Youth, presence and agency: the case of Kigali’s motari

ABSTRACT This paper concerns young men who drive motorcycle taxis in Kigali, the capital city of Rwanda. Through an ethnographic account of the livelihoods of these motari, it seeks to account for their continued presence in a city whose authorities are openly hostile to their business, yet in which they remain a significant social force. I argue that it is not either by the exercise of ‘agency’ that motari achieve a social presence in Kigali, but through the social relations in which they are engaged. These relations immobilise them and effectively prevent them from mounting any concerted political challenge to hostile city authorities. However, I suggest that this lack of agency is one reason for their significance, since it makes them available as a resource for the schemes of others. I use this case study to argue for a rethinking of the notion of agency in the anthropology of youth. Rather than celebrating autonomous action or the creative, subversive play of the young, I propose instead a relational understanding in which the capacities and opportunities presented by groups of people in social relations grounds their social significance. It may be the very fact of young people’s domination that makes them socially significant.


Introduction
This paper has two aims. One is ethnographic: to make an account of the way in which a specific group of young people, motorcycle taxi drivers, achieve a social presence in Kigali, the capital of Rwanda. An ethnographic account is appropriate in this regard since my account will depend heavily on what riders and other people interested in their business say and do in their practical engagement with the motorcycle taxi business, rather than on the way in which this business works in principle (Malinowski 1926). While an intensive ethnographic perspective on a problem is necessarily limited in scope compared to the use of statistical data, for example, it nevertheless has the capacity to elicit details of social relations inaccessible by other means (Geertz 1973). The second aim is methodological: to propose ways in which anthropologists of youth might approach their project through closer attention to social relations; this is intended as an antidote to the focus on issues of agency, identity and cultural production, which have served as the impetus for the production of the anthropology of youth as a sub discipline up until this point. I make my argument through an exploration of the profession of motorcycle taxi drivers, motari, in Kigali, the capital city of Rwanda. This profession is locally called ikimotari, literally 'the thing of the motorcyclists'. It employs around 10,500 young men in Kigali. My argument is based on data gathered from an eight month ethnographic study in Kigali's Nyarugenge district, conducted between January and September 2012. 1 Here I provide contextual detail and indicate the direction of the analysis that I pursue in the following sections.
The government of Rwanda defines 'youth' as people between the ages of 14 and 35 2 ; this demographic makes up around three quarters of the country's population (National Institute of Statistics 2012a). However, youth, and especially the burgeoning urban youth population, is almost completely neglected in social studies of Rwanda. The only substantial study of Rwandan youth is Marc Sommers' book Stuck (Sommers 2012). Strictly ethnographic accounts of Rwandan youth are, however, almost completely lacking: Rwandan studies are broadly populated by genocide survivors and perpetrators (often as ethic proxies) (e.g. Mamdani 2001), peasants, and the elites which staff government, development agencies and businesses (e.g. Gready 2010). While considerable efforts have been made to give Rwandan peasants a profile as political actors (Clark 2014;Ingelaere 2010Ingelaere , 2014Newbury and Newbury 2000), the same has not been done for youth.
The same neglect of youth is evident in public policy in Rwanda. Youth are referred to as urubyuriko, which refers to the state of being young but is commonly extended to young people as a direct translation of the English usage of 'youth'. This term dominates policy debates on youth. Young men, who are commonly conflated with 'youth' as a whole, can be described as abasore or more modishly, abajene from the French jeunes. Responsibility for youth rests with a Ministry for Youth, Culture and Sport and the Ministry of Education. In Rwandan political discourse, youth appear as students at school or university, although only a narrow elite continue through secondary or tertiary education (Sommers 2012); youth are idealised as the dynamic entrepreneurs of the new Rwanda; or else youth are feared as potentially anti-social elements. In short, Rwandan political discourse shows all the hallmarks of modern representations of youth internationally: youth is where modernity 'hides its dreams' (Foucault 1976); yet it is also a 'terror incognita' threatening the margins of polite society with oedipal images of chaos and disorder (Comaroff and Comaroff 2005). There is an almost total lack of effective youth services or governmental engagement with young people, which is the responsibility of the overworked volunteers who staff the National Youth Council offices in local neighbourhoods (Sommers 2012). The social condition of Rwanda's youth was almost completely unknown until Sommers' study. Both in scholarship on Rwanda, and in local politics, Rwandan youth are either neglected completely or co-opted for other people's politics. This is exactly the kind of observed predicament which stimulated the anthropology of youth and its desire to recover the young as a subaltern group for anthropological study. The anthropology of youth in its current form arose out of a desire amongst scholars to recover the lost voices of the young (Bucholtz 2002). The narrative runs like this: anthropology has always been dominated by adult perspectives; where youth has been studied, it has usually been as the state of adolescence or as a life stage hedged about by rituals controlled by senior community members. Youth in itself, or more importantly, young people in their own right, have been almost completely neglected. As a result, anthropologists of youth claim, there is a need to attend to the young as people with something to say, worthy of anthropologists' attention (e.g. Cole and Durham 2007). I am wholly sympathetic with this general approach.
