Research undertaken for this paper was supported by the European Science Foundation.
1This paper analyses the distribution of petrol in the Goma-Gisenyi borderland of the Democratic Republic of Congo (DRC) and Rwanda. The site has been plagued by violence, and has proven to be a hotspot for regional conflicts over the past two decades. Petrol is essential to everyday life throughout the region, while subject to unreliable supply and prone to exacerbating political instability (see Anderson and Browne 2011). At a local level, it is a practical lens through which to examine the interaction between state and non-state actors.
2Goma and Gisenyi form a single conurbation in a narrow corridor of land connecting DRC and Rwanda (Map 1). The international border dividing the two cities empowers local actors. With respect to gasoline distribution, this empowerment is concentrated on the Congolese side, where Goma forms a node in the web of trading routes spanning East Africa. Non-state trading networks in the city thrive due to a combination of lax regulation and significant market demand. In recent years, the Association des Petits Pétroliers du Nord Kivu (A.P.PE.NO.KI) has begun to monopolize Goma’s second-economy gasoline trade. The organization has adopted a sophisticated structure resembling that of local state institutions and operating parallel to them. I argue that interactions between A.P.PE.NO.KI and local officials reinforce forms of local statehood in a manner that further fragments the overall cohesion of the Congolese state.
Map 1. Location of Goma and Gisenyi on the DRC-Rwanda border
Source: OpenStreetMap licensed under the Open Data Commons Open Database License and modified by the author.
3This paper is organized into three main sections. The first offers an overview of research methods and of theory concerning state authority and political economy in borderland regions. It outlines the theoretical tools deployed in navigating the borderland context. Section two focuses specifically on gasoline, and on the organizations that distribute it in Goma and Gisenyi. This includes data on petrol pricing, as well as on where, how, and by whom it is distributed. A.P.PE.NO.KI is identified as a key player in the trade, and the organization’s structure is outlined in detail. The third section unpacks local narratives on cross-border trade. Goma’s political economy is shown to accommodate and empower the A.P.PE.NO.KI organization. This section considers the nature and political implications of interactions between A.P.PE.NO.KI and Congolese state representatives. It is followed by a summary and some concluding remarks highlighting themes that warrant further investigation.
4Research for this paper relied primarily on 15-20 minute interviews conducted with fuel truck drivers, petrol station managers, border police, and A.P.PE.NO.KI representatives, alongside a number of other miscellaneous informants. These discussions were supplemented with longer structured interviews held in more private locations.
5Individuals described in detail the role that they understood the border to play in empowering organizations that distribute petrol. The local trade is intimately tied to processes specific to the borderland context, and these narratives offered insights into how A.P.PE.NO.KI derives its political influence. Further research utilized a combination of secondary academic literature on the site, local news outlets, maps, and government documents.
6Interviews took place in a combination of Kiswahili, Kinyarwanda, French and English with the help of translators from the Université Libre de Kigali. Although residents of Gisenyi, these assistants were dual Congolese-Rwandan citizens, and had lived for extended periods on both sides of the border. They were able to move and communicate freely throughout the twin cities. On occasion, these assistants travelled to areas of Goma that were difficult for me personally to access.
7The body of research presented here was conducted between April and July 2011. Although I have returned to the site on several occasions since, political instability in 2012 and 2013 limited the opportunities for follow up research. Goma is a city that changes overnight. But it is also a place in which entrenched practices have a remarkable endurance. Even as the individual actors have changed, the political processes that are the focus of this paper have not lost their relevance.
8Due to the nature of certain topics covered (cross-border smuggling and informal trading in particular), a significant proportion of respondents declined to be named or recorded. Subsequent political upheavals in the region have made me hesitant to publish names even where consent to do so was acquired two years ago. I recognize how credibility can suffer from a heavy reliance on anonymous interview material. Nevertheless, since the bulk of informants were those engaged directly in the everyday practice of the petrol trade and not high-profile spokespersons, I am not sure what their names would contribute other than a small degree of unwanted exposure. Any contested or seemingly anecdotal evidence was triangulated to the best of my ability, or else discarded.
9What follows is undeniably a snapshot. It offers a close-up image of the petrol trade in Goma and Gisenyi as it operated during my presence there in 2011-12. Information was gathered with care, but is nonetheless limited in its scope. Accounts of local trade will always be biased by the specific prejudices held by respondents and participants, and are subject to constant re-evaluation due to the dynamism of the two cities. Despite these limitations, however, the interactions and negotiations identified in local narratives do engage directly with ongoing debates on the manifestation of political power in the Goma-Gisenyi borderland. They speak in particular to themes of state sovereignty, and the manner in which it permeates the region.
