CULTURAL ANTHROPOLOGY, Vol. 35, Issue 3, pp. 462-486, ISSN 0886-7356. DOI: 10.14506/ca35.3.05

THE RIGHT TO THE REMAINDER: Gleaning in the Fuel Economies of East Africa’s Northern Corridor

AMIEL BIZE

University of Bayreuth and Cornell University

https://orcid.org/0000-0001-7213-9712

---

Pulling into Gaakwen,¹ one of Kenya’s busiest truck stops, a driver might reach a hand out of the window of his truck’s cabin and make a gesture—a rotation of the wrist, like someone holding a bottle by the neck and swishing it around. This hand movement signals to brokers waiting among the densely parked tractor trailers that the driver wishes to sell fuel. It sets off a series of interactions: a broker approaches the truck and strikes a deal with the driver; a crew arrives with jerricans and plastic tubes; they siphon anywhere from ten to forty liters of petrol or diesel from the truck’s tank. When conducting research in Gaakwen, I took note of the gesture as an element of a symbolic world particular to these transactions, a means of demonstrating to brokers that the trucker was in the know. But what exactly it symbolized was not initially clear to me. The motion did not obviously refer to fuel, nor did it mimic any of the actions or objects of the siphoning itself—what, then, did it signal? After some time, I began to interpret it as a reference to a leftover: the gesture, I determined, indexed the fuel remaining in the tank, the share sloshing around and available for sale. More, I came to understand that the idea of a leftover was key to the entire transaction: truck drivers described what they were selling as “leftover” fuel, and dealers understood their profits in terms of a marginal “extra” they were able to gain. Despite the illegality of the exchange, drivers and dealers saw themselves as conducting an entirely legitimate activity: gathering and exchanging residual bits of a commodity. Rather than an illicit economy, then, this was an economy of remainders.

This article traces my effort to understand the roadside fuel trade and the role of the remainder within it, as well as to understand what marginal exchanges organized around leftovers might tell us about the larger liberalized economy in which they are embedded. The Kenyan fuel sector was liberalized in 1994, with the goal of allowing more players into the market—fuel was to become more accessible through competition that would lower prices, and through the entry of new distributors who would help fuel reach underserved areas. Observers widely agree that the strategy has failed—top suppliers continue to dominate the fuel market, and distribution remains concentrated along certain pathways (IEA 2000; Gichuru 2007; Kojima 2010). As a result, the movement of fuel into more peripheral and rural areas often takes place through unofficial means, and the roadside trade proves key among these. Thus, while these roadside exchanges are small, they are not insignificant: they account for a substantial share of fuel distribution within Kenya.

In this article I explore the roadside fuel trade and the concept of the remainder in two ways. First, I draw inspiration from an ancient practice—gleaning, the right of the marginalized to gather harvest leftovers—and use it to think about what exchanges organized around remainders might tell us about entitlement and (re)distribution at the margins of capitalism. Emerging from ancient and feudal societies where redistributive obligations obtained within a context of established inequality, gleaning proves surprisingly apropos for examining distribution at the edges of liberalized economies, where access to goods and notions of economic justice are often embedded in explicitly unequal exchange relationships. Second, I engage in a detailed ethnographic examination of the practices of valuation in the roadside fuel economy, which are organized around a remaindered share rather than a purely price-based conception of profit, to suggest that this conception of value reflects understandings of entitlement and exchange often overlooked even by critics of liberal economics. I thus propose that the idea of the remainder can offer insight into “actually existing” liberalization (Brenner and Theodore 2002), as well as into alternative value concepts arising not necessarily in opposition to capitalist practices, but alongside them.

In asking what the remainder can tell us about how marginalized communities engage with capitalist circuits, I also consider how one might study practices of valuation. I build here on the work of economic anthropologists attentive to the cultural logics of calculation and measurement (Strathern 1992; Weiss 1997; Guyer 2004; Peebles 2015), and especially to “the social, semiotic, and material processes whereby substances get utilized, unitized, and numericalized” (Kockelman 2016, 91). As Jane I. Guyer (2004, 19) writes, to do this, one must adopt an approach that “puts transactions into slow motion and lingers on every frame . . . objects, numbers, scales, the modalities of exchange, relationships.” This article thus lingers on details. Looking at the fuel trade from the perspective of both drivers and dealers, it examines the minutia of transactions: terminology, unit sizes, performative practices—even container preferences. With attention to what such details tell us about value definition and the creation of meaning more generally, I aim to show how visions of justice and even social order become reflected in the practical and conceptual tools of measurement and exchange.

* * *

My interest in economies of remainder derives from an ethnographic study of post-agrarian rural transformations in Kenya, in which I tracked shifting understandings of value and risk among rural populations whose sources of livelihood were no longer predominantly agricultural. Gaakwen exemplified the way that, increasingly, economic opportunities were found on the road rather than on the land. Initially a market center for the surrounding villages, Gaakwen grew into a bustling settlement in the early 2000s when truck transport overtook the railway as the preferred option for long-distance cargo and truckers began stopping there for the night.² By the time I began fieldwork there in 2013, Gaakwen had become a hub of entrepreneurial activity in the midst of rural space. Restaurants, guest houses, bars, and other shops and services lined the highway and formed the heart of a settlement that spread into former farmland on either side of the main road. Having literally sprouted out of agricultural land, Gaakwen was also a site at which the infrastructure of the transnational trade route came into contact with marginalized rural economies. Day and night, trucks and tankers bearing commodities passed through the truck stop, ascending west toward Kenya’s western counties and the landlocked countries of East Africa, or returning east, often empty, toward Nairobi and the coast. Gaakwen’s bustling rest-stop economy thrived on servicing and surveilling this transport. Truckers stopped to eat or spend the night, leaving behind some of their “mileage” (spending money for the trip) and creating a market for products and labor drawn from surrounding rural areas.