Motari fit this general pattern extremely well. Motari are overwhelmingly youngmy informants had an average age of around 29, and many had begun working at 16 or even younger. When sharing the results of my research with Kigali City Council, I was directed to the office of the Vice Mayor for Youth, Culture and Sport. In the media and in their everyday lives, motorcyclists are most commonly identified as a social problem, threatening 'security', umutekano, and have most intense relations with the police and the disciplinary wings of their own organisations. In these relationships, the state of youth, urubyuriko, is cited as a cause of 'bad behaviour' amongst motorcyclists: young people are often regarded by those in authority as impetuous, ill-disciplined and lacking the maturity to make appropriate decisions; they require control and guidance (c.f. Sommers 2012). The youthfulness of motorcyclists therefore contributes to their reputation as a group who are hard to manage, noncompliant with regulations and policy, or even resistant. Motari themselves sometimes identify as 'youth', urubyuriko, and more often as unmarried men, abasore these are importantly negative marks of failure: specifically, their failure to secure an income that would enable them to build a house and marry key markers of social maturity in Rwanda. 3 Indeed, as Sommers (2012) shows, in many contexts, to be young in Rwanda is to occupy a subordinate position and be subject to restrictions and various forms of exclusion. Motari are young, and their youth is a mark of other deficiencies and threats; the way in which motorcyclists are approached by the authorities in Kigali is importantly influenced by this status.
While an identity as youth has little positive value to the social lives of motari, this does not mean that, as a population of young people, they do not have an enormous impact on life in the city. The question, of course, is howespecially in a city which is, as I shall show, overwhelmingly hostile to the ikimotari. Posed in this way, the question becomes less an issue of how motorcyclists elaborate identities as youth, and much more about the social relational systems (which include, but are not limited to the symbolic relations of identity) give motari a presence in the citya social facticity perhaps, which prevents them simply from being legislated or wished away by the powerful (the key insight from Durkheim 1938).
In taking this approach, I locate the issue at hand firmly in the (post?) modern moment in urban Africa (Rakodi 2006). A major concern in the literature on contemporary African cities has recently been the way in which disorganised, marginal and often destitute populations can conjure forms of socially significant action out of the apparent chaos and uncertainty of their lives (Boeck 2004(Boeck , 2011Simone 2004). A dialectic of order and disorder, visibility and invisibility figures the resourcefulness of populations whose urban lives might otherwise be understood to have 'failed'. At issue here is the question of how people whose marginal positionrelative to the state, development agencies, or elitescan nevertheless be parlayed into a certain social presence as salient figures in the plans and calculations of others, and thus achieve certain kinds of agency. My suggestion here is that one effective way of exploring the presence of such marginal bodies of people is to attend to the political economy of their livelihoods (Staples 2007), the resources they deploy and those they present to others to demonstrate how, although often disorganised, impoverished or excluded, such groups figure in the plans and schemes of others, and thus exert influence.
This approach is entirely different to that taken by most anthropologists of youth. The contemporary anthropology of youth starts out with the assumption that young people are agents, while youth is a social construction (see for example Amit and Wulff 1995). These basic commitments give the contemporary anthropology of youth a very definite focus, a product of these pre-theoretical commitments (Moore 2004). In many respects, agency and social construction are difficult to keep in one frame since the socially (that is, compulsorily) constructed nature of the person or subject might be seen to undermine her agency, if this is interpreted as a capacity for free action (Butler 2005; see also Laidlaw 2002). In a previous generation of anthropology, this antimony might have been expressed in the contrast been individual and social structure (Fortes 1963). There is thus a clear tension between the terms, in that prima facie it is not clear whether or to what extent agency can be maintained, as 'freedom' in the face of social construction (Stern 2000;Merlan 2016;Asad 2000;Mahmood 2001). Only certain theoretical directions will allow agency and social construction to operate in a complimentary way. Most important of these are broadly post-modern or conservative post-structural positions associated with the 'interpretive turn' in anthropology (Asad 1998;Clifford 1986). These perspectives locate agency in the capacity to define identities; social construction becomes evidence of human action (as opposed to impersonal Durkheimian forcessee Said 2003). As such the anthropology of youth focuses on the cultural positions taken up by the young, the symbolic resources they deploy to maintain them, and their effect on the overall symbolic economy of their societies. The twin commitments of the anthropology of youth, to the agency and social construction of young peoplecan be maintained only so long as the basic focus of the sub discipline is on identity and 'cultural production' (Bucholtz 2002). Likewise, agency can only, really, manifest itself in young people's capacity to modify and rework the symbolic terms on which youth are represented and represent themselves (Honwana and Boeck 2005).
