Jump to ContentJump to Main Navigation
Arbitrary StatesSocial Control and Modern Authoritarianism in Museveni's Uganda$

Rebecca Tapscott

Print publication date: 2021

Print ISBN-13: 9780198856474

Published to Oxford Scholarship Online: June 2021

DOI: 10.1093/oso/9780198856474.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 07 December 2021

Whither the State?

Whither the State?

Surveillance, Crime Preventers, and Potential State Presence

(p.126) 6 Whither the State?
Arbitrary States

Rebecca Tapscott

Oxford University Press

Abstract and Keywords

The chapter studies the complex relationship between state and society, drawing on scholars including Timothy Mitchell and Joel Migdal who see the distinction between state and society as produced through practice. It looks at how Uganda’s ruling regime manipulates the relationship between state presence and absence, such that citizens are sometimes categorized as outside the state, sometimes as agents of the state, and—most often—placed in a liminal space where their standing vis-à-vis the state is ambiguous. The chapter examines Uganda’s flagship community policing programme, Crime Preventers, described as a ‘floating population’ that works as the regime’s ‘eyes and ears’ across the country. By mobilizing crime preventers, the regime fostered the possibility of state presence while keeping crime preventers themselves in a liminal space from which they could make few claims on state authorities.

Keywords:   Crime Preventers, community policing, elections, electoral violence, Gulu, state presence, surveillance

Despite the apparent weakness of the Ugandan state, as measured by indicators like water and sanitation, control of corruption, and government effectiveness, the possibility of state presence is felt across the country.1 Even in remote and rural Uganda, where the state has been dismissed as absent, its interventions have been likened to the ‘dry season rains … occasional and potentially destructive’ (B. Jones 2009, 3). In my research, ‘the state’ was made real and potentially present neither by brick-and-mortar initiatives nor by seeking to extend and regularize its control. Instead, its potential presence emanated from a fluid distinction between state and society. State actors sometimes hold that there is no difference between state and society, considering the state as embodied in Ugandan citizens. They thus transfer responsibility for state performance to citizens, committing citizens to police their own neighbours and transmit intelligence to higher authorities. The state exists wherever citizens are present. Other times, state authorities uphold the division between state and society, drawing a bright line between the two entities. Citizens are outside of the state, and thus subject to state surveillance and discipline as members of society. More often, authorities place populations and individuals in a liminal space between state and society, with the possibility of determining their status at any future time, thereby keeping their rights and responsibilities in flux. These scenarios reflect a fluid opposition between state presence and absence, whereby at times the state is clearly present or absent, at times it is hard to tell, and at times there is no meaningful way for the state to be absent, as it is embodied in ordinary citizens.

Key to this form of state presence was the widely shared conviction that government spies are omnipresent. As one sub-county official explained:

[People say that the government has eyes and ears everywhere in Uganda—is it true?] Yeah, that one is very true. Actually, everybody is like an informer of the other. It’s just too much now. You will not know who is who, who is taking the information where … There are so many spies, even in police, those intelligences … you may (p.127) think sometimes they work together with the army. That’s why security in Uganda is so tight. Everybody is a spy. They have also been using women—you will not know who is who in Uganda [laughs].

(Sub-county official, Moroto, 19 February 2018)

The perception that government informants are everywhere both comes from instability in the opposition between state presence and absence, and further destabilizes it. Ugandans widely report government informants ‘deep down in the village’, and conclude that one can never know to whom one is speaking. The added possibility that a person may act as an informant in the future creates further uncertainty and fear of betrayal. Countless unofficial informants report to isolated, fragmented, and obscured chains of authority, which are commonly thought to reach the Statehouse. Given that citizens believe the regime is informed and has the capacity to intervene, non-intervention appears to be as much a choice as intervention. In this way, potential state presence—the third factor of institutionalized arbitrariness—transforms the regime’s overwhelming capacity for violence from an episodic reality to an omnipresent imaginary that governs citizens’ daily lives.

The perception of possible state presence can be achieved in diverse ways. Joel Migdal—who famously described contemporary states in the global South as ‘weak’ and overrun by society—argued that the state and its capacity can only be understood through ‘the process of interaction … with those whose actual behaviour they are vying to control or influence’ (Migdal 2001, 23). Timothy Mitchell emphasizes that the state’s boundaries are drawn and maintained through internal interactions between people acting as members of the state or society. The boundary is made to appear external and authoritative as a ‘distinctive technique of the modern political order’ (Mitchell 1991, 78). The state must therefore be understood not as a structure ‘but as the powerful, metaphysical effect of practices that make such structures appear to exist’ (Mitchell 1991, 94, emphasis added). The presence of the state, then, is not simply about its manifestation. It is also about the production of its limits. That is, citizens’ sense of state presence is produced by divisions that actors draw and reify between state and society. Building on these insights, my research shows how drawing and redrawing these divisions destabilizes the imaged opposition between state presence and absence, creating the perception that the regime could potentially be present at any time.

Uganda’s NRM regime has consistently adopted programmes and policies that rely on and reinforce the fluidity and complexity between state and society. This chapter focuses on one such programme: Uganda’s Crime Preventers, nominally a community policing initiative that recruited tens of thousands of underemployed youth shortly before the 2016 presidential elections. These crime preventers would be the regime’s ‘eyes and ears’ and ‘help with the elections’. Paired with unofficial status, this created the impression—and perhaps the reality—that crime preventers could be anywhere and everywhere, working as government cadres and (p.128) reporting to the regime. The programme dramatically expanded the state’s surveillance apparatus and quite literally embedded it in local communities. And yet, unpaid and often disowned by the regime, crime preventers also remained at times meaningfully distinct from the regime, reliant on their families and neighbours to survive. A veteran police officer described the programme as a ‘parallel group’ designed to spy on the police, a ‘sort of political force which works for the benefit of’ the regime, who are ‘used to boost numbers during campaign rallies … Buses of them. Those are artificial crowds. The floating population’ (Veteran police officer, Soroti, 29 January 2018). Having been identified as willing volunteers, they were on standby for the regime, ready to engage in ad hoc work of a wide-ranging nature and able to be disowned at any time.

