youths were arrested, and around 10 of them were hanged.105 In another instance, in 2007, the Taliban discovered a US-paid agent – a poppy farmer – was acting as a handler for several other Afghan agents. The Taliban demanded that he summon all of his agents together and said that if they repented, their lives would be spared; instead, everyone in the vicinity was killed.106

Insurgents and civilian communities often kill noncombatants who are suspected, but actually inno- cent, of informing. This phenomenon is due to the problem of identification of unlawful combatants. Additionally, the inaccuracy of punitive violence simultaneously serves a deterrent purpose; punishing a suspected agent not only eliminates the threat, it also scares witnesses from acting in any way that even vaguely resembles spying for fear of false accusation. For this reason, it is noted that preventive and punitive violence overlap.

Evidence in all three cases indicates that noncombatants are often killed without proper investigation/ trials because of insurgent and community suspicion. This was the case with Rabhi Badawi (described in the Preventive Violence section), who was taken from prison prior to his trial and executed.107 Human Rights Watch condemned HAMAS for conducting a ‘summary execution’ of 18 suspected agents on 22 August 2014.108 Hughes and Tripodi observe that suspicion killings are also extremely common in OEF-A and OIF cases. They write: ‘the experience in Iraq or Afghanistan would suggest that if a Western soldier stops to talk to a local, he may easily arouse suspicions that leads to the latter’s murder’.109

Fearing that paid agents direct NATO strikes, the Taliban in Afghanistan has repeatedly performed violent purges of suspected spies without trial or using sham trials.110 They have performed summary executions of tens of agents at a time. For example, one Taliban commander in Afghanistan claimed: ‘we have arrested about 60 or 70 spies, and we killed many of them’.111 Other insurgents claimed to have hanged 17 suspected spies in just the previous week.112 It is estimated that the counterintelli- gence Khurasan unit in the FATA has performed at least 250 assassinations and public executions of suspected spies.113 An American military official observed: ‘when we conduct a raid on a Haqqani leader … [the Khurasan] go in and massacre people’.114 The danger of mistaken punitive violence is reflected in Waziri public opinion; Frau observed that there is a pervasive fear among Waziri civilians of being falsely accused of spying by local Taliban.115

Notably, an Iraqi mayor describes the problem of community-driven inaccurate elimination of agents, while speaking with Washington Post reporter Shadid:

INTELLIgENCE AND NATIONAL SECURITy 119

Sitting in a chair and holding a cup of sweet tea, the mayor expressed frustration. Suspicions have become so common that more than 100 Muslim clerics met last week and issued a statement that not all Iraqis working with US forces should be considered informers. ‘Whenever somebody talks to the Americans’, he said shaking his head, ‘they think he’s an agent’.116

Residents were carrying out punitive killings of suspected agents with unsettling inaccuracy in his town. The inaccuracy of punitive violence caused by the ‘identification problem’ of agents is a common theme across all three cases, among both insurgent organizations and civilian communities.

Conclusion

Through an exploratory analysis of COIN HUMINT activities in OEF-A, OIF, and the Israeli–Palestinian conflict, this research has identified six patterns of violence against noncombatants associated with COIN agent networks – coercion, covert and clandestine activities, unauthorized crime, misdirected targeting, preventive violence, and punitive violence – that follow an incentive structure embedded in the intelligence war framework.

While agent networks are essential tools for COIN operations, in some ways they indirectly motivate corollary violence against the civilian population. In OEF-A, OIF and the Israeli–Palestinian conflict, agent networks have functioned as crucial HUMINT collection resources, insurgent identification and location tools, proxies for covert and clandestine activities, and social control mechanisms. At the same time, they have inadvertently incentivized all the conflict actors – the incumbent, agents, noncombatants, and insurgent group – to employ violence against noncombatants in the competition for access and intelligence.

Collateral damage is an unfortunate, unavoidable necessary evil in COIN operations, particularly in a complex civil war environment in which the civilians are often peripherally or integrally involved in the conflict. HUMINT operations are necessary to build a detailed intelligence picture and find, fix and finish the insurgents, but operations in pursuit of this political objective do not always align with the principle of civilian protection. HUMINT collection is an ugly business that sows mistrust within pop- ulations, generating first, second, and third degree effects that must be carefully identified, assessed, and balanced against strategic advantages of agent network use.