However, a starting point in material social relations (as opposed to 'social construction') does not place the same limits on how we can think about agency (which I refuse to assume at the outset, see Laidlaw 2002). It gives us the opportunity to rethink agency relationally and to ask, in what relationships and contexts do young people compel otherseither deliberately or unwittinglyto take account of them? The 'agency' revealed here is less the act of a subject, than the particular historical outcome of an intersubjective system (Toren 1999). To paraphrase Butler, there is no need for a unitary, prior 'doer' behind deeds (1990, 1995a, 1995b) which are effective exactly because they are committed with (or compelled by) 'another in mind' (Strathern 1988, 272). To state the point simply, whereas a focus on identity foregrounds the 'resistance' of people against a system, a relational focus encourages an analysis of the co-production of people and the system in which they live, including the modes of agency which will appear in that system (Ingold 1995).
I begin by outlining the socio-economic basis of the ikimotari at the level of individual motorcycle owners, riders and their customers. I go on to show how these artisanal workers are implicated in organisational and regulatory regimes which aim to constrain and control them. From these frameworks, I turn to the relationship between motorcyclists and the city authoritiesa relationship which has historically been fraught, occasionally violent, and is premised on the city's unwillingness to permit the ikimotari to continue in Kigali. While motorcyclists' capacity to resist the pressures placed on them by elites and organs of state is extremely limited, they have nevertheless survived and prosper. As a partial explanation for the continued presence of the ikimotari in Kigali, I explore the ways in which the profession presents others with opportunities for rent-seeking, enrichment and political influence. These political contests, both visible and hidden, place motorcyclists in the city and define their ability to pursue their livelihood. That is, they demonstrate the agency of this group of young people, defined as their capacity to figure in the plans and schemes of others, to have value and social 'weight'what I speak of here as 'presence'.

The ikimotari in Kigali
In common with East Africa's other major cities, motorcycle taxis are a ubiquitous feature of Kigali, the capital city of Rwanda. Locally called taxi-moto or simply moto, these taxis congregate at major intersections, bus stations and in marked areas throughout the city but can be hailed anywhere. A trip on a moto costs between RWF 300 and RWF 2000 (about £0.30-£2.00). Taking a moto is almost always the fastest way to move around Kigali, and for the majority of people without their own vehicles, is usually the only means of transport available in the early mornings or at night.
Motorcycle taxis are not the only form of public transport in Kigali. The cheapest and most widely used form of transport is provided by small minibuses called matatu with a capacity of either 18 or 36. These ply set routes, usually from the town centre outwards into the suburbs. Bus fares range from RWF 100 to RWF 300. Taxicabs are also available, although they are expensive and rarely used by most people. Taxi drivers complain that they find customers only on special occasionsweddings, funerals, for trips to the airport or a big night on the town. Recently, large buses familiar from European transport systems have been introduced into Kigali, operated by Kigali Bus Services (KBS) (Oz Architecture and Kigali City Council 2013).
Motorcycles are valued because of their capacities to transport people fast, relatively cheaply, door-to-door and at all hours. Whereas buses work on designated routes, with some neighbourhoods a considerable distance from a bus route, motos are available everywhere. Unlike buses, they also drop people exactly where they want to go and operate after 9 pm, serving bars and clubs in town. These are issues of importance in a context in which to walk long distances, sweating in the heat is felt to be demeaning and in which male social life revolves around drinking in bars. Taxicabs offer a similar service, but their fares are simply too high for most people. Motos are not cheap, but even people on relatively modest incomes can afford to use them when the need arises. They also have the advantage of being able to negotiate standing traffic, a considerable issue in Kigali on arterial routes at peak times. We should not allow ourselves to lose sight of these basic dimensions of the moto sector in the analysis which follows.
The value of motorcycles for transport is in spite of their low social prestige. Motari operate in the context of an explicit cultural hierarchy of vehicles, in which operatingor better still owninga vehicle confers considerable prestige. Bicycles used to transport goods and people occupy the bottom rung of this hierarchy, followed by small mopeds used for personal transport and as taxis. Motorcycles are regarded as a step up from either, but almost all of the motorcyclists I spoke to aspired to drive taxicabs in spite of the lower incomes they offer, minibuses or even trucks. The aspiration to drive a taxi was often explicitly made an issue of 'respect' (agaciro) by informants. Privately owned carspreferably luxurious sedans or four-wheel drive models occupy the top of this hierarchy. 4 Not incidentally, this ordering of vehicles correlates closely with age: working as a motari is a young man's profession, since it requires considerable strength. Correspondingly, driving a moto is not a prestigious livelihoodall the less as riders are usually dirty, dusty and dressed in layers of old jackets and sweaters at workbut one appropriate for youths and other people of low social status. Many motorcyclists complain of the 'disrespect' (agasuzuru) they receive from customers and officials.