Gaakwen’s networks of brokers took advantage of this interface to facilitate the diversion of cargo out of its “predestined pathways” (Appadurai 1986) and into the rural spaces that provided alternative markets for it. The Mombasa-Malaba highway that passed through Gaakwen was part of the Northern Corridor—one of East Africa’s main commodity transport routes and a notoriously leaky one. As trucks pulling sealed containers traversed the Kenyan countryside, commodities would often escape: through accidents, pilfering, or a combination of both sometimes called an “intentional accident” (ajali ya makusudi). Among the variety of escaped goods that were bought and sold in Gaakwen, a particularly sophisticated and established trade had grown up around siphoned fuel. Crews of siphoners would regularly extract fuel from the tanks of tractor trailers and into twenty-liter jerricans, load them onto pickup trucks or motorcycles, and sell them into what the government called, with suspicion, “local markets.” Fortunes were made from the roadside trade, and its illegality did not undermine the sense that it was a legitimate and attractive business. Indeed, the practice of siphoning was accompanied by meaningful conversions that turned fuel-qua-commodity into fuel-qua-remainder, and thereby into something available for appropriation.

My ethnography revealed complex dynamics of legitimacy and entitlement playing out in the gaps of the formalized economy. Not only trucking companies but also the energy regulatory authority, the police, and the oil majors viewed the roadside fuel trade as extremely dangerous. The circulation of fuel through unregulated channels and at prices lower than those determined by the regulatory body undermined the authority of the set price and the collection of levies subsumed into fuel prices. Further, fuel dealers often “extended” siphoned petrol by diluting it with kerosene, which is taxed at a lower rate and therefore less expensive—the government estimates that 80 percent of kerosene bought in the country is used to adulterate petrol, though this figure remains disputed (Okoth 2018). For authorities, then, siphoning self-evidently constituted a crime, one exacerbated by questions of tax, price, and the material consequences (for engines) of using adulterated fuel.

Roadside communities, however, saw the appropriation differently. Despite the illicit status of fuel dealers’ work, my respondents in Gaakwen regarded fuel siphoning as an established trade. One interlocutor, an employee of a trucking company, described fuel dealing as a “sector” (sekta) of Gaakwen’s economy. As he put it, the fuel trade was simply too established to be understood as illegitimate: “The fuel trade is illicit (haramu), but it is licit (halali) for those who are invested in this sector.”³ Indeed, fuel dealing was explicitly distinguished in Gaakwen from the theft of fuel, understood as siphoning without the driver’s knowledge or failing to pay after having agreed to buy. A long-term Gaakwen resident described fairly draconian measures that the township had taken to curtail thefts of both fuel and cargo, measures he saw as essential to protecting Gaakwen’s reputation as a safe place to park and to sell fuel. Fuel dealers in Gaakwen had also organized themselves into a trade association, with representation in the township’s self-governance structure,⁴ which regulated prices and prosecuted thieves. Dealers saw themselves as providing a much-needed resource to off-road populations, creating income opportunities and engaging in a highly profitable enterprise. They did not see these goals as necessarily in tension with one another.

It is not at all surprising that in contexts of evident inequality, and along a trade route not (yet) fully enclosed, resources will find ways to escape. Nor is it surprising that this escape will be legitimized as a form of “social banditry” (Hobsbawm 1965). Indeed, scholars have often discussed such appropriation of commodities through the lens of the “moral economy,” by pointing to the way that resources are embedded in structures of meaning, obligation, and justice that extend beyond the realm of financial transaction (Thompson 1991; Scott 1997). In this article I take a complementary approach by focusing on the concept of the remainder. Against the charge that their actions constituted property theft, those engaged in the trade framed fuel siphoning as collecting a “leftover” (salio). This was, I argue, not merely an attempt to legitimize taking fuel—rather, it revealed a specific mode of valuation that we should take seriously. When truck drivers who sold fuel and fuel dealers who bought it constructed the thing they were transacting as a leftover, they were not only claiming their right to take fuel (because the leftover is a fraction available for appropriation) but also describing a relationship to the commodity flows that traveled along the highway—a relationship embedded in the form of valuation itself. As I will show, in rewriting commodities through the concept of the leftover, Gaakwen’s traders produced an understanding of value that distinguished between “necessary” and “remaindered” shares, and thus created an entitlement to fuel that—while claimed as a right—also recognized the trucking companies’ claim. Traders asserted their right to profit from fuel as long as their profit remained marginal, in the sense both of small and of peripheral. Attending to the explicitly marginal nature of this claim can help us see beyond the dichotomy of profit motive and moral usage that often marks discussions of moral economies (Palomera and Vetta 2016; Carrier 2018). I begin, then, by situating marginal claims within the historical and metaphorical space of liberal circulation.

FLOW AND OTHER METAPHORS

Fuel’s movement across Kenya and even East Africa was shaped by the technological and regulatory realities of transport and the fiscal processes that structure distribution through price, taxation, and (de)regulation. The liberalization of the fuel sector in 1994, part of a broader effort to open markets in Kenya, sought to facilitate fuel’s distribution around the country. But its reliance on models of market distribution did not adequately take into account the country’s uneven development, which was materially entrenched in infrastructure patterns that affected distribution in enduring ways. Thus we see a familiar disjuncture between the optimistic metaphors of liberal economics and the realities of exchange on the ground. In this section, I examine this contrast by exploring the difference between liberalism’s liquid metaphors—particularly “flow”—and those used in the roadside economy.