This programme interchangeably cast crime preventers as members of society and members of the state security apparatus. Sometimes, government officials declared that crime preventers were no more than concerned citizens volunteering information to state security services. Other times, officials declared that all Ugandans should be called ‘crime preventers’ simply because they are citizens. The latter framing collapsed the imagined distinction between state presence and absence, placing the burden of responsibility for crime prevention on citizens. Finally, government officials sometimes claimed that crime preventers were granted special access to and knowledge of the state, and thus were subject to particular rules restricting their rights to talk to the press, to discuss politics, and to protest government initiatives. Though they were volunteers and civilians, they were threatened with martial law if they disobeyed these orders. These continual reframings produce a fluid opposition between state presence and absence that regularly recalibrates the political playing field to favour those already in power. A focus on crime preventers further helps reveal how Uganda’s NRM approaches controlling its massive youth population, which in 2020 constituted approximately 80 per cent of the population.2 Uganda’s Crime Preventers programme therefore reveals a balancing act in which the regime distributes just enough resources to just enough citizens to produce a convincing enough image of state presence.

1. Producing Crime Preventers as Ambiguous Actors

While media frequently framed crime preventers as a political militia, state actors ultimately formed and maintained the programme by alternately and interchangeably deploying multiple justifications for their existence. Contradictory (p.129) narratives made crime preventers difficult to discern—both for society and for crime preventers themselves. Many youths joined the programme because they believed the 2016 elections would return the regime to power, and that supporting this result would give them access to regime patronage. Their calculation seemed conservative—for decades, the NRM has controlled distribution of resources and economic opportunities. At the same time, one of the regime’s basic justifications for the Crime Preventer programme was the regime’s limited capacity and need for volunteers to help keep peace during the potentially tumultuous electoral period. This paradox—of a state that controls distribution of economic, political, and social opportunities and also needs to recruit uneducated, impoverished youth to ensure its continuation—reflects a dilemma. The regime needed to gain support from youth, but it also needed to prevent youth from organizing politically and demanding benefits, jobs, or more representation. The following section illustrates how crime preventers were produced as ambiguously state and non-state actors, with two important effects. First, their activities further cultivated the state’s potential presence at large; and second, the ambiguity of their role prevented crime preventers from threatening the regime in their own right.

1.1 An Opportunistic Repurposing: From Community Police to Political Tool

Like many of Uganda’s informal militias, crime preventers operate awkwardly between state and society, taking on responsibilities associated with formal state security providers while reaping few of the benefits. Authorities often justify such ambiguity through references to the UPDF Act of 2005 and the Constitution, which together provide for all able-bodied Ugandans to undergo state-provided military training and for trained civilians to serve as auxiliary forces.3 In 2018, an army spokesman explained that anyone can join the Reserve Forces; he encouraged his interlocutor, a newspaper reporter, to join, noting that reserve forces support the nation in diverse activities when needed, and not just in times of war (Bagala 2018a). Similarly, many state authorities framed crime preventers equally as state agents and civilians. As ambiguous actors, crime preventers were particularly effective at intimidating ordinary citizens without requiring access to power that would allow them to make effective claims on the regime.

(p.130) According to Bruce Baker, the first crime prevention panels in Uganda were established as early as 1993 to bolster policing capacity at a local level.4 The programme reportedly trained tens of thousands of participants in:

[T]he nature of community policing and crime prevention; the differences between criminal and civil cases; the importance of preserving evidence at the scene of a crime; the institution of criminal proceedings; the LC judicial structure and the cases that they should and should not handle; summons and warrants; road safety; community service; bomb threats; sexual offences; human rights; constitutional rights; domestic violence; laws as they relate to children; marriage and divorce; and mob justice.

(Baker 2005, 30)

While the programme Baker described had no apparent political role, this had changed by 2011, when crime preventers were recruited to help provide security for the presidential elections. The programme was nominally voluntary; participants formally received no payment. In some locations, recruits received three months of training and were provided with uniforms and batons and instructed to patrol public spaces. Some were accused of intimidating and beating members of the political opposition. One former crime preventer explained his role as similar to that of a polling assistant:

We also helped a lot with the voting—escorting the votes. We would take the presiding officer with the ballot box up to the polling station. We would make sure the votes aren’t stolen by opening the box before voting started to show everyone it was empty. Then we would make sure people vote only once by marking their finger with ink when they leave the polling station. Then we would take the ballots up to the sub-county and they would be counted from there.

(Former crime preventer, Lira, 7 November 2014)

Some of those who helped with the 2011 elections reported receiving a one-time payment of 300,000 shillings (100 US dollars) and additional training to be hired as special police constables, or police officers on contract. Reportedly, some crime preventers were able to climb the ranks and become fully incorporated into the police force, while others were retrenched, often without notice or explanation.

Broadly speaking, Ugandans described the Crime Preventer programme as little different from other government programmes including mayumba kumi (‘ten houses’), chaka mchaka (military training for civilians), LDUs, and other militias, such as home guards and arrow boys, established in response to specific security (p.131) threats. An official working for the ISO5 told me in late 2014 that all local vigilantes had been transformed into crime preventers. Although such a transformation was not universally recognized—nor presumably implemented—his statement reflects the reality that many were unclear on the specifics of the Crime Preventer programme, viewing differences among informal security arrangements as purely semantic.

Part of this confusion can be explained by an organizational shift that occurred in 2014. The then IGP Kale Kayihura, in what appeared to be an opportunistic move, organized and trained young elites, and gave them the helm of what had formally been a local community policing initiative. This group of young elites—mainly university students in Kampala, coalesced in January 2014, when they requested self-defence training from the police to prevent on-campus crimes. The students reported that Kayihura ‘picked interest’ and supported their cause, training 700 students at the Police Training School at Kabalye (Bagala 2015).6 By layering a formal, top-down institutional design on top of the Crime Preventer programme’s original localized structure, the regime was able to quickly construct a national programme with name recognition, local legitimacy, and grass-roots connections.

The elite students took on leadership roles in the formal structure, named the National Crime Preventers Forum (NCPF), and maintained close relationships with the president and then-IGP Kayihura. Rumours circulated that the top leadership of the NCPF were the IGP’s son and daughter.7 Photographs posted on Facebook in 2015 pictured the top leadership of the NCPF with Museveni and Kayihura at Museveni’s private home (on file with author). In the months before the election, the top leadership of Crime Preventers publicly acknowledged their own support for the president and the NRM, explaining that Kayihura and Museveni were patrons of the programme (NCPF leadership, Gulu, 4 February 2016).8 When I visited the NCPF offices in Kampala in December 2015, it appeared they were in the midst of moving in. New laptop computers were stacked on a table in the lobby, and posters of Museveni’s face, inscribed with ‘The Father of Our Country’, adorned the walls.