In late 2012, there were around 10,500 registered motorcycle taxi drivers in Kigali, 5 a city of some 1.2 million people. This is almost certainly an underestimate of their true numbers, as in some areas and at certain times, large numbers of undocumented riders work, especially away from the city centre and on dirt roads. Their sheer numbers give motari importance in the city's economy. The average household size in Kigali is 4.7 people (National Institute of Statistics 2012b). Interviews with motorcyclists in 2012 suggested that on average they support close to this number of people, around 4.5. That means that between 4% and 5% of the city's population is supported by the ikimotari, or as motari put it, as a motorcyclist 'you feed many people' (utunze abantu benshi). 6 Moreover, the sector is growing rapidly. In the three years to 2011, the number of registered motorcycle taxis in Rwanda increased by more than 250% (Rwanda Utilities Regulatory Agency 2011). This growth has been driven in part by increased registration of motorcyclists, but also, motari claim, by a relaxation of the driving test and a considerable increase in the opportunities for testing. The growth in motos reflects a countrywide expansion in both candidates for driving tests and awarded licenses in recent years. Between 2009 and 2011 147,000 licenses were awardedmore than 70% more than there are vehicles on Rwanda's roads (Asiimwe 2012). Learning to drive is clearly a widely held aspiration amongst Rwandans, especially male youth. Driving a motorcycle is seen by many as a route out of povertyand reasonably so: a motorcyclist can expect to take home between |RWF 1000 and RW 5000 a day, a good income for the poorly educated and unskilled youth who make up the bulk of Kigali's motari.
The Rwandan National Institute of Statistics collects poverty data on a household basis, so it is difficult to see how these figures compare to young people in other situations. However, the national poverty line at the time of my research was RWF 118,000 per adult per year (National Institute of Statistics 2012b). Assuming that motari belong on average a household of between four and five people, even if that household had no other workers, a motorcyclist earning just RWF 1700 a day would be able to maintain all of its members above the poverty line. The conclusion that motorcyclists are not poor is corroborated by Sommers' (2012) account of urban youth poverty. Although his data was collected in 2008 and prices have since risen, it is striking that the young people he interviewed regarded an income as adequate if it enabled them to secure lodging and buy one hot meal a day. In 2012, a one room house, often shared with at least one other working person, could cost as little as RWF 25,000 per month, while a hot meal could cost as little as RWF 500. Even a relatively unsuccessful rider can easily command these amounts of money. Of course, the livelihood of a motorcyclist is by no means as comfortable as that of the city's emergent middle class. 7 Riders were very well aware of this and extremely realistic about the relative monetary worth of their profession. For example, someone working as a bank teller or as a salesperson in one of the upscale supermarkets in towna job for educated young peoplewould expect to earn three or four times as much as a motorcyclist, and of course undergo much less risk in the process.
The ikimotari is a strictly artisanal business. Most riders do not own the bikes they ride. Rather they either rent or lease them from small-scale entrepreneurs, ababosi or abakoresha, 'bosses' for around RWF 5000 per day. These are not relationships of employment, even though motari sometimes speak as though they were. Rather they are relationships of clientship and patronage, normally between a senior person and a younger man. Bosses can make considerable amounts of money from their relationship with motari. Over twelve months, a motorcycle leased or rented at RWF 5000 per day can return almost RWF 300,000 on an investment of RWF 1,300,000, or about 20% profit. Where bosses rent or lease bikes they have acquired second handfor as little as 500,000the possible profits are much higher.
These relationships are highly individual. In some cases a rider's boss is a neighbour or friend, perhaps someone with an interest in the young man and who introduces him to the business in order to advance his prospects while making a profit. In other cases, riders barely know their bosses and enter into purely monetary transactions with them. In these cases, other riders usually make the introduction or occasionally riders are connected to bosses through their co-operative organisations. A minority of riders are owner-operators, most having acquired their first machines through a lease arrangement with a boss, a few using commercial credit. Another, larger minority have no regular access to motorbikes at all and, lacking a regular boss, rely on borrowing motorcycles from other motari at the same RWF 5000 fee. In any case, riders gain access to motorcycles in individuated and sometimes closely guarded networks of friendship and patronage.
There are no established social systemsbased on ethnicity, region, kinship or any other factorwhich I could discern organising motari into wider networks beyond their particular friendships, immediate family and their bosses. Indeed, motorcyclists' social circles are likely to be localised rather than based on their occupation. Riders I became friendly with were likely to move amongst other young people, usually in insecure employment or with no fixed source of income: market carriers, builders' helpers, petty hustlers, car washers and drug dealers figured prominently in the peer groups of riders I knew. These friendship groups seemed more often than not to be based on where a rider lived, rather than anchored in circle of colleagues. 8 Social interactions amongst motorcyclists are in any case rather fleeting and unstable given their mobile work and the insecurity of their livelihoods. Likewise, most of the riders I knew well commuted to work, in many cases from outside the city.
At the most basic level, we have to take into account two key dimensions of the ikimotari as a livelihood. One is the nature of the business itselfproviding individual, door-todoor transport at relatively low cost, twenty-four hours a day. The other is the social relations by which individual motari gain access to motorcycles. These relationships constitute the ikimotari not as a cohesive group organised around some shared interests, but as a collection or assemblage of people, all practicing the same profession, but supported by particular and atomised networks of patronage and dependence. Motorcyclists are organised, but as we shall see in the next section, this organisation contributes little to their capacity to advance their interests or to make their presence felt.

