State Capacity and Elite Enrichment in Uganda's Northeastern Periphery

Introduction

Uganda's successive governments made little effort to project state power in Karamoja, a remote and often violent region in the northeast of the country, until 2006, when the authoritarian National Resistance Movement (NRM) regime of President Yoweri Kaguta Museveni launched a disarmament campaign that resulted in substantial, but curiously uneven, expansion of its capacity to govern the area. Deployment of thousands of soldiers curtailed livestock raiding and interethnic conflict in Karamoja, ¹ albeit at the cost of grievous human rights abuses. In the wake of disarmament, the Ugandan government complemented military presence with an extensive administrative apparatus that effectively penetrates Karamojan society but, costly investments in physical infrastructure aside, provides it with few public goods. Limited state involvement in service delivery – undertaken largely by international governmental organisations (IGOs) and non-governmental organisations (NGOs) – notwithstanding, the regime has expressed its professed commitment to improving the wellbeing of the largely pastoralist local population through active promotion of sedentarisation and, despite generally poor prospects in semi-arid Karamoja, crop cultivation, including on communal land taken over by members of the Ugandan political elite, who also own mining operations in the region. These ramifications of newfound state interest in Uganda's long-neglected northeastern periphery seem incongruous; in reality, however, they evince a coherent strategy that the NRM regime has pursued both in Karamoja and elsewhere in the country to maintain its power and advance the interests of the political elite.

I argue that the regime has strategically prioritised development of the capacity of the government bodies needed to exercise power over Uganda's population and territory, taken advantage of its consequent political dominance to reward elite loyalty to President Museveni with access to opportunities for personal enrichment, only provided public goods conducive to the realisation of these political and material objectives, limited investment in other government functions, and delegated their performance to IGOs/NGOs. The speed and scope of Karamoja's incorporation – that is, substantive integration into the Ugandan political system – in the wake of the 2006 disarmament drive throw these key characteristics of NRM governance into particularly sharp relief. The deployment of the Ugandan state's well-developed coercive apparatus has allowed the regime to extend its ability to project power in the region, establish control over Karamojans, and enforce their compliance. The regime has also built capable administrative bodies that further extend government reach and help to detect and contain popular disaffection, including through suppression of indigenous livelihood practices. Improvements in security conditions and state capacity have enabled the construction of physical infrastructure that facilitates both military operations and natural resource exploitation by members of the NRM elite, who have also repeatedly appropriated funds intended for public services, mostly paid for by donors and provided by IGOs/NGOs, rather than the local bureaucracy.

Based on analysis of archival, interview, and news media data, this article makes a threefold contribution to the scholarly literature. First, by documenting the NRM regime's differential investments in specific state apparatuses, I extend the scholarship on state power and state capacity. Second, my examination of the logic of Karamoja's incorporation helps to make sense of the apparent contradictions – which I discuss in the next section of the article – of the contemporary political order both in the region and elsewhere in Uganda. In particular, and third, the article explains the decision to incorporate the country's northeastern periphery and provides original evidence of its consequences, including elite enrichment.

Karamoja and “the Rest of Uganda”

Along with neighbouring borderlands of Ethiopia and Kenya (Mosley and Watson, 2016: 453, 460–462) – and, indeed, political peripheries across sub-Saharan Africa (Herbst, 2000) – Karamoja long “held no attractions for government” (Barber, 1962: 111). “Nearly all government policies towards Karamoja in both the colonial and post-colonial periods aimed to minimise negative spillover effects of internal upheaval” (Stites, 2022: 9) on adjacent parts of Uganda, although successive disarmament campaigns had little long-term effect (Mirzeler and Young, 2000: 415–17). In these conditions, “[g]overnment institutions and structures for enforcement were either absent or extremely weak. State authority existed only in towns and only during the day” (Stites et al., 2019: 23).

Scholars of the region have provided extensive evidence of its historical marginalisation. They also recognise the transformative impact of the 2006 disarmament drive (Czuba, 2019: 559; Kandel, 2018: 15; Stites and Howe, 2019: 138). This broad consensus does not, however, extend to the conditions in place after the campaign gradually wound down in the late 2000s. One strand of scholarship holds that disarmament “made it possible for a previously ‘absent state’ to return to Karamoja” (Stites et al. 2019: 23). As a result of “[t]he state's success in building up local government capacity” in the region, “[f]or the first time in history, Karamoja is gradually experiencing substantive political incorporation into Uganda” (Kandel, 2018: 15). Other scholars emphasise the continuing weakness of the Ugandan state in the region, “its authority largely circumscribed to military and police units” (Tapscott, 2021: 152), in part because of the low capacity of other government bodies (Hopwood et al., 2018: S146). The literature also presents the state as both distant from (Tapscott, 2021: 152) and – thanks to local intermediaries simultaneously accountable to fellow Karamojans and beholden to the NRM regime – embedded in society (Czuba, 2019). In scholarly accounts, the government inspires fear and resentment (Arasio and Stites, 2022: 32; Hopwood et al., 2018: S151) as well as gratitude for post-disarmament security improvements, which prior to a resurgence of violence in the early 2020s (Arasio and Stites, 2022) were “nearly uniform across locations in Karamoja” (Stites and Howe, 2019: 140). However, even at its most stable in the mid-2010s, the region only saw “relative peace”: the government successfully curtailed raiding but did little to stop small-scale theft and domestic violence (Ibid.).