(p.132) The NCPF adopted an institutional structure that paralleled that of the police, with coordinators at the village, parish, sub-county, district, and sub-regional levels. The extent to which the NCPF coordinated with the Uganda Police Force remained unclear. Crime Preventer coordinators—or ‘commanders’, as they were sometimes called—explained that they frequently received instructions from the NCPF headquarters, which only sometimes appeared to have been shared with local police. Between 2014 and 2016, the number of citizens trained as crime preventers surged, and by November 2015, the government claimed to have recruited 30 citizen volunteers in each of Uganda’s 56,000 villages for an estimated total of 1.5 million (Gaffey 2016; Uganda Police Force 2015). This would have constituted nearly 4 per cent of the country’s population. Many argue that these numbers were inflated to intimidate the political opposition; however, more accurate tallies are difficult to come by since Crime Preventer coordinators dutifully filled village rosters with the 30 required names—whether or not those listed committed to participate actively. Crime preventers were often trained by police or former military officers and asked to support the work of the police.

Structurally, the role and identity of crime preventers was ambiguous. Crime preventers were interchangeably accountable to various masters: they were formally and institutionally accountable to the NCPF; informally accountable to central government authorities, such as the president and the then IGP; and personally and socially accountable to the communities where they lived and worked. This ambiguity extended the programme’s foundational contradiction: on one hand, state representatives emphasized the state’s potential absence, its fragility, its need for additional capacity, and its inability to finance crime preventers; and on the other, many participants decided to join the programme based on their belief that they would be rewarded (with a job or payment) when the government was successful in the 2016 elections.9

1.2 Training Crime Preventers: Political Education and Military Spectacle

Training for crime preventers not only inculcated obedience and subservience to the ruling regime, but also repeatedly revealed to recruits the military and economic strength behind the NRM complex. Localized training, typically run by active or retired police officers, taught crime preventers about state law as well as the achievements of the NRM regime. Although training sessions were (p.133) generally modelled on other police and military training, there were no formal instructions for what should be taught or how, leaving each individual trainer to determine his own curriculum. Training at village and sub-county levels emphasized marching (Figure 6.1), but also included other military drills and culture, such as songs, Swahili commands, and saluting, as well as ad hoc lessons in patriotism and law enforcement. Some crime preventers mentioned tough punishments for being late or failing to take training seriously. However, because crime preventers worked on a voluntary basis, commanders could not be too harsh lest they quit. Training was often held in public places, such as sport fields or sub-county headquarters, where passers-by could observe them marching in formation. These performances of militarism distinguished crime preventers from ordinary citizens in their communities. In some cases, crime preventers reported that they were increasingly estranged from their communities, where people saw them as spies and government informants.

Whither the State?Surveillance, Crime Preventers, and Potential State Presence

Fig. 6.1 Crime preventers at pass-out ceremony with President Museveni presiding

The regime bolstered crime preventers’ loyalty through direct efforts at indoctrination. One LC1 chairman, who was himself a long-time supporter of the NRM, explained his views on party indoctrination through chaka mchaka, a military training course designed for civilians.

In the cadre course, they teach what they call patriotism. They have a lecture given by experienced politicians. They start way back with the history of Uganda, before colonisation. They talk of the good and the bad things that the (p.134) government does and then you are given the freedom to discuss … Then you try to compare the past and the present … It’s like a debate. The government in power will always praise itself more. They explain that before [the NRM was] there, things were like this or that, but now, we have UPE [universal primary education], USE [universal secondary education], better security, roads, and so on. During the lectures, they are also very tricky. The lecturer will tell you the good things. Then there will be another one to tell the bad things. That’s when you’ll hear a lot of questions. Then you will know who to focus on and how to convince them. That’s how you can learn how to really support the party. I think that the crime preventers are also getting these lessons.

(LC1 chairman, Gulu, 9 February 2016)

Patriotism and nationalism are common elements in Uganda’s military training. For example, in chaka mchaka, ‘political education’ or ‘ideology’ was a key part of the training. Recruits were urged to become ‘transformation agents’ of Uganda by participating ‘active[ly] in economic and productive development and … act … as the instigators and promoters of government programs in their communities’ (Verma 2012, 104). Patriotism became entwined with self-interest. The LC1 explained, ‘The training on patriotism is a way of giving recruits the wisdom of the good of the government and how to convince people that the government in power is the best’ (LC1 chairman, Gulu, 9 February 2016). Respondents were convinced that, on the whole, the indoctrination successfully convinces recruits that the government is responsible for the good developments they observe in society. Sessions on ‘patriotism’ and ‘nationalism’ were regularly included in the curriculum at the Police Training School at Kabalye, where many recruits went for more advanced training.

National-level training further reinforced crime preventers’ perceptions of a personal relationship with the ruling regime. At such training, much of which was reportedly held at the Police Training School at Kabalye, crime preventers learned that their role was to support the state and protect the peace—for example by disbanding protests. In Uganda, where state, government, and party are synonymous, it was difficult for even the most thoughtful participants to distinguish activities that were partisan from those that served the public interest. Simply being incorporated into an institution affiliated with the police was sufficient to win the rhetorical support of many crime preventers, even though material rewards were minimal for most. One member of the NCPF explained that this was because crime preventers felt the NRM regime was ‘the only one looking out for them’ (NCPF leadership, Gulu, 4 February 2016). Many respondents commented that the armed forces always vote for their ‘boss’, Museveni, and several alleged that the military has been responsible for voting fraud in the past. Militarization also contributes to support for Museveni. One regional police commissioner explained:


We teach [crime preventers] discipline—for example, when I say, ‘Stand easy’, you don’t ask ‘Why?’; when I say ‘Turn right’ you don’t ask, ‘Why?’ And we teach them rudimentary military skills, especially parade. You know us security people like parade. And how to greet and pay compliments to authorities. Basically, respect for the forces.

(Regional police commissioner, Kampala, 6 November 2015)

Crime preventers are taught to respect command hierarchy, do as they are told, and ask questions later, if at all. The training also built a relationship between the state and recruits, as one crime preventer explained:

With the force, once you join, they tell you that the first priority is to keep secrets and be disciplined. With the force, it is command. That is the most important thing. When the government gives you that knowledge, they will never leave you. You cannot leave the army, because they have given you all the government secrets.

(Crime preventer, Gulu, 2 February 2016)

High-ranking government officials, including the IGP and the president, attended these retreats personally to meet the trainees and bestow gifts.