Motorcyclists' organisations
Although a motorcycle taxi rank can look chaotic, motorcyclists in Kigali are well organised and highly regulated. Motari are officially registered with the Rwanda Utilities Regulatory Authority (RURA) and are required to pay business taxes to the city. More importantly, in order to operate they must become members of co-operative or syndicate organisations. There are two such organisations operating in Kigali, FERWACOTAMO and SYTRAMORWA. FERWACOTAMO is a co-operative federation established under the 2007 Law of Co-operatives, formalising and replacing its predecessor, ASSETAMORWA, an informal association in 2008. SYTRAMORWA replaced ATAMIMORWA in 2012, formalising another informal association as a trade union under regulations promulgated by the Ministry of Labour.
The motorcyclists' organisations, however, do not allow motari to represent themselves but extend forms of control over them. The main impact of these organisations on motari livelihoods is to provide 'security' (umutekano). This security is very much for the general population, with motorcyclists constructed as a threat; their organisations provide security by guaranteeing riders' identity, through documenting them and enforcing the use of individually numbered jackets as well as policing their behaviour and issuing punishments. These activities are funded from fees of between RWF 300 and RWF 500 per day and fines, normally RWF 3000 levied for a variety of offences against co-operative rules. These organisations also operate as conduits for directives from the city council, central government and police through branch, district and provincial meetings which members are compelled to attend. In this regard it is important that both organisations are para-statal entities: FERWACOTAMO is ultimately answerable to the Ministry of Economics and Finance via the Rwanda Co-operatives Agency (RCA), while SYTRAMORWA's chain of command runs through the Ministry of Labour. Both have close ties to the relevant departments in the City Council. Moreover, they frequently co-operate with the police and other authorities in campaigns against illegalities and 'anti-social behaviour' amongst motorcyclists. Like most 'civil society' organisations in Rwanda (Gready 2010(Gready , 2011Longman 2011), those governing the ikimotari are clearly designed to group and array people for control rather than to operate effectively in their interests (Goodfellow 2014;Goodfellow and Smith 2013).
The organisation of motorcyclists does not help to promote them as a cohesive 'group' capable of voicing shared concerns. Organisation is designed to afford control or 'security' and paradoxically ensures that members are not capable of organising in their own interests. Indeed, their organisations often regard their own members as a social problem, and collude with the police and other authorities in their campaigns against the ikimotari, which I outline in the next section.

Relations with authority
Motorcyclists have a tense relationship with city authorities. This is true both of their presence in medium and long-term policies, and in the day-to-day aspects of policing and public order management in the city. Officials commonly represent motorcyclists as a problem, perhaps a necessary evilcertainly something they hope to get rid of as 'development' takes hold in Kigali. 9 They regard the ikimotari as dangerous for other road users, and there is reason to suspect that they also think that 10,000 ill-educated youth occupying a strategic niche in the city's infrastructure poses other kinds of dangers too.
Planning documents and transport strategy show that the city aims to eradicate motorcycles from the main roads of the city altogether over the coming years. The flagship Kigali Conceptual Masterplan (Oz Architecture and Kigali City Council 2007), which charts the planned development of the city over the next three decades has no place for motos. Nor does the completed Nyarugenge District Masterplan (Surbana Architecture 2010)its transport and infrastructure section devotes only three sentences to motorcycle taxis. The city plans a 'rapid transport system' based on large commercial companies running big buses, such as those already operated by KBS in the city, which will replace motorcycles or at least displace them from the black road. One transport official phrased the city's plans as follows: As our public transport system continues to be betterbecause as you see results in improvements: we are getting newer buses, bigger busesso the general problem is scheduling them and we are working on it. So the moment we get a good public transport system, then we'll slowly fade motorcycles out.
So that's why we are encouraging people to invest in public transport, putting there incentives for them so they can invest in public transport and then we can get rid of the motorcycles.
Conflict between the City and motorcyclists was crystallized for motari by an event in 2006: KCC banned motorcyclists altogether from the downtown area. One KCC official described the moto ban of 2006 as follows: [We stopped motorcyclists] because of accidentsmore accidents in the city, bad driving and at that time there was a big statistic of accidents … That's why we stopped it for a small period like I think it was one week. But it helped in the change of mind, of disciplining them … Motorcyclists' recollections of the events of 2006 suggest that the ban was enforced with considerable levels of violence. One driver recalled the ban in the following terms: It was in 2006. That time was a tough period. We had a communiqué saying, no motos -on the radio and the TV, saying no motos in Kigali, we don't want to see motos in Kigali. And then motari were shot because of violating that rule. And also, the police hit them with police cars. It was about two weeks, because we spent one week without working, and then the next week, we went back to the road [in violation of the ban]. So it was like that. Some of us kept working, but the police were hitting those motos. And the police also said to those people who take motos, 'if you take motos, we won't be responsible for your safety' … Then after that, the President, His Excellency, found that that was a problem, and he decided to bring us back onto the road.