Financial and physical infrastructures in Kenya have long colluded to undermine an ethics of active redistribution. In the immediate post-independence period (as before independence), fiscal policies focused on creating an environment attractive to private enterprise—as instantiated in a seminal economic document, Sessional Paper N.10 of 1965. This blueprint proposed a trickle-down approach that focused investment on already advantaged areas, where it would “yield the largest increase in net output” (Republic of Kenya 1965). The expectation was that this approach would create financial and human resources that, by expanding the national economy, would produce the fiscal space for investment in marginalized areas. In reality, the policy and its successors generated development patterns marked by a concentration of services and capital-intensive infrastructure along a narrow corridor, leaving large parts of the country underserved.

The idea that established distribution patterns had entrenched unequal access to petroleum was one of the advertised justifications for the fuel trade’s liberalization. Its supporters argued that the oil majors—Chevron, Mobil, Total, Shell, and BP—concentrated on large urban markets and left unprofitable rural hinterlands underserved. By undoing a regulatory regime that put products out of the reach of many, petroleum-sector liberalization would advance the circulation of fuel into marginal areas. In the 1990s, then, reform efforts removed price controls on fuel and opened up procurement, distribution, and pricing to market regulation. Controls were replaced with an open-tender system through which companies won import permits for crude. Ultimately, however, this system allowed the oil majors to continue to dominate the national market (Gichuru 2007).⁵ Nearly three decades after these free market reforms, the spatial logic of preliberalization distribution persists. A map of fuel stations shows a concentration along a corridor stretching from Mombasa, past Nairobi, and ultimately forking with one branch to Lake Victoria and another to the Uganda border. This same trajectory of service distribution, pioneered by the railway, had been taken by Kenya’s oil pipeline, highways, high-voltage electricity pylons, schools, and hospitals. As with distribution, so it was with prices: the bulk of the country, falling more than fifty miles north and south of this corridor of development, also suffered the highest fuel prices—a difference of up to 20 percent in cost.⁶

Liberalization did alter the terrain of fuel access, however, in one unexpected but significant way: it intensified the decline of the state-run railway and promoted transport by road. As part of the structural adjustment programs implemented in Kenya, the transport sector also underwent liberalization. The state cut back its support for the railway, further undermining an ageing infrastructure badly in need of capital investment (Ogonda 1992).⁷ More fuel started moving by road, in two different ways. Not only did tanker trucks take over fuel transport from the railway but also more fuel was used to power road transport—because more goods were moving by truck. Increased road transport gave roadside populations expanded opportunities to access fuel and other commodities, albeit through unofficial channels. Thus, while liberalization did not make fuel more accessible through its price/competition mechanisms, it inadvertently brought it closer to rural populations along the highway. The emergence of Gaakwen, stimulated by trade between the road and the roadside, itself lent evidence to this newfound proximity. The stop’s bars, restaurants, and accommodation, as well as its commodity traders, mechanics, and sex workers facilitated exchange between the highway and rural spaces, while various kinds of surveillance, like trucking-company checkpoints and security contractors tracking containers, sought to curtail it.

Trucking companies expressed anxiety about their inability to keep the trade route segregated from the rural networks of the roadside. One company boss articulated the sentiment as a fear of “leakage.” Virginia Roberts, the director of a trucking company headquartered near Gaakwen, told me that addressing fuel siphoning was one of the biggest challenges of her job. Her trucks “leaked” fuel, she said, in much the same way her husband’s cattle ranch “leaked” milk: “He has to make sure that milk goes into the mouths of his calves and not off the farm in plastic bottles.” Drawing a perhaps unintentional analogy between the plastic jerricans with which siphoners took fuel and the plastic bottles in which hired workers smuggled milk off the farm, the director made the tapping of fuel a wrong equivalent to that of denying a calf its mother’s milk: not just pilfering, but a kind of moral injury. It was intriguing, then, that fuel dealers also linked milk and fuel: siphoners were called wakamuaji, “milkers.” But where for the director this metaphorical rapprochement moralized the crime of fuel theft, for dealers the metaphor made obvious why one should have a right to fuel. Milk and fuel were both seen as essential resources—when large landowners and company owners had plenty while others had none, bits tended to find escape routes, with the help of plastic bottles.

What Roberts called “leakage” should be situated in relation to Gaakwen’s role as an interface between the highway and the countryside, and in relation to the uneven development that shaped the meaning of this interface both before and after liberalization. Gaakwen was located amid highly visible imbalances: on the one hand, the highway created a stark division between developed infrastructure (along which goods traveled, enclosed in surveilled containers) and the underdeveloped hinterland. On the other, Gaakwen was surrounded by vast farms owned by former settlers, the political elite, and multinational corporations. In this context, many perceived fuel siphoning as a mechanism of tacit redistribution. As Nikhil Anand (2015) argues in discussing the “leakiness” of water infrastructure in Mumbai, what is considered leakage by one party can be resource provision for another. In diverting resources, Gaakwen’s fuel dealers understood themselves to play a role similar to those who made illegal water connections in Anand’s account. By transporting fuel to the “interior” (the vernacular for parts of the country not served by a major road), and selling at lower-than-pump prices, they facilitated its movement to places where it was needed for tractors, generators, motorcycles, and pickups—machines used on farms and rough roads. Non-siphoned fuel could also find its way into these spaces, but fuel dealers’ prices proved attractive, and their distribution networks were particularly suited to the terrain and its socioeconomic conditions. As one dealer put it, his trade provided a “service” to rural populations located far from sanctioned fuel facilities.