10For the purposes of this paper the term borderland refers to the territorial margins of individual nation states: spaces in contact with international boundaries (either by physical proximity or else owing to the convenience of transport infrastructure). In any borderland, this border-contact is a defining feature of local economic and political practices. Borderlands, unlike borders, refer first and foremost to communities located near states’ territorial limits, not to the international relations of nation state governments (Stoddard 1989, Coplan 2012).
11The political theory of state boundaries has been shaped above all by their paradoxical capacity to both divide and unite (Kopytoff 1987, Nugent and Asiwaju 1996). Cultural and institutional differences develop over time and combine with regulatory discrepancies in a manner that produces significant cross-border imbalances. For the most part, the greater these imbalances (be they social, economic or geographic), the greater the opportunities they provide. Cross-border discrepancies have been likened to potential difference in a battery, inducing both people and goods to flow back and forth (Jackson 2006). The result is that neighboring populations living in proximity to an international divide are often brought closer together by their privileged access to the ‘best of both worlds’ (see Anderson and O’Dowd 1999, Feyissa and Hoehne 2010).
12In as much as the border is a site of opportunity, it is also a source of power (Das and Poole 2004, Zartman 2010). The existence of a border gives rise to a variety of roles: border guard, customs agent, smuggler, cross-border trader and so on. Between these roles, unique relationships are established that organize local power structures and produce networks and hierarchies in the surrounding area. This idea challenges an international relations model in which the ideal-type border would form a clean break between two political ‘like units’ (Waltz 1979: 96). A state’s controlling influence very rarely extends evenly to these territorial limits (Bierschenk and Olivier de Sardan 1997, Hansen and Stepputat 2001). Instead, modes of governance completely divorced from state control flourish around borders, emerging in competition with powers at the center (Raeymaekers 2010, see also Tilly 1985: 169-171). In extreme cases, the state in the borderland comes to act as just one political institution among many, and can be harnessed, or “colonized”, to serve the interests of non-state actors (Das and Poole 2004).
13Borderlands give rise to a ‘polyform scenario’ of local governance that, as Roitman (1990: 695) argues: “articulates itself via the intersection of various agents”. The empowering opportunities provided by these regions often exaggerate the contest between state and non-state groups competing for political authority. In order to interpret broader power structures, it becomes necessary to examine in detail the specific points of contact and exchanges that occur between the different political spheres.
14The empowerment of non-state brokers in African borderlands carries significant implications for the nature of state sovereignty. Central authorities often see diminishing capacity the further they are extended into territorial margins (Rosenau and Czempiel 1992, Albert and Brock 1998). With state incapacity, sovereignty proliferates and is commodified. In some cases it is reduced to a crude adhesive, holding together otherwise unwieldy institutions simply by permitting more individuals to share in the benefits of a sovereign role (Englebert 2009: 99). This produces a multiplicity of institutions which share in the exercise of public authority. In various forms, branches of public institutions interact with government-recognized “traditional institutions” and other hierarchies of authority emerging from the border, making them, to varying degrees, state-sanctioned (Lund 2006: 685).
15Public authority emerges from this melting pot of political actors only as: “the amalgamated result of the exercise of power by a variety of […] [institutions] conjugated with the idea of the state” (Lund 2006: 686, author’s emphasis). Existing between state and society, such contesting political authorities have been labelled “twilight institutions”. It is the responsibility of the sovereign authority to define the political margins these groups inhabit, and countless examples show how destabilizing failure to do so can prove (Zeller 2009: 140, see also Asad 2004). Twilight institutions are by no means unique to the borderland environment, but the empowering economic opportunities conferred by local cross-border economic practices function to exaggerate their influence.
16Cross-border trade is a defining feature of life in Africa’s borderlands. A number of recent studies have shown borders not so much as obstacles but as economic resources exploited by borderland residents (Feyissa and Hoehne 2010). To discuss power in the territorial margins requires an understanding of how political authorities relate to patterns of trade and networks of traders.