This research provides a theoretical foundation for better understanding these direct and indirect effects on civilians, focusing on the issue of violence. There is an urgent need for further research into the sociological and psychological effects of COIN HUMINT operations, and how an incumbent (along with intervening militaries) may tailor intelligence policies and practices to alter incentive structures and diminish noncombatant-targeted violence associated with agent networks.

Notes 1. Riechmann, “Coalition Informant Plays Both Sides.” 2. Lynfield, “Israel–gaza Crisis.” 3. See Costa, “Phoenix Rises Again”; Miska, “Protecting Soft Networks”; Macgaffin, “Clandestine Human Intelligence”;

Hung, “Optimization-Based Selection of Influential Agents”; gallagher, “Human Intelligence in Counterinsurgency”; Beall, “The HUMINT Heresies”; Baker, “HUMINT-Centric Operations”; Berman et al., “Can Hearts and Minds be Bought?”; Butler, “Killing Cells.”

4. Moran, “Evaluating Special Branch”; Moran, From Northern Ireland to Afghanistan; Cohen and Dudai, “Human Rights Dilemmas in Using Informers.”

5. Intelligence term for agent recruitment. 6. Intelligence term for running (training, financing and operating) an agent after he/she has been recruited. 7. Defined as systems of motivating influences. 8. In this paper, the product of noncombatant-targeted violence is noncombatant casualties, defined as killed and

injured noncombatants and noncombatants-turned-agents. 9. In this way, this research confronts the same research challenge faced by the first academic to study political

informers, gary Marx. See Marx, “Thoughts on a Neglected Category.” 10. For the purposes of this paper, information refers to raw data (unanalysed, unprocessed, and unvalidated) from

any source. Intelligence refers to analysed, processed and validated information from any source. Additionally,

120 B. STIME

intelligence is generally classified due to the nature of collection methods, security of the source, and the potential impact on national and local security.

11. See greer, Supergrasses; Hewitt,  Spying 101, 33; US Army, Human Intelligence Collector Operations; UK Army, Counterinsurgency Operations; UK Ministry of Defense, Joint Doctrine Publication (JDP) 2–0.

12. See Kalyvas, The Logic of Violence in Civil War. 13. ’Single-event’, ‘one-time contact’ or ‘volunteer’. 14. ’Multiple-event’, ‘formal contact’, ‘developed source’ or ‘long-term penetration source’. 15. Including, inter alia: insurgent groups, criminal groups, and foreign governments. 16. galula, “Counter-Insurgency Warfare,” 1–12; Kalyvas, The Logic of Violence in Civil War, 89–91; Tomes, “Relearning

Counterinsurgency Warfare,” 17–18; Duyvesteyn and Angstrom, Rethinking the Nature of War, 101; Kitson, Low Intensity Operations, 71.

17. US Army, Human Intelligence Collector Operations, s.3–1; UK Army, Counterinsurgency Operations, s.B-6–12; Mumm, “Crowdsourcing,” 59; US Army, Intelligence and Electronic Warfare Support, s.5–1; McFate, “Anthropology and Counterinsurgency,” 37; Hoffman, “The Logic of Suicide Terrorism,” 6–7; gallagher, “Human Intelligence in Counterinsurgency,” 1.

18. See Khatchadourian, “The Kill Company”; youssef, “Fear of Informants has Stoked Climate”; Finer, “Informants Decide Fate of Iraqi Detainees”; Shadid, “For an Iraqi Family”; Ware, “Chasing the ghosts.”

19. See ‘block control’ strategy in US Army, Counterinsurgency Operations, s.204:3–31, 3–7; Israeli agent networks in Cohen and Dudai, Human Rights Dilemmas in Using Informers, 231–3; Hughes and Tripodi, “Anatomy of a Surrogate,” 8; Hewitt, Snitch!, 6, 95–122.

20. UK Army, Counterinsurgency Operations, s.10–30; Kirk-Smith and Dingley, “Countering Terrorism in Northern Ireland,” 565; Moloney, A Secret History of the IRA, 315.

21. Hewitt, Snitch!, 98. 22. Marx, “Thoughts on a Neglected Category,” 402–42; Kirk-Smith and Dingley, “Countering Terrorism in Northern

Ireland,” 556, 564–5; Hewitt, Snitch!, 7–18, 33–4; UK Army, Counterinsurgency Operations, s.A-4–9; Moran, “Evaluating Special Branch,” 20; Burds, “Agentura,” 129–30; Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 237.