Anand (2015, 308) argues that leakage is essential to the system—it diffuses the political pressure that could build up from the complete deprivation of a vital resource. For this reason, the state accepts a certain amount of leakage. The useful insight here, echoed by those involved in Gaakwen’s fuel trade, is that authorities tacitly permit a certain amount of illicit redistribution, thereby alleviating the pressure to make necessary infrastructural provisions to the margins. Thus, in light of Kenya’s entrenched distribution patterns, the roadside trade in siphoned fuel could indeed be seen as leakage—it diverts resources from the pathways that both organize commodity circulations and serve as “structures of confinement” (Tsing 2005, 6). But if leakage was the operative concept from the perspective of those who sought to enclose a resource, I found that those who liberated it used other concept-metaphors—central among them that of the leftover or remainder.

A closer look at the metaphors may prove helpful in understanding why, despite the liquidity of their commodity, fuel dealers did not employ the concept of leakage. As a metaphor for extracting value, “milking” suggests a far more active, entitled, and (as I will show) unit-specific process than “leakage,” which describes a passive outflow of resources.⁸ Indeed, leakage is something like a shadow concept of flow, a metaphor that has dominated ideas around distribution and circulation in capitalist contexts (Khan 2006). Flow draws an analogy between the movement of liquids and the circulation of capital to instill a sense of the naturalness of the market—when commodities and capital are said to flow, the word implies that they will organically find routes to where they need to go. It is for this reason that the metaphor is so frequently mobilized in discussions of liberalization, where regulation is perceived as an unnecessary interruption to the “natural” course of circulation. Central to this metaphor is an elision of the material form and purpose of resources themselves, as the specificity of things, anchored in their materiality, is replaced by abstracted value. Thus abstracted, the concept of flow ignores the labor and the desires that, in enacting circulation, inevitably generate a very un-liquid “friction” (Tsing 2005; Larkin 2013). So leakage works like flow in that it naturalizes diversions that in fact require work—practical and meaningful labor. This naturalization proves particularly problematic because it implies that nothing needs to be done to help resources reach underserviced areas: liquid metaphors have underpinned “trickle-down economics” precisely because they imply that distribution happens on its own (thanks to gravity or osmosis), without the need for more active intervention.

Contrast such metaphors of effortless, abstracted circulation with the way Gaakwen’s traders understood fuel and its (re)distribution. The diesel and petrol “milked” in Gaakwen were routinely described to me as what was “left over” from a driver’s trip. Of course the fuel was, literally, left over, in the sense that it remained in the tank. But contained in the idea of leftoverness as expressed by drivers was a more complex claim: that the share only came into existence under particular material conditions, and that access to it was marginal to some other use. The share taken was understood as qualitatively distinct from that which was essential to the journey. Interestingly, this distinction emerged from trucking companies’ practice of fuel dispensing. Companies fueled trucks at the beginning of a trip in an amount intended to last until the trip’s end. The precise quantity of fuel was variable—it depended on the length and quality of the route, the type of vehicle, its age, and the kind of cargo it was carrying—but the idea that trucks received fuel sufficient for their journey was consistent. The trip became the unit in which fuel was dispensed (an indivisible unit determined by a specific purpose) and it was in relation to this unit that the leftover was defined.

The significance of the unit form in shaping valuation can be helpfully elucidated by Paul Kockelman’s (2016) work on household economies in Guatemala. Describing economic practices in which one use value is substituted by another, rather than exchanged for another, he shows that measuring one thing against another to see if it is a suitable replacement turns, in part, on a system of evaluation within which unit sizes emerge intuitively out of use. In replacement economies, he writes, unit sizes often seem “natural” (one day’s labor, one meal’s worth of food, etc.); resources fall into these units as though “inherently formed.” He juxtaposes this against an exchange-value economy in which price or other abstracted and standardized representations of value ensure that commodities can be valued and exchanged not just in whole units but in fractions or multiples (Kockelman 2016, 118–19).⁹ This is significant because only in relation to such intuitive, indivisible, units does a remainder become sensible, as leftover food only makes sense in relation to the meal.

So paradoxically, drivers’ notion that fuel could be left over derived from trucking companies’ dispensing practices, as, partly, did drivers’ sense of entitlement to it. Their claim was enhanced by the fact that the fuel required for a trip could vary. Companies drew on fleet-management experience to determine the number of liters (just) sufficient for a particular trip, and they were suspicious if usage did not align with their expectations. But they recognized and anticipated a measure of variability, and driver reliability thus held great importance. Indeed, companies avoided putting young drivers on long-haul trips because, not yet knowing how to conserve fuel, these drivers were more likely to run out of fuel mid-trip.¹⁰ Drivers, in turn, were acutely aware of their role in managing fuel usage under unpredictable circumstances. As several explained, it was impossible to say exactly how many liters of fuel were required even for a familiar stretch of road, because driving conditions like traffic and police presence could significantly affect the trip. The irregularity gave rise both to the promise of a leftover and the fear of shortfall—if a driver ran out of fuel before the end of the trip, he would have to negotiate with the company for top-ups, and sometimes even pay for it himself. As I explore in more detail below, the very variability of the essential share made drivers responsible for their fuel usage and established the leftover as a product of their actions, indeed, their skill, rather than of happenstance.

So the remainder is a particular kind of share: it comes into existence in relation to a mode of valuation that links unit with purpose, and relies on the idea that portions remaining beyond that purpose are “extra”—available for fresh claims. Where trucking companies imagined a fraction that could be seamlessly reincorporated into their working capital, those involved in the roadside fuel trade saw a distinct share that they could take. The morality of the remainder did not inhere in a more familiar contrast that pits the appropriation of resources for use against profit—fuel traders were seeking profit just as much as trucking companies. Its moral charge lay, rather, in troubling how surpluses were represented and to whom they were thought to belong. The remainder, then, was a conceptual mechanism through which surpluses—not always easily given—could be actively diverted from circuits of reinvestment. To explore this further, I turn now to gleaning, and to the drivers and dealers who trade fuel in Gaakwen.