17The commercial node is a central concept in the analysis of borderland economies (Dobler 2009, Zeller 2009). These are sites where well-trodden international trading routes intersect with state boundaries. Such nodes punctuate the chains of commerce that connect Central Africa to the global market. They produce dynamic sites, starkly different from their surrounding hinterlands, where local activities come into contact with processes that originate great distances elsewhere (Zeller 2010). Frequently, what emerges in these sites is a class of transnational, informal traders; externally supported and capable of exerting significant political influence (Raeymaekers 2010: 581).
18Smugglers thrive in border towns, circumventing or renegotiating economic regulations imposed by centralized authorities. Nevertheless, it would be wrong to portray these deregulated economic spaces as wholly chaotic. The economic exploitation of borders cannot be explained away as the short-term product of the delayed expansion of state control from the center to the periphery. There are countless examples of the sophisticated and enduring rationality of borderland economic activities, whether they take place within or outside of state regulation. Regarding these informal economic activities, Roitman’s (1990: 677) concept of straddling is informative. In her words: “straddling describes the way in which agents are implicated in production processes that lie both within and outside of state control”. Formal and informal spheres are not taken as separate, but are instead seen to constitute a single system of “production, distribution and exchange” (Roitman 1990: 679). Informal activities frequently involve state representatives. These activities need not be illegal, but are defined by the manner in which they deprive the central authority of revenue it would otherwise accumulate.
19Economic activities that openly subvert the state can simultaneously uphold it through the implicit recognition of its functions and a constant interaction with its agents (see also Rothschild and Chazan 1988, Chalfin 2001, Coplan 2012). It is quite often the case that individual state representatives find themselves performing seemingly contradictory roles in borderland areas. In the borderland context, this process allows state authorities to remain relevant despite limited capacity and powerful rivals (see also Titeca 2012).
20To sum up, borderlands represent zones that facilitate interaction between trading networks (both local and trans-national), between neighboring states, and between state and non-state modes of governance. Among their most defining features is their dynamism, as these relationships are shaped and reshaped by the constant interaction between inhabitants (Paasi 1996). Various social actors in borderlands come to “imagine and instantiate” state sovereignty in ways that do not always align with the unitary, territorial logic of the central state (Chalfin 2001: 202). Others make direct use of the legal command that sovereignty confers in order to accumulate wealth and resources (Englebert 2009: 80). Analysis of power-relations in these regions must account for both the plurality of different political spheres and the blurred distinction between formal and informal economic activities.
21Due to poor transit routes, political instability, and the greater financial incentives of the international market, very little of the oil extracted and refined on Africa’s West coast is shipped East across the continent. As a result, the economies of the Great Lakes region suffer from a “chronic shortage” of indigenous energy supplies, and a near total reliance on the foreign import of gasoline (Johnson 2003: 6). Despite the recent oil discoveries in Lake Albert in Uganda and Lake Turkana in Kenya, this situation is not set to change in the near future (Anderson and Brown 2011). As things stand, fuel destined for Goma and Gisenyi arrives into the ports of Dar es Salaam and Mombasa, where it is refined and shipped over land. From the coastal refineries, fuel is trucked to Eldoret in Western Kenya, and then on by pipeline to Soroti in Uganda before passing through Rwanda. On arrival at the DRC border it is distributed by local petrol stations and networks of traders.
22The initial phase of this research was concerned with establishing differences in how petrol is distributed in Goma and Gisenyi. The aim was to determine the specific opportunities conferred by the border, with a view to how these opportunities are exploited, and ultimately to how structures of power are organized around them. This meant first gathering factual data regarding government regulations and local pricing, as well as to whom, by whom and where gasoline is bought and sold. Official government statistics offer at best a partial image of the trade, and are particularly patchy on the Congolese side of the border. Even where available, these data are often inaccurate, and on occasion deliberately misleading or simply false (Tegera and Johnson 2007: 5). As a result, information about pricing and sales was taken directly from discussions with traders, fuel truck drivers, and petrol station managers.
23Rwandan authorities strictly police the distribution of gasoline in the country. Warehouse managers in Gisenyi claimed that the government stockpiles large quantities of gasoline as a precaution in Kigali against nation-wide shortages, which are uncommon as a result. The sale price is kept consistent across the territory, and distribution is handled by those in the direct employ of the major petrol companies: Kobil, Shell, Total and Engin. Although the price does fluctuate significantly over time, it does so only following extensive negotiations between distributors and the Rwandan Minister of Trade and Industry.