23. See Bellaby, “What's the Harm?,” 115–16. 24. Focusing only on the left side of this framework does not imply that insurgents are only reactive to incumbent

activities, nor does it imply a moral justification of insurgent activities as such. As depicted in the diagram, the incumbent and insurgent group(s) alike act both proactively and reactively.

25. Kalyvas, The Logic of Violence in Civil War. 26. The reason for this will be explained in the Punitive Violence section. 27. For a full description of HCTs, see US Army, Human Intelligence Collector Operations. HCTs are part of Special Forces

and intelligence units from groups such as the British Joint Support group (JSg) (an elite group within the British Army Intelligence Corps and the successor of the FRU in Northern Ireland), the American Joint Special Operations Command (JSOC), and the Central Intelligence Agency (CIA).

28. See Elite UK Forces, “Elite Intelligence Operatives, Joint Support group”; Central Intelligence Agency, “On the Front Lines”; US Army, Human Intelligence Collector Operations, s.3–21.

29. This was one of the primary purposes of ‘Operation Close Encounters’ and census conduction in Iraq. See Crider, “Inside the Surge”; Keirsey, “Reflections of a Counterinsurgency Company Commander,” 74.

30. See Clark, “The Vital Role of Intelligence in Counterinsurgency Operations”; Hoffman, “The Logic of Suicide Terrorism”; Crider, “Inside the Surge,” 8.

31. A common theme in the literature is concern over insurgent infiltration of HN law enforcement and intelligence agencies. This endangers agents and creates supply lines of misinformation that decreases overall HUMINT reliability and endangers SSFs. See gallagher, “Human Intelligence in Counterinsurgency”; Otterman, “IRAQ.” In such circumstances, intervening forces set up new, reliable intelligence organizations and identified locations for new police outposts. See Kitson, Low Intensity Operations, 74; Clark, “The Vital Role of Intelligence in Counterinsurgency Operations.”

32. Agent recruitment operations are referred to as agent networks ‘human source contact operations’ (SCOs) or ‘source operations.’ See US Army, Human Intelligence Collector Operations; US Army, Tactics in Counterinsurgency.

33. For example, see the poppy farmer anecdote in the Punitive Violence section of this chapter. Also, Crider writes: ‘this man was already running a network of 10 other individuals in Doura and areas south of Baghdad … this source provided information to us … even better, the information that he and his sub-sources provided allowed us to detain 10 targets.’ See Crider, “Inside the Surge,” 13.

34. Block Control: dividing a community or populated area into zones where a trusted resident reports on the activities of the population. See US Army, Counterinsurgency Operations FM (Interim) 3–07.22, 3–31.

35. Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 231–3. 36. See Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 233–4; Hewitt, Snitch!, 70–1, 133; Marx,

“Thoughts on a Neglected Category,” 414. 37. Coercive recruitment is widely considered ineffective for long-term agent management; thus, it is usually a last

resort, once other voluntary recruitment methods are deemed implausible.

INTELLIgENCE AND NATIONAL SECURITy 121

38. See Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 238; Perry, “Repugnant Philosophy.” 39. See Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 238. This case is further described in a

subsequent section of this paper: Incumbent Covert Action for Network Operation. 40. Burkett, “An Alternative Framework for Agent Recruitment,” 11; Marx, “Thoughts on a Neglected Category”; Charney

and Irvin, “A guide to the Psychology of Espionage.” 41. Security Awareness Division, “Security Awareness in the 1980s,” 13; Cohen and Dudai, “Human Rights Dilemmas

in Using Informers”; Perry, “Repugnant Philosophy.” 42. Marx, “Thoughts on a Neglected Category,” 414. 43. Cook, “Spotlight Shines on Palestinian Collaborators.” 44. O'Connor, “gay Palestinians are Being Blackmailed.” 45. Cook, “Spotlight Shines on Palestinian Collaborators.” 46. Hewitt, Snitch!, 133–4. 47. See Kirk-Smith and Dingley, “Countering Terrorism in Northern Ireland,” 556, 564–5; Hewitt, Snitch!, 7–18, 33–4;

UK Army, Counterinsurgency Operations, s.A-4–9; Moran, Evaluating Special Branch, 13–14, 20; Burds, “Agentura,” 129–30; Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 237.