1.3 Motivating Crime Preventers: Batons and Bikes

Though the regime promised to reward crime preventers, the distribution of these goods was often opaque and unpredictable. Recruits were asked to show that their loyalty was unwavering by sticking with the programme despite numerous disappointments, broken promises, and wasted time.10 Even then, only some participants were rewarded with promotions, payments, or praise. These rewards were used both to cultivate personal ties between crime preventers and the ruling regime, and to create examples that could be held up to motivate other crime preventers. For instance, crime preventers who were eventually integrated into the police became narrative lore. More concretely, in 2015, crime preventers at the sub-county and district levels were given motorcycles, nominally to facilitate their movements and intelligence gathering. However, many crime preventers used them for their personal travel or to start a motorcycle taxi business. (p.136) One crime preventer explained that the motorcycles were framed as a gift to reward crime preventers for all the unpaid work they had done to support the regime:

The time of election [in 2011] we went to police school in Kabalye … We worked day and night. Day and night! They can pick you even around 3am, moving, patrolling. We worked, I said, ‘It’s ok.’ Until the time of elections, [then] they gave us motorcycles. [It was on] 24 December 2015.

[Like a Christmas present?] Yeah. As a gift from the president for my work. We moved with those motorcycles. We were very sharp in getting information … You know if the president gives you a gift that is a gift—it is yours.

(Former crime preventer, Soroti, 31 January 2018)

The comments of this crime preventer highlight that though these motorcycles were publicly justified as an investment in the programme, within the programme they were recognized as patronage.

Promises made to crime preventers often relied on the regime’s survival: police promised crime preventers that they would get their reward after ‘the big man’ won re-election; payment for the special police constables would occur after the election results were announced. If Museveni was defeated, crime preventers expected to return home empty-handed. They were thus implicated in the regime’s success both professionally and personally. The IGP and the ruling regime reinforced this narrative, suggesting in rallies and public speeches that crime preventers should be prepared to fight for the government if the peace was disturbed (The Insider 2016). Rumours of armed crime preventers making arrests in Gulu and other districts were circulated constantly in the months before the 2016 elections. In response to such rumours, citizens assumed that crime preventers would be instrumental in enforcing the president’s electoral success, whether with veiled or overt violence.

Crime preventers were regularly tasked with activities that demonstrated their allegiance to the regime and distanced them from their communities—without decreasing their dependence on those communities. In this way, they became agents of regime-led projects of social control, carrying out a type of intimate and personal governance that is distinct from neopatrimonialism. For example, crime preventers were often tasked with arresting gamblers. Gambling is an illegal but popular pastime. One crime preventer lamented that he no longer felt safe in his community because of the work he did for the police.

Even my friends, they don’t like me [anymore] because they say for us [crime preventers] we are capturing people, gamblers. Whether I am doing it or not, they say I am the commander, I am instructing them [the crime preventers] to go and do the work. I am not happy every day. The work which I’ve entered in is not (p.137) good at all. My life is not safe … I have [moved] from the place where I used to sit always because when I go there they are just saying ‘ah, this one is not a good guy’.

(Crime preventer, Gulu, 3 February 2016)

Many community members and elites such as lawyers, NGO workers, and politicians described crime preventers as uneducated ex-rebels and criminals—youth who have no option but to be used as tools of the ruling regime. At best, such assessments were sympathetic, but more frequently they were dismissive. Despite such denigration, many crime preventers decided to stick with the programme, reasoning that they had already committed to it and their friends and families would understand their choice as a potential pathway to a future livelihood.

Those who refused to follow commands faced retribution or replacement, as one commander explained: ‘If you say you don’t like [the command]—immediately, I have to replace you. Because an order is an order. But not by force. Not even putting that person in too much pressure. Saying, “You just go out.” I have replaced them but not reported them’ (Crime preventer, Gulu, 3 February 2016). Dismissing crime preventers seems innocent enough; however, many had made significant sacrifices in terms of time and personal relationships hoping that their dedication would eventually yield a reward.

The relentless demands of the programme culled crime preventers, and conditioned them to have high hopes and low expectations. Those who were not desperate, patient, or committed dropped out over months of broken promises. The occasional distribution of gifts and select promotions created examples that leadership could point to when seeking to motivate crime preventers. These hopes, in turn, encouraged competition among recruits, which helped balance the camaraderie developed in training such that recruits’ strongest commitments were to the regime, not to each other. Thus, crime preventers remained fragmented, protecting the state from an organized interest group of young men that could otherwise have demanded representation or remuneration.

2. Ambiguous Activities, Uncertain Identities: Fostering Potential State Presence across the Country

During the 2016 elections, crime preventers took on an ambiguous role as at once ordinary citizens and, simultaneously, state agents. The following examples describe how crime preventers were used to bolster attendance at rallies, to police the elections, and to spread rumours about electoral interference. Their activities contributed to the perception that the state might be present anywhere at any time.

(p.138) 2.1 Demonstration at Karuma Bridge

During the 2016 election, crime preventers were called upon to attend rallies, protests, and demonstrations, producing the appearance of a ground-swell of support for government initiatives, programmes, and candidates. The programme served as a network through which those in need of crowds could recruit bodies in numbers. Because of the ambiguity surrounding the programme, the government was able to brand participants either as crime preventers or as ordinary citizens, often selecting post hoc the identity best suited to its publicity needs.

In 2015, regime insider Amama Mbabazi joined the presidential race to challenge Museveni. Mbabazi had formerly been Museveni’s prime minister and a founding member of the NRM. Many initially saw him as a formidable opponent to Museveni, someone who knew the regime’s strategies and could use them to unseat the president, or at the very least peel off some of his support. It was thus a surprise when, on 10 July 2015, an MP for the opposition—the honourable Odonga Otto representing Aruu county in northern Uganda—organized a protest against Mbabazi. The protest was planned to meet Mbabazi’s convoy at Karuma Bridge over the Victoria Nile river. The location was strategic both symbolically and practically: the bridge divides Uganda’s northern region from the south, and it provides a choke point that cannot be bypassed. The protest photographed well (Figure 6.2): over 400 youth mobilized, publicizing the message that Mbabazi had stolen 1.4 billion shillings from the Peace, Recovery, and Development Plan, a programme designed to rebuild northern Uganda after the LRA conflict (Ocungi and Okaba 2015). Local and national news media covered the demonstration; however, there was no confrontation as Mbabazi had delayed his trip to northern Uganda, reportedly for unrelated reasons (Etukuri and Semakula 2015).