24The situation in DRC is very different. The cost of gasoline imports alone in North Kivu almost doubles the province’s official exports, contributing significantly to its overall trade deficit (Tegera and Johnson 2007: 22). At the same time, roughly 75 per cent of North Kivu’s public revenue comes through import duties, with tax on gasoline a major component. Still the actual rate of taxation is relatively low, and many fuel truck drivers claimed to be able to negotiate favorable deals at the Congolese border posts. As a result of these informal tariffs, gasoline entering the city is often first transited through Rwanda with special permits that avoid the higher Rwandan import duties. The deregulated Congolese market is highly susceptible to pricing fluctuations based on market and political forces along its lengthy supply chain.
25North Kivu imports over $44 million worth of gasoline annually (Tegera and Johnson 2007: 26). Due to insecurity and poor infrastructure on the route north to Ishasa (located on the DRC-Uganda border), its importation through Gisenyi into Goma has risen substantially in recent years. There are two main forms: ‘Super’, from Mombasa, and ‘Origine’ from Dar es Salaam. Gasoline arrives at the Grande Barrière (the largest border post, located on a main road on the lake front) by fuel truck, and is distributed to the official petrol stations in the center of town. Although the same major companies are responsible for providing gasoline to both Goma and Gisenyi, the price at the pump is markedly higher on the Rwandan side of the border. Gasoline prices averaged 1.26 USD per liter in Goma and 1.92 USD per liter in Gisenyi during the period of field research. The fifty per cent higher gasoline price on the Rwandan side of the border is testament to North Kivu’s low tax rates on gasoline importation as well as Goma’s status as a node in international chains of commerce.
26In Gisenyi, the sale of gasoline is handled exclusively by petrol stations run by Kobil, Shell, Total and Engin. These are clustered close to the main road by the lake front and provide for both local vehicles and for those crossing into and out of the DRC. As a result of its lower sale price in Goma, petrol having passed through Rwanda is frequently brought back into the country in private vehicles that cross the border simply to refuel. Although the petrol is not sold on once these vehicles return to Rwanda, the practice significantly lowers the personal cost of operating a vehicle in Gisenyi. This was repeatedly cited as a reason for slow trade in petrol stations in Gisenyi, which are undercut by those in operation just across the border. Rwandan petrol station managers complained of the lack of profitability that resulted from Congolese competition.
27Large petrol stations in Goma (again run by Kobil, Shell, Total and Engin) sell on gasoline in bulk to privately owned stations and individuals. Vehicles in Goma can be refueled directly from traders with jerry cans operating on street corners across the city. The practice is common, and those involved in the trade are known locally as ‘Kadhafis’ after Colonel Gaddafi of Libya (a principal supplier of petrol in recent decades). Based on estimates offered by individuals involved in the trade, there are between 600 and 800 Kadhafis working throughout the city.
28It became apparent early on that disparities in gasoline distribution between Goma and Gisenyi were symptomatic of more persistent differences in the political economies of the two cities. Using Doevenspeck’s (2011) analysis of local perceptions of the border as a springboard, further discussions were catered towards eliciting narratives on the border that related to the gasoline trade. Interviewees were extremely forthcoming on this subject, which can be attributed to the everyday importance of the subject matter.
29Interviewees on both sides of the border made the claim that in Goma more people are able to benefit – even with very small profit margins – from the trade in necessary commodities. In Gisenyi, trade in gasoline involves essentially two parties: petrol station managers and car owners. In Goma, there seems to be a much wider proliferation of these and similar goods. Different parties buy up petrol at subsidized rates and distribute it in smaller quantities, with proportionally smaller profits achieved at each stage of dissemination, all the way down to Kadhafis operating on the streets. Similarly, where local government in Gisenyi regulates gasoline pricing to ensure that it stays consistent, in Goma traders unfailingly exploit even the smallest territorial price discrepancies. As an example, in Mugumba (a district just north of the city) Kadhafis charge 1700 CDF per liter, 10 per cent higher than those operating around Petrol Station Simba just three kilometers away.
30The political economy of Goma was repeatedly referred to as one of Kavoyu. The term translates roughly as ‘disorder’, but does not carry the same negative connotations (of chaos and inefficiency) as in English. Examples I was offered equated to descriptions of individual rational actors pursing their own self-interest unabated. When applied to life in Goma, Kavuyo was used to describe the economic opportunities of the city. Reflecting on this environment, one interviewee commented that Goma is a place where: “anyone can do anything, tomorrow […] but at their own risk” (Interview, A.P.PE.NO.KI trader, Birere District, Goma, 21 June 2011).