48. Jamieson and McEvoy in Dudai, “Informers and the Transition in Northern Ireland,” 34. 49. See Walsh, “Drone War Spurs Militants”; Callam, “Drone Wars.” 50. Cohen and Dudai, “Human Rights Dilemmas in Using Informers,” 237. 51. Tait, “Family of Palestinian Man.” 52. Ibid. 53. i.e., insurgent informers who are actively involved in the crimes and/or war activities on which they themselves

are providing information. 54. Moran, “Evaluating Special Branch,” 12, 14. 55. See Stevens in Moran, “Evaluating Special Branch,” 13–15, 20; Brown, Into the Dark; Bamford, “The Role and

Effectiveness of Intelligence,” 600–2; Ingram and Harkin, Stakeknife, 82–93; Moran“Evaluating Special Branch,” 14;“Police Ombudsman for Northern Ireland."

56. See, for example, the case recounted by the CIA officer in Afghanistan, recorded by Mayer in “The Predator War.” 57. As opposed to intelligence packages for non-kinetic targeting, such as for influence operations. 58. Scahill and greenwald, “The NSA's Secret Role”; Rowley, Dirty Wars. 59. See Callam, “Drone Wars”; Major (Redacted) to the Commander of 1st BCT, 10th MTN, Camp Liberty, Iraq; Finer,

“Informants Decide Fate of Iraqi Detainees”; Mayer, “The Predator War”; Kahl, “How We Fight,” 83, 85; Walsh, “Drone War Spurs Militants.”

60. Kalyvas, The Logic of Violence in Civil War, 174; gallagher, “Human Intelligence in Counterinsurgency,” 1–2; Bamford, “The Role and Effectiveness of Intelligence,” 590; Mayer, “The Predator War”; Callam, “Drone Wars.”

61. Marx, “Thoughts on a Neglected Category,” 420. 62. Kalyvas, The Logic of Violence in Civil War, 178. 63. Ibid., 176. 64. Ibid., 173. 65. Finer, “Informants Decide Fate of Iraqi Detainees.” 66. Callam, “Drone Wars.” 67. “How a Boy Named Steve-O Became a Secret Weapon.” 68. Riechmann, “Coalition Informant Plays Both Sides.” 69. Myer, “The Predator War.” 70. Human Rights Watch, “gaza.” 71. Associated Press, “Rights group.” 72. King, “After Israeli Airstrikes.” 73. Kahl, “How we Fight,” 83 74. Finer, “Informants Decide Fate of Iraqi Detainees.” 75. Semple, “US Forces Rely on Local Informants.” 76. Major (Redacted) to the Commander of 1st BCT, 10th MTN, Camp Liberty, Iraq. 77. Mayer, “The Predator War.” 78. Callam, “Drone Wars.” 79. Mayer, “The Predator War.” 80. Ibid. 81. Ibid. 82. Amnesty International, Northern Ireland; Amnesty International, Will I Be Next? 83. Shah, “Drone Blowback in Pakistan.” 84. Kalyvas, The Logic of Violence in Civil War, 192. 85. Rudoren, “Collaboration in gaza Leads to grisly Fate.” 86. King, “After Israeli Airstrikes.” 87. Rodriguez, “Pakistani Death Squads.”

122 B. STIME

88. Ibid. 89. Rivera et al., “Militants Turn to Death Squads.” 90. Ibid. 91. Shadid, “For an Iraqi Family.” 92. Walsh, “Drone War Spurs Militants.” 93. Ibid. 94. youssef, “Fear of Informants has Stoked Climate.” 95. Shadid, “For an Iraqi Family.” 96. Harris, “In gaza, Hamas Targets Palestinian Informants.” 97. Walter, Terror and Resistance, 23. 98. See Akerström, Betrayal and Betrayers. 99. Walsh, “Drone War Spurs Militants.” 100. Shadid, “For an Iraqi Family.” 101. For a theory about community secrecy and identity that is commonly cited in literature on agents and betrayal,

see Simmel, “The Sociology of Secrecy,” 37. 102. Shirwood, “Palestinian Collaborator.” 103. Riechmann, “Coalition Informant Plays Both Sides.” 104. Riechmann, “Coalition Informant Plays Both Sides”; Shadid, “For an Iraqi Family.” 105. Moreau, “Taliban Seeks Vengeance in Wake of WikiLeaks.” 106. Smith, “Insurgents Blame Paid Informants.” 107. Harris, “In gaza, Hamas Targets Palestinian Informants.” 108. Lynfield, “Israel–gaza Crisis.” 109. Hughes and Tripodi, “Anatomy of a Surrogate,” 9. 110. youssef, “Fear of Informants has Stoked Climate.” 111. Smith, “Insurgents Blame Paid Informants.” 112. Ibid. 113. Rivera, “Militants Turn to Death Squads.” 114. Ibid. 115. Frau, “The 'Living Under Drones' Report.” 116. Shadid, “For an Iraqi Family.”