Whither the State?Surveillance, Crime Preventers, and Potential State Presence

Fig. 6.2 Protesters walking toward Karuma Bridge; some carry MP Odonga Otto (centre)

The following day, crime preventers from Gulu complained that they had been ‘tricked’ into joining the rally. They said that police officers had notified them that there would be a three-month training at the Police Training School at Kabalye, and those who wished to attend should report immediately to the Central Police Station in Gulu and board the transportation provided. Instead, the crime preventers were driven 65 kilometres from Gulu and told to disembark and don T-shirts with a red ‘x’ through Mbabazi’s initials (JPAM) and the slogan ‘Why buy a Benz of 600 million with PRDP [Peace, Recovery, and Development Plan] money?’ They were then instructed to join the other protesters in marching the remaining 10 kilometres to Karuma Bridge. The day after the protest, some of the crime preventers were disgruntled at having been misinformed about the purpose of their travel, coerced to march in the heat without water or food, abandoned with no means to return to Gulu Town, and paid nothing for their time. These crime preventers brought their complaints to the district and regional police (p.139) commanders, who reportedly responded that they themselves had been similarly misinformed.

Because the Crime Preventer programme had a loose and undefined mandate, politicians were able to mobilize them for overtly political activities and then claim that the crime preventers were acting of their own volition. In turn, the crime preventers were willing to obey orders even when there was significant evidence to suggest that they were being manipulated. For example, several crime preventers reported that they were first put on a lorry that was in such bad shape they doubted it would be able to make the trip to Kabalye. Then, when they were given the T-shirts, they donned them without asking questions, despite not knowing what ‘JPAM’ (Mbabazi’s initials) meant. They continued to follow the orders to descend from the bus, join the rally, and march to Karuma Bridge. They only challenged these orders in retrospect, when it became clear that they had gotten a raw deal. The police rejected the crime preventers’ rights to make claims.

A variety of evidence suggests that the police who informed the crime preventers about the sham training did, in fact, know that they were really being taken to the demonstration. One retired police officer reflected that someone in the police force must have given permission: ‘Because you can’t come from nowhere and pick someone who I’m looking after [as a member of the police force], and take [him] away’ (Retired police officer, Gulu, 16 October 2015). (p.140) Others speculate that the police offered the services of crime preventers to please the NRM party leadership, whether directly commanded to do so or not, because ‘That is how you get promoted—do something to please the president’ (Locally elected politician, Gulu, 21 September 2015). One crime preventer who facilitated training explained that he suspected foul play and initially refused to go:

I refused to go to Karuma—the CID [criminal investigations director] came to talk to me, and said to me ‘Let me take these people, and then I’ll provide transport for you to come back.’ The police knew that if I did not go the others also would not. [In this way] I was forced to go to Karuma.

(Crime preventer, Gulu, 29 September 2015)

One district-level politician further asserted that there was evidence that the NRM paid the organizing MP, Otto, to stir up resentment against Mbabazi in northern Uganda (District-level politician, Gulu, 19 February 2016).

[After Mbabazi announced his candidacy] the Movement got in touch with some individuals in the opposition in northern Uganda—one is honourable Odonga Otto. Otto was given money [by the NRM] … Where did he get money to hire six buses from, trucks for carrying people? He took over 400 youth. He fed them. He gave them over 20,000 [shillings] each to return on. Where does he get over four million to spend? … The people he took were crime preventers. The police lied to crime preventers that they were taken for further training in Masindi. They [the crime preventers] went on radio; they were very bitter … The most disturbing part is, how do the police mobilise such people, don’t tell them the truth and allow a political leader to use them in a wrong way?

(Elected official, Gulu, 25 September 2015)

He posited that this was why Otto—an opposition politician—meddled in internal NRM party politics, despite facing discipline from his own party for doing so (The Insider 2015). The protest was thus a political manoeuvre in which crime preventers were collateral damage, instrumentalized by the NRM and Otto for their own ends.

The crime preventers continued to seek redress: they threatened to march to Gulu’s Central Police Station in protest, but were told that if they did, they would be tear-gassed. Instead, the crime preventers went to journalists from the major newspapers, who ran a number of articles with headlines including ‘We were tricked to join anti-Mbabazi demo—Crime Preventers’ (Otto 2015), and ‘Gulu Crime Preventers hoodwinked into joining anti-Mbabazi demos’ (Ocungi 2015). They also went to a human rights NGO, which advised them to go to the Department of Labour to complain about unpaid work. Reflecting on this (p.141) recommendation, another crime preventer said: ‘But they are not government employees, so they were left hanging.’

The police, who had previously been unresponsive, contacted the complainants and threatened them with jail time for going to the press. When asked why crime preventers are not allowed to talk to the press, the police public relations officer for the region explained that it was for their own protection: ‘It’s risky [to talk to the media]! If you report on your friend who has committed a crime, what will stop him from doing something bad on you? So, it’s like you’re an intelligence officer and you need to be protected’ (Police public relations officer, Gulu, 23 November 2015).

Seven months after the protest, in February 2016, one crime preventer declined the opportunity to become a constable because he feared retribution for having spoken to the press about the Karuma Bridge demonstration. Although the press coverage of the protest noted 40 disgruntled crime preventers, my interviews suggest that, in fact, crime preventers were bussed from various districts in northern Uganda, including Pader and Nwoya. Others may not have complained because they were satisfied with the amount they were paid (reportedly between 5,000 and 20,000 shillings, the equivalent of 1.67 to 6.67 US dollars). Or perhaps they realized the potentially high costs of publicizing their situation, as alluded to by the police public relations officer.

Crime preventers were used for rallies on other occasions, either to control crowds or to increase attendance. For example, on the day of the president’s nomination, buses full of youth dressed in yellow—the colour of the NRM party—filled the streets of Kampala, whooping and yelling. One crime preventer explained:

We went to Kololo for the president’s nomination. We went as supporters, and we all put yellow [the colour of the NRM party]. The DPC said the president wants to talk to the crime preventers. I thought, ‘The president needs to tell us something important.’ They provided transport from and to. [We were told] those who went will get 500,000 shillings each. Then we were told to stand for the rally and listen to the speech. He was for nomination. They told us, ‘you should be happy, you will be paid for it. Sing, dance and wave—you’re 500,000 [shillings] is coming’.

(Crime preventer, Gulu, 20 November 2015)

This strategy, although blatant, creates the appearance of massive support. Because there is no transparency about who is there to get paid and who genuinely supports the candidate—often there may be no substantive difference—the ploy appears to be effective at bolstering a candidate’s popularity and augmenting the perception that there is widespread public support for the regime.