31Such narratives are illustrative of the case made by Douglas (1966: 95) in Purity and Danger, who states: “disorder by implication is unlimited, no pattern has been realized in it, but its potential for patterning is unlimited. We recognize that it is destructive to existing patterns; also that it has potentiality. It symbolizes both danger and power”. Narratives on the border frequently reflected Douglas’ analysis. Gisenyi was regarded as orderly in opposition not to Goma’s disorder, but to its potentiality. Notably, despite its use in describing the economic contrast between two cities, Kavuyo is considered first and foremost to be a political term. When asked on the source of Kavuyo, – of profitable disorder – in Goma, the answer from respondents was almost uniformly “the state” (l’Etat). Goma’s make-or-break economy (of which A.P.PE.NO.KI is a key beneficiary) may be a symptom of the city’s Kavuyo, but the nature of state involvement was most commonly understood to be the cause.
32It seems clear that the economic advantages brought about by the DRC-Rwanda border at Goma-Gisenyi lie predominantly on the Congolese side. Despite its reputation for insecurity, Goma sits in a privileged economic and geographic position. Its status as provincial capital, the proximity of substantial mineral deposits and a large NGO presence are all responsible for attracting wealth from overseas. Simultaneously, the city benefits from the proximity of Gisenyi, which represents a neighboring stable haven, as well as a profitable cross-border trading partner.
33Although Gisenyi certainly benefits from its own geographical location, its Congolese neighbor mediates its access to the international forces that permeate the borderland. In ways that Gisenyi does not, Goma represents a node in a chain of international commerce that connects it to the Africa’s East coast and beyond. As Goma begins to recover from economic downturns caused by open war and the destructive eruption of Mount Nyiragongo in 2002, economic actors are increasingly organizing into powerful collectives. My findings suggest that when it comes to the distribution of gasoline and kerosene, one major beneficiary of this environment of profitable disorder is A.P.PE.NO.KI, which has grown to dominate the city’s second economy gasoline trade.
34The Kadhafis of Goma have adopted a complex, formal organization. Groups of ten to twenty individuals working in each small sector appoint a street level ‘president’, most often their eldest. These presidents appoint one of their number to manage wider districts of town, who in turn vote for sector-chiefs and a single president to run the Association des Kadhafi de Goma (AKANGO). AKANGO is a subsidiary of A.P.PE.NO.KI, which, under its current leader Bentin Kanega, forms the single largest network involved in second economy gasoline distribution across the province. The formal organization of these groups is a relatively recent phenomenon, which members date back to around 2002. They have expanded rapidly over the past decade. Today the command structure is rigid, and the president in each sector has the final say on all issues. Their responsibility, referred to as kubavugira, involves speaking for and representing the interests of the whole group. This kind of formal organization of gasoline distributors is only apparent on the Congolese side of the border. Trading collectives (Copera) do exist in Gisenyi, but these tend to be localized and specific to groups of stalls rather than to individuals trading in a particular commodity. Conversely, many characteristics that I identify in the A.P.PE.NO.KI organization are shared by other civil society actors in Goma. The organization is not a closed system, and ranking A.P.PE.NO.KI representatives are also involved directly in networks that deal in other commodities.
35Approximately 80 per cent of Kadhafis in the city are men, who work closer to petrol stations than their female counterparts. This geographical segregation of men and women involved in the trade was attributed to the physical nature of competition for passing vehicles in areas of high demand. Most young men involved are in their early twenties. A minimum age of 18 has been imposed in an effort to reduce the number of Kadhafis engaging in solvent abuse. Traders come from a range of different backgrounds, but most commonly joined the Kadhafis having been students, artisans, cross-border traders, smugglers or shop owners.
36Several respondents spoke of increased selectivity in accepting individuals into the Kadhafi organization in recent years. It is approaching a critical mass, above which there is a fear of internal competition. A formal process requires newcomers to be ‘adhered’ into the town sector in which they wish to work. The network offers a stable source of income and a platform for group influence that attracts both Rwandans and Congolese alike. As things stand, roughly one in ten Kadhafis are either Rwandan or dual citizens. When interviewed, these individuals commented that national prejudices were inconsistent and shaped by changes in the wider political environment. They claimed that residents of Gisenyi were free to join, but that Kiswahili was mandatory due to a lasting suspicion of Kinyarwanda speakers in Goma. This suspicion was greatly exacerbated by 2009 Nkunda mutiny and the M23 rebellion in 2012-13. A.P.PE.NO.KI’s increasing exclusivity contributes to its emerging monopoly of the gasoline trade in Goma, as non-members are prohibited from trading publicly.