“Absence of state provision of public goods” (Hopwood et al., 2018: S146) and “inadequate policy or programmatic support for the lives or livelihoods in the region” (Stites et al., 2019: 31) have further contributed to local people's illbeing. Even though the government has launched multiple development initiatives, “there is very little if any evidence to suggest they achieved their objectives in Karamoja” (Nakalembe et al., 2017). These interventions’ impacts have nonetheless been considerable. The NRM regime has worked hard to realise its vision of Karamoja as a “settled community of commercial crop farmers, wage workers, and commercial livestock farmers” (Muhereza, 2017: 7). Expansion of the land area dedicated to these economic activities, a result of both “radical enclosure of the commons by the government” (Filipová and Johanisova, 2017: 20) and “land-grabbing” by well-connected individuals (Czuba, 2019; Rugadya et al., 2010), has undermined pastoral production – recognised as more suitable to the shock-prone semi-arid environment (Nakalembe et al., 2017) – increasing Karamojans’ reliance on food assistance provided, like most other public goods in the region, by IGOs/NGOs (Stites and Akabwai, 2010: 40).

The post-disarmament Ugandan state in Karamoja depicted in the scholarly literature is simultaneously strong and weak. The NRM regime has set out to transform socioeconomic conditions but leaves the provision of basic services to other organisations. And government actions both promote and compromise the wellbeing of the correspondingly appreciative and disaffected local population. Karamoja's commonly emphasised distinctiveness from the “rest of Uganda” (e.g. Barber 1962, 111; Hopwood et al., 2018: S142; Tapscott, 2021: 166; cf. Krätli, 2010) notwithstanding, this seemingly paradoxical state of affairs is far from unique to the region. On the contrary, it typifies the political system that the NRM regime, in power since 1986, has built across Uganda.

The scholarly literature on Ugandan politics portrays the regime as entrenched, centralised, and personalised, with essentially all political power in the country vested in President Museveni, who exercises his hegemony through both the Ugandan state and the NRM, the ruling party (Fisher and Anderson, 2016: 69). The NRM is an impressive political machine that distributes patronage to supporters, coopts opponents, and – following the transition from a “no-party” system to multiparty elections in 2006 – regularly delivers victories in the presidential and legislative contests crucial to maintaining the regime's considerable, if declining, popular legitimacy (Golooba-Mutebi and Hickey, 2016: 602; Wilkins, 2021: 150; Reuss and Titeca, 2017: 5–8 and 14; Vokes and Wilkins, 2016: 589–591). At the same time, the party's weak internal coherence fuels tensions within the regime (Wilkins, 2021), which the president has, despite his dominant position, rarely tried to contain (Vokes and Wilkins, 2016).

The Ugandan state, “the bedrock of Museveni's power” (Wilkins, 2021: 150), replicates the NRM's discordance. The president maintains a well-funded, equipped, and trained coercive apparatus made up of the Uganda People's Defence Force (UPDF), Uganda Police, and various paramilitary organisations, which effectively project state power across the country (Reuss and Titeca, 2017: 10). The increasingly frequent, if unpredictable (Khisa, 2021; Tapscott, 2021), application of force has helped to embed NRM dominance in the Ugandan political culture (Vokes and Wilkins, 2016: 593) in addition to providing this coercive apparatus with valuable hands-on experience – bolstered by participation in peacekeeping missions abroad – which has further increased its capabilities (Fisher and Anderson, 2016: 67).

The regime has also constructed a multilevel administrative apparatus that extends throughout Uganda's territory, connects the Museveni-led national political elite with local subordinate elites responsible for the system's operation, and helps the president and his inner circle to exercise power (Czuba, 2019; Tapscott, 2021: 54–55). As a result, “the possibility of state presence is felt across the country” (Tapscott, 2021: 136). Capable “pockets of effectiveness” within the national government support this administrative structure and effectively perform crucial governance functions such as the administration of petroleum extraction (Golooba-Mutebi and Hickey, 2016: 607).

In contrast, the Ugandan state's ability to collect taxes and deliver public services, necessary to realise the ambitious development agenda that occupies a prominent position in official rhetoric (Hickey, 2013), remains low (Andrews and Bategeka, 2013: 14–22; Tapscott, 2021: 136) despite decades of generous, and assiduously pursued, donor support (Mwenda, 2006). In these conditions, IGOs/NGOs have assumed responsibility for the – also donor-funded – provision of important services, especially in the agriculture, education, and water, sanitation, and hygiene sectors (UBOS, 2022b: 246). Donors have often preferred to direct funding to IGOs/NGOs due to well-documented misappropriation, which appears to have become increasingly common over time (Golooba-Mutebi and Hickey, 2016: 606), of public resources by the NRM elite, including members of the Museveni family (Vokes and Wilkins, 2016: 588–589).

The Ugandan State

As “[a] form of organised domination that delivers order and public goods,” the Ugandan state ought to “rise above the variety of private interests […] centralising the use of coercion and extracting resources that can then be used to provide public goods” (Centeno et al., 2017: 2). President Museveni has successfully delivered order and centralised the use coercion, but his creation primarily extracts resources to serve the private interests of the NRM elite instead of providing public goods. This combination makes for uneven performance of government functions; in this section of the article, I show that it also adds up to a coherent strategy that helps to maximise state power and advance elite and regime interests while limiting effort and expenditure that do not support these objectives.