Acknowledgements Special thanks to Dr. Jon Moran and Professor Mark Phythian of the University of Leicester, Dr. Andrew Thomson of Queen’s University Belfast, Colonel James F. Powers Jr. (US Army, Retired) of the Joint Special Operations University, Dr. David Spencer of the US National Defense University’s William J. Perry Center for Hemispheric Defense Studies, and the Oxford Research group’s Remote Control Project.

Notes on contributor Britta Stime is a defense and communications consultant with the Rendon group. This paper reflects her personal analysis and research, and does not represent the work or opinions of the Rendon group or its clients. She holds a master’s degree in Violence, Terrorism and Security from Queen’s University Belfast, and a bachelor’s degree in International Relations and Spanish from gonzaga University.

References Akerström, Malin. Betrayal and Betrayers. New Brunswick: Transaction, 1991. Amnesty International. Northern Ireland: Time to Deal with the past. London, England: Amnesty International, September

12, 2013. Amnesty International. Will I Be Next? US Drone Strikes in Pakistan. London, England: Amnesty International, October 22, 2013. Associated Press. “Rights group: Israel Violated Laws of War in gaza.” Associated Press. Accessed February 13, 2013. http://

cnsnews.com/news/article/rights-group-israel-violated-laws-war-gaza Baker, Ralph. “HUMINT-Centric Operations: Developing Actionable Intelligence in the Urban Counterinsurgency

Environment.” Military Review 87, no. 2 (2007): 12–21. Bamford, Bradley. “The Role and Effectiveness of Intelligence in Northern Ireland.” Intelligence and National Security 20,

no. 4 (2005): 581–607. http://dx.doi.org/10.1080/02684520500425273.

INTELLIgENCE AND NATIONAL SECURITy 123

Beall, David. “The Humint Heresies: The Disposition of Human Intelligence Collection in Counterinsurgency.” Military Intelligence Professional Bulletin (2009): 32–37.

Bellaby, Ross. “What’s the Harm? The Ethics of Intelligence Collection.” Intelligence and National Security 27, no. 1 (2012): 93–117. http://dx.doi.org/10.1080/02684527.2012.621600.

Berman, Eli, Jacob Shapiro, and Joseph Felter. “Can Hearts and Minds Be Bought? The Economics of Counterinsurgency in Iraq.” Journal of Political Economy 119, no. 4 (2007): 766–819.

Brown, Johnston. Into the Dark: 30 Years in the RUC. Dublin: gill & Macmillan, 2005. Burds, Jeffery. “Agentura: Soviet Informants’ Networks & the Ukrainian Underground in galicia, 1944-48.” East European

Politics & Societies 11, no. 1 (1996): 89–130. http://dx.doi.org/10.1177/0888325497011001003. Burkett, Randy. “An Alternative Framework for Agent Recruitment: From MICE to RASCLS.” Studies in Intelligence 57, no. 1

(2013): 7–17. Butler, Michael. “Killing Cells: Retooling Human Intelligence Collection for global Decentralized Terrorism.” Paper presented

at ISA’s 50th Annual Convention “Exploring the Past, Anticipating the Future”, New york City, February 15, 2009. Callam, Andrew. “Drone Wars: Armed Unmanned Aerial Vehicles.” International Affairs Review 18, no. 3 (2010). http://www.

iar-gwu.org/node/144. Central Intelligence Agency. “On the Front Lines: CIA in Afghanistan.” Accessed November 21, 2012. https://www.cia.gov/

about-cia/cia-museum/experience-the-collection/text-version/stories/on-the-front-lines-cia-in-afghanistan.html Charney, David, and John Irvin. “A guide to the Psychology of Espionage.” Association of Former Intelligence Officers, August

28, 2014. Clark, David. “The Vital Role of Intelligence in Counterinsurgency Operations.” USAWC Strategy Research Project (2008).

http://www.strategicstudiesinstitute.army.mil/pdffiles/ksil309.pdf. Cohen, H., and R. Dudai. “Human Rights Dilemmas in Using Informers to Combat Terrorism: The Israeli-Palestinian Case.”