(p.142) 2.2 Special Police Constables and the 2016 Elections

One of the few publicly stated rationales for recruiting crime preventers, aside from combatting crime in a general sense, was to help ensure peace during the 2016 presidential elections. This included the prevention of violent protests, management of election rallies, and keeping order at polling stations. A few weeks before the elections, the police reportedly recruited 36,000 police constables (Kato 2016), many of them drawn from the crime preventers. The widespread presence of crime preventers at polling stations bolstered the perception of state presence at the very moment when the regime’s continued survival required electoral support. Training and short-term employment also gave the crime preventers a taste of how they might access the state’s patronage after the election, helping ensure their loyalty at this key moment.

Although the selection process was not transparent, crime preventers believed that they were recommended for work at polling stations by sub-county crime preventer coordinators and selected based on the loyalty and commitment they had shown to the programme over the preceding months. Reportedly, in Gulu District, nearly one-fifth of recruits were dropped without explanation after initial selection; this appears to have occurred in other districts as well. According to police officials in Masaka District in the Central Region, some were cut because they failed interviews and others because they were physically weak or did not have the minimum educational requirements (Ssenkabirwa and Kisekka 2016). In Gulu, one crime preventer speculated that those dropped did not make it through security checks, which were conducted in Kampala after the recruits submitted their fingerprints. Another guessed that cuts had been made because of the government’s limited resources.

Those who were selected participated in a one-week residential training in Gulu Town. A significant part of the training focused on disciplining the new recruits. One recent recruit explained to me some of the tactics used to teach the recruits discipline and respect for authority:

Say you are sitting with a colleague after hours. An instructor comes and looks at you. ‘Get up, come. Go back. Come. I told you to come. Go back and sit. Come!’ He’s seeing how you are responding. It will be put in your notes, you’re someone who can follow command, which is needed in the force.

There’s a lot of lies in training. They call them ‘sweet nothings’. They even brought a very big fat cow. They say, ‘Today, it’s for you guys.’ They took it behind the kitchen and hid the cow. Guys were happy … Guys started washing their dishes, looking for pepper … When it came to lunch time, the whistle was blown, and everyone started fighting to get into line. They dish beans and posho [a staple food]. So when you come and you get you are expected to say ‘thank (p.143) you’. When you don’t say ‘thank you’ [that’s indiscipline]. In the force, you are not supposed to initiate anyone to support you … There’s no riot in the force. [If there is a problem] go alone and say, ‘please, this was not good’. So, you can also be handled alone.

(Newly recruited special police constable, Gulu, 13 February 2016)

In addition to instructions on their duties, crime preventers were trained on the institutional organization of the police force, including their role as special police constables. They were also provided with a list of activities prohibited on election day and another list of ‘uniform rules’. The duties of the special police constables included:

  1. 1. Know we are the special police constables, not election constables [who are tasked with managing the voting process].

  2. 2. Support the police in patrol during night and day.

  3. 3. Manage violence and enforce the law.

  4. 4. Be ready to do duties of emergency in case of a bad situation arising.

  5. 5. We should be ready to work with the army security agency, including prisons, army, and intelligence agents.

  6. 6. Discipline: we should not put on [political] party shirts, or flash any [party] slogan. Anyone who does that will be charged with the Police Act.11 That person will be sent to prison.

The uniform rules focused on how special police constables should present themselves as part of the force, rather than as crime preventers. Special police constables are told that they must not be partisan; however, this is difficult in a context where the regime and state are functionally fused. A few days before the election, I asked one recent recruit what he would do if he observed violations on polling day. He explained:

As a polling constable, there are certain limits—my hands are tied. If anything happens, I have to note it down. So when you [the victim of injustice] go to your candidate to make a complaint and they call me, then I can give the information in court. Or, if I cannot go to court [because of my own security] I can give the report [to someone else to deliver]. With the government having a longer arm, people want to be on the safe side.12 I’ll take note. I’m not going to court to be on the safe side.

(Special police constable, Gulu, 19 February 2016)

(p.144) The respondent felt that he could help improve the system through participation, although he feared that making public statements in court about voter intimidation, fraud, or other irregularities might put his own safety at risk. He also explained that although special police constables were clearly instructed to keep partisan attitudes to themselves,

At the end of the day, all the big people will come [to the training]. What they do is say you have to vote wisely if you don’t want to go back to the bush or back to the IDP [internally displaced people] camps. Of course, what they mean is to vote for Museveni. It influences a lot of guys, but not all.

(Special police constable, Gulu, 19 February 2016)

Other researchers have noted that such security training programmes are infused with pro-government propaganda. For example, a 2016 Human Rights Watch report cited a Crime Preventer training programme that stated ‘Every good thing you are seeing around is as a result of good NRM governance’ (Human Rights Watch 2016).

On election day, special police constables were often indistinguishable from long-serving police officers, though some were identifiable by their ill-fitting or incomplete uniforms. I identified one special police constable by his boots, which were pink and lined with faux fur rather than the black military boots that most police officers wear. Outside the tallying station, a colleague whispered in my ear that the officers with white lapels were crime preventers. The degree to which they blended in, however, reflects a comment that a female employee at a human rights NGO made to me:

When the [special police constables] came back from six months of training they were uniformed and you could not tell who was who … When circumstances call for it, they just change the uniforms around … Unless someone tells me who is behind the uniform, we can’t know.

(Human rights NGO employee, Gulu, 29 September 2015)

The crime preventers who worked as special police constables were told they would be paid 11,000 shillings (3.67 US dollars) each day for 14 days of work, starting with the presidential elections on 18 February 2016. Reportedly, the police asked some crime preventers to return their uniforms before paying them, which resulted in riots quelled only when the police ‘explained to them the police procedures and they [the crime preventers] understood our position and their response to our order is now good’ (Bagala 2016).

Thus, young men who became crime preventers were introduced to the NRM system, however cursorily, through a process of training and the hope of future employment. The Crime Preventer programme served the regime’s need to bolster (p.145) its perceived security capacity in the eyes of citizens, without substantially increasing costs or training youth to the extent that they themselves could threaten the regime. The programme had the added benefit of regularly exposing crime preventers to the NRM’s military and economic strength, cultivating their belief that the regime has access to wealth and sovereign violence that these young men, if they played their cards right, could potentially benefit from.

2.3 Rumours and Suspicious Activities

The many rumours circulating around the Crime Preventer programme helped create uncertainty about whether or not crime preventers were agents of the NRM. For example, in December 2015, crime preventers were reportedly deployed across the country to check the voter registration list. This deployment and Mbabazi’s call for an investigation into it were reported in various newspapers (Mugume 2015; Musinguzi 2015; NTV 2015; Segawa 2015; Sserunjogi 2015). Several crime preventers and civilians explained to me that crime preventer coordinators for each village were given the voter registration list and asked to verify it door to door, checking off individuals who were correctly registered, marking deceased ‘D’, and those who had moved ‘DR’.