37According to the rules of the organization, there are strict penalties for smoking or drinking on the job, as well as for diluting the petrol that is sold. These rules are enforced internally, without recourse to the police. Breaching them can result in fines or outright exclusion from the organization. Once accepted into A.P.PE.NO.KI, Kadhafis must contribute a monthly fee of 2000 CDF ($2.25 USD) to their local branch (AKANGO in Goma). This money goes towards maintaining the organization and negotiating favorable deals with state representatives and the major gasoline suppliers. It also adds to a communal welfare fund, which individuals can draw on in times of misfortune (hospitalization was a common example, as well as cases in which jerry cans had been stolen or confiscated and individuals were in need of emergency funds).
38Kadhafi pricing is based on complicated agreements negotiated between petrol station managers and sector presidents in the A.P.PE.NO.KI organization. Station managers take steps not to be undercut by Kadhafis working in their district, but also rely on the bulk sales that they make to the AKANGO organization. When station managers need money to pay taxes, it is common practice to seek cash from ‘their’ Kadhafis in return for further price reductions. A similar situation arises when fuel truck drivers arrive at the Grande Barrière. Truck drivers often phone Kadhafis in Goma to come to the border and assist in upfront payments of the import tax. In this way, AKANGO does not operate in a manner that is wholly exempt from taxation, and instead contributes indirectly through assistance to its suppliers. Despite these forms of assistance, however, the relationship with private petrol stations is skewed in the Kadhafis’ favor. A.P.PE.NO.KI regularly boycotts stations that do not comply with its demands. According to a number of separate accounts from both sides, it was said to be unlikely that petrol stations near the city center could continue to operate without Kadhafi support.
39A.P.PE.NO.KI has no official party political alignment in the region, and does not require its members to engage directly in political activities. As one street-level president stressed in the context of party political associations: “Dans l’association, c’est libre d’aller où tu veux” (Interview, A.P.PE.NO.KI Street President, Birere District, Goma, 25 June 2011). This is not to say that the organization does not engage directly with the state. Respondents spoke of constant interaction with state officials at every level of their operations. Negotiations occur with customs agents as petrol crosses the border and with local police officers as it is sold in petrol stations and on street corners. Higher up the chain of command, deals are brokered between sector presidents and local burgomeisters, and between the organization’s president and local politicians.
40Somewhat counter intuitively, many Kadhafis claimed to relish periods of gasoline shortage, during which they could significantly hike their prices while station pumps run dry. During these periods, the trade becomes significantly more lucrative, and as a result the Kadhafis are able to strike deals and obtain gasoline from private sources with whom they share in the increased profits (Interview, Hotel Serena, Gisenyi, 4 July 2011).
41To sum up, A.P.PE.NO.KI exerts significant control over the gasoline trade across Goma. The organization is selective, and members abide by a system of rules within an internal political structure. They enjoy a stable income and a degree of financial security through a communal welfare fund. Although its representatives defend the organization’s claim to neutrality in local party politics, they interact constantly with Congolese state representatives in the course of their operations. Unlike Uganda’s ‘OPEC Boys’ studied by Titeca (2006), few Kadhafis have past employment with the military. When asked, individual members suggested that the political influence of the organization stems more from its structure and the necessity of the traded commodity than from their physical threat to Goma’s security.
42A.P.PE.NO.KI typifies a twilight institution, controlling a significant stake in the distribution of commodities essential to everyday life. Local narratives suggest that the organization’s power stems from the nature of the border and the manner in which it shapes economic and political practices in the city. Encompassing a wide network of actors involved in the gasoline trade, A.P.PE.NO.KI effectively controls the availability and pricing of gasoline throughout Goma.
43The President of A.P.PE.NO.KI holds authority over the head of AKANGO, who in turn commands district heads, warehouse managers, sector heads, street-level presidents and the ultimately small groups of Kadhafis found on most street corners. The Congolese state, meanwhile, operates through “the mayor, burgomeisters, chiefs of neighborhoods (quartiers), chiefs of cells and street chiefs” (Englebert 2009: 63). The structural parallels are clear and are no coincidence. A.P.PE.NO.KI’s local AKANGO branch functions as a political unit, enacting internal discipline, drawing revenue up the chain of command and providing welfare back down it.