Terrorism and Political Violence 17, no. 1–2 (2005): 229–243. Cook, Jonathan. “Spotlight Shines on Palestinian Collaborators.” Al Jazeera. Accessed February 17, 2014. http://www.

aljazeera.com/indepth/features/2014/02/spotlight-shines-palestinian-collaborators-201421611151157762.html Costa, Christopher. “Phoenix Rises Again: HUMINT Lessons for Counterinsurgency Operations.” Master’s diss., Naval War

College, 2006. Crider, Jim. “Inside the Surge.” Center for a New American Security. Accessed June, 2009. http://www.cnas.org/files/documents/

publications/CNAS_Working%20Paper_Surge_CriderRicks_June2009_ONLINE.pdf Dudai, Ron. “Informers and the Transition in Northern Ireland.” British Journal of Criminology 52, no. 1 (2011): 32. http://

dx.doi.org/10.1093/bjc/azr055. Duyvesteyn, Isabelle, and Jan Angstrom. Rethinking the Nature of War. London: Frank Cass, 2005. Elite UK Forces. “Elite Intelligence Operatives, Joint Support group.” 2014. http://www.eliteukforces.info/joint-support-

group/. Finer, Jonathan. “Informants Decide Fate of Iraqi Detainees.” The Washington Post. Accessed September 13, 2005 http://

www.washingtonpost.com/wp-dyn/content/article/2005/09/12/AR2005091202040.html Frau, Robert. “The ‘Living under Drones’ Report: A Critical Appraisal.” European University Viadrina Frankfurt. Accessed

October 9, 2012. http://chrgj.org/wpcontent/uploads/2012/10/Living-Under-Drones.pdf gallagher, Michael. “Human Intelligence in Counterinsurgency: Persistent Pathologies in the Collector-Consumer

Relationship.” Small Wars Journal (2011): 1–12. galula, David. “Counter-Insurgency Warfare: Theory and Practice.” The Journal of Politics 26, no. 4 (1964): 1–12. greer, Steven. Supergrasses. Oxford: Clarendon Press, 1995. Harris, Emily. “In gaza, Hamas Targets Palestinian Informants in Crackdown.” NPR News (Washington, DC). Accessed June

4, 2013. http://www.npr.org/blogs/parallels/2013/06/12/186975387/in-gaza-hamas-targets-palestinian-informants-in- crackdown

Hewitt, Steve. Spying 101. Toronto: University of Toronto Press, 2002. Hewitt, Steven. Snitch! New york: The Continuum International Publishing group, 2010. Hoffman, Bruce. “The Logic of Suicide Terrorism.” The Atlantic Monthly Group 291, no. 5 (2003): 1–10. Hughes, geraint, and Christian Tripodi. “Anatomy of a Surrogate: Historical Precedents and Implications for Contemporary

Counter-Insurgency and Counter-Terrorism.” Small Wars & Insurgencies 20, no. 1 (2009): 1–35. Human Rights Watch. Gaza: Palestinian Rockets Unlawfully Targeted Israeli Civilians. New york: Human Rights Watch. December

24, 2012. https://www.hrw.org/news/2012/12/24/gaza-palestinian-rockets-unlawfully-targeted-israeli-civilians. Hung, Benjamin. “Optimization-Based Selection of Influential Agents in a Rural Afghan Social Network.” Master’s diss.,

Massachusetts Institute of Technology, 2010. Ingram, Martin, and greg Harkin. Stakeknife: Britain’s Secret Agents in Ireland. Madison, WI: University of Wisconsin Press, 2004. Kahl, Colin. 2006. “How We Fight.” Foreign Affairs, Nov/Dec. https://www.foreignaffairs.com/articles/iraq/2006-11-01/how-

we-fight. Kalyvas, Stathis. The Logic of Violence in Civil War. Cambridge: Cambridge University Press, 2006. Keirsey, Jim. “Reflections of a Counterinsurgency Company Commander.” Company Command: Building Combat-Ready

Teams (July 2008): 67–74. http://cc.army.mil/pubs/armymagazine/docs/2008/CC_0807-Keirsey2.pdf.

124 B. STIME

Khatchadourian, Raffi. “The Kill Company.” The New Yorker. Accessed July 6, 2009. http://www.newyorker.com/ magazine/2009/07/06/the-kill-company?currentPage=all

King, Laura. “After Israeli Airstrikes, Hamas Kills 18 Suspected Informers.” Los Angeles Times (Los Angeles, CA), August 22, 2014. Kirk-Smith, Michael, and James Dingley. “Countering Terrorism in Northern Ireland: The Role of Intelligence.” Small Wars &

Insurgencies 20, no. 3–4 (2009): 551–573. Kitson, Frank. Low Intensity Operations: Subversion, Insurgency and Peacekeeping. London: Faber, 1971. Lynfield, Ben. “Israel-gaza Crisis: Hamas Says 18 Suspected Collaborators Executed.” The Independent (London, England).