Crime preventers expressed conflicting interpretations of this activity. Some argued that the information would be used to bias the election in favour of the NRM. One crime preventer said that they were supposed to remove known members of the opposition from the registry. Others suggested that checking the list would intimidate voters; still others that the deceased would be made to vote for the NRM. Another crime preventer pointed out that this task put him in a difficult position: ‘Should I do anything stupid with [the voter registration list], it will backfire on me. [The community members] know me from my childhood. It’s very risky to do anything’ (Crime preventer, Gulu, 4 February 2016). Others insisted that the exercise was intended to make the list more accurate or prevent the opposition from rigging the election. A police officer in Gulu Central Police Station explained:

The voter registrar of the Electoral Commission is the EC’s responsibility. Each presidential candidate is given the voter registration to cross check if it’s okay. He can use any method. The crime preventers are members of the community … Some presidential candidate [might have] decided to use them to check [if the voters are existing or dead] because they [crime preventers] are many … This helps you know the number of people who are registered. The ones who are alive, you can know the number, and then you can know if the number of votes given is (p.146) more or less. When they add the votes up, it should be slightly less than the overall list. This helps them to know there was no rigging.

(Police officer, Gulu, 13 February 2016)

Others denied knowledge of this activity, despite reports in the media, and by community members and crime preventers.

Reportedly, each village coordinator was paid 5,000 shillings (1.67 US dollars) to verify the list, while sub-county coordinators were paid 150,000 shillings (50 US dollars). One sub-county coordinator explained that although the village coordinators were supposed to do the work, he was responsible for completing the task:

I leave it with village coordinator. If it becomes difficult for them, I fill it. Me as a commander, I have to do the work to make sure the form is full. If they [the Crime Preventer leadership] say, go and plant for me the maize, I cannot say, ‘I’m tired.’ I have to finish the work … For them they know [which community members are alive, dead, or have moved]. If they don’t know, you have to ask someone who knows, so that he helps you to mark those people. But secretly. When people know [what we are doing], that is another problem again. That is why they are talking on us every day.

(Crime preventer, Gulu, 3 February 2016)

Despite the hundreds of individuals involved in checking the list across the country, the purpose of the exercise remains obscured. A leader from the NCPF said he did not know what happened with the revised lists, but implied that they made their way all the way to the Office of the President. ‘I don’t know what [the president] did with it [the updated voter registration list]. I don’t know where they took them. We came back to do our work’ (NCPF member, Gulu, 4 February 2016). Respondents—including journalists, crime preventers, and politicians both in and out of power—generally seemed unconcerned about this activity. Upon further questioning, I interpreted this indifference as rooted in their belief that the entire system was fixed. Thus, they saw little value—and lots of risk—in investigating the minutiae of how.

These rumours show how crime preventers were cast in an uncertain light: they were agents of the NRM and at the same time ordinary citizens simply pursuing a potential livelihood. Their activities in the community brought this ambiguity to the doorsteps of ordinary Ugandans. Together, the three examples illustrate how crime preventers were leveraged for political ends throughout their recruitment, training, and deployment. Both the police and politicians continually redefined crime preventers’ relationship to the regime, sometimes framing them as agents of the state and sometimes as citizen volunteers.

(p.147) 3. Analysis: Whither the State?

There is a saying in northern Uganda, gamente c’inge bor—which translates as ‘the government has a long hand (or forearm)’. The saying is akin to the English phrase, ‘the long arm of the law’. In an interview, a community member explained:

You see, the government starts with me and you who are seated right here provided we talk what is consistent with what the government has laid down in the laws … The strength of the government depends on us who are here at the grass roots. Without us there is no government.

(50-year-old male community member, Gulu, 5 March 2015)

In this telling, the state is not only omnipresent, but actually embodied in Ugandan citizens. The regime’s ability to collapse the state–society division and to create the perception that all Ugandans are part of the state can be traced, at least in part, to Uganda’s history as a no-party state. Many scholars note that Uganda’s no-party system was akin to a one-party state. But the emphasis on all Ugandans being part of the Movement—and a big tent approach to politics that nominally encouraged meritocratic competition and even critique and reform—helped fuse the NRM to the Ugandan state and citizenry (Carbone 2008). In my interviews, many Ugandans linked contemporary political turmoil to the return to multiparty politics. One LC councillor elaborated on this perceived fusion between citizens and the state:

Nowadays we are all part and parcel of government … whether central or local government. Whenever there is something going on wrong, we at the grass root leadership try our best to report to the responsible authority [so they can] respond and rescue the situation. It has become easier [now] that we do not have to move very long distance to report cases of crime since there nearby police post. The state has consolidated its authority up to the grass roots.

(LC1 Chairman, Gulu, 5 March 2015)

In his view, the Ugandan state is ‘consolidated … up’ to the local level in part because of an historically embedded notion that citizens are the state, but also that they are subject to the state. The co-existence of these notions allows the regime to regularly reframe the identity of citizens, and concomitantly their rights and responsibilities, thereby keeping the state–society boundary fluid and unconsolidated. This fluid boundary makes it appear that the state could be present at any time, embodied in ordinary citizens.

The use of the Crime Preventer programme was opportunistic: the ruling regime repurposed an existing community policing programme to extend the (p.148) party’s patronage system and earn the support of tens of thousands of unemployed and desperate youth. The regime kept rules about crime preventers vague, allowing sufficient space for powerful politicians to manipulate the programme for political ends. The regime and the police used a combination of promises and threats to keep recruits in a precarious situation; as a result, crime preventers allied themselves with the regime, not with each other. Crime preventers could not reliably predict whether state authorities would act in their personal or formal capacity. Importantly, decisions by state authorities—whether personal or formal—were backed by the dual threats of exclusion from the regime’s system of resource distribution and of symbolic and material coercion, such as arrest and detention. The resultant unpredictability undermined the ability of citizens to act strategically or hold state actors accountable. Aspects of this system reflect theories of neopatrimonialism. But the Crime Preventer programme also demonstrates how internalized obedience to an abstract idea of state power is an immediate force in individuals’ everyday lives. Patron–client ties can exist in this system, but they are contingent and uncertain, framed by the idea of the state’s power.