44A second parallel between A.P.PE.NO.KI and the state can be observed in the proliferation of gasoline and of domestic sovereignty down their respective chains of command. As Englebert (2009: 63) argues, sovereign roles within the Congolese state apparatus are “parceled out down to the lowest and most local levels of officialdom”. This is not a deliberate strategy on the part of state authorities in Kinshasa, but rather a product of their incapacity to centralize sovereign authority (see Raeymaekers 2010). In a similar fashion, gasoline products are bought up in bulk and sold in smaller quantities by members operating at different tiers of the A.P.PE.NO.KI organization, ranging from warehouse managers that stockpile thousands of gallons of gasoline, to individuals who sell it by the liter on street corners. These twin proliferations (of sovereignty and of gasoline) perform an important function in sustaining both the state and A.P.PE.NO.KI, as members of both exploit the profit margins available at their personal tiers of authority. Although the incentives on both sides differ, recent years have nonetheless witnessed a marked convergence in the way state authority and gasoline distribution are organized. While on the one hand A.P.PE.NO.KI is adopting a sophisticated political structure, on the other, local state authorities have come to resemble an extended trading network, converting domestic sovereignty into personal profit.
45A.P.PE.NO.KI members engage in constant interaction with Congolese state officials. Representatives of the Kadhafi organization push for lower taxation, and for improved arrangements regarding their personal security, the enforcement of credit arrangements, and the supply of gasoline. Meanwhile, police and other officials regularly negotiate favorable deals regarding the cost of petrol and the personal revenue they can generate from import fees. At times, local state representatives may even make use of the Kadhafi network to sell their own fuel reserves and profit personally from the transaction. This relationship with the state is markedly different from accounts of the OPEC Boys further north on the DRC-Uganda border (see Titeca 2006). Where the influence of the OPEC boys is driven by threats – to local security, gasoline provision and the political support of politicians – A.P.PE.NO.KI relies more on economic incentives in its negotiations with the state. This appears more conducive to a lasting partnership than the interaction between state officials and the OPEC Boys in Arua, where periods of cooperation are punctuated by frequent defections by both sides (Lecoutere and Titeca 2007).
46Comparing the two parallel structures, representatives of A.P.PE.NO.KI appears to “lie just below” state representatives at each tier of authority (Interview, petrol station manager, La Corniche, Gisenyi, 6 July 2011). Thus Kadhafis trading on the streets must negotiate with, and are ultimately subject to the authority of, local police officers. Meanwhile petrol station and warehouse managers answer to burgomeisters or police cell-chiefs. Further up the chain, the Kadhafi leader for Goma, who negotiates with the local police chief, borrows for himself a degree of authority that he can exercise over the chief’s subordinates. Once negotiations take shape, Kadhafi positions come to be state sanctioned, and their occupants engage in mutually beneficial partnerships with state officials. In this manner, the structures of the two organizations have become integrated.
47The kind of partnership outlined above typifies a “colonization of the state” by actors operating in its margins (see Das and Poole 2004). It is one clear example of a non-state network harnessing the state to its own ends while in the process blurring the state – non-state distinction. What is particularly striking about the A.P.PE.NO.KI case is the manner in which local Congolese state representatives actively encourage this process. State officials in Goma exert influence over the crucial gasoline trade only through their capacity to integrate the Kadhafis’ organizational structure into their own. This suggests an extension of state control not through the threat of legal exclusion, but rather through the de facto, institutional inclusion of powerful non-state actors. It amounts to a proliferation of domestic sovereignty, which leaks out of the state hierarchy and into that of the Kadhafis. Ranking members of A.P.PE.NO.KI are empowered both with a degree of indirect control over the state’s agenda, as well as at times a direct authority over state agents in lower ranking positions. Although police escorts are not uncommon for civil society presidents in the DRC, a number of interviewees emphasized this as an example of A.P.PE.NO.KIs integration into the state hierarchy. Indeed, reflecting on the increasingly prolonged periods of cooperation between A.P.PE.NO.KI and the state, one interviewee remarked candidly that police officers in Goma “answer to the president of AKANGO” (Interview, Birere District, Goma, 20 June 2011).