GLEANING THE ROAD

As I have noted, the drivers I spoke to consistently referred to fuel sold as what they had “saved”: a share that they could claim as their own. To explain, let me recount a story told to me by my friend and sometime host, William—a community leader and keen observer of the life of the truck stop. A parable about drivers caught selling fuel by police, the story both articulates drivers’ sense of entitlement and locates the fuel trade within the heightened surveillance of the Northern Corridor.

As William had it, one day “Macho Nne,” the founder of a trucking company called Ngong Works, was tipped off that his drivers were selling fuel. The company had its base only a few kilometers from Gaakwen, and truckers coming back in a convoy had stopped by Gaakwen to sell their remaining fuel, since they were nearing the end of their journey.¹¹ Patrolling police officers caught them in the act but proposed to look the other way if financially induced to do so. The truckers agreed, but when it came time to pay up, they only gave the officers a small payout: 500 shillings (about US$5). The policemen, estimating how many liters had been sold, thought this seemed like far too little. But the drivers refused to give more, asking, “Why should you deserve more, kwani ni yako?” (rudely: “What, you think the fuel is yours?”), and insisting that they were giving the cops a bribe, not a cut. That is, they were offering the police not a percentage of the fuel sold, but a one-off payment. In saying this, they implied—as William did, telling the story—that the fuel was actually theirs to sell, that they had a right to profit from, quantify, and share it as they saw fit, in ways the policemen did not.

The policemen threatened to turn them in, but the drivers did not budge, so in the end the police accompanied the drivers to their base and reported them to the boss, Macho Nne. Here the drivers argued that the policemen were harassing them because they had refused to pay a bribe, since Ngong Works had a policy of not bribing. To demonstrate their innocence, they asked Macho Nne to check the fuel reserves in their tanks. When the gauges showed as much fuel as he expected after their trips, it was the policemen who left defeated.

William told this story with great relish, fully aware of the irony of drivers asserting a claim to the fuel and denying the policemen’s claim, although both sides were engaged in what would seem, from the outside, to be equally illegal activities. Embedded in the story was also an explanation of how the drivers asserted a sense of ownership over the fuel, as against the policemen’s and Macho Nne’s. Essentially, the argument ran that if Macho Nne did not know he had lost any fuel, the fuel belonged to the drivers. This could work in two different ways, William explained, according to how they managed to bring the truck home with the “right” amount of fuel in the tank. Either they replaced what they sold with kerosene, or they managed to save fuel by driving skillfully, most notoriously by “freewheeling” (coasting) down and up hills, but also through other virtuosic techniques, such as keeping an even speed or a well-calibrated use of gears.¹² In the first case, the fuel could be said to belong to them because no one was the wiser: overlookable amounts were available for guiltless appropriation. In the second case, the fuel was theirs because they employed great skill and potentially risked life and limb to save fuel—conjuring a share, an “extra,” that could be sold without eating into the revenue expected by their bosses. While drivers understood that their actions limited the ability of trucking company owners to realize a surplus (that is, to profit off the truckers’ labor), they recoded what they sold as a leftover that they had created, to establish a right to it that did not directly compete with the trucking company’s right.

Although drivers lamented their low incomes and compared these to their high-value cargo and the profits their bosses made, those I interviewed did not anchor their claims to fuel in the language of need-based redistribution. Instead, their positions often mimicked or extended the logic of (entrepreneurial) capitalism. Their right to claim fuel, they argued, resulted from their skilled manipulation of their vehicles, and from the risk they bore. Transport company owners made the bulk of the profit, but it was drivers who carried the bulk of the risk, absorbing irregular loading schedules and road conditions into their bodies by denying themselves sleep when necessary, and exposing themselves to the accidents that plagued the corridor. They spoke of an intimate knowledge of their trucks and their limits. Their claim on leftover thus required a slight revision of classic understandings of the “moral economy.” The drivers did not juxtapose traditional use rights and embedded economies against the impersonal profit orientation of the market. Rather, they claimed the right to gain something from their skilled management of driving conditions. Intriguingly, this “feel” for the truck and its fuel consumption also allowed a distinction between essential and leftover fuel to preexist the fuel’s actual use. As one Gaakwen resident explained, experienced drivers sometimes sold fuel near the beginning of their trip, projecting how much they would “save.” The leftover was thus an effect of the material relationship between truck and fuel, but also of an artful negotiation of drivers’ position as artisans of the road, which, according to them, gave them a right to a marginal share. The margin emerged both from the contingencies of material and through a preexisting social negotiation.

The simultaneously contingent and planned aspects of the remainder in William’s account resonate in intriguing ways with the ancient ethics of harvest leftovers. To articulate this, let me digress briefly into a discussion of gleaning, a practice that could be glossed as “the marginal’s right to remainders.” In ancient and feudal societies around the world, gleaning was a right largely reserved for the “poor and alien.” Villagers—particularly the indigent, the young, the old, and women—were allowed to enter a field after the crop had been harvested and pick up what had been left behind. The Hebrew Bible mentions and endorses gleaning multiple times, and it was practiced quite widely through the early modern period. In Europe in the seventeenth and eighteenth centuries it contributed substantially to household economies: one researcher argues that gleaning could provide English peasant households with as much as 10 percent of the grain they needed for the year (King 1991). The practice nearly died out during the nineteenth and twentieth centuries as a result of increasingly exclusionary enclosure practices and because more efficient agricultural technologies—especially the combine harvester—gradually reduced the wastage of the harvest.