Nor has this confrontation really ended. Rather it has taken on new forms. Between 2011 and 2012, the Traffic Police prosecuted an aggressive campaign of impounding motorcycles. Motorcyclesbut no other kinds of vehiclescould be confiscated for a range of offences from speeding to being washed outside designated areas. Impoundment was in addition to other fines. It does not appear to have been a formally legal measure. Many motari regarded this as a tactic to 'discourage' them and drive them off the road. They told stories of arbitrary penalties levied for 'worn tyres' and often accompanied by demands for inzoga y'abagabo -'beer for the witnesses', money to keep the bike on the road. Motorcyclists commonly complained that the police were 'hunting' them for their own gain, with rumours that traffic officers had quotas of fines to fill. 'The police are our enemy now,' one rider told me, echoing a common refrain.
Motorcyclists' capacity to resist any of these measures is extremely small. Uneducated, disorganised, without shared economic networks, or close regional ties to one another (Sommers 2012 offers essentially the same diagnosis), they are locked into organisational structures which are inclined to work with the agencies and interests which threaten them. Even when a regular rider goes missingmaybe jailed, or out of work because his bike has been takenhis colleagues may not know where he has gone. More or less reluctant compliance and exploiting their inherent mobility to avoid encounters with police are the major strategies by which motari get by. The social dimensions of their livelihood, within the political context of their organisations and generally prostrate civil society in Rwanda prevent Kigali's motari asserting their presence through any kind of political activism or resistance.
Nevertheless, the hostility of the city does make motorcyclists present to other people in definite ways. Motorcyclists appear to crystallize un-modern forms of livelihood in opposition to radical modernising plans for the city. The government of Rwanda aims to make Kigali into a vibrant economic hub for the East African Community (Oz Architecture and Kigali City Council 2007; Goodfellow and Smith 2013) as a central plank in its Economic Development and Poverty Reduction Strategy (Republic of Rwanda 2007), a plan which involves (on paper at least) the replacement of almost every structure in the city over the next thirty years as well as an overhaul of transport and infrastructure. In this context, motari as largely uneducated and relatively poor youth, populating the city's streets with a most 'African' form of transport are often used as a figure of continuous resistancea 'constitutive other' to modernising development (Simone 2004). Motorcyclists inject 'disorder', akajagari into the 'programme' of development in a context in which centralised 'policy' is regarded by the powerful as final and not open to debate (Ingelaere 2014). Motorcyclists represent the 'cheap life' of ordinary people, as one FERWA-COTAMO official put it and as such are out of place in the new modern (expensive?) city. A city transport official echoed his sentiments: Most of them didn't have good upbringingbecause some of them are street children, others from villages or unruly children who come to sit and maybe learn how to drive a motorcycle and get a permit … You can see him, when he sees a passenger across the road, he makes a turn without checking whether there is a vehicle coming behind him and all of a sudden he's knocked down. So: we are having all of those problems with themwe try to tell them to keep on the right side of the road but they are not doing itso we have many problems with them.
Moreover, it is clear that the authorities' hostility to the ikimotari is in no small measure connected to the fact that most motorcyclists are young. A senior police officer made this connection explicit when he remarked to me in an interview: You know, they are a young group of people who are in this kind of business. Most of them have a low level of educationand this makes a big difference … So what I'm saying is that their behaviour is the common behaviour that you always find in youththat one is a bit commonthey are stubborn … they are young people most of them. Now, given that their background of education, their background up to this level, it's not also easy to influence the way they are behaving … And because of putting them together, working in this environment, they tend to be a bit difficult to manage along the road. They are not like a senior person like you. They are not like very responsible old people who drive carefully. 10 In many ways the continuation of the ikimotari is under threat. Yet it remains as a growth sector in the city's transport infrastructure, and a constant figure of the undesirable, unmodern in contrast to official modernities (Scott 1998;Holston 1989). Moreover, motari succeeded in winning out in their conflict with the city in 2006. It is uncommon for the Rwandan state to back down in a conflict with its own citizens (Straus and Waldorf 2011) and we should remark quite how effective motorcyclists were in opposing their exclusion from the city. Indeed, considering the pressure from the City Council to replace all forms of owner-operated and small-business based transport from the main urban area (Oz Architecture and Kigali City Council 2013), we must question how and why motorcyclists continue to operate in such a hostile social and political environment.
The ikimotari as a resource for others One possible basis of motorcyclists' social facticity or presence in the city is the way in which their numbers, organisation and the profitability of their business enables them to operate as an opportunity for rent-seeking activities which reach high up the political and commercial hierarchy of Kigali. The most dramatic example of such rent-seeking connections is perhaps the implementation of a programme to provide motari with hygienic head covers to passengers inside their helmets. These covers, called akanozasuku in Kinyarwanda, are similar to the hygienic caps worn in food processing. Made of lightweight cotton material and individually packed, they are intended to be used once and thrown away. The official explanation for the introduction of the head covers was that they protected passengers from 'skin diseases' which might be contracted through sharing helmets.
Providing akanozasuku to all passengers became a requirement for motari in 2011, announced by the Ministry of Health on the television, radio and in the press and notionally enforced by the co-operatives. They were purchased in bulk by co-operatives in FER-WACOTAMO, which had the responsibility of selling them on to their members in packages of 25 for RWF 1000. Police, district and city leaders repeatedly insisted that motorcyclists were not allowed to raise their fares to compensate for the price of head covers, citing the falling cost of fuel which had not resulted in lower fares. Motari were quoted fiercely opposing the akanozasuku programme in the press and radioa stance of political opposition not normal in Rwanda.