There are “two quite different senses in which states and their elites might be considered powerful” (Mann, 1984: 188): despotic and infrastructural. Given the NRM regime's effective insulation from societal pressure, the despotic power – “the range of actions which the elite is empowered to undertake without routine, institutionalised negotiation with social groups” (Ibid.: 188) – of the state it controls is largely unconstrained (Hanson and Sigman, 2021: 1). The regime can arbitrarily “reinforce its interventions with overwhelming and unaccountable violence” (Tapscott, 2021: 3), using “unpredictable assertions and denials of authority” (Ibid: 4) to express its dominance, without necessarily exerting direct control. Such control requires infrastructural power, “the capacity of the state to actually penetrate civil society,” and project government authority across space (Mann, 1984: 189). Frequent scholarly emphasis on their oppositional relationship notwithstanding, the two types of state power are to a large extent complementary (Ibid.; Hanson and Sigman, 2021; Soifer and vom Hau, 2008). If despotic power gives Alice in Wonderland's Red Queen the “ability to order one's head to be cut off,” infrastructural power “captures the ability of the Red Queen to hunt down Alice and enforce her decapitation” (Soifer and vom Hau, 2008: 223). To penetrate society and hunt down adversaries, states need capable government bodies that can operate throughout their territories. The NRM regime has built such bodies. In this respect, the infrastructural of the Ugandan state is substantial. However, the regime has only invested in those infrastructural capabilities that help to extend its despotic power, assure its survival, and benefit the NRM elite.

This prioritisation of regime and elite interests accounts for the Ugandan state's uneven capacity. State capacity refers to governments’ ability to perform multiple core functions: coercion and protection, administration, goods provision, and extraction. Scholars have increasingly recognised the need to disaggregate the concept and identified multiple dimensions of state capacity, often linked to states’ coercive, administrative, and extractive functions (Centeno et al., 2017; Hanson and Sigman, 2021). Museveni's differential investments in state capacity show that it can vary across these dimensions.

The president has built a capable coercive apparatus he can readily deploy to contain security challenges. The administrative capacity of the Ugandan state is, however, only well developed to the extent to which it serves regime and elite interests. The administrative apparatus effectively governs crucial sectors of the economy and penetrates society. Its many agents keep tabs on the population to detect disaffection, work to address it, and penalise opposition activities, including by mobilising the coercive apparatus. In contrast, performance of administrative functions – notably provision of public services – that do not help to advance regime and elite interests is consistently weak. Although similarly low overall, the extractive capacity of the administrative organs of the government offers additional insights into regime priorities. While collection of revenue from Uganda's small tax base has attracted limited regime attention, the NRM elite has successfully used pockets of effectiveness and external support to extract resources needed for the performance of the government functions it values and delegate responsibility for those it does not to donors and IGOs/NGOs. The Ugandan state's infrastructural power is thus considerable insofar that it supports the exercise of despotic power, while infrastructural capabilities that do not serve the NRM regime and elite are poorly developed, with corresponding functions performed by other organisations.

The NRM elite has also used its control over the Ugandan state to engage in resource extraction for another purpose: to build private wealth. As long as NRM dignitaries only dip into money intended for government functions not essential for regime survival, misappropriation of donor funds and other resources is consistent with Museveni's narrowly focused state-building efforts. By bestowing access to wealth, the president can also reward elite loyalty and limit intra-regime disaffection with his hegemonic position. As such, personal enrichment encourages elite cohesion, protects presidential dominance, and facilitates regime survival in addition to helping Museveni's lieutenants improve their socioeconomic status.

The key aspects of Karamoja's post-disarmament political order – strong coercive capacity, uneven administrative capacity, delegation of key government functions to other organisations, and elite enrichment – are, therefore, present elsewhere in Uganda. Indeed, they define the country's political system. Scholarly attribution of this combination to simultaneous pursuit of multiple “contradictory strategies of political rule” (Golooba-Mutebi and Hickey, 2016: 605) highlights the internal tensions that have at times riven the system. For instance, in 2011 short-term electoral calculations drove Museveni to commandeer resources from the pockets of effectiveness (Ibid.: 610–613). Most of the time, however, as the foregoing discussion indicates, the individual “strategic modes” (Ibid.: 613) employed by the president complement one another. Accordingly, rather than distinct – much less incompatible – strategies, such modes constitute integral and interdependent components of a single coherent strategy responsible for the regime's long-term survival. ²

Methodological Approach

The intensity of Karamoja's incorporation into the contemporary Ugandan political system illuminates the strategy that the NRM elite has developed to advance and protect its interests. At the same time, situating developments in the region in the context of the regime's broader objectives helps to make sense of the abrupt decision to initiate disarmament in 2006 and the seemingly incongruous properties of Karamojan politics in subsequent years. The region is not different from the rest of Uganda because it is exceptional; rather, it features extreme values of the key characteristics of the country's political system. As such, post-disarmament Karamoja can be thought of as an extreme case in Seawright and Gerring's (2008) typology.

To glean the insights offered by this case, I draw on data from government, donor, and IGO/NGO documents, news media sources, and interviews. Official publications and reportage offer valuable, but limited, evidence concerning political developments in the region; they also shed little light on the rationale behind government operations. For this reason, I have conducted sixty-eight interviews with politicians, administrators, civil society leaders, and IGO/NGO staff uniquely well positioned to fill evidentiary lacunae and share otherwise inaccessible information due to their insider status, connections, and consequent extensive knowledge of politics in Karamoja and across Uganda. Most of the interviews that inform the article took place in Kampala and in Karamoja's Amudat, Moroto, and Nakapiripirit districts between September and December 2016; I subsequently complemented those in-person conversations with online interviews. I only disclose non-identifying respondent characteristics to conform with the data anonymisation requirements of the study's institutional review board approval. ³ To account for response bias, multiple interviewees and/or other sources have corroborated the findings I share in the next three sections of the article, which detail government activities corresponding to the coercive, administrative, and extractive state functions, respectively. Online supplementary material provides further information about these activities. In particular, Appendix 1 documents the frequency of media coverage of Karamoja, Appendix 2 reports the human rights violations that the UPDF committed during and following disarmament, and Appendix 3 details NRM elite involvement in exploitation of the region's natural resources.

Disarmament and the Coercive Apparatus of the Ugandan State in Karamoja

Karamoja's incorporation required substantial expansion of the presence and capacity of government bodies charged with projecting the NRM regime's power in the previously neglected periphery. Until the mid-2000s, conflict precluded investment in the administrative and extractive capabilities that support the realisation of elite objectives. According to a 2007 government report, “in Karamoja, the UPDF [did] not have the total monopoly over the means of coercion because of the widespread proliferation of small arms and ammunitions” (OPM, 2007: 4). The Chief of Defence Forces General Aronda Nyakairima declared:

We are facing up to Karamoja. That area has lagged behind in development because of the lawlessness there. We cannot tolerate this. […] Only forceful disarmament will secure the lawless area for the development projects. (Wasike, 2006)

Accordingly, incorporation began with the deployment of the Ugandan state's capable coercive apparatus.

In May 2006, President Museveni directed the UPDF to commence forcible “cordon and search” disarmament operations, which involved the formation of perimeters around settlements followed by hut-by-hut searches for weapons and detainment of their alleged owners in addition to skirmishes with warriors and aerial bombardment (Mkutu, 2008; Stites et al., 2007: 67–70). In contrast to previous initiatives – hindered, as the government acknowledged, by “inadequate planning, ineffective coordination, [and] insufficient funding” (OPM, 2007: viii) – the new disarmament drive was backed by strong political commitment and financial resources sufficient for extended deployment of thousands of soldiers and militia members (equipped with helicopter gunships, tanks, rockets, and rocket-propelled grenades) to search for guns; locate, confront, apprehend, and interrogate suspected raiders; man a large network of barracks established throughout Karamoja; and guard livestock corralled in enclosures adjacent to UPDF encampments (Stites et al., 2007: 67–70). At the height of the campaign, the number of security personnel stationed in the region might have reached 60,000. ⁴ In 2010, as disarmament was coming to an end, the UPDF sent 17,000 of the 40,000–45,000 troops in its ranks to Karamoja (IISS, 2010; Wanyama, 2010b). These numbers dwarf earlier military deployments in the region: 4,000 in 1999, 7,000 in 2001, 3,000 in 2002, and 6,000 in 2005 (Agence France-Presse, 1999; New Vision, 2001 and 2002; Isaur, 2005). Media mentions of military presence in Karamoja reached all-time highs in 2006, 2007, and 2010 (see Plot 2 in Appendix 1).

By 2010, the government had achieved most of its disarmament objectives. The UPDF's spokesperson wrote at the time that it had “carried out a successful disarmament programme” (Kulayigye, 2010). Brigadier Patrick Kankiriho, the commander of the Karamoja-based UPDF 3rd Division, predicted: “Very soon we shall declare Karamoja free from illegal guns. Then we can embark on integrated development and total pacification” (Wanyama, 2010a). At a 2011 event held to celebrate disarmament's success, General Nyakairima declared:

There is stability in the area. […] The army has contained road ambushes and killings with criminal successfully being prosecuted. […] The UPDF intervention in the disarmament is close to the end. The UPDF has successfully recovered over 30,000 guns (Kazungu, 2011).

Some of those weapons had been confiscated prior to disarmament. In particular, Karamojans had voluntarily surrendered as many as 10,000 guns in 2001 and 2002 (Kasasira, 2010) before the UPDF turned to coercive measures; it recovered a further 1,000 weapons from 2003 to 2005, 4,000 in 2006, 3,000 in 2007, and 2,000 in 2008 (Edyegu, 2008). Media coverage of gun confiscation in Karamoja increased sharply in 2006 before tapering off in subsequent years (Plot 4 in Appendix 1). By 2010, the UPDF estimated “the number of illegal guns in the hands of [Karamojan] warriors at 700, a figure that shows substantial reduction in the illicit arms in Karamoja” (Kasasira, 2010). Amid “total pacification,” the insecurity that had long limited the Ugandan state's ability to project power in the region came to a halt. Media reports of violence and raiding in Karamoja peaked in 2006 and declined steadily thereafter (Plots 5 and 6 in Appendix 1).

Effectively under military occupation, post-disarmament Karamoja was firmly in government hands (Kazungu, 2011; Wanyama, 2010a). ⁵ A member of parliament (MP) describes the region (in the 2010s) as a “place of security. Soldiers patrol everywhere. If you raid, you will be punished. You cannot do anything without the government knowing. The government is now in charge.” ⁶ In subsequent years, the UPDF ensured stability through enforcement of the Moruitit and Nabilatuk resolutions – which specify penalties for raiding and hold communities accountable for their payment (Stites and Howe, 2019: 150–151) – as well as targeted actions against individuals able to procure guns. ⁷ The police, which at the time of disarmament the government described as “skeletal” (OPM, 2007: 54), also grew. By 2021, over 1,000 police officers worked out of 61 new stations and posts in the region (OPM, 2021: 6; also see Plot 3 in Appendix 1). In the words of an influential Karamojan politician, “now you see the policeman walking with the local people, the soldier walking with the warrior.” ⁸ From the government's perspective, this display of power, infrastructural as much as despotic, constituted an unqualified success. Both for Karamojans and for the Ugandan state, its cost was, however, very high.

In the course of disarmament, the UPDF routinely committed serious human rights violations. Since to my knowledge no effort has been made to calculate the total human toll of these abuses, in Appendix 2 I present a list of incidents recorded in the scholarly literature, IGO/NGO reports, and the media. Although fragmentary, the data permit a cautious estimation of the number of victims of human rights abuses committed by Ugandan military personnel in Karamoja: at the very least around 1,000 – and conceivably several thousand – deaths and many thousands of cases of beating, injury, sexual violence, torture, displacement, and extrajudicial detention can be attributed to the UPDF. Media coverage of human rights violations in the region increased precipitously in 2006 and peaked in 2007 (Plot 7 in Appendix 1). Disarmament also contributed to the erosion of pastoralism: confiscation of guns and creation of kraals restricted mobility, necessary for pastoralist livelihoods (Stites and Akabwai, 2010: 32), and many Karamojans lost their livestock to not only armed neighbours but also UPDF personnel. ⁹

The cost of disarmament was also high for the Ugandan state. First, local resistance to the brutality of UPDF operations led to deaths of military personnel. According to the UPDF, which has likely undercounted the casualties, 269 soldiers were killed and 289 injured in Karamoja between 2001 and 2011 (Kolyangha, 2011). Second, although the government has never disclosed disarmament's financial cost, military operations in Karamoja consumed a significant proportion of its security budget, which in the mid-2000s constituted approximately 10 per cent of Uganda's overall budget (Kuteesa et al., 2006: 11). Furthermore, and third, the human rights violations the UPDF committed in 2006 and 2007 attracted some scrutiny from the international community that the regime has otherwise usually diligently courted (HRW, 2007; OHCHR, 2006).

Coming after decades of apparent disinterest in Karamoja, the government's determination to effect disarmament despite its high cost and the possibility of negative repercussions represented an abrupt policy reversal. This shift provides the first glimpses of the region's place in the broader NRM strategy, a civil society leader suggests:

The government had intentions of making sure it's in control of [Uganda's] boundaries. As a politician Museveni was challenged that there are parts of the country where he was not in control. […] And even before NRM came to power, they knew there were resources here that can contribute to the central plate. That is one other reason, a reason of greed. ¹⁰

The regime signalled its commitment to establishing control over Uganda's entire territory as early as in the 1980s. To this end, it made several attempts to disarm Karamojans but initially lacked requisite coercive capacity. “When Museveni took over in 1986,” a former MP observes, “he wanted to use the same rudimentary approach as [his predecessors] Obote and Amin, and he had it rough. A lot of soldiers were killed.” ¹¹ The Ugandan state's despotic power was already considerable at that time, but the regime lacked the infrastructural capabilities necessary to exert control over the region; its impressive coercive apparatus was only developed gradually in the course of decades-long struggle to contain serious security challenges, including multiple rebellions and military interventions in the Democratic Republic of Congo (DRC). Even after the coercive apparatus's capacity had increased, the NRM government had to, therefore, consider its priorities:

The question was: which war do you fight? Karamoja wasn’t a war. […] Karamoja guns weren’t such a big issue for the government, but once guns began to fall quiet elsewhere in Uganda, the government decided to make Karamoja part of Uganda. ¹²

It was only after withdrawal from the DRC and gradual winding-down of the deadly Lord's Resistance Army insurgency in northern Uganda that the regime could reallocate resources and complete the task that it had set itself two decades before. By 2007, a Ugandan intelligence officer noted, “apart from Karamoja and Garamba [in the DRC], there [were] no foreseeable internal and external threats against Uganda” (Allio, 2007). In contrast, insecurity in Karamoja challenged “the ability of the UPDF to fulfil its constitutional mandate of securing the international borders” (OPM, 2007: 4). “To consolidate the authority of the state in the region,” the government's stated goal, “these challenges [had] to be systematically addressed” (OPM, 2021: 14).

According to several interviewees, the political elite's concern with personal enrichment through exploitation of Karamoja's natural resources was equally important. ¹³ I reproduce two especially insightful passages from my conversations with two of them, a senior LC5 official and a civil society leader, below:

My perspective is that the NRM spent so much money on disarmament because … you pacify the area, then you get access to resources. ¹⁴

I have uncovered no direct evidence of such an avaricious motivation for Karamoja's incorporation, but post-disarmament elite involvement in mining and acquisition of land in the region offers indirect support for this claim.

The feasibility of resource extraction in Karamoja depended on continuous exercise of regime power. A civil society leader explains post-disarmament security presence in the region thusly: “They are protecting the very many things they own here. They are protecting the investments they brought here. The reason is not insecurity, but to protect what they came here for in the first place.” ¹⁶ For much of the 2010s, local bodies of the Ugandan state's coercive apparatus often relied on mere threat of decisive punishment of non-compliance, taking advantage of the fresh memories of disarmament-era brutality. ¹⁷ Over time, however, application of force has become more frequent, especially after conflict – attributable to poverty, hunger, and Karamojan disaffection with “malevolent actions on the part of the security forces, such as the indiscriminate impounding of animals” (Arasio and Stites, 2022: 33) – returned to the region in 2020. Following several years of decline, the frequency of media reports of human rights violations in Karamoja began to increase in 2017 (Plot 7 in Appendix 1). In the course of one year in 2021 and 2022, armed forces killed 465 and arrested 11,264 Karamojans accused of causing unrest (The Independent, 2022).

Government Administration and Service Provision in Karamoja

The well-developed coercive capacity of the Ugandan state allows the regime to swiftly and decisively respond to Karamojan discontent. Although ready to use force when necessary, the NRM elite nonetheless recognises the benefits of subtler methods of control. For this reason, the regime has devoted considerable attention to the construction of a capable administrative apparatus that helps it to penetrate local society, anticipate and contain disaffection, preclude effective popular resistance, and facilitate natural resource extraction as well as security operations.

The Local Council (LC) system, in place across Uganda, is the central component of this administrative apparatus, which, much like the UPDF and other security agencies, effectively exercises aspects of infrastructural power complementary to despotic power. Prior to disarmament, LCs – established in the 1980s as “resistance councils” and given their current name in 1997 – only had “a phantomal presence although limited authority” in Karamoja, especially at the LC1 (village) level (Mirzeler and Young, 2000: 424). Although in some areas the system remained “effectively defunct” until the early 2010s (Hopwood et al., 2018: S148), many LC1 chairpersons’ importance increased during disarmament, when the UPDF required that they help to coordinate cordon and search operations in their settlements and issue documents needed to sell livestock. ¹⁸ Elected by members of their communities but accountable to the government, which obligated them to settle communal conflict, minor crimes, and misdemeanours and to report more serious discord and offenses to the military and police, in subsequent years, the 3,005 Karamojan LC1s (EC, 2020b) – by the mid-2010s operational throughout the region – emerged as key local representatives of the Ugandan state. ¹⁹ “You have soldiers in the barracks, yes, but the LC is in every village. So, the government is in every village. And the LC can report to the government,” a Moroto elder explains. ²⁰ The also elected higher-ranking LC3 and LC5 officials perform analogous functions in Karamoja's 113 sub-counties and nine districts (up from two in 2001: Abim, Amudat, Kaabong, Karenga, Kotido, Moroto, Nabilatuk, Nakapiripirit, and Napak). ²¹ Each LC comprises a chairperson and a variable number of councillors; in the 2021 elections, Karamojan voters filled 2,223 LC positions (EC, 2020a). (This number does not include LC1s, elected separately in 2018.) The LC5 chairpersons are also supported by district executive committees (for example, the one in Moroto District has three members) and extensive district bureaucracies run by central government-appointed chief administrative officers (CAOs). ²² Since the days of the no-party system, NRM members have held nearly all elected offices. ²³ A parallel hierarchy of unelected administrators – 482 parish chiefs and 113 sub-county chiefs (the latter assisted by the same number of community development officers [CDOs]) as well as nine presidentially-appointed resident district commissioners (RDCs) and their staff – works alongside the LCs. ²⁴ Overall, “the system of governance is almost covering everybody,” notes a senior LC5 official. ²⁵

The scope of governance functions performed by this system matches its reach. Local officials, both elected and unelected, resolve disputes, penalise transgressions, promote and help to coordinate the implementation of government policies, work to ensure popular compliance, notify superiors of signs of disaffection, and report governance challenges beyond their purview for resolution by other agencies. ²⁶ A parish chief describes her role thusly:

My work is to mobilise people for meetings. As they come to meetings, I open the meetings, and welcome officials. NUSAF [Northern Uganda Social Action Fund] 3 was piloted in my parish. I mobilised people to learn about it. […] I’m the secretary for the LC3 council. […] I also write monthly and quarterly reports. I submit them to the sub-county chief, and then she reads them and forwards them to the RDC. […] When an LC1 reports a major dispute or problem, he brings it to me. Then the case goes to the CDO, then to the sub-county chief, then to the police. ²⁷

The senior LC5 official adds: “Should any problem happen in the village, it is up to the LC1, and then LC3, and LC5, and parliamentarians, then to the president. The chain goes up and comes down.” ²⁸ In addition to transmitting decisions for implementation at the lower levels of the government hierarchy, this chain of command allows local officials to convey the wishes and concerns of the members of their communities to superiors. In this respect, they serve as useful intermediaries between the local population and the state. Karamojan officials’ autonomy and influence within the government is, however, limited. A civil society leader explains:

Power is centred. The president appoints the RDC, the CAO. The president makes sure the MP or LC5 wins the election. What the government has done is that you and I are locals of this place, but we’re spying on each other. Officials have to make sure government programmes succeed. They tell you: this programme is great, it will make us developed. But they will also have you arrested if you complain. ²⁹

“Powerless in the face of the military presence” (Carlson et al., 2012: 13), local government functionaries have few incentives to complain. Instead, they assist the dedicated Ministry for Karamoja Affairs, between 2009 and 2016 headed by the president's powerful wife Janet Kataha Museveni, in the implementation of key policies it has pursued since disarmament: promotion of agricultural production and sedentarisation and infrastructure development. ³⁰

Since, due to low and highly variable rainfall, most of the region cannot support reliable and sustainable agriculture, government efforts to transform Karamojan pastoralists into settled agriculturalists have had no impact on crop yields or food security. However, they have contributed to the erosion of indigenous livelihood practices; by the late 2010s, only a minority of households in the region owned enough livestock to permit viable pastoral production (Catley et al., 2021; Nakalembe et al., 2017: 2). Although detrimental to Karamojans’ wellbeing, decline of pastoralism helps to secure regime and elite interests in the region. ³¹ First, as long as Karamojans own livestock, they need to protect themselves from raids and restock following the frequent shocks inherent to the region's ecosystem. Availability of weapons acquired for this purpose, responsible for the high incidence of raiding before disarmament (Mirzeler and Young, 2000: 408; Mkutu, 2008: 100), threatens stability and the Ugandan state's hard-earned ability to project power in Karamoja. The heavy-handed response to the post-2020 resurgence of conflict speaks to the government's concern about such challenge to its authority. Karamojans could also turn their guns against agents of the state, as they did during successive disarmament campaigns. Second, economic autonomy from the Ugandan state made possible by successful continuation of pastoralism might, similarly, help Karamojans challenge elite appropriation of their region's natural resources and improve their bargaining position vis-à-vis the NRM regime more broadly. Conversely, the local population's poverty and weakness facilitates maintenance of government control over Karamoja and elite resource extraction.

Road construction and extension of Uganda's electric grid to the region serve a similar function. A tarmac road that connects Moroto, Karamoja's largest town, with the neighbouring Teso region was completed in 2020 at the cost of 646 billion Ugandan shillings (UGX; approximately USD 175 million; Wanyama, 2020). The Moroto–Nakapiripirit highway, tarmacked for UGX 184 (USD 50 million), is also being extended outside the region for another UGX 478 billion (USD 130 million; Daily Monitor, 2018; Nabwiiso, 2013). Other transport corridors, including roads from Kaabong through Kotido to Moroto and from Kosiroi through Amudat to Nakapiripirit have been upgraded. The government has also connected Moroto, Matany, Nakapiripirit, and a few smaller trading centres as well as some government institutions and mining operations to the power grid (New Vision, 2013). Electricity supply is important for mining operations and the comfort of their workers, businesspeople, and government officials. ³² The new roads have meanwhile facilitated troop movements, helping the government contain security threats, and improved access to major natural resource extraction areas. In the words of a Karamojan MP, “some of those big people in the government have interest in the region. They know that if they don’t invest in good infrastructure they can’t tap into this wealth. This is the key reason the government is investing.” ³³ The Uganda National Roads Authority has also indirectly attributed road construction to resource extraction: “now that Karamoja is home to marble and limestone, the road [to Teso] has been upgraded to […] bear heavy load” (Wanyama, 2020). Security considerations are no less important: after conflict returned to the region in the early 2020s, “the government commenced the construction of security roads in Karamoja to ease the mobility of the security forces” (Eyoku, 2023). These costly investments enhance government control and facilitate resource access but offer no comparable benefits to most Karamojans, few of whom can afford to travel or consume goods imported from outside the region; the power grid only reaches 1 per cent of Karamoja's population (UBOS, 2021).

More generally, except for the welcome, but temporary, reduction of insecurity after disarmament, improvements in local people's living conditions have been at best negligible. The Ugandan newspaper The Observer has offered a pithy summary of Janet Museveni's accomplishments as the Minister for Karamoja Affairs:

A close look at key areas in Karamoja will tell you that it is only the occupants of her 15 or so vehicle motorcade, politicians, and the not-so-analytical that will say Karamoja is different from then, and even so, as a direct result of her work. A visit at any time to Naro Apotiyaro village in the backyard of the plush State [L]odge in Morulinga, Napak [D]istrict, will shock you with signs of death from hunger and poverty. (Longoli, 2012)

Statistical data corroborate The Observer contributor's suspicion that despite the pro-development rhetoric poverty alleviation and service provision – the aspects of infrastructural power that do not support the exercise of despotic power – are not a priority for the NRM regime. According to the 2019/2020 national household survey, at the time of data collection 75 per cent ³⁴ of the region's inhabitants were food insecure (the average across all Ugandan districts was 39 per cent); 65 per cent ³⁵ of Karamojans had incomes below the poverty line (45 per cent more than elsewhere in the country); the adult literacy rate stood at 25 per cent ³⁶ (just over a third of the Ugandan average); and only 42 per cent ³⁷ of children were enrolled in (very poor-quality) primary schools (compared to 80 per cent across Uganda) and 12 per cent ³⁸ in secondary schools (the national figure was 27 per cent; UBOS, 2014, 2018, and 2021). Not only is Karamoja the poorest part of Uganda, but the absolute level of its inhabitants’ poverty has barely changed even as conditions elsewhere in the country have improved markedly (UBOS, 2022a). Devastating effects of the COVID-19 pandemic have only served to compound the region's plight; in 2022, half a million Karamojans, one-third of the population, faced critical food insecurity (Cullis and Arasio, 2022).

Furthermore, while the government provides some public services in most of Uganda (UBOS, 2022b), those available to the region's inhabitants are overwhelmingly funded by donors and administered by IGOs/NGOs (Daily Monitor, 2012a; Irish Aid, 2016). ³⁹ “Karamoja was handed over to the NGOs,” argues a former MP. ⁴⁰ In 2022, donors committed USD 61 million for projects, of which 82 per cent were implemented by IGOs/NGOs, in Karamoja. That same year, the Ugandan government allocated UGX 139 (USD 37 million) for the region's administration, with only UGX 30 billion (USD 8 million) earmarked for development activities – some of them donor-funded – including the promotion of crop cultivation and sedentarisation (KRSU, 2022). The effectiveness of these activities is dubious at best, notes a civil society leader:

There are very many projects to uplift the people. But most of these programmes don’t reach the intended beneficiaries. We’ve had NUSAF 1, 2, and 3, all of them targeted at uplifting the living standards of the people, but the life of the people has not improved. People are never consulted on programmes, on what their needs and priorities are. Programmes are just imposed. ⁴¹

Between repression, suppression of indigenous livelihoods, and poor public service provision, most Karamojans have derived few benefits from the region's incorporation. In contrast, the Ugandan political elite has profited handsomely.

Elite Extraction of Karamoja's Resources

Karamoja offers the NRM elite outstanding opportunities for personal enrichment. Elite Ugandans’ pursuit of wealth in the region has taken two forms: direct misappropriation of funds intended for public goods provision in Karamoja and exploitation of natural resources.

Theft from public purse has only been documented on a few occasions. In 2012, government officials stole as much as UGX 150 billion (USD 40 million) from the donor-funded Peace, Recovery, and Development Programme (PRDP), which operated in Karamoja and northern Uganda (Daily Monitor, 2012a). A decade later, the vice president, prime minister, two deputy prime ministers, and multiple ministers allegedly pilfered thousands of iron sheets earmarked for low-cost housing in Karamoja (Kazibwe, 2023). However, such brazen misappropriation of public resources has attracted considerable backlash: several donors paused aid provision following the PRDP scandal, while the theft of the iron sheets, which Museveni labelled a “political mistake,” caused such public outcry that the president found it necessary to discipline some of the perpetrators, including the two serving Karamoja Affairs ministers Mary Goretti Kitutu and Agnes Nandutu, who were arrested for their role in the scheme. Given these risks, elite Ugandans appear to have generally preferred to build wealth through extraction of the region's natural resources.

Although exploration remains in early stages, Karamoja contains substantial deposits of gold, silver, copper, iron, titanium, manganese, niobium, tantalite, chrome, rare earth and radioactive minerals, precious and semi-precious stones, and limestone and marble (Hinton et al., 2011). The region is also an excellent setting for the creation of private livestock ranches and, in the western “green belt,” agricultural production. ⁴²

To harness these opportunities, members of the NRM elite have acquired exploration and mining licenses that cover most of Karamoja as well as titles to particularly valuable tracts of land. ⁴³ The weak legal protections for communal land owners and government decision to de-gazette 54 per cent of the region, previously protected as wildlife conservation areas, have facilitated resource extraction, which expanded rapidly following disarmament (Rugadya et al., 2010: 2). ⁴⁴ In 1996, only thirteen companies held mining or exploration licenses in Karamoja. By 2010, thirty-eight licenses covering 6,897 square kilometres (a quarter of the region) had been granted (Ibid.: 19–20). Over 140 mining companies operated in Karamoja by 2015 (Ariong, 2015). Although precise figures are not available, large-scale land acquisition appears to have followed a similar trajectory. ⁴⁵

My interviewees believe that most major investors in mining and land in the region belong to the political elite: they are members of the Museveni family, cabinet ministers, other senior government officials, and well-connected businesspeople. ⁴⁶ I briefly discuss a number of such cases below. In Appendix 3, I provide further details and outline the strength of evidence relating to each case.

The Museveni family's involvement in resource extraction in Karamoja goes back to the 1990s, when the president's brother General Caleb Akandwanaho, known as Salim Saleh, established a gold mine at Lopedo in (what is now) Kaabong District (Africa Analysis, 2002; Abadaka, 2009). The operation continues to be known as “the first family's mine.” ⁴⁷ Since around 2014, General Muhoozi Kainerugaba, Museveni's son, has allegedly operated another gold mine on Mount Moroto. ⁴⁸ In 2012, two companies apparently controlled by Salim Saleh's wife Jovia Akandwanaho acquired land titles to 6,130 hectares of land at Kamacharin and 2,001 hectares at Kakomongole in Nakapiripirit District (Daily Monitor, 2012b). ⁴⁹ Unidentified members of the Museveni family have also supposedly acquired land in Karenga in Kaabong District. ⁵⁰ In addition, an NGO worker notes that “there were a lot of rumours about Janet [Museveni]'s corruption, but being in power means you insulate yourself from accountability, you are in control of things.” ⁵¹

Other members of the NRM elite have also benefitted from exploitation of Karamoja's natural resources. Tororo Cement – co-owned by Sam Kutesa, the long-time Minister of Foreign Affairs (and Muhoozi Kainerugaba's father-in-law) – sources most of the limestone it needs to manufacture its eponymous product from Mount Moroto. ⁵² In addition, Jan Mangal, a company co-owned by the State Minister for Housing Sam Engola, operated a gold mine at Nakiloro and Nakibat in Moroto District from 2012 to 2015 or 2016 (HRW, 2014: 55–60). ⁵³

Conclusion

Elite pursuit of wealth through appropriation of Karamoja's resources is a function of the power of the NRM regime. It also simultaneously contributes to the regime's stability and survival by limiting intra-elite tensions. These twin objectives – personal enrichment and regime dominance – have informed the development of a coherent strategy responsible for the seemingly contradictory characteristics of political order across Uganda, but put into unusually stark relief by the intensity of Karamoja's incorporation. The NRM elite has consistently invested in the capacity of state bodies that protect its material and political interests, neglected government functions aimed at improving societal wellbeing, delegated responsibility for the performance of those functions to other organisations, and accumulated personal wealth.

This strategy's crystallisation in the wake of the 2006 disarmament campaign helps to make sense of the decision to extend state control over Karamoja, subsequent government actions in the region, and the organisation of the political system the regime has built in the country. It also shows that elites that control entrenched authoritarian regimes such as Uganda's can strategically exercise their despotic and infrastructural power and prioritise investments in the capacity of some state apparatuses. Scholarly recognition of the importance of disaggregating the concepts of state power and state capacity notwithstanding, empirical evidence of differential investments in state capacity and their relationship to types of state power remains limited. In Uganda, the NRM regime has only built capable government bodies where extension of infrastructural power has supported its exercise of despotic power. Examination of the drivers of uneven performance of government functions in other settings – democratic and hybrid as well as authoritarian – presents an intriguing opportunity for future research. Such research may benefit from consideration of extreme cases such as Karamoja and the inferential advantages they offer. Conversely, while the region's rapid political transformation distinguishes it from the rest of Uganda, its incorporation belies African states’ historical reluctance to project authority away from centres of political power, suggesting the possibility, which also merits further research, of analogous processes in other long-neglected political peripheries, including across the borderlands of East Africa and the Horn.

Supplemental Material