As both Leviticus and Deuteronomy make clear, the leftovers gathered in the fields were intentional leftovers: the landowner was obligated to leave them. Here is Deuteronomy 15:8:

When you reap your harvest in your field and have forgotten a sheaf in the field, you shall not go back to get it; it shall be for the alien, for the orphan, and for the widow.

And Leviticus 23:22:

When you reap the harvest of your land, you shall not reap to the very edges of your field, or gather the gleanings of your harvest. You shall not strip your vineyard bare, or gather the fallen grapes of your vineyard; you shall leave them for the poor and the alien: I am the LORD your God.

Gleaning is an odd sort of redistribution, one in which the landowner’s obligation is fulfilled almost accidentally. Even the meaning of the gleaned share is ambivalent—it sits somewhere in between wealth and sustenance. In medieval England, this ambiguous status is apparent in bylaws regulating gleaning, which cannot agree on the question of whether gleaned produce was subject to the Church’s 10 percent tithe (Ault 1961): is it income, or not? One contemporary scholar notes that economists have overlooked gleaning because it does not involve money or prices (Bardi 2015). But it is precisely this incommensurate and contingent quality that allows redistribution to persist without directly challenging the system of rights represented by the landowner’s ability to realize a surplus. Gleaning exists within a framework of established inequality. In being excluded from the realm of exchange—by the very fact of being remaindered—the transfer of goods is explicitly not a transaction between equals. So both the framing as remainder and the gesture toward contingency simultaneously make the transaction possible and limit its scope.

This ancient ethics of redistribution becomes surprisingly helpful for thinking about fuel siphoning and contemporary market capitalism more generally—both for the ways it resonates with contemporary practice and for those in which it diverges. Contemporary scholars have mobilized gleaning in thinking about the right to use copyrighted material for “harmless” purposes (Gordon and Katyal 2009) or the right to gather digital “resources” and digitally distribute alms (Bostock 2005; Cherkaev 2019). Similarly, food justice activists draw extensively on the language of gleaning when asserting the right to salvage both farm produce and processed foods—things that would otherwise go to waste. These contemporary examples help us understand the gleaned share as, crucially, one that can be taken without disturbing the dominant (landowner’s) claim. Gleanings are taken, as Xenia Cherkaev (2019) writes, “in ways you won’t even notice.”

Tellingly, many of these contemporary examples of gleaning focus on the right to take a remainder (and the tacit acceptance of this taking) more than the obligation to give one. Returning to Gaakwen, we can see that an “intentional remainder” lies at the heart of William’s story, but it is clear that truck owners are no longer ethically mandated to redistribute. Instead, the drivers of the story, like those I spoke to, articulated their right to remainders by defining themselves at once as subjects of redistribution, as skilled workers, and as risk-bearers. This combination reflects a broader shift in which redistributive obligations become rerouted through work relations—as historians of early capitalism describe, in-kind payments for artisans progressively replaced customary rights of access (before being criminalized during industrialization as pilfering and then theft) (Linebaugh 2006; Allen and Barzel 2011). Contemporary examples across the world reveal that workers exhibit a sense of entitlement to leftovers, even if employers disagree: numerous ethnographies show how extras or scraps from factories and farms are appropriated as “in-kind supplements” to wages (Ditton 1977; Harries 1994; Callebert 2017; Cherkaev 2018):¹³ These takings share a commonality: defining them as leftover allows workers to fill gaps in their livelihoods without troubling the landowner’s claim to the field itself and without a sense of wrongdoing. Equally, when fuel dealers argue that the fuel they are trading is left over, they are stating their right to access it in a way that does not quite contest the trucking company’s right to it. As one respondent put it, “let me say that everyone benefits, directly or indirectly.” This is then a moral economy in its classic sense: an ethical formation that structures socioeconomic relationships across inequality. Morality here is a language for negotiating with power and reminds us that claims are shaped by relationships of marginalization and hierarchy, however much these may not align with liberal aspirations to equality.

Of course, anthropologists have long paid attention to forms of obligation and dependency that clash with such liberal aspirations (Ferguson 2013). For Kenya, studies of patronage have shown how older and wealthier relatives are expected to redistribute across their kin and social networks, just as politicians are expected to hand out money during campaigns and to give projects to their constituencies (Atieno-Odhiambo 1987; Haugerud 1997; Smith 2008; Blunt 2010). Rhetoric reflecting hierarchical relationships of obligation also proved prominent in Gaakwen. Drivers often referred to truck owners simply as Tajiri—“rich man”—suggesting a personalized hierarchy defined by wealth rather than employment structure. And beyond the fuel trade, residents praised large landowners who occasionally invited villagers to enter their land and cut overgrown brush for firewood. But depictions of patronage do not fully capture the entitlement represented in the claim to leftovers. First, fuel gleaners did not rely on social or kin ties to assert obligations; similarly, their claim took shape not in relation to age-based or gendered forms of inequality but rather directly addressed the question of who controlled the surplus. In rewriting this surplus as remainder, fuel traders both engaged with an extractive understanding of labor and value and turned it to their ends. Second, the claim was expressed in the nature of the share itself, rather than through reference to a social position. In this sense, it reflected the way that unit sizes in popular economies take on a kind of morality, as in the distinction made once by a maize seller who told me: “We”—those exchanging in a more personalized economy—“sell by the debe [tin can],¹⁴ but they”—those involved in Nairobi’s estranged economy—“sell by the kilo.”¹⁵ Both were standardized measures of exchange, but the debe was popularly linked to units of daily use, where the kilo represented alienated exchange. Moreover, those selling by the debe were expected to include a bit of extra—a trader who leveled the top of the pile instead of leaving it heaped up was considered stingy. Thus, where the kilo was an abstracted standard, the debe was more like the harvest—a concrete form embedded in social life. In this sense it lent “qualitative values” to “quantitative measures,” as Brad Weiss (1997, 352) writes regarding currencies. Similarly, the leftover as a share shows us how social and moral values can be articulated in systems of measurement—in the units of use and exchange as much as in social categories of personhood and status. To explore this last point, I turn now to fuel dealers and their profit calculations.