When I arrived in Kigali in 2012, the authorities appeared to be working to roll the programme out. Special bins were being installed on the streets, painted black and marked 'Akanozasuku'. Stickers were to be seen on motorcycle windshields, marked 'Akanozasuku kamwe, ku mugenzi umwe, inshiro imwe' (one head cover, for one passenger, one use). Repeated exhortations from the police, and continued resistance from motari to buying the head covers, which began costing RWF 100 each before reducing in price to RWF 40 kept the issue in the media, even though there was little sign that motorcyclists were adopting akanozasuku.
Emanuel, an experienced motorcyclist, dismissed akanozasuku as a means for the wellconnected of making money from the ikimotari. They were, he said, not an 'interest for motari' but for others -'no interest for me,' he said. He insisted, as he had before, that the head covers were just a ruse to make money. There was no evidence that sharing helmets caused people to get sick, Emanuel saidsomething other motari also maintained (the 'skin diseases' or 'skin related diseases' akanozasuku were supposed to prevent were never publically identified by the Ministry of Health). This was just a way to 'find money for straw,' he claimed. There were rumours that this business belonged to FERWACOTAMO who were exploiting motari in order to get rich. He identified a businessman called Dodoni who was acting as a middleman for the Chinese company that produce the head covers. 'He makes 1% on each cover sold, so he stands to get very rich. This guy was only making a profit for himself so he had to be chased.' He was no longer in the business, Emanuel thought, but the government still gives him money. Emanuel identified akanozasuku as a device for well-placed businessmen to seek rents from the ikimotari. For him, this was a symptom of Rwanda, where money is 'like rain falling in the forest'imvura igwa mu ishyambait doesn't reach all the way down to the forest floor, but makes the tops of the trees wet.
The gist of this story was confirmed for me by Claude, a senior official in one of the motorcyclists' organisations. He identified the key businessman involved in the akanozasuku project as Bayigamba Robert. Bayigamba is an ex-minister for youth, explained Claude. He is currently the chairman of EWSA and was previously the chairman of the Private Sector Federation (he is also head of Aegis Trust Rwanda Chapter which runs the national genocide memorial at Gisozi and is the MD of Manumetal, a high-end furniture company). He was also central to the privatisation campaign in 2002, especially in the telecommunications sector. He has close links to the President, Claude claimed. Bayigamba is, said Claude, the link to the Chinese who supply the head covers. He owns 51% of the akanozasuku supply company; one of motorcyclists' organisations owns 20% and the remaining 29% is owned by the East Africa Health and Services Co-operative which is made up of private share-holders, at least some of whom are also senior members of the motorcyclists' organisations. Bayigamba is, Claude suggested, a 'big boss', using the organisation of motorcyclists to convert this large, yet disparate body of young men into an opportunity for profit through rent-seeking.
The akanozasuku programme demonstrates motari's capacity to become objects of other peoples' schemes through the organisation of their work. Atomised in their working relations, but kept in place by a series of regulations and a pervasive system of security and control, motari are forcibly integrated into organisations which have some capacity to compel their compliance. As such motorcyclists easily become a captive population, which others can make use of for their own gainall the more so as they are threatened and excluded by the City authorities. Yet although this is a story of exploitation and rent-seeking, it is also one way in which the ikimotari comes to be of value to people of power and influence. Motorcyclists can act as a field in which various interests can come to exploit them through regulatory and non-regulatory means. In the case of the akanozasuku project, the ikimotari becomes a fiefdom, a means of extracting rents and diverting wealth from an artisanal sector to elites by the invention of fees and charges. This is only possible because, owing to the structure of their livelihood, they are entrapped in regulatory mechanisms of security which threaten their independence and capacity to represent themselves. Yet these forces of coercion and exploitation are exactly what gives the ikimotari some value to the powerful and insulates them from the hostility of city authorities. 11 It is possible that without this exploitation, the livelihood of the motari would not exist in Kigali. 12

Conclusion
The point I want to make in conclusion is that the importance of youth in any given social formation is not necessarily the same as 'agency' conceived as the ability to control or refigure received social reality. In my reading, motari are not significant in Kigali because they engage in these kinds of symbolic struggles. Rather they are of importance because of the ways in which their livelihoods present resources or capacities, both for themselves (in terms of their incomes, for example) and others (as a means of transport, as a source of rents). Their 'being there' in the urban environment either forces or invites others to act with respect to them, triangulating what they do with the fact of motorcyclists' physical existence and the capacities which they reveal in different relationships. Hence for a late-night bar patron, motorcyclists' (obvious) capacity to provide convenient 24-hour transport shapes what is possible in terms of other relationships. Conversely, for some business people, and indeed, the motorcyclists' own organisations, these young men have the (less obvious) capacity to be controlled, immobilised, monitored and subjected to certain kinds of exaction. Again, what is possible for such people is very much predicated, insofar as their schemes touch on motorcyclists, on who motari are and how their livelihoods operate.