Accessed August 22, 2014. http://www.independent.co.uk/news/world/middle-east/israelgaza-crisis-hamas-says-18- suspected-collaborators-executed-9686162.html

Macgaffin, John. “Clandestine Human Intelligence.” In Transforming U.S. Intelligence, edited by Jennifer Sims and Burton gerber, 79–95. Washington, DC: georgetown University Press, 2005.

Major (Redacted). Major (Redacted) to the Commander of 1st BCT, 10th MTN, Camp Liberty, Iraq. US Army 14168-14206. FOB Justice, Iraq. April 5, 2006.

Marx, gary. “Thoughts on a Neglected Category of Social Movement Participant: The Agent Provocateur and the Informant.” American Journal of Sociology 80, no. 2 (1974): 402–442.

Mayer, Jane. “The Predator War.” The New Yorker. Accessed October 26, 2009. http://www.newyorker.com/ reporting/2009/10/26/091026fa_fact_mayer

McFate, Montgomery. “Anthropology and Counterinsurgency: The Strange Story of their Curious Relationship.” Military Review 85, no. 2 (2005): 24–38.

Miska, Steve. “Protecting Soft Networks.” CTX Journal 3, no. 4 (2013): 5–11. Moloney (ed.). A Secret History of the IRA. New york: W.W. Norton, 2002. Moran, Jon. “Evaluating Special Branch and the Use of Informant Intelligence in Northern Ireland.” Intelligence and National

Security 25, no. 1 (2010): 1–23. Moran, Jon. From Northern Ireland to Afghanistan. Burlington, VT: Ashgate, 2013. Moreau, Ron. “Taliban Seeks Vengeance in Wake of WikiLeaks.”Newsweek (New york). Accessed August 1, 2010. http://www.

newsweek.com/taliban-seeks-vengeance-wake-wikileaks-71659?from=rss Mumm, Nick. “Crowdsourcing: A New Perspective on Human Intelligence Collection in a Counterinsurgency.” Small Wars

Journal 3, no. 3 (2012): 1–10. O’Connor, Nigel. “gay Palestinians are Being Blackmailed into Working as Informants.” Vice. Accessed February 19, 2013.

http://www.vice.com/en_uk/read/gay-palestinians-are-being-blackmailed-into-working-as-informants. Otterman, Sharon. “IRAQ: Infiltration of Iraqi Forces.” Council on Foreign Relations. Accessed February 28, 2005. http://www.

cfr.org/iraq/iraq-infiltration-iraqi-forces/p7828#p5 Perry, David. “Repugnant Philosophy: Ethics, Espionage, and Covert Action.” Journal of Conflict Studies 15, no. 1 (1995):

92–115. Riechmann, Deb. “Coalition Informant Plays Both Sides of Afghan War.” Associated Press. Accessed June 7, 2011. http://www.

cecildaily.com/news/article_d7944b54-9135-11e0-979b-001cc4c03286.html?mode=jqm Rivera, Ray, Sharifullah Sahak, and Eric Schmitt. “Militants Turn to Death Squads in Afghanistan.” The New York Times (New

york). Accessed November 28, 2011. http://www.nytimes.com/2011/11/29/world/asia/haqqani-militants-use-death- squads-in-afghanistan.html?pagewanted=all&_r=0

Rodriguez, Alex. “Pakistani Death Squads go after Informants to U.S. Drone Program.” Los Angeles Times (Los Angeles, CA). Accessed December 28, 2012. http://articles.latimes.com/2011/dec/28/world/la-fg-pakistan-death-squads-20111228

Directed by Rowley, Richard. Dirty Wars. London, England: Serpent’s Tail. DVD, 2013. Rudoren, Jodi. “Collaboration in gaza Leads to grisly Fate.” The New York Times (New york). Accessed December 2, 2012.

http://www.nytimes.com/2012/12/03/world/middleeast/preyed-on-by-both-sides-gaza-collaborators-have-grim-plight. html?pagewanted=all&_r=0

Scahill, Jeremy, and glenn greenwald. “The NSA’s Secret Role in the US Assassination Program.” The Intercept. Accessed February 9, 2014. https://firstlook.org/theintercept/2014/02/10/the-nsas-secret-role/.