EXCEPTIONAL VALUATION

After brokers approached a driver and ascertained his interest in selling fuel, they would call a mnunuzi, a “buyer.” Brokers were go-betweens, while buyers were the chief entrepreneurs; they rarely appeared on the roadside, preferring to conduct their business by cell phone. When the deal was struck, the buyer would send a crew of wakamuaji, “milkers,” to extract the fuel by sucking it through a plastic tube into plastic jerricans, which served at once as vessels for the fuel and as the unit of exchange. Once siphoned, the driver received payment for the fuel and the broker who had arranged the trade, a commission. Finally, young men called punda, “donkeys,” whose primary job was to carry the full containers, would take the fuel off the roadside and into warehouses where it was stored for sale. All transactions were made in cash and prices were consistent—fixed at an amount slightly lower than the price set by the state regulator.

I was fascinated by the mix of metaphors in this trade—especially given the constellation of cattle, banditry, agriculture, and wage labor that marked Gaakwen’s history and which I do not have space to address here at length.¹⁶ These metaphors were remnants of an older moral economy, organized around cattle and farming, that was in the process of being supplanted in Gaakwen around the time of my research. In this way, they linked contemporary fuel dealing to forms of agrarian and pastoralist life associated with positive values: reproduction, life, growth. When a fuel dealer called the action of siphoning “milking,” the dead (and even poisonous) qualities of fuel were overwritten by associations with lively and productive value derived from cattle-keeping. Moreover, ideas of both “milking” and “bleeding” cattle were explicitly associated by one of my interlocutors with fuel when speaking of how to make extraction sustainable. One can only take small amounts, he pointed out, because you have to be careful not to kill the cow (or its calf).

But the analogy with milking also points to something else, which becomes visible in the equivalence between truck and cow. Cattle are uniquely suited to thinking analogically about the relationship between liquid and the units in which it circulates and is extracted. As such they reveal something important about exchange and distribution. To return to the earlier discussion of flow, an analogy between truck and cow makes visible that containment is an essential, albeit overlooked, aspect of circulation (Shryock and Smail 2018). Whether in the truck’s tank or the cow’s udder, liquid commodities moved within containers. And when siphoners milked a truck for fuel, they extracted a liquid out of one container (tank/udder) and moved it into another. The extraction created a new kind of enclosure, the jerrican, which in turn shifted the unit of valuation. These containers and the form of calculation they underpin are the focus of this final section.

The jerrican was the unit of extraction, exchange, and calculation in the roadside fuel economy. Called mitungi (sg., mtungi)—derived from the Kiswahili word for “pot”—jerricans were ten- or twenty-liter plastic containers first introduced to East Africa by the German military in the form of steel receptacles for storing and transporting fuel. During my fieldwork, new and recycled plastic mitungi were used across Kenya for fetching, ferrying, and storing all kinds of liquid. They served not only as containers but also as ways of tracking and measuring use. Household water usage, for instance, was often calculated in mitungi. Indeed, the most common mtungi size—twenty liters—was a widespread unit of measurement in popular economies, so much so that it functioned as what Guyer (2004, 56) calls a “threshold number.” One counted in single liters up to twenty—from there, the twenty-liter unit became the base number, and one counted in mitungi. This showed that they were not just useful as vessels but also provided an evaluative function—the containers’ function inhered in their standardization as much as in their ability to contain liquid.

The centrality of this unit within the fuel trade was tied to its utility for both siphoning and transport, reflecting the wider fact that unit sizes in popular economies are determined not only by how they measure or what they symbolize but also by how they facilitate movement. Fuel dealers were able to divert fuel because they traded in amounts small enough to be overlooked and to be accessible to economically marginal buyers. Using mitungi, they repackaged fuel in quantities more portable across uneven surfaces (by motorcycle or pickup) than the quantities required to fill a fuel station’s tank. This transfer freed fuel from the confines of corridor infrastructure and served as an inversion of the large-scale distribution mechanisms linked to economies of scale—in these economies of unscale, distribution pathways were defined by the ability to offer small quantities and to traverse uneven terrain.

Gaakwen’s dealers preferred the thin plastic containers used in water dispensers to the heavier plastic jerricans, because the thinness of the plastic allowed them to expand the containers. (These thinner containers were also called mitungi and also came in ten- and twenty-liter sizes.) Dealers altered their containers’ capacity by pouring fuel into them and then placing them under a hot tin roof. The heated fuel and vapors would expand the plastic, so that the container could hold more fuel than the measurement associated with it. Thus, when they extracted “twenty liters,” showing a driver the full container as an indication of the amount extracted, they were in fact claiming more. This was one of the ways they created what Guyer (2004) would call a “marginal gain.” Using the flexibility of one of modernity’s materials, plastic, traders destandardized the standards themselves. Drivers knew there was some degree of alteration taking place, which gave the jerrican an even more central role in the transaction. During negotiation, the crew would show the driver the container they planned to use, and fuel dealers’ skill involved their ability to present their containers as fair. The drivers’ awareness and acceptance of some variability dovetailed with trucking company bosses’ knowledge that some fuel was being lost en route. Within reason, and so long as it remained a remainder, a difficult to quantify leftover, some diversion could be overlooked.