Agency as an idea has not been withdrawn completely from this analysis. Rather it has become something other than the property of individuals, held in opposition to others. Agency here is built out of other people or out of relations. In making this claim, I draw heavily on recent anthropological critiques of the epistemological basis of social science. Notable amongst these have been those advanced by Strathern (1988Strathern ( , 1991 and Latour (1993), both of whom have argued that the relentless focus on the relation between individuals and a systemtypically conceptualised as society or cultureplaces serious constraints on our ability to comprehend social life. For both theorists, the 'purification' (Latour 1993) of a social field to reveal individuals as points of autonomous agency is simultaneously to render invisible the relationships that make that agency possible. Hence for Latour, a focus on human agency neglects the agency of non-human systems in constituting it. Similarly for Strathern, attributing an act to a specific author erases the chain of action that made that authorship possible (Strathern 2001). In both cases, those elements of a situation that are erased in the effort to identify agency are relegated to a non-agentive or constraining context or system against which people appear to act (Strathern 2003). In this vein, Graeber (2001) has argued for a reconceptualization of social science that would remove completely the distinction between individual and society, and thus between agency and structure, in favour of a vision of social life as a system in operation, where agency and events are multiply constructed out of the actions of others. In short, to act is always to stand in relation to someone or something else as the condition of acting.
Naturally, there is nothing specific to studies of youth about these theoretical positions. They become pressing insofar as anthropologists and other scholars of youth oppose the young to other sections of society, social structures or cultural systems that constrain or control them and against which their agency is manifest. In fact, the entire architecture of 'youth cultural agency' outlined in the introduction to this paper stands in stark contrast to the kind of theoretical positions mapped out by Strathern, Latour and Graeber. Hence, at a methodological level, this reading of agency suggests that it is a mistake to isolate youth from other domains of social life. It is common, for example, to distinguish 'youth cultures' from 'adult society' and then to enter into an analysis of the way in which youth cultures demonstrate the agency of youth in the appropriation and modification of 'adult' norms and values (Blackman 2005). Yet the distinction between youth and adult social spheres can never be more than an artefact of an analytical perspective. It is just as fruitless to try to define 'women's social lives' from the lives of their menfolk: these things, viewed at a wider angle, are necessarily intertwined and mutually constitutive (Strathern 1980(Strathern , 1988Abu-Lughod 1990). This is not to say that either young people, or women are unimportant to a given social formation. It is rather to recognise that even the subsidiary elements in a system shape that system in significant ways (Scott 1998). How women relate to men is generally recognised as more significant than the fact that there are women (Moore 1999); likewise, the relation of young people to their seniors should surely by granted more significance than what young people do in their putatively independent cultural worlds.
Thus, just as feminist anthropology has effectively shifted its methodological stance from a focus on women to a focus on gender relations, it may be time for the anthropology of youth to abandon the notion that attending to young people and their actions is an adequate basis for accounting for the various social phenomena we term 'youth'. Very often it is the lot of the young to be dominated and exploited by people older and better connected than they are. This does not mean that young people do not matter, or that we must strive to isolate this neglected population in opposition to others; it simply calls for the acknowledgement that the agency of youth is manifest more in the ways they are coopted for the schemes of others, perhaps, as here, related to the patterns of their livelihoods, than in their independent exercise of will. That is, young people are more significant in their presence than in their capacity for independent and unconstrained action.
Notes middle class (Burger et al. 2015) and often teeter on the brink of a return to poverty (Visagie 2015). They are also subject to patterns of vulnerability and humiliation that are not consistent with a recognisably 'middle class' lifestyle (Scott 1990). 8. See Sommers (2012) for a corresponding account of the atomised and isolated lives of Kigali's poor youth. 9. On the authoritarian quality of development in Rwanda see Straus andWaldorf (2011), Chemouni (2014) and Booth and Golooba-Mutebi (2012). 10. Of course, there is nothing to say that motorcyclists would not be subject to great restrictions if they were middle aged: Rwandan peasants of any age are certainly placed under very tight forms of supervision in many contexts (Ansoms 2009). However, it remains the case that officials in various contexts use the relative youth of motorcyclists as a rationale for extending close control over them. 11. There are indications that motorcyclists, their customers and bosses are also of importance as a constituency for the ruling party, RPF Intokanyi. Motorcyclists, as a compliant and wellorganised group, are commonly mobilised for party rallies in stadia. It is also instructive that, in their accounts of the 2006 ban, many motorcyclists referred to the intervention of the President as a crucial factor in lifting the ban. It is possible that the party operates partly against the interests of commercial elites in this regard. 12. It is significant to note that this observation stands in direct opposition to Scott's (1990) contention that such an organisation of exploitation, which he regards as being central, along with structures of humiliation, to systems of power, produce resistance as a countervailing force (see Palmer 2014 for a review of the use of this theory in Rwanda). The implication of my argument is that the 'resistant' nature of the ikimotari is in an important sense the result of its effective incorporation into exploitative systems. This argument, however, must be reserved for a future publication.