Semple, Kirk. “U.S. Forces Rely on Local Informants in Ferreting Rebels in West Iraq.” The New York Times (New york). Accessed December 10, 2005. http://www.nytimes.com/2005/12/10/international/middleeast/10protectors. html?pagewanted=all&_r=2&

Shadid, Anthony. “For an Iraqi Family, ‘No Other Choice’.” The Washington Post. Accessed August 1, 2003. http://www. washingtonpost.com/wp-dyn/articles/A10829-2003Jul31_3.html

Shah, Aquil. “Drone Blowback in Pakistan is a Myth. Here’s Why.” The Washington Post (Washington, DC). Accessed May 17, 2016. https://www.washingtonpost.com/news/monkey-cage/wp/2016/05/17/drone-blow-back-in-pakistan-is-a- myth-heres-why/

Shirwood, Harriet. “Palestinian Collaborator: ‘I am a Traitor. I Sold My People. But for What?’” The Guardian (London, England). Accessed May 17, 2011. http://www.theguardian.com/world/view-from-jerusalem-with-harriet-sherwood/2011/may/17/ israel-palestinian-territories

Simmel, georg. “The Sociology of Secrecy and of Secret Societies.” American Journal of Sociology 11, no. 4 (1906): 441–498. Smith, graeme. “Insurgents Blame Paid Informants for Increasingly Accurate NATO Strikes.” Globe and Mail (Toronto, Canada).

Accessed February 28, 2007. http://army.ca/forums/index.php?topic=49908.245;wap

INTELLIgENCE AND NATIONAL SECURITy 125

Tait, Robert. “Family of Palestinian Man Found Dead in Israeli Jail Claims He was Tortured.” The Telegraph (London, England). Accessed February 25, 2013. http://www.telegraph.co.uk/news/worldnews/middleeast/israel/9893494/Family-of- Palestinian-man-found-dead-in-Israeli-jail-claims-he-was-tortured.html

Tomes, Robert. “Relearning Counterinsurgency Warfare.” Parameters 34, no. 1 (2004): 17–18. UK Army. Counterinsurgency Operations (Strategic and Operational Guidelines) . Volume 1, Part 10. London, England:

UK Ministry of Defense, 2007. UK Ministry of Defense. Joint Doctrine Publication (JDP) 2–0: Understanding and Intelligence Support to Joint Operations.

Shrivenham, England: The Development, Concepts and Doctrine Centre, 2011. United States Department of Defense Security Institute. Security Awareness Division. Security Awareness in the 1980s:

Feature Articles from the Security Awareness Bulletin, 1981 to 1989. Richmond, Virginia: Department of Defense Security Institute, 1989.

US Army. Intelligence and Electronic Warfare Support to Low-Intensity Conflict Operations. Department of the Army. FM 34-7. Washington, D.C.: Department of the Army, 1993.

US Army. Counterinsurgency Operations. FMI 3-07.22. Fort Leavenworth, Kansas: Department of the Army, 2004. US Army. Counterinsurgency Operations FM (Interim) 3-07.22. Washington, DC: US Department of the Army, 2004. US Army. Human Intelligence Collector Operations. Department of the Army. FM 2-22.3. Washington, D.C.: government

Printing Office, 2006. US Army. Tactics in Counterinsurgency. Department of the Army. FM 3-24.2. Washington, D.C.: government Printing Office,

2009. Walsh, Declan. “Drone War Spurs Militants to Deadly Reprisals.” The New York Times (New york), December 29, 2012.

http://www.nytimes.com/2012/12/30/world/asia/drone-war-in-pakistan-spurs-militants-to-deadly-reprisals. html?pagewanted=all&_r=0.

Walter, Eugene Victor. Terror and Resistance. New york: Oxford University Press, 1969. Ware, Michael. “Chasing the ghosts.” Time Magazine, September 18, 2005. http://reidreport.com/TIME-chasing_the_ghosts.

htm. “Police Ombudsman for Northern Ireland.” Policeombudsman.org. Last modified 2014. http://www.policeombudsman.org. “How a Boy Named Steve-O Became a Secret Weapon.” Age, October 5, 2004. youssef, Nancy. “Fear of Informants has Stoked Climate of Fear in Baghdad.” McClatchy DC, February 24, 2006. http://www.

mcclatchydc.com/2006/02/24/13446/fear-of-informants-has-stoked.html.