The extra capacity made for more than a form of trickery. It came to define the entire transaction. Dealers sold fuel by the mtungi, not by the liter, and the extra space they were able to create in their collection containers was an important element of their profit calculation. Their profit was based on a remainder—one reckoned not primarily through price, but through the accumulation of extra amount. To explain, let me recount how a dealer described his calculation. When I asked him how much he earned on an average transaction, he hesitated, not having a number to hand. Finally, he responded by first telling me the price at which he usually bought “twenty liters” (one expandedmtungi) of fuel, then explaining that that containers he used actually held more like twenty-five liters, then telling me for how much he sold twenty liters (one not-expanded mtungi). The approximately five liters created by the mtungi’s expansion was how he represented his profit. But he could not—or did not—articulate this profit in terms of price, nor even in a precise number.

As it moved out of trucks and into dealers’ networks, then, the fuel’s value underwent a transformation—from a calculation based on price per liter, to one based on the extra capacity in a swollen mtungi. We might remember, of course, that in economics, profit itself is defined as a marginal calculation: it is what is left over after production costs have been accounted for. But what was unique about this calculation is that the mechanism of profit generation was not a price markup, but rather the generation of an unnamed share—that extra space in the mtungi. Moreover, the roadside transaction, semihidden between the bodies of tractor trailers, using performative techniques of persuasion and illusion, worked with repertoires of exchange and calculation that preserved a sense of marginality and smallness. Embedded in seemingly banal questions of unit size, then, was a cultural logic in which measures also constituted meanings.

As Guyer (2004) writes, marginal transactions are both historically and culturally shaped: their repertoires reflect context-specific socioeconomic logics. So much so, she explains, that some kinds of transactions get named and institutionalized, “their components become conventionalized” (Guyer 2004, 97). The widespread recognition of the fuel trade’s performances and categories shows these to be established elements of a local repertoire. Attending to this conventionality might help us avoid a certain scholarly binarism in discussions of marginal economies—in which the criminalization of practices like fuel siphoning is countered by scholars who sympathetically describe illicit trades as a “survival” practice (see Fennell 2016 for a similar point¹⁷). Such accounts of “surviving,” “hustling,” and “getting by and making do” (Simone 2004; Newell 2009; Peterson 2014) capture important aspects of a contemporary crisis in livelihoods, but the framing results in scholars viewing these practices as external to the logic of the dominant economy. What my ethnographic research suggests, in contrast, is that claims to remainders are considered positive entitlements (not simply need-based “survival”) and are embedded in systems of evaluation that simply do not align with dichotomies of formal/informal or licit/illicit. Rather, the roadside fuel trade is grounded in an alternative logic: one linked to the needs of rural residents, to the exigencies of off-road transport and the conventions of popular economies, and to vestiges of pastoralist and agrarian economies—all which take shape in dealers’ repertoires of calculation, measurement, and exchange.

THE REMAINDER

In The Moral Economy of the Peasant, James C. Scott (1977, 6) argues that risk-averse peasants will prioritize subsistence over profit—they focus, he says, on “what is left rather than what is taken.” “What is left,” for Scott, refers to the basic requirement of subsistence and the source of peasant ethics. Thinking about what remains “forms the standard against which claims to surplus by landlords and the state are evaluated” (Scott 1977, 6). Engaged in a debate around peasant economies and rationality, Scott argues that activities not oriented toward profit-maximization can nevertheless be rational; at the same time, he shows how economic decisions are socially and morally articulated.

The remainder in Scott’s (1977) discussion operates somewhat differently from the remainder in my account, but I mention it here for its attention to peasants’ sense of entitlement, expressed in terms of a particular kind of share. Through the language of “subsistence,” peasants make a complex claim: it is at once ethical and economic, but it does not derive from a presumption of equality. The conceptualization of remainder that I tracked in my observations of Gaakwen also serves this complex function: simultaneously ethical and economic in ways difficult to disengage, it asks us to think about the “entanglement of values” (Palomera and Vetta 2016) that marks economic practice. Crucially, it asks us to widen our framework for thinking about how livelihood claims are articulated in contexts of entrenched inequality: the claim to remainders asserts that the accumulation of surplus has its limits; that there can and should be something left over.

Gaakwen’s traders, I suggest, are reimagining the category of “profit margin” from the perspective of the marginalized. The organization of value around “remainders” more than “price” constitutes a moral economy not organized in opposition to profit or to capital, but one that nevertheless decenters it. With this article I hope to suggest that attending to remainders helps us go beyond dominant frameworks for understanding economic relations: neither liberal discourse nor empathetic assessments of the poor’s supposed survival practices give a complete sense of how claims to a living are asserted. Attending to remainders asks us to think about how, within the gaps of a liberalized economy, alternative forms of valuation and distribution are established and made ethical.

ABSTRACT

MUHTASARI

NOTES

Acknowledgments This article is based on research supported by the Wenner-Gren Foundation, the Social Science Research Council, and the National Science Foundation. Basil Ibrahim and Xenia Cherkaev read and commented on many drafts: I am ever grateful for their insight. I also received very valuable advice from Brad Weiss and no fewer than six anonymous reviewers for Cultural Anthropology. My particular thanks go to Dan Rono, Musa Rama, and many other interlocutors in Gaakwen who not only facilitated research but spent hours offering me their thoughts.

CULTURAL ANTHROPOLOGY, Vol. 35, Issue 3, pp. 462-486, ISSN 0886-7356, online ISSN 1548-1360. © American Anthropological Association 2020. Cultural Anthropology journal content published since 2014 is freely available to download, save, reproduce, and transmit for noncommercial, scholarly, and educational purposes. Reproduction and transmission of journal content for the above purposes should credit the author and original source. Use, reproduction, or distribution of journal content for commercial purposes requires additional permissions from the American Anthropological Association; please contact permissions@americananthro.org. DOI: 10.14506/ca35.3.05