The Crime Preventer programme was able to mobilize and organize unemployed and underemployed young men, while in the main keeping them competing with one another rather than coordinating to make claims on the regime. Those who became crime preventers were generally very poor; their families needed food, shelter, and education. As young men, it was their designated social role to provide (Dolan 2011; Tapscott 2018). Those who complained about the terms of work were made into examples—excluded from the programme or even threatened. They were thus thrust outside the NRM’s patrimonial structure, potentially losing access to resources in both the short and long term. One crime preventer coordinator explained his conundrum:

The opposition doesn’t care about us. We need the money of the government because we are the jobless people. [The government] say[s] they care about us, but [I think] what we’re doing is not good. But we went [to be crime preventers] because of the money.

[What do you think about that?] Their point is not bad. But when you support [the government], there is nothing we shall gain from them. Their word is always very sweet like that. But there is nothing to gain. You have to pray to God to help you … but not these politicians. They’re all the same. Whether they send them there, they will talk like they will give you something. But when they go for parliament, you will see them after five years.

(Crime preventer, Gulu, 18 November 2015)

Local communities intensified these dynamics, viewing crime preventers as unemployed, uneducated former rebels who were using the title and access that (p.149) came with being a ‘crime preventer’ to become ‘crime promoters’. In this way, they wrote off crime preventers and ignored many of their activities, estranging crime preventers from their communities and making them even more precarious.

4. Conclusion

By examining the experience of young men recruited into the Crime Preventer programme, this chapter has shown both how they contributed to the governing strategy of institutionalized arbitrariness and how they were governed by it. While the implementation of the programme was disorganized and contradictory, its governance effects were in some ways neat. Because authorities sometimes narrated crime preventers as integral to the regime, their presence offered ordinary citizens evidence of state presence. Ordinary people saw crime preventers training in public spaces, standing to attention for ceremonial ‘pass outs’ that marked the successful completion of their initial training, apprehending suspects in their communities, and guarding detainees at local police stations. Their participation in rallies and demonstrations swelled crowds, physically demonstrating support for the regime. Their recruitment in even the most remote villages revived the perception of the state’s grass-roots surveillance capacity; their review of the voter registration list seemingly brought the state’s eyes and ears to the doorsteps of citizens across the country. At the same time, state authorities also disowned crime preventers and categorized them as ordinary citizens. Crime preventers had no uniform, no appointment paper, and no identification. They were unpaid and disrespected. Their training was on offer to all Ugandans; they were doing their duty as citizens. In such situations, powerless to pull strings or make claims on state authorities, their experiences revealed and reinforced a stark division between individuals inside the NRM tent and those left at the door.

By keeping the role of crime preventers ambiguous, the NRM regime also transformed crime preventers from an apparently innocuous community policing intervention into a tool to efficiently and effectively extend existing systems of resource distribution to a potentially troublesome segment of the population—underemployed, marginalized youth who might otherwise have joined the opposition. Overall, the Crime Preventer programme appears to have helped the regime win votes and placate Ugandans during what might otherwise have been a contentious and potentially violent election cycle. Although quite distinct from the practice of encouraging local security initiatives, the Crime Preventer programme worked on much the same logic. By keeping the purpose of the programme fluid and undefined, the regime maximized its benefits while limiting its responsibility for crime preventers’ activities and needs.

If the state is informed about everything and has the capacity to intervene, then its lack of action appears to be as deliberate as its actions. This casts the (p.150) appearance of state intention even in its absence. The perception of potential state presence causes citizens to self-police, constantly trying to calculate the risks and rewards of their actions. The relationship between state and society is constitutive of any governing strategy. In democratic polities, a robust and secure civic space theoretically allows citizens to make certain claims on the government—for example, for better schools, lower taxes, or human rights protections. Instead, in Uganda, the blurred and dynamic relationship between state and society creates a possibility of state presence—one that is real enough to impose meaningful threats, and at the same time, fleeting and difficult to pin down.


(1) In 2008, a Brookings report ranked Uganda as the 27th weakest state out of 141 countries in the developing world (Rice and Patrick 2008).

(2) In 2014, an estimated 75 per cent of Ugandans were below 30 years of age (Uganda Bureau of Statistics 2016). Born after the regime took power, Uganda’s youth are less enamoured of the regime’s liberation narrative; moreover, an estimated 64 to 83 per cent of them are unemployed (Reuss and Titeca 2017, 2354).

(3) Section 6(1)(c) of the UPDF Act 2005 states that the sources of the Reserve Forces shall include ‘auxiliary forces, state security organisations and such other citizens of Uganda as have undergone military training under Article 17(2) of the Constitution’. That article states: ‘It is the duty of all able-bodied citizens to undergo military training for the defence of this Constitution and the protection of the territorial integrity of Uganda whenever called upon to do so; and the State shall ensure that facilities are available for such training.’

(4) A Ugandan journalist dates the Crime Preventer programme’s first ‘pass out’—or completion of initial training—to 1994 (Bagala 2015).

(5) The ISO is the country’s national counterintelligence agency.

(6) Although a press release noted, ‘The public is also invited at any stage to come and witness these programs at the Police Training School to help appreciate its value to all’ (Enanga 2014), I did not find this to be the case. When I visited the Training School, I was turned away because I did not have a letter from the office of the IGP.

(7) In my interviews, this rumour was both denied and affirmed by people with first-hand knowledge of the IGP’s family. While lower-level crime preventers unanimously believed the rumour, higher-level crime preventers denied it and said it was merely a reflection of the close relationship between the IGP and these ‘brilliant’ young leaders.

(8) One respondent, a leader in the NCPF, claimed that crime preventers were non-partisan, while also explaining, ‘We are mostly being facilitated by the sitting government. We cannot do work without them’ (NCPF leadership, Gulu, 4 February 2016).

(9) The idea of the Ugandan government as both strong and weak has also been observed in the area of national security, notably by Jonathan Fisher in his study of the Uganda’s manipulation of its status as a ‘fragile’ state (Fisher 2014b).

(10) For example, respondents in Gulu told me that on numerous occasions they were asked to travel to Kampala, they believed to participate in training. Crime preventers travelled from various districts to the Central Police Station in Gulu Town. Upon arrival, they were told to go back home and return at a later time, then asked to wait for hours on end with no food, water, or shelter. On one occasion, limited transportation meant that many who came were sent home with nothing. Those who remained were bussed to political rallies for the NRM in Kampala.

(11) The Police Act specifies that any ‘member of a security organisation placed under the command of the inspector general for the performance of police duties’ is subject to the police disciplinary code of conduct (‘The Police Act’ 1994, pts. VI, Section 44).

(12) This saying is akin to the phrase ‘the long arms of the law’ (Tapscott 2017b).