In search of chiefly authority in ‘post-aid’ Acholiland: transformations of customary authorities in northern Uganda

ABSTRACT This paper investigates the complex relation between protracted donor interventions and the production of customary authority. More specifically, the paper analyses the impact of post-conflict donor interventions (and their withdrawal) on the position of customary chiefs in the Acholi region in northern Uganda. As important brokers between international aid agencies, the Ugandan government and Acholi communities, customary chiefs became key actors in post-conflict peacebuilding programmes. Using the concepts of extraversion and development brokerage, the paper demonstrates how dwindling access to external donor funds has strongly affected Acholi customary authority. To secure their authority and legitimacy, customary chiefs re-shifted from an ‘outward’ to an ‘inward’ orientation, a process that we call ‘introversion’

In the wake of the conflict pitting the Ugandan government against the rebels of the Lord Resistance Army (LRA), northern Uganda has been the operational ground for a large humanitarian and post-conflict reconstruction 'industry'. 1 This industry consisted of a diverse set of actors such as international advocacy groups, non-governmental organisations, donor agencies and church-based charity organisations. These external actors occupied a central place in local politics, economies and society in northern Uganda and engaged in diverse fields of humanitarian assistance, development, transitional justice and post-conflict reconstruction. Although some research has investigated the transformative impact of this international aid presence on local patterns of governance 2 , a thorough understanding of its impact on the production of public authority remains largely absent.
One entry point to look at this impact is investigating the changing position of customary chiefs 3 in northern Uganda, more particularly the Acholi region. As important brokers occupying a strategic position in northern Uganda's post-conflict society, situated between international donors, the Ugandan state and 'local communities' 4 , their authority has been strongly affected by external intervention dynamics. After their authority had been undermined from the war and devastating effects of protracted forced displacement 5 , chiefs saw their position partly restored during the immediate post-conflict period. Both international donors and the Ugandan state recognised customary chiefs (locally called rwodi) as key actors in post-conflict transitional justice as well as in conflict mediation and resolution. 6 This recognition strongly increased their power and public authority, and largely shifted their base of legitimacy from the community to external actors.
We will use the concepts of 'extraversion' and 'development brokerage' to analyse customary chiefs' strategies to position and reposition themselves in the changing aid-landscapes of northern Uganda. Bayart's notion of 'extraversion' (referring to the deliberate agency of African actors in creating and maintaining dependency on external actors) 7 enables us to analyse rwodi's attempts of 'capturing aid'. The concept of 'development brokers' referring to a new social category of intermediaries between 'local communities' and aid agencies 8 , offers a useful analytical lens to investigate the ambivalent position of rwodi in today's 'post-post conflict' context of aid withdrawal. This context is characterised by a significant pull-out of international aid agencies, mostly those with a focus on humanitarian intervention. Our paper demonstrates how the current phase of aid withdrawal forces rwodi to reposition themselves to re-establish their authority.
We argue that the decrease in donor funding reduced rwodi's ability to 'capture' external aid, which eventually affected the ways they sought legitimacy and tried to establish their authority. When analysing the production of customary authority and legitimacy, we mainly do so from the perspective of the chiefs themselves, focusing on the collective as well as individual mechanisms they deploy to seek legitimacy and secure their authority. The change in mechanisms we observe is strongly informed by a shift from 'institutionalisation' towards '(re-)-individualisation', and by a shift from an external towards an internal orientation. These shifts led to a redefinition of brokerage positions, and chiefs' connections to the state, communities and the remaining donors. Customary chiefs who formerly derived most of their public authority from an explicit connection with international donors now deliberately reinvest in internal sources of legitimacy with their local 'communities'. In reference to Jean-François Bayart's notion of 'extraversion', 9 we will identify this process as a mechanism of 'introversion'.
The paper is based 6 months of fieldwork in Acholiland between 2015 and 2018. Interviews were held with rwodi from 13 chiefdoms with different relationships with KKA (Ker Kwaro Acholi, the Acholi cultural institution) and different levels of (past) donor engagement. While some of them belonged to the royal lineage, others did not. Additional respondents (for interviews and focus group discussions) were staff members from international organisations, clan leaders, community elders, local NGO staff, KKA representatives, religious leaders and government officials. (Participatory) observations took place of chiefly activities and practices (mediation meetings, cultural festivals, ordinary afternoons at the chiefdom palace or home). 10 The main researcher held a position between 'insider' and 'outsider'; being Ugandan but not from the region. She therefore had limited access to Acholi customary systems and norms, rendering the collaboration of Acholi researchers and translators indispensable.

Customary authority, brokerage and extraversion in changing political orders
The history of Acholi rwodi throughout the conflict and post-conflict periods in northern Uganda illustrates chiefs' continuous search for legitimacy, navigating 'changing political orders'. 11 By this search, customary authority 12 is produced, reproduced, constructed and deconstructed; depending on both structural factors and their individual agency. 13 As such, customary chiefs 'constantly move in and out of the capacity to exercise public authority'. 14 Customary leaders' legitimacy and authority is derived from their particular position between the state, local communities and external actors. The literature on the customary chief as intermediary is extensive, with a particular focus on the role of chiefs in indirect colonial rule. 15 As Hoffmann et al. also argue in this issue, from this historical intermediary position, customary chiefs have been able to access and appropriate different forms of political, economic and moral power and authority. 16 For the (colonial and post-colonial) state administration, Acholi chiefs acted and continue to act as key figures to connect to clan and lineages elders, as such integrating them into their (neo-) patrimonial mechanisms of power and control. 17 For their communities, historically, Acholi chiefs gained moral and financial support by acting as regulators and mediators of social life and by drawing on a set of cultural norms and values on morality, identity and the maintenance of tradition. 18 For international donors, as we will demonstrate, they acted as key figures, believed to provide their interventions with local and international legitimacy by 'involving the grassroots or local communities'. 19 One particular strand within the literature on customary authority focuses on customary chiefs as 'development brokers', a social category of intermediaries between donors and potential beneficiaries of aid. 20 It documents how in some parts of Africa, aid has become an important resource for chiefs, and capturing aid an important characteristic of customary authority and legitimacy. The analytical lens of development brokerage enables us to better understand the opportunities for social, political and economic mobility, power and authority that lie within this particular position. Development brokers' authority is derived from the 'inside' as well as the 'outside'. This tension between chiefs' 'upward' vs 'downward' legitimacy and accountability has been discussed extensively. 21 The post-conflict authority and legitimacy of Acholi customary chiefs was very much 'upward' legitimacy, depending on their close connection to external donors through peacebuilding and post-conflict reconstruction programmes. This strong dependency on external resources is revealed by chiefs' current repositioning as prompted by the withdrawal of donors in northern Uganda.
Where the concept of the 'development broker' underlines the crucial agency of customary chiefs in 'capturing aid', the notion of 'extraversion' goes even a step further. Africanist scholars have used 'extraversion' to describe the process by which 'groups or individuals employ their dependent relationship with the external world to appropriate resources and authority'. 22 This is a historical process, by which customary chiefs and other types of 'elites' accustomed themselves to making their dependence on the colonial metropoles and donors both productive and advantageous. 23 The notion of extraversion thus underlines both the historical component of development brokerage and the power to 'mobilise, appropriate and redirect foreign resources and agendas' and 'to render their own subjection and dependence into a deliberate strategy or mode of action'. 24 In the construction of legitimate customary authority, the provision of services and chiefs' position in patronage networks is of equal importance. Or in other words, customary authority, like other forms of public authority, is built on practices/performances as well as resources. 25 These resources include both symbolic resources or what we call 'repertoires' in this paper, and material resources. In the context of northern Uganda, public goods provision has strongly been informed by dynamics of civil war and its aftermath. In the post-conflict setting, influenced by the humanitarian industry, public goods were very much framed in repertoires or registers of 'healing', 'reconciliation' and 'peace'. They took the form of the provision of (transitional) justice, protection, reintegration, mediating land conflicts, and restoring moral order. 26 On the other hand, in a context of neo-patrimonial rule, an important source for legitimacy is to be found in patronage networks. As development brokers, chiefs occupy a key position in these patronage networks, in which they are able to 'capture aid' to be channelled again through these personal networks, in ways that grant them prestige and legitimacy. Aid-withdrawal has presented a shift in northern Uganda's political economy, contributing to a shift in what Comaroff and Comaroff call the political geography of customary power. 27 The economy of donordriven post-conflict interventions has passed its peak and humanitarian and peacebuilding donors are increasingly disinterested in the region. In this new setting, chiefs are not only confronted with fading symbolic registers (of reconciliation, peace and traditional justice) from which they draw their authority, but also with fading material resources connected to these registers.
As an effect of this changing 'regime', we observe how customary chiefs now increasingly have to reach out again to their 'base', from which they had largely become alienated. We will call this process a strategy of 'introversion'. These observations corroborate Kapidžić's assertion that the process of legitimation of (customary and other) public authority in northern Uganda remains 'inherently local', and connected to locally informed and accepted practices, behaviours and norms. 28 Acholi chieftaincy and its historical trajectories Customary authority in Acholiland has been created and recreated through a long historical process shaped by internal and external dynamics. Drawing on registers and symbols of tradition, culture and morality, rwodi have continued to play an important role in Acholi society, even during phases of weakening, fragmentation, violent oppression or abolishment. Chiefly authority has been shaped in interaction with Arab traders, colonial administrators, the Ugandan nation-state, armed rebellion and donor interventions. 29 Carving out their space and chiefly position strongly relied on chiefs' claim to the cultural and 'traditional' realma register that in itself has been subject to continuous processes of transformation.
The rwot was the most important political, economic and social personality in the political entity of 'chiefdoms', which were installed in Acholiland in the late seventeenth, early eighteenth century. 30 Chiefly authority was however not absolute, as power was shared with lineage elders (ladit kaka) and political authority was mainly vested in the clan heads. 31 This changed during the colonial period, when the process of institutionalisation of political chieftainship was used to effectively bring Acholiland under British control. Acholi chiefs were classified into those who 'resisted' and those who 'acquiesced' to colonial rule. The former would be punished and replaced by new (appointed) chiefs, while the latter would be rewarded (with for example privileges and guns). 32 As a result, their political powers were reoriented; while previously derived from patronage and ritual and symbolic power, now authority was derived from more coercive forms of power. 33 To distinguish the 'original', hereditary rwodi from the new British appointees, people would refer to the former as rwodi moo ('chief of the oil') in opposition to the latter, rwodi kalam ('chief of the pen'). 34 Rwodi kalam, who represented what Mamdani called 'manufactured' forms of chiefship 35 , were unable to rely on historical lineage connections or spiritual powers and to mobilise registers and symbolic resources of a customary past. As such, they never enjoyed the same legitimacy in the eyes of their subjects as the rwodi moo. Their legitimacy largely depended on the recognition and support of external actors, in particular the colonial administration. Until the present, the distinction between the two remains an important factor in debates on Acholi chiefs' historic legitimacy. Several chiefs we met during our study invested significant efforts in documenting their histories (in booklets) with chiefdom histories going back to the pre-colonial period, to 'prove' their claims of being an authentic rwodi moo.
Not long after independence, the position of customary chiefs and kings in Uganda was formally suspended. Yet, being deprived of explicit political authority did not mean rwodi were left powerless. According to Hopwood, customary law, for example in land governance, remained an important arena of customary authority. 36 Postcolonial dynamics of violent conflict, militarisation and forced displacement in northern Uganda 37 have also strongly impacted rwodi's position and authority. 38 The LRA conflict, which lasted for over two decades, left Acholiland with fragmented governance structures, institutional breakdown, increased poverty levels and detrimental physical and psychological effects on communities. The legitimate 'order of authority' was decimated during the war and worsened with the displacement of people into camps. 39 The profound social disruption resulting from the forced encampment of the Acholi population in crowded internally displaced peoples' (IDP) settlements has been described as a form of 'social torture'. 40 Many rwodi were displaced themselves or had left their clans to find refuge in Gulu town (the main urban centre of northern Uganda) or Kampala (the capital city of Uganda). 41 Life in the camps undermined the status and power of customary authorities, who suddenly shared the same (humanitarian) space with their 'subjects', standing in the same line for food distribution or other kinds of assistance. 42 Other authority structures entered the scene, such as camp administrators, local councillors (LCs, state administrators within the tiered system at the decentralised level, such as districts, villages and parishes) and military actors. This further weakened the space for customary chiefs to exercise power. 43 Also, the social problems caused by the war were just too great for the chiefs to handle. 44 At the core of this crisis was the fact that people had been cut off from their lands, and thus from their most important source of livelihood. This crisis and the disturbed relation between people and their land was never fully restored after the war ended, and has remained one of the most severe challenges in post-and post-postconflict Acholiland. 45 Confronted with the inability to access land, and its effects in the form of structural food insecurity and poverty, rwodi stood powerless. Much of their historic roles were embedded precisely in brokering people's access to land (as the most important gate-keepers), and ensuring food-and other aspects of security.
In the midst of their structurally impoverished and vulnerable communities, many rwodi were left disempowered, lacking the resources to facilitate their work and establish their authority. With their clan members depending on humanitarian assistance, chiefs could no longer receive gifts or food from their 'subjects'. 46 In case people needed to call upon customary services such as the performance of a ritual, the needed resources (such as a goat) could neither be provided by themselves nor by the rwot. 47 This way, throughout the violent conflict, 'Acholi elders and chiefs largely lost their power of social regulation, their role of mediating conflicts and much of their legitimacy'. 48 Interestingly, precisely this narrative of 'social disruption' and 'cultural breakdown' would later actively be mobilised by customary chiefs themselves as well as by international donors to stress the urgent need for rwodi's re-emergence in Acholiland to 'reconstruct' the war-torn society. 49 Donor intervention and the 'search for the local': the institutionalisation of Acholi customary authority In a report published in 1997 (in the midst of the LRA war) by Dennis Pain, an anthropologist contracted by the NGO International Alert, Acholi customary chiefs were identified as the ideal leading actors in transitional justice mechanisms. 50 A clear distinction between rwodi moo and rwodi kalam was made in the report, and the term 'credibility' was used several times to stress rwodi moo's legitimacy as not being 'compromised by government's influence'. 51 In this way, rwodi moo's marginalisation turned into an opportunity and potential strength, as they were considered to have 'become independent of the political and administrative processes'. 52 Finally, the report explicitly mentioned that international donor funds should be the main resource to support these rwodi and to install 'traditional mechanisms' of reconciliation, such as rituals. 53 The influential report by Dennis Pain was the start of an intensive involvement of customary authorities in reconciliation programmes. The programmes, known for the performance of 'traditional' reconciliation rituals such as mato oput, 54 have been documented extensively and have been the subject of much controversy and critique. 55 Indeed, the ways in which transitional justice was organised in Acholiland at the end of the LRA conflict form a clear example of a donor driven 'celebration of localism' in post-war peacebuilding. 56 Customary leaders came to play a crucial role in providing an 'authentic' environment for 'local participation' in the peacebuilding process. To ensure a systematic 'grassroots approach' carried out by reliable 'local stakeholders', international donors were looking for a suitable 'broker' who could translate both the communities' as well as the donors' needs. As supposed community representatives with historical experience in collaborating with external actors, rwodi were well positioned to become these brokers.
Through donor support, customary chiefs could re-enter the public sphere and regain some of the public authority they had lost during the war. Following up on Dennis Pain's call, ACORD (Agency for Cooperation and Research in Development), with funding from the Belgian government, identified, anointed and re-instated 54 customary chiefs. These 54 chiefs all claimed to be rwodi moo. The 'ACORD process' took place in the form of mediatised coronation ceremonies. A former ACORD staff member explained that this reinstatement process was aimed at boosting people's confidence in customary leadership. 57 According to several informants, the ACORD process formed the incentive for clan elders to declare themselves rwodi. 58 To distinguish rwodi moo from rwodi kalam, documentation of historic trajectories of chiefdoms was being consulted, but these documents based on oral histories often contradicted each other. 59 Throughout the turbulent history of post-colonial regimes, besides rwodi kalam and rwodi moo, lineage heads also came to occupy chiefship positions. Material 'evidence' of rwotship, such as the possession of customary regalia like the royal drum, had sometimes been destroyed or lost during the years of war. 60 The identification process caused debates and contestation between clans, many of which still linger until today. 61 People often refer to this ACORD process as very delicate, and characterised by several flaws. 62 For international donors to be able to fully use the potential of customary authority in the peacebuilding programmes, customary authority not only needed to be revived, but institutionalised as well. As such, an Acholi Paramount Chief was installed with an official ceremony in 2000, who acted as the main representative of customary leaders. The position of the Paramount Chief was from the start a contested issue amongst Acholi rwodi and clan elders and often came up in discussions during the fieldwork. It is seen by some as a 'violation' of Acholi traditions, given the historical absence of a centralised and hierarchical customary power structure. 63 The Paramount Chief was appointed as the head of a 'cultural institution' called Ker Kwaro Acholi (KKA), a kind of umbrella organisation under which the 54 recognised chiefdoms of the Acholi subregion were united. In an earlier article in this journal, we have demonstrated how the creation of KKA turned customary authority into an urban-based, heavily sponsored NGOlike institution, with a considerable voice in local governance. 64 Between 2003 and 2013, KKA positioned itself as a key institution in the aid economy. For example, in 2005, the institution was appointed as the main actor in the CRCM component (Community Reconciliation and Conflict Management) of the World Bank funded NUSAF programme (Northern Uganda Social Action Fund). KKA coordinated rwodi to perform rituals or take part in NGO programmes and workshops. Furthermore, it was through the institution that donor funding would be passed on to individual rwodi. Apart from funds, KKA also channelled information. 65 Calls for projects from donors and NGOs would only reach the rwodi through the institution. 66 As such, KKA played a central role in chiefs' access to donor funds, and the prestige that came with it.
Hierarchically structured with a paramount chief, prime minister, a cabinet of ministers and a council of chiefs, KKA functioned as a 'workable' format for international donors to channel funds to 'beneficiaries'. In parallel with the 'manufactured' chiefs of the colonial period, post-war customary authority in northern Uganda had thus transformed into an increasingly 'fixed' and hierarchical centralised structure of customary authority.

Rwodi become development brokers
After the UN Under-Secretary-General for Humanitarian Affairs and Emergency Relief Coordinator, Jan Egeland, referred to Uganda's conflict as the world's 'most neglected humanitarian emergency', the presence of international humanitarian-, peacebuildingand development organisations expanded massively (mostly around 2003 and 2004). 67 Apart from performing donor-funded reparation and reconciliation ceremonies, chiefs were trained in several other domains within the humanitarian and peacebuilding industry. Through financial support and workshops, Acholi customary chiefs were 'empowered' in, for example, the fields of gender, health, land conflict managements and human rights. 68 Clan elders were trained in UNWOMEN gender and peace-building programmes and KKA implemented a gender programme initiated by War Child Canada and Oxfam. 69 Training reinforced the position of customary chiefs as development brokers, turning them into what Bierschenk has referred to as 'masters of developmentalist jargon'. 70 After having been intensely consulted and approached by international donors and NGOs, customary chiefs positioned themselves more assertively and used extraversion strategies by approaching donor agencies themselves. Part of the strategy of development brokers is to valorise their own activities and make themselves indispensable. 71 Chiefs involved in the KKA secretary and close allies of the Paramount Chief had an advantaged position in this regard. For example, the rwot of Lamogi chiefdom evolved through the ACORD process and the creation of KKA as a close ally to the Paramount Chief Acana, becoming his deputy. This position enabled him to get direct access to donor connections and support. During an interview, he explained how he had been working with organisations such as CARITAS, SAFER WORLD and GWED-G and how, until today, he has been approaching several NGOs to fund community projects in his chiefdom, such as community training on environment and reconciliation ceremonies. 72 In their position as indispensable brokers, chiefs became influential figures in the postconflict intervention landscape. Just as Schiltz and Büscher have demonstrated with regards to humanitarian research brokers in Gulu, these brokers are far from passive operators and instead actively shape humanitarian discourses and practices. As such, they strategically reproduce the particular social order in which they obtain their own privileged provision. 73 Customary chiefs became specialists in humanitarian narratives on local 'needs' and participatory solutions of reintegration, reconciliation and other buzzwords within the post-conflict reconstruction vocabulary. 74 In this way, driven by on the one hand, international donors' 'search for the local', 75 and on the other hand, by rwodi's deliberate agency in the form of mechanisms of extraversion, customary chiefs were established as key actors of public authority.
The donor-driven institutionalisation of customary authority had complex outcomes in terms of rwodi's legitimacy. Although their broker position 'worked' in two ways, the status they obtained from this position was unbalanced, and mainly situated at the donor end. In general, the process boosted their 'external' or 'outward' or 'upward' legitimacy but affected their 'internal' or 'inward' or 'downward' legitimacy negatively (since chieftaincy came to be perceived as an alienated, westernised institution). This ambiguity in customary authority is referred to in the literature on customary authority as a 'tragedy': 76 it is precisely this inward legitimacy, on which their outward legitimacy is based, that prompts external actors to work with them in the first place. To the community members, this transformation and 'NGO-isation' of customary authority was problematic in several ways. First of all, many clan members contested KKA's status as a representative institution of customary authority, instead referring to it as an 'artificial creation'. 77 As one clan elder stated: Ker Kwaro is an NGO just like any other. The Paramount Chiefs acts like a king who can rule over Acholi chiefdoms, while he is not, he is not a king, we don't have that thing here. The secretary is taking all support that is supposed to be to the benefit of the community. 78 But not only was the institution's legitimacy questioned, the legitimacy of individual rwodi was also strongly challenged throughout the process. An often-heard concern was that the training organised by international and local NGO partners kept them occupied and they no longer had time to conduct their traditional chiefly duties. Chiefs who were part of KKA in Gulu were sometimes perceived as 'corrupt'. This perception relates to the dependency of chiefs on NGO support, to corruption scandals within KKA, and to the general disinterest and disengagement from chiefs with their communities 'back home' in the villages. The following quote from a priest further reveals the complex position of customary chiefs between 'tradition' and 'modernity': Ker Kwaro's image of an NGO somehow contradicts the traditional ways of living the chiefs are advocating for (…) It is supposed to be about tradition and culture, but apart from their traditional clothing nothing in the secretary looks like traditional. It is bureaucratic. Their modern houses do not have shrines anymore. 79 Some of the 'new' chiefly roles were at times locally perceived as contradicting 'original' Acholi traditional views, such as their training with regards to Western concepts of human rights, women's emancipation, etc. Further, in the eyes of community members, the chief's position as a development broker only results in increased legitimacy if the chief is able to channel the extracted resources into the community. Yet, clan members sometimes suspect their 'leaders' of only acting in their own interests, and thus feel abandoned.
Many chiefs became corrupt with KKA. They spend their days in the office in Gulu, they eat donor money, but this money does not come to the community. 80 The fall of KKA: extraversion under pressure Currently, more than ten years after the end of the LRA conflict, Acholiland is experiencing a withdrawal of international humanitarian and peacebuilding agencies. In 2010, NGO Forum-Gulu (an umbrella organisation that unites all agencies in Gulu) had a membership of 120 organisations, while at the time of fieldwork in 2017, this number was reduced to 60, with about 40-45 still operational. 81 Following the most recent episode of the war in South Sudan, from 2014 this withdrawal had been speeded up as humanitarian organisations moved away from Gulu to install themselves in Juba (the capital city of South Sudan). Yet, other factors equally lie at the basis of this process, such as corruption scandals within the Peace Recovery and Development Programme (PRDP, the government programme coordinating all post-conflict reconstruction interventions), 82 the global financial crisis as well as the shrinking international legitimacy of president Museveni. 83 The effects for rwodi were felt on different levels: on an individual level, their brokerage position was challenged, and on the institutional level, KKA gradually experienced a dwindling relevance and authority. Not only did the narrative of 'bringing stability and peace' with which they had crafted their authority position lose its immediate relevance, more importantly, without donor funding it was hard to continue their operations, maintain their patronage position and keep their status and prestige.
In addition to closing down NGO offices, some remaining donors also ended their collaboration with the customary institution KKA after it came into disrepute for the mismanagement of funds. After several donors had become increasingly frustrated over KKA's lack of financial accountability, in 2014 a series of corruption scandals caused a sudden disengagement from international NGOs and embassies. 84 At a certain point, the paramount chief was even being asked by the Embassy of the Netherlands to refund the sum of 230 million shillings that had been donated to KKA. 85 These events had huge effects on the public image and integrity of KKA which are still felt today.
According to the KKA prime minister, 'rwodi did not develop internal systems of sustaining themselves, so when donors left, everything collapsed'. 86 To a certain extent, KKA had become just like any other local NGO that emerged in Acholi post-conflict community answering to international donors' need for local partners. Donor exit led to instability within KKA and a gradual breakdown of its functioning, as it failed to remunerate its staff. 87 During our research, the KKA office was often found empty, and KKA's main activity had been reduced to the organisation of traditional conflict mediation gatherings. To continue to function, just like in the pre-KKA period, rwodi imposed a fee of UGX. 60,000 (approx. US $ 17) on the people requesting them to mediate in a conflict (such as land conflicts, witchcraft allegations, domestic violence, or spiritual attacks). According to KKA members, this fee was insufficient to run the institution, which largely depended on voluntary engagement since funding ceased around 2014. 88 The fact that KKA was simply not able to provide its core services anymore caused the authority and legitimacy of the chiefs to dwindle. Moreover, it contributed to their failure to redistribute funds and provide symbolic resources. The crisis KKA found itself in was complete, being not able to provide upward nor downward accountability.

Securing customary authority 'after aid': from the institutional back to the individual
The process of the institutionalisation of customary authority under KKA had reinforced the hierarchy amongst Acholi chiefs. With the downfall of KKA, the position of the paramount chief was strongly affected. In general, KKA's disintegration left customary leaders fragmented and divided. As one ARLPI staff member explained: There is an increasing division between the chiefs, with chiefs as rivals and strong competition today. They are divided. This started with their revival, as Ker Kwaro does not recognise certain of the chiefs. But their internal rivalry only grew. 89 The downfall of KKA represented a new phase in Acholi customary authority in which hierarchies were again redrawn. For example, the crisis of Paramount Chief Acana gave rise to other influential rwodi to emerge, such as rwot Yusuf Adek 90 , creating room for alternative constructions of customary authority. We observe how, today, Acholi chiefly authority is equally vested in concrete actions (such as mediating land conflicts, regulating cases of witchcraft, organising ceremonial burials or festivals) and charisma as in a hereditary position. Today one finds rwodi with very diverse backgrounds, young as well as old, drawing on different material and symbolic resources and using different repertoires (for example 'investment', 'sustainability' or 'social balance'). Most rwodi encountered during the fieldwork are engaged in different income-generating activities since their financial support has now been largely suspended. For instance, the rwot of Pagak is a software developer working in Kampala, the rwot of the Ariya clan is deputy principal at a national college and the rwodi of Lamogi and Alokolum are both involved in large-scale farming.
Rwodi themselves stress the fact that in 'the past', 91 this was unheard of. Rwodi often depend on very personal, individual capacities, histories and strategies to construct their authority and be successful or respected as chiefs. Personal capital, skills, relations and abilities or personality are very important, sometimes overlooked elements in chiefly authority. 92 Re-adjusting to the changing post-post conflict setting, rwodi's brokerage position depends on the connection with their communities, with the state and with external actors.
Re-connecting to remaining donors by mobilising the right repertoires: to walk the walk and talk the talk Welcoming, cleansing and reconciliation rituals at the return of ex-LRA members (often strongly mediatised and largely attended) conducted by customary chiefs were crucial 'moments' of visibility of chiefly authority and reinforced their legitimacy. 93 These rituals have gradually become exceptional, since the 'reception centres' (through which returnees would pass to be assisted with their reintegration) have closed their doors, and LRA members no longer massively 'return' from the bush. The registers of reconciliation and peaceful reintegration on which chiefs had established their position have gradually lost their relevance.
One of the strategies of customary chiefs to navigate this changing context was to adapt to discursive shifts in international peacebuilding approaches, similar to NGOs adjusting to donor agendas and narratives. In northern Uganda, these narratives shifted from socalled 'invisible children' 94 during the war to justice and reconciliation right after the war, 95 to land conflict, gender issues and environment today. With no alternative resources available to fill the vacuum left by the departure of humanitarian NGOs, and with no clear registers to which rwodi could explicitly relate (apart from the register of land mediation), external aid agencies remain an important point of reference, both for KKA and individual chiefs. KKA has tried to adapt by shifting focus from justice and reconciliation to topics that are more 'fundable' at present such as land conflict, women's empowerment and nature conservation. During the fieldwork, we observed for example many chiefs involved in exploring the topic of 'environment and sustainability'.
After the closing of the IDP camps, many international NGOs shifted their interest towards the land conflicts that arose when people on return could no longer access their land. Land conflicts in Acholi are mostly to be situated around issues of interclan and household boundaries. 96 Yet, new forms of larger scale land conflicts have been on the rise since the Acholi sub-region has opened up for investment and 'development' by the central government. Rwodi have been involved in negotiating these land conflicts, because land tenure in Acholiland largely remains under customary law, which is recognised by the government and allows chiefs to settle disputes over customary tenure. As pointed out by Kapidžić, mediation remains a crucial register of customary authority (and a legitimising procedure of public authority in general) in Acholiland. 97 For Hopwood & Atkinson, what they have called 'deep knowledge' 98 that chiefs possess on land practices remains an important customary resource. However, chiefs' legitimacy is not a given, and rwodi are sometimes being criticised by community members as corrupt in their engagement in land conflict mediation. 99 Although NGOs interested in land have often approached rwodi as the main actors in charge, it is important to point to the fact that their role in land governance has historically rather been indirect. 100 Other customary elders, such as rwodi kweri and rwodi okoro used to play a much more prominent role in the organisation of customary land. 101 However, chiefs that have collaborated with international NGOs with regards to land have been 'empowered' and 'sensitised' by training on, for example, land laws and on 'best practices' of settling disputes.
The ability to shift to relevant registers and project priorities is a typical part of brokerage and extraversion mechanisms in an aid context. 102 The engagement of customary chiefs with gender issues is another illustration of this. This engagement is contested by some, arguing that their customary position, grounded in patriarchal practices and gender-roles, is in contradiction with the discourse of women empowerment they have been advocating in the context of internationally funded projects. 103 During a 'sensitisation' gathering on gender-based violence organised in Unyama sub-county (in December 2017), we observed this contradicting reality. Only (elderly) women were being 'mobilised' to attend, and during the three hours' interactive discussion between the chief, council members and the women, the main focus was on women's own responsibility in avoiding their husbands treating them violently. Through his discourse, the rwot's interpretation of 'empowerment' and 'agency' was clearly something very different from the one advocated by the donors that provided him training and that appeared on the poster we later found in the hut where we shared a meal afterwards. This case illustrates how rwodi appropriate new repertoires by reinterpreting them to fit within their customary registers.
Unlike ten to fifteen years ago when donors searched for rwodi to partner with, some of these chiefs have now become full-time donor-hunters trying to 'capture' remaining aid. Rwodi offer their expertise in reconciliation ceremonies, community development and cultural revival ceremonies like dancing or cooking competitions. These dynamics, according to the prime minister of KKA, almost amount to begging: 'But how can a chief beg? A chief in Acholi is not supposed to beg let alone be held accountable or reprimanded'. 104 From our observation of the daily activities of rwodi, some of them spend a lot of their time on the writing and submission of project proposals to donors as well as to the Ugandan Government.

Re-connecting with the Ugandan state and the risks of politicisation
The historical relationship between customary authority and the Ugandan state has been well documented. 105 In Acholiland, this relationship has changed with the emergence and withdrawal of international aid agencies. With regards to KKA, a dual rapprochement could be observed, with KKA reaching out to the government, and the government trying to get KKA under its control co-opting it in its patronage networks. Involving customary chiefs in government programmes was a way to reconstruct the state's post-war public authority and legitimacy in Acholiland. 106 Where ten years ago KKA was given an important role by the government in their Peace Recovery and Development Programme (PRDP), KKA now reaches out with project proposals to the government. KKA's prime minister explained that when doing so, the institution deliberately shifts its discourse from being donors' preferred partners to the government's preferred partners, arguing that KKA is 'carrying out state responsibilities of development and should be recognised for this'. 107 The institution operates under the Ministry of Gender, Labour and Social Development, but there is increasing discontent about the way the Ministry treats the institution. As the KKA deputy prime minister explained: We were in a government project on HIV AIDS with them, a project of three years, we did all the work, the mobilisation, the preparation of the communities, and then they put another organisation for implementation, like that, without consulting us (…). Afterwards they even asked us to write the report but we refused. They proclaimed they worked with us, to make their project look right, community based, but they were not! We will never write for them! 108 With the general shift from humanitarian to development interventions, the Ugandan state has increasingly centralised donor interventions in northern Uganda and has become the dominant actor in the current post-conflict reconstruction arena. 109 To a certain extent, this has pushed rwodi back into their more 'traditional' brokerage roles as customary authorities, translating between the state administration and the local 'communities'.
On local levels of the state administration, relationships between state councillors and customary chiefs strongly vary. For example, in it is not uncommon to find the LC1 being at the same time a member of the chiefdom elders' committee. Or for instance, the LC1 of Koro also serves as the rwot's secretary. Some rwodi explained how their status and prestige derived from their position as 'donor-darlings' reinforced their connections with the LCs. 110 Some chiefs work in government positions, and others have close connections to politicians in parliament.
While a close collaboration with the state apparatus at the local level is in general perceived as an accepted aspect of rwodi's brokerage position, their connection to 'politics' is not. Whereas donor support is generally perceived as 'neutral' and 'technical', government support is sometimes perceived as 'politically corrupt'. Donor dependence shaped customary chiefs' image as 'corrupt' and easy to bribe', and the politicisation of some rwodi is being said, also by chiefs themselves, to have a strong impact on the perceived integrity and legitimacy of customary authority amongst their 'subjects'. 111 For example, one rwot had been accused of involvement in a land grabbing scandal in which he allegedly took bribes from General Oketta (government official and senior military officer). Additionally, he was strongly involved in Museveni's electoral campaign in 2016, for which he received a new car from the president himself. 112 Studies on the role of rwodi in customary land tenure also mention complains about corruption, greed and bribery. 113 During a participatory observation of a 3-day 'cultural festival' organised by KKA in December 2017, the presence of the Ugandan state was striking; as was the absence of the international donor agencies. Huge donations (and attendance) of politicians to the festival, and the presence of President Museveni himself as 'chief guest', were all illustrations of the rapprochement between the state and (the remnants of) the customary institution. 114 It was at the same time the subject of much controversy in Acholiland, where Museveni has historically been unpopular.
In 1997, the report by Dennis Pain connected rwodi's legitimacy to their disconnection from politics and the fact that they had not been corrupted by the Ugandan state. Twenty years later, we see that chiefs' legitimacy is boosted more by their position within government patronage networks, than by their connection with international humanitarian or development agencies.

Re-connecting to the communities? Strategies of 'introversion'
Legitimate public authority is not only based on material, but also on symbolic resources. In order to access the latter, rwodi increasingly reach inwards. In present-day Acholiland, sustainable legitimacy of customary authority cannot fully be realised without connection to the chief's 'base', his 'subjects' or his 'community'. This level seems to be the most difficult one to re-connect to. The historical role of the customary chief to fully represent 'his people' and be held accountable by them, had largely been hollowed out throughout the post-conflict period. To a certain extent, this was the outcome of rwodi strategically positioning themselves within narratives and registers developed by international donors. And although these registers were developed in order to answer to the 'needs of the communities', they mainly represented donors' interpretation of those needs, and not necessarily corresponded to the diversity of needs of post-conflict communities in Acholiland. 115 Apart from reconciliation and peacebuilding, rural as well as urban communities needed structural poverty and food insecurity to be addressed. The government-led PPRD has been criticised for lacking a participative approach, designing 'development' in a way largely disconnected from the local communities, in the form of large-scale infrastructural interventions and agricultural reform. 116 After the 'donor adventures', fulfilling the role of true representatives of their communities appears not an easy task. Even those chiefs who were less involved in donor projects are observed today struggling to connect with their communities and to be respected authorities, since peoples' faith in their cultural leaders has been seriously undermined. At the institutional as well as the individual level, different attempts are made by customary leaders to restore this relationship. In this process, which we call 'introversion', rwodi are increasingly pushed to reach back to their communities; or we could say 'reach in', as this implies a shift from an external orientation to an internal orientation.
Alienation was not only an outcome of extraversion strategies, but also of the process of 'personalisation' of customary authority. Historically, Acholi customary leadership was never purely shaped by individual agency; rwodi never acted 'alone' and were always surrounded by a committee of clan elders. 117 Yet the empowerment of individual chiefs through workshops and training largely overlooked these broader customary structures. This demonstrates how customary authority, throughout its history of colonial and postcolonial institutionalisation, had increasingly become embodied by one single person. This disconnected the chief from his broader 'customary entourage'. During the post-conflict period, the nature of rwodi's entourage changed radically. Rwodi were no longer surrounded by their clan elders (ladit), or other customary actors like traditional healers (ajwaka), but instead by NGO staff, policy makers, journalists, researchers and sometimes politicians. 118 While this new entourage also generated respect, the alienation from their historical customary entourage weakened claims to symbolic resources such as cosmological knowledge and to registers of morality and customary wisdom. Customary authorities (in plural) take the form of, for example, councils or zonal leaders, who are often responsible for the practical, daily functions of the chiefdom. The regional context of displacement and the related process of urbanisation in Acholiland 119 resulted in several chiefs living today in Gulu town (or even Kampala) and not residing within their chiefdom anymore. This further reinforced the disconnection with their communities, as chiefdoms need to be run from a distance. In this case, as we observed during our research, community members usually complain that they do not 'see' the chief. Sometimes, this means that they do not know the chief, they have no idea where he is, what he does, how he looks like. Rwodi are sometimes blamed for being unapproachable, or for being 'passive chiefs', 'simply staying at his palace, not do sensitisation and simply wait for the people to take their issues to them'. 120 Chiefs are supposed to be actively engaged in their chiefdoms, as public authority must consistently be practiced or performed by those claiming it. 121 Successful introversion (as locally legitimised customary authority) thus strongly relies on performative and visible chiefs. During the post-conflict aid boom, chiefs were strongly visible and performative, due to the peacebuilding and transitional justice programmes. Yet current everyday chiefly practices such as intervening with ritual sacrifices in cases of suicide, disease or impotence are rather 'invisible', compared to mediatised rituals such as mato oput. These interventions are much more localised and small-scale, less political and receive almost no attention from newspaper or radio media.
The case of the rwot of Koro is illustrative of these dynamics of introversion. The rwot of Koro used to work with external organisations occasionally, but has radically shifted back his focus towards the local community. He is for example not involved in project writing anymore. His chiefly tasks are being conducted at his home. With his committee, he holds court sessions every two weeks to handle cases such as murder, compensation, witchcraft, theft and land wrangles. While some of the youth in his chiefdom do not know him, he is very much respected by the older community members because of his interventions and rituals, for example at a time when there had been increasing numbers of suicide cases among youth in the chiefdom. 122 This rwot is being referred to as having 'remained inside the community' instead of having left 'abroad' for trainings, workshops or spending his days in KKA offices in Gulu town. 123 Rwodi need to prove their presence, vision and capacity through practices that speak to the everyday demands of the communities. These demands are informed by the current political and socio-economic challenges people face in their everyday lives in post-post conflict Acholiland. If introversion is about being able to properly and effectively provide answers to community needs, land is definitely at the centre of these demands. Since land is a very valuable resource, the stakes in brokering access to it can be very high. 124 Reconnecting to their 'base' for many rwodi meant disconnecting from KKA. By referring to KKA as 'corrupt', and the paramount chief as 'a puppet of Museveni', 125 some chiefs position themselves as the 'true' alternatives of customary authority, deeply rooted within their communities. This process is very much embodied in the figure of rwot Yusuf Adek of Pageya chiefdom. His position as rwot, and even the status of his chiefdom itself is heavily contested and controversial has even never officially been recognised by KKA which refers to Adek as an illegitimate and self-proclaimed chief. 126 For his part, Adek refers to KKA chiefs as 'NGO created chiefs', and 'while Acana is with the donors, I am with the community, with the people'. 127 His claims to legitimacy, based on the register of authenticity and 'true custom', are in a way ironic, given his self-proclaimed chiefly status. This has however not prevented Adek from becoming a very influential customary authority in the region. To the contrary, during the time of our research he was one of the most vocal, outspoken and 'public' of all Acholi customary chiefs. His public authority is strongly vested in economic resources, his visibility in the public space, his controversial opinions about KKA and his charisma. This illustrates how Acholi customary authority is highly diverse today and does not correspond to the simplified label of 'the customary chief' nor to a simplified customary position as applied by Dennis Pain and subsequently by the international community.

Conclusion: a difficult balance
Historically, customary chiefs always needed to adapt to changing regimes. 128 Their authority is extremely dynamic, and takes the form of a process that is constantly 'in the making' rather than a given, or an 'emerging property', always in production and never definitively formed. 129 The current changes in northern Uganda's post-post conflict donor landscape confront chiefs with a situation of profound instability. In trying to navigate this instable situation, rwodi have to find a balance between mechanisms of extraversion and introversion. Reconnecting rwodi to their communities does not lie in a total break with donor or NGO registers and practices; it rather lies in finding a right balance, in navigating new and old registers, resources and practices. Chiefs are thus caught up in a complex balancing act, being engaged in an ongoing effort to 'carve out a space' for themselves in the current post-post conflict constellation 130 and to try and secure their position. In employing mechanisms of 'introversion', some chiefs use the rhetoric of 'returning' to their prewar positions. However, there is of course no such thing as a pre-war society that reemerges after the conflict and when the humanitarian crisis has come to an end. The present (which looks radically different from the past after decades of war and external intervention) proves difficult to navigate. Many chiefs appear somewhat stuck, unable to produce new narratives or mechanisms that are adapted to the changing context.
'Transitioning' moments for rwodi have been multiple in Acholi history of customary authority. Take the end of colonial rule, for example, when the colonial administration withdrew, and chiefs were forced to forge alliances with newly emerging post-colonial political elites. 131 The present situation presents a similar moment of transition. The postconflict humanitarian context in northern Uganda, which legitimised chiefs' strong engagement with aid agencies and their reconciliation roles, seems to have come to an end. The Museveni regime actively promotes the image of northern Uganda as a 'postpost conflict' zone and emerging field for development and investment. 132 In this postpost conflict arena, the most influential actors in northern Uganda are the state, investment companies, the World Bank, and societal actors such as born-again churches. Several of these actors do not seem to have an explicit or public interest in rwodi for the provision of their services and the realisation of their legitimacy, as was the case with the donors before. Today, maybe even stronger than ever, the key to customary legitimacy thus lies at the level of the local community. It their current attempts to re-establish themselves as public authorities, finding the right balance between extraversion and introversion or between investing in material and symbolic resources seems key. It is the register of moral authority, as a historic aspect of rwotship and key for its legitimation, which seems to be at the heart of this difficult balancing exercise.