48It is undeniable that the A.P.PE.NO.KI – state partnership has a subversive impact on state authority throughout the borderland, as the sovereignty embodied in state-sanctioned positions is proliferated and diluted. This is especially pronounced considering the manner in which the Kadhafis straddle the formal and informal economic spheres (see Roitman 1990). By integrating the Kadhafi organization into their own hierarchy, state representatives become directly complicit in second economy activities that ultimately deprive the central Congolese authorities of revenue. Indeed, the extent to which these practices occur constitutes one of most significant political discrepancies between Goma and Gisenyi. Put simply, there is a much higher degree of concurrence between the legal modes of operation in Gisenyi and Kigali than between Goma and Kinshasa. Since the interests of Gomese officials only partially align with those of the Congolese state more broadly, the situation challenges monolithic readings of state involvement in borderlands. By the time that localized forms of statehood are generated they are not just out of sync with the centralized authority; they often directly contradict it. Owing to a disjointed vertical distribution of domestic sovereignty, law is both interpreted and implemented differently at every link down the chain.
49To argue that the interaction between state and non-state actors in the DRC’s eastern borderland has undermined central control of the region presupposes the extension of that control in the first place. This is not my intention. Indeed, the history of Congolese state authority in North Kivu has been mixed, varying according to the regime in Kinshasa and the political relationships with neighboring states (see Keith 1919: 186, Nzongola-Ntalaja 2002: 66, Mamdani 2002: 234-244). Nevertheless, this process certainly contributes to the ongoing regional fragmentation of state authority in the DRC. In a region where Kinshasa’s control has been repeatedly contested (through both violent uprisings and habitual disobedience in everyday practices), the state authorities in Goma have been forced to adapt in order to stay relevant. Their partnership with groups such as A.P.PE.NO.KI extends their influence through the integration of powerful non-state actors into state-sanctioned positions of authority. Individual negotiations with Kadhafi representatives at every tier of officialdom rest on the ongoing recognition of both the function and the authority of the local state officials involved. This all takes place at the expense of the Congolese state’s internal cohesion. The net result is that A.P.PE.NO.KI promotes and upholds forms of localized statehood in a manner that perpetuates the fragmentation of state authority within the wider Congolese territory.
50Gasoline is essential to everyday activities in Goma and Gisenyi, and has the power to shape and reshape the political and economic orders of the two cities. Its local distribution contributes to the reconfiguration of state sovereignty, concentrated in Goma, and ultimately to the ongoing fragmentation of Congolese state authority throughout its territory.
51Despite their physical proximity and the extensive cross-border trade that exists between them, Goma and Gisenyi exhibit polarized political and economic environments. The twin cities lie on a crucial trading corridor connecting East and Central Africa. Still it is Goma and not Gisenyi that represents a commercial node on this chain. Goma is considered a site of wealth, benefiting more than its Rwandan neighbor from the international forces that permeate the borderland. Government regulations in Goma are lax and often negotiable, and the city exhibits an environment of ‘profitable disorder’, known locally as Kavuyo.
52This paper has offered a detailed account of A.P.PE.NO.KI, a network of traders and warehouse managers that has come to dominate Goma’s second economy gasoline trade. Research at the site showed a sophisticated hierarchy that allows this network to efficiently exploit the opportunities presented in the borderland. Its structure mirrors that of local Congolese state institutions, and interactions with state representatives occur at every tier of authority.
53Ongoing negotiations between A.P.PE.NO.KI and the state give rise to mutually beneficial partnerships that constitute the de facto institutional integration of the two. The impact of these partnerships on state authority in the borderland is multi-faceted. On the one hand, they function to extend state control by integrating powerful non-state actors into state-sanctioned positions of authority. On the other, these partnerships constitute a dilution of state sovereignty that further undermines the Congolese state’s internal cohesion. Through their close partnership with A.P.PE.NO.KI, local authorities in Goma become directly complicit in activities that deprive the state institutions in Kinshasa of revenue that would otherwise be accumulated. The process reifies and upholds a state presence in Goma that contradicts the idea of a unified, Congolese political authority.
54One line of enquiry that warrants further investigation relates to A.P.PE.NO.KI’s potential lifespan in Goma. The gasoline industry is prone to volatility, and the organization’s resilience remains to be seen. It would be interesting to establish whether A.P.PE.NO.KI’s structure can (i) endure the kinds of upheavals that have plagued the site’s recent history, and (ii) persist in the same form regardless of changes in the individuals occupying different positions in its hierarchy. If it can, then the organization’s structural integration into the state could well become normalized. This would make the implications for state authority highlighted in this study all the more